opportunities for agricultural biotechnology in the eu

Transcription

opportunities for agricultural biotechnology in the eu
Doctoraatsproefschrift nr. 908 aan de faculteit Bio-ingenieurswetenschappen van de K.U.Leuven
OPPORTUNITIES FOR AGRICULTURAL
BIOTECHNOLOGY IN THE EU:
POLICIES AND NOVEL CONSTRAINTS
Koen DILLEN
Supervisor:
Prof. E. Tollens, K.U.Leuven
Members of the Examination Committee:
Dissertation Presented in Partial
Prof. R. Merckx, K.U.Leuven, Chairman
Fulfillment of the Requirements for
Dr. M. Demont, AfricaRice
the Degree of Doctor of
Prof. E. Mathijs, K.U.Leuven
Bioscience Engineering
Prof. J. Vanderleyden, K.U.Leuven
Prof. J. Wesseler, Wageningen University
June 2010
©
2010
Katholieke
Universiteit
Leuven,
Groep Wetenschap
&
Technologie,
Arenberg
Doctoraatsschool, W. de Croylaan 6, 3001 Heverlee, België
Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd en/of openbaar
gemaakt worden door middel van druk, fotokopie, microfilm, elektronisch of op welke andere wijze ook
zonder voorafgaandelijke schriftelijke toestemming van de uitgever.
All rights reserved. No part of the publication may be reproduced in any form by print, photoprint,
microfilm,
electronic
or
any
other
means
Legal Deposit Number: D/2010/11.109/25
without
written
permission
from
the
publisher.
Preface
And he gave it for his opinion, "that whoever could make two ears of corn, or two
blades of grass, to grow upon a spot of ground where only one grew before, would
deserve better of mankind, and do more essential service to his country, than the
whole race of politicians put together." Jonathan Swift, Gulliver's Travels--Voyage to
Brobdingnag
Four years…four exciting years leading up to this doctoral degree. Four years of
ample intellectual freedom and curiosity. Four years that led me to interesting places
and situations. Four years with ups and downs. Four years that profoundly enriched
my life. But above all, four years in which I met, interacted with and relied on
fantastic people. These are the people I want to dedicate this preface to.
First of all, I would like to thank my promoter, Professor Eric Tollens, for his trust
in the fact that I would make this doctorate work. Something I wasn’t sure of myself
at the beginning. During the growth process of this doctorate, he proved to be a
valuable mentor, guided me around in the miraculous world of academics, and
convinced me of the practical implications of our work.
A second word of gratitude goes to the members of the Examination Committee.
Professors Mathijs, Merckx and Vanderleyden, for proofreading my manuscript and
providing their valuable comments. A special thank goes to Professor Wesseler as he
was there during the whole process of the doctorate. We attended the same
conferences, discussed research topics and he welcomed me at the University of
Wageningen for workshops and meetings.
Now, before you think I forgot you, the time has come to thank you. Yes Matty,
I’m addressing you in second person to capture your attention. During the first eight
months of the doctorate, I had the honor to occupy the office next to Dr. Demont. You
introduced me to the interesting world of agricultural biotechnology and the
methodologies for impact assessments. Near the end of your stay at the Division for
Agricultural and Food Economics, this cooperation was so intense that I spend
perhaps more time in your office than in my own. During these first eight months,
there were enough ideas in the pipeline to fill two doctoral dissertations. Even after
his move to Senegal, Matty played a continuous role in my research. Through mutual
visits or Skype, ideas (the one even more crazy than the other) were exchanged and
elaborated. Thanks to this splendid cooperation, the EUWAB project, founded in
1999 by Matty himself, grew to be an authority on the analysis of the economic
impact of GM technology in the EU. By the latest count, the project produced, among
other things, 22 international journal publications, something to be proud of. Thank
you Matty for the trust, the warm welcome in Leuven and Saint-Louis, and the more
than pleasant interaction.
Of course during these four years at the Divison, a whole group of colleagues, all
with their own personality and background came by. Hopefully I’m not forgetting
anyone if I’m thanking Annelies, Abebe, Cristal, Zaid, Fredu, Josaphat, Henry,
Kidane, Mesfin, Erwin, Tinne, Isabel, Ellen, An, Basil, Anneleen, Greet, Alice,
Wouter, Jorge and Benny for the funny, serious and other moments we shared. A
special thank you goes to Professor Maertens who had to endure my endless questions
about econometrics and who gave me her wise vision on the problems I faced.
Finally, I would like to thank Josee, Godelieve and Sofie for the logistic and
administrative support.
Next I would like to thank the people at Monsanto for their support. In particular
Ivo Brants who showed me the practical consequences of GM crops and introduced
me to people all over the world that deepened my understanding of the technology.
Another group of people that deserve great acknowledgment are the people at the
University of Missouri (e. g. Professor Kalaitzandonakes) and IFPRI (Dr. José Falck
Zepeda) who made my research stay in the US a fruitful one. A warm thanks to
professor Paul Mitchell from the University of Wisconsin for the cooperation which
grew from a casual meeting at a conference and resulted in two publications.
Finally a special thank you to the people all over the world that opened their door
and shared their experience with me in the last four years. It is just impossible to list
you all in this acknowledgement but that does not mean there is any less gratitude.
Tenslotte nog een woordje in het Nederlands. Ik wil jou bedanken Lotte voor het
geduld dat je met me hebt gehad als ik weer eens in het buitenland was of weer eens
over economie aan het ratelen was. Voor de steun en houvast die je bood als er zich
eens een grotere hindernis voordeed. Wat de toekomst zal brengen is per definitie
onzeker (gelukkig maar), maar ik weet alvast dat ik er samen met jou wel geraak.
Natuurlijk mogen mijn ouders en familie niet ontbreken in deze ondertussen al
lange lijst van bedankingen. Zij creëerden de omgeving voor mij om zelf mijn weg te
ii
Preface
gaan en de wereld te ontdekken. Ik heb deze kans met twee handen gegrepen en zal
dat blijven doen.
Verder ook nog een woordje van dank voor de andere mensen die daar waren om
me te steunen, ontspannen, etc. Ik denk aan Karel, Simon en Wim voor de pauzes in
het Geo-instituut. Ik denk aan Kris, Veerle, Bram, Sarah en Karel voor de vele
(ont)spannende momenten. Joren en Willem voor hun enthousiasme en nuchtere kijk.
Ik denk aan de Woodies voor alle vroege en late uurtjes. En natuurlijk denk ik aan alle
andere vrienden die er tijdens deze 4 jaar gewoon waren.
Vermits jullie nu natuurlijk zitten te popelen om de rest van dit boekje te lezen nog 1
keer…..
VAN HARTE BEDANKT!
Koen
iii
iv
Samenvatting
Ondanks het feit dat de commerciële introductie van genetisch gemodificeerde (GM)
gewassen meer dan een decennium geleden plaats vond, is het debat omtrent de
impact en het nut van GM gewassen nog zeer actueel, zeker binnen Europa.
Doorheen de tijd heeft het debat zich toegepast op verschillende thema’s gaande van
de risico’s voor het leefmilieu, over voedselveiligheid tot economische effecten. De
aandacht voor de economische effecten kan worden verklaard door het feit dat GM
gewassen vaak beschermd zijn door intellectuele eigendomsrechten die de
welvaartscreatie en vooral de verdeling ervan beïnvloeden. De laatste tijd gaan er
stemmen op om een socio-economische evaluatie onderdeel te maken van de
bioveiligheid en deregulatie procedure van GM gewassen. Echter, het introduceren
van socio-economische evaluaties in deze procedure is niet eenvoudig omwille van
legale, conceptuele en methodologische redenen. In deze thesis wordt geprobeerd
progressie te boeken in enkele van de methodologische problemen. De
methodologische oplossingen worden vervolgens gebruikt om inzicht te verwerven in
enkele recent veranderde beleidsmaatregelen en omstandigheden in de Europese
landbouw.
Een socio-economische evaluatie van een technologie voor ze de markt bereikt, ex
ante, leidt tot het endogeen probleem van imperfecte en onvolledige data. Er is geen
informatie beschikbaar over het segment van adopterende boeren en de effecten op
typische adoptieparameters zijn niet detecteerbaar. De voorgestelde methodologie
probeert dit probleem op te lossen door gebruik te maken van parametrische modellen
en expliciet rekening the houden met de heterogeniteit die er bestaat in de
landbouwpopulatie. Deze aanpak suggereert dat heterogeniteit een cruciale rol speelt
in ex ante evaluaties omdat ze de prijszetting door de innovator en de hieruit
voortvloeiende de welvaartscreatie bepaalt. Het expliciet in rekening brengen van
heterogeniteit vermijdt mogelijks vertekende resultaten door een homogeniteits- of
prijszettingsafwijking. De voorgestelde methodologie wordt vervolgens toegepast op
het geval van herbicide tolerante (HT) suikerbieten op het niveau van de boer en op
het geaggregeerde niveau door middel van het EUWABSIM model, om de empirische
consequenties te vatten. Deze analyse wordt verder aangevuld door een Bayesiaanse
analyse die expliciet rekening houdt met verschillende irreversibele effecten die
samengaan met de introductie van GM technologie.
v
Door middel van de ontwikkelde modellen wordt vervolgens onderzocht hoe
recente (voorgestelde) veranderingen in het Europese beleid de potentiële adoptie en
deregulatie van GM gewassen beïnvloeden. Eerst wordt onderzocht hoe de recente
hervorming van de Europese suikermarkt leidt tot een wijziging van de determinanten
van innovatie in de sector. De resultaten tonen aan dat deze effecten significant zijn,
maar op zo’n manier dat de oorspronkelijke doelstelling van de hervorming, de
competitiviteit van de sector verhogen, niet wordt aangetast. Vervolgens wordt de
impact van veranderingen in de commitologie procedure van de EU op de kans tot
deregulatie van HT suikerbiet geëvalueerd. Het verschuiven van de autoriteit over
deregulatie naar het nationale niveau zorgt voor een betere afstemming van de
economische drijfveer met de kans op deregulatie.
Tenslotte wordt de focus gericht op de maïswortelboorder, een recent in Europa
geïntroduceerde maïsplaag. Door middel van een enquête bij Hongaarse boeren wordt
de bedreiging doordeze insectensoort in kaart gebracht. Gebruikmakend van deze
resultaten en data uit de Verenigde Staten, wordt er vervolgens een bio-economisch
model opgebouwd gebaseerd op de eerder gepresenteerde bouwstenen. De toepassing
van het model op acht Centraal Europese landen toont de economische bedreiging
door de soort aan, en geeft inzicht in de werking en competitiviteit van de
verschillende strategieën om schade te beperken. Tegelijkertijd berekent het model de
potentiële waarde, zowel monetair als niet-monetair, van het dereguleren van een
wortelboorder resistente maïsvariëteit.
vi
Abstract
Despite the commercial introduction of genetically modified (GM) crops over a
decade ago, the debate about the impact of GM crops in society is still very much
alive, especially in Europe. Through time, the discussion has hovered over various
themes such as environmental risk, food safety, consumer protection and economic
effects. Within the frame of economic concerns, the welfare and distributional effects
of a technology protected by intellectual property rights play a key role. Recently,
voices were raised to include the economic impact of GM crops in the biosafety and
deregulations procedure, both at the global and the European level. However,
including socio-economic assessments in this procedure is not straightforward due to
legal, conceptual and methodological reasons. This dissertation highlights some of the
methodological problems and suggests possible solutions. Moreover, the proposed
concepts are used to shed light on some recent policy changes and novel constraints
in European agriculture.
A socio-economic assessment of a technology before its introduction, ex ante,
creates the endogenous problem of imperfect and scarce data. No information is
available on the segment of adopting farmers and the effects on typical adoption
variables cannot be observed. The proposed framework tries to reduce this constraint
through the use of parametric models and the explicit incorporation of heterogeneity
in the farmers population. The framework suggests that heterogeneity plays a crucial
role in ex ante impact assessments as it affects the optimal corporate pricing strategy
and the resulting welfare distribution of the technology among stakeholders. Hence
accounting for heterogeneity in ex ante impact assessments eliminates both a potential
pricing and homogeneity bias. In order to fully understand the consequences, the
framework is applied to impact of the potential introduction of herbicide tolerant (HT)
sugar beet in the EU, at the farm level, and at the aggregate level through the
EUWABSIM model. Moreover, the possible irreversible effects related to the
introduction of HT sugar beet are included in the model through a Bayesian approach.
Secondly, the developed models were employed to assess the effect of recent
(proposed) changes in European policies that may affect the adoption and
deregulation of GM crops. First, the effect of the recent change in the European sugar
policy on the innovation incentives in the sugar sector is assessed. The results show
that the innovation incentive is significantly altered, but in a way it supports the initial
vii
aim of the policy reform to increase the competitiveness of the sector. Secondly, the
impact of changes in the commitology procedure of the EU is assessed. Results show
that shifting the approval decision to the national level increases the convergence of
economic incentives with the likelihood of deregulation.
Finally, the threat of the invasive species Western corn rootworm (WCR) for
European agriculture is assessed. First, the results of a survey among Hungarian
farmers to elicit their perceptions of the threat are presented. Combining these results
with data from the USA resulted in a bio-economic model based on the earlier
described framework. The model output for eight central European countries
quantifies the potential threat for European maize production and sheds light on the
appropriateness and competitiveness of the different crop protection strategies.
Moreover, the model estimates the potential value, both pecuniary and non-pecuniary,
generated by deregulating a WCR resistant maize variety.
viii
Table of Contents
Preface ............................................................................................................................ i
Samenvatting ................................................................................................................ v
Abstract .......................................................................................................................vii
List of Abbreviations ............................................................................................... xiii
Chapter 1.
Introduction .......................................................................................... 1
1.1 Socio-economic experiences with GM crops in the world ......................... 3
1.2 GM crops in the European Union ............................................................... 7
1.2.1
Coexistence between GM and conventional crops ....................... 8
1.2.2
The role of ex ante socio-economic impact assessments in the
regulatory process ..................................................................................... 11
1.2.3
Chapter 2.
Dissertation’s hypotheses ........................................................... 14
Corporate Pricing Strategies with Heterogeneous Adopters: The
Case of Herbicide Tolerant Sugar Beet .......................................................... 19
2.1 Introduction ............................................................................................... 20
2.2 Model ........................................................................................................ 21
2.3 Application ................................................................................................ 26
2.3.1
Data ............................................................................................. 31
2.4 Results and Discussion.............................................................................. 33
Chapter 3.
European Sugar Policy Reform and Agricultural Innovation ...... 39
3.1 Introduction ............................................................................................... 40
3.2 Model ........................................................................................................ 42
3.2.1
The Old Market Organization for Sugar ..................................... 46
3.2.2
New Common Market Organization for Sugar........................... 50
3.3 Data and model calibration ....................................................................... 52
3.4 Results ....................................................................................................... 55
3.5 Discussion ................................................................................................. 60
Chapter 4.
Global Welfare Effects of GM Sugar Beets under Changing Sugar
Policies .............................................................................................................. 63
ix
4.1 Introduction ............................................................................................... 64
4.2 Model ........................................................................................................ 65
4.3 Data and Model Calibration ...................................................................... 66
4.4 Results and Discussion.............................................................................. 68
Chapter 5.
The Barroso Proposal of Nationalizing GM Approval: A Look at
HT-Sugar Beets under Changed European Sugar Policy............................. 73
5.1 Introduction ............................................................................................... 74
5.2 Herbicide Tolerant Sugar Beet .................................................................. 76
5.3 The Economic Model ................................................................................ 78
5.3.1
Social reversible effects .............................................................. 81
5.3.2
Social irreversible effects............................................................ 82
5.3.3
Calibration of the MISTICs ........................................................ 82
5.4 Model Results ........................................................................................... 84
5.5 Voting Assessment .................................................................................... 87
5.6 Discussion ................................................................................................. 91
Chapter 6.
Are Farmers Willing to Pay for a Maize Variety Resistant to
Diabrotica virgifera virgifera Damage? ......................................................... 95
6.1 Introduction ............................................................................................... 96
6.2 Methodology ............................................................................................. 99
6.3 Data ......................................................................................................... 101
6.4 Results ..................................................................................................... 106
6.5 Discussion ............................................................................................... 108
Chapter 7.
On the Competitiveness of Diabrotica virgifera virgifera Crop
Protection Strategies in Hungary: a Bio-economic Approach ................... 111
7.1 Introduction ............................................................................................. 112
7.2 Control Options ....................................................................................... 113
7.2.1
Crop rotation ............................................................................. 113
7.2.2
Chemical control ....................................................................... 114
7.2.3
Bt maize .................................................................................... 115
7.3 Bio-economic Model............................................................................... 116
7.3.1
x
Data ........................................................................................... 120
7.3.2
Simulations and Results ............................................................ 124
7.4 Discussion ............................................................................................... 131
Chapter 8.
The Western Corn Rootworm, a New Threat to European
Agriculture: Opportunities for Biotechnology? .......................................... 133
8.1 Introduction ............................................................................................. 134
8.2 Maize Production and WCR in each Country......................................... 135
8.2.1
Austria ....................................................................................... 135
8.2.2
Czech Republic ......................................................................... 135
8.2.3
Poland ....................................................................................... 136
8.2.4
Romania .................................................................................... 136
8.2.5
Serbia ........................................................................................ 137
8.2.6
Slovakia .................................................................................... 138
8.2.7
Ukraine...................................................................................... 138
8.3 Bio-economic Model............................................................................... 139
8.4 Results ..................................................................................................... 144
8.5 Discussion ............................................................................................... 150
Chapter 9.
Conclusions and Further Considerations ...................................... 153
Hypothesis 1: Heterogeneity among farmers has no effect on corporate pricing
strategies of proprietary technologies and resulting welfare effects. ..................... 154
Hypothesis 2: Parametric approaches do not have the potential to complement
the inherent scarce data in ex ante impact assessments. ........................................ 158
Hypothesis 3: EU policies not directly related with GM crops do not affect the
potential adoption and deregulation of GM crops in Europe. ................................ 160
Hypothesis 4: The invasive species Western Corn Rootworm does not presents
an economic threat for Central European maize farmers....................................... 164
Hypothesis 5: EU farmers cannot gain from the introduction of Bt maize
resistant to WCR damages. .................................................................................... 166
List of References ..................................................................................................... 169
List of Publications .................................................................................................. 189
International Scientific Journals ...................................................................... 190
Book Chapters .................................................................................................. 191
xi
xii
List of Abbreviations
AFC
Auto Financing Constraint
CARA
Constant Absolute Risk Aversion
CDF
Cumulative Density Function
CE
Certainty Equivalent
CMO
Common Market Organization
CP
Crop Protection
CV
Contingent Valuation
EBA
Everything But Arms
EC
European Commission
EU
European Union
EUWABSIM
Specific partial equilibrium model simulating the introduction
of HT sugar beet
GM
Genetically Modified
HT
Herbicide tolerant
IPM
Integrated Pest Management
IPR
Intellectual Property Right
LDC
Least Developed Countries
LSR
Land Supply Response
MISTIC
Maximal Incremental Social Tolerable Irreversible Cost
PDF
Probability Density Function
ROW
Rest Of the World
SEACER
Spatial Ex Ante CoExistence Regulation
UK
United Kingdom
USA
United States of America
WCR
Western corn rootworm
WPN
World Price Non responsive
WPR
World Price Responsive
WTP
Willingness To Pay
WTO
World Trade Organization
xiii
xiv
Chapter 1. Introduction
Agricultural biotechnology is a broad concept which ranges from animal cloning
(Faber et al., 2003), to crops producing pharmaceutical products (Moschini, 2006).
This dissertation focuses on the most common application of biotechnology in
agriculture, genetically modified (GM) crops. GM crops have been inserted with
genes, from within or outside of the species, that transcode for desirable properties
initially not present in the species. Three distinct groups of GM crops can be
differentiated. First generation GM crops are designed to overcome constraints in the
agricultural production process, such as pests and weeds. Second generation GM
crops are targeted to consumers, delivering extra quality traits such as improved
nutritional value. Third generation GM crops produce end products previously not
considered an output of agriculture, such as pharmaceuticals. GM crops as a group
have a huge potential to ensure the supply of sufficient food and raw material in the
upcoming decades, when constraints such as climate change and population growth
will affect agricultural production and food security (von Braun, 2007).
The first GM crops were commercialized in the mid-1990s. Since, adoption
increased steadily, both in industrial and developing countries. In 2009, GM crops
were grown on 134 million ha worldwide or about 10% of arable land (James, 2010).
Area (million hectares)
140
120
100
80
Total
60
Industrial
40
Developing
20
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
0
Year
Figure 1: Worldwide adoption of GM crops (1996-2008)
The area planted with GM crops in the world and disaggregated between industrial and developing
countries
Source: James (2009)
Up to now, only first generation GM crops have been commercialized. Only a limited
number of crops and traits are available in the market. Herbicide tolerant soybeans
2
Chapter 1. Introduction
make up 53% of the global GM area. Maize, both herbicide tolerant and pest resistant,
accounts for another 30% of the area, or 24% of the maize grown worldwide (James,
2009). Other important GM crops include cotton and oilseed rape. Cotton resistant
against bollworms and butworms is important for developing countries, e.g. the
adoption in India and its introduction on the African continent. Some other crops have
been cleared for commercial use but are only grown on relatively small areas, e.g.
virus resistant papaya in Hawaii and herbicide tolerant sugar beet in the USA.
1.1 Socio-economic experiences with GM crops in the world
As these first generation GM crops were developed to overcome agricultural
production constraints the adoption can be explain by a high farmer’s demand. A
logical explanation for this demand would be the economic profitability at the field
level. Most of the available GM technologies have been developed by the private
sector and therefore are protected by intellectual property rights (IPRs). These IPRs
grant market power to the innovator (Weaver and Wesseler, 2004) allowing the
private company to charge a premium, or technology fee, for the use of the
technology. 1 This market structure raises concerns about the distribution of benefits
between the downstream and the upstream sector and became part of the societal
debate surrounding GM crops. Qaim (2009) reviews the economic literature both at
the farm level and the aggregate level published in the first decade of
commercialization. In this introduction some of this literature is discussed and the
important issues in relation to the dissertation’s research questions are highlighted. As
the economic effects of herbicide tolerant crops and pest resistant crops are different
we look at them separately.
Herbicide tolerant (HT) crops are resistant to a broad spectrum herbicide which
changes the agricultural production system. As yield effects from the technology are
limited from the trait as such, the main benefit lies in the reduced expenditures on
herbicides, labor and fuel. However, at least part of those benefits accrues to the
innovating sector through the technology fee. As the technology fee is of similar
magnitude as the cost reduction, average benefits are generally low or even negative
(e.g. Fernandez-Cornejo and McBride, 2002; Fulton and Keyowski, 1999; Phillips,
1
This assumption only holds in countries where institutions and enforcement mechanisms to protect IPRs exist.
There is some evidence (e.g. Qaim and de Janvry, 2003 for Argentina) that markets and economic responses
are different. However, in this dissertation we focus predominantly on agriculture in developed countries with
strong institutions.
3
2003). Despite the low monetary value, adoption of these crops has been successful.
In the USA, over 95% of the soybean acreage possessed the HT trait in 2008 (NASS,
2009). Two reasons explain this observation. First, farmers are heterogeneous and the
average value does not exclude a group of farmers to gain from the technology.
Secondly, the management ease of the broad spectrum herbicide and the increased
weed control have a non-pecuniary value to farmers (Marra and Piggott, 2006). These
non market valuated returns may sometimes increase monetary income through the
use of excess labor in other markets, especially in family farms (Useche et al., 2009).
Fernandez-Cornejo et al. (2005) for example show that off farm income indeed drives
the adoption HT soybean in the USA.
Insect resistant crops affect agricultural production in a different way. The
commercialized varieties are inserted with a gene from Bacillus Thurgeniensis, in
order to produce a protein toxic for specific lepidopteron and/or coleopteran insects.
Hence, the use of Bt varieties replaces the application of chemical control agents for
that pest. The economic effect is twofold. First there is a cost reduction from the
reduced chemical use (similar to HT crops). It is important to note that Bt crops do
not eliminate insecticide use, as not all pests are targeted by the Bt variety. Secondly,
Bt crops may increase the actual yield by avoiding previously uncontrolled pest
damage. These two effects are interlinked. Farmers previously applying optimal
chemical control will have cost reductions and resource poor farmers with
uncontrolled damage will experience yield increases after adoption (Qaim and
Zilberman, 2003). A body of literature has focused on the farm level effect of Bt
crops in developed and developing countries. Return on investment of the technology
fee depends on the country and the crop considered. For Bt cotton the increase in
gross margins ranges from $23/ha in Argentina (Qaim and de Janvry, 2003) to
$470/ha in China (Pray et al., 2002) while for Bt maize this ranges from $12/ha in the
USA (Naseem and Pray, 2004) to $70/ha in Spain (Gomez-Barbero et al., 2008).
To address the distributional concerns, a more aggregated approach is needed.
Both partial equilibrium models and general equilibrium models have been used in
this context. Partial equilibrium models are used to assess the impact of one specific
crop ceteris paribus. Demont et al. (2007b) perform a meta-analysis on these models’
results to discover the welfare sharing between farmers, consumers and input
suppliers. They discover that, despite the variability in results from different countries
4
Chapter 1. Introduction
and crops, about one third of the value accrues to input suppliers and two third to
farmers and consumers, both in developing and developed countries.
A general equilibrium approach takes into account substitution effects between
markets and regions, and is generally used to assess the global value of GM crops.
Estimates for Bt cotton for instance range from $0.7 billion (Anderson et al., 2008) to
$1.8 billion (Elbehri and Macdonald, 2004). Of course, those absolute values depend
on the adoption rate while the distributional effect is assumed to be constant over
time. The effect for consumers is often not straightforward as agricultural policies
affect the price elasticity to changes in supply. A recent study by Brookes et al. (2010)
using a computable general equilibrium model suggests that world prices of maize,
soybeans and canola would be, respectively, 5.8%, 9.6% and 3.8% higher compared
to the actual situation in 2007 if no GM crops would have been adopted.
The aforementioned economic assessments only consider the net reversible effects
of the introduction of GM crops. A limited number of studies include the irreversible
social costs related to the introduction through the use of real options. This approach,
explicitly taking into account uncertainty, allows the calculation of the maximal
incremental social tolerable irreversible costs (MISTICs) to justify the immediate
release of the technology (Demont et al., 2004; Scatasta et al., 2006; Wesseler et al.,
2007). Hence, these studies include environmental effects in the socio-economic
impact assessments.
All of the studies discussed present an economic rationale for the adoption of GM
crops. Despite this apparent economic benefit for both farmers and consumers,
adoption has been limited to 25 countries 2 with three countries (USA, Brazil and
Argentina) accounting for 80% of the area grown. It seems that once a variety is
commercialized farmers’ demand is substantial, but the farmer’s pull is not strong
enough to push the hurdle of deregulation. The reason can be found in the opposition
against GM crops in the public perception. Especially in Europe, the anti-GM feelings
and the interest groups against the technology are strong (European Commission,
2006a). Over the years, the focus of the debate has shifted over diverse frames of
reference as freedom of research, environmental risk, food safety, consumer
protection, bioethics, economic policy and international trade (Scholderer, 2005).
These concern are especially present in Europe but generate important spillover
2
Official number, anecdotal evidence suggests the use of illegal GM seeds in other countries such as Vietnam,
Thailand, Pakistan and Ukraine.
5
effects to other countries through trade regulation, lobbying groups and
misinformation (Gruère and Sengupta, 2009).
The most dominant factor in the debate seems to be the perception of
environmental risks, e.g. super weeds, impacts on non-target organisms, decreased
biodiversity, and the potential health effects. These risk are indeed important and have
to be closely monitored in the future. However up to now, no negative externalities
have been documented. Moreover, extensive reviews by e.g. Shelton et al. (2009) and
Sanvido et al. (2007) have found no negative effects of the introduction of GM crops
so far. These risks have to be considered and closely monitored in the future but for
now do not seem to lead to problems. On the other side some positive environmental
and health benefits have been documented, including:
•
Reduced pesticide use in Bt crops (Barfoot and Brookes, 2007; Hu et al.,
2006; Qaim et al., 2006; Qaim and Zilberman, 2003)
•
Increased no till practices in HT crops, leading to reduced fuel and better
erosion management (Kalaitzandonakes, 2003)
•
Reduced toxicity of the broad spectrum herbicide production system
(Devos et al., 2008a)
•
Potentially increased biodiversity as breeding through biotechnology can
be more easily backcrossed to local varieties (Zilberman et al., 2007)
•
Decreased mycotoxins in maize (Wu, 2006)
•
Decreased exposure of farmers to insecticides (Huang et al., 2005).
The literature points out that the actual presence of risks is probably not the main
reason of the public opinion against GM crops. Therefore a school of research has
focused on different possible determinants of public opinion making. The influence of
information dissemination and media was highlighted as biasing the view of
consumers (Marks et al., 2002; Marks et al., 2003; Moon and Balasubramanian,
2004). Gaskell et al. (2004) argue that it is not the perception of risks that influence
the opposition but the absence of clear benefits for the consumer. Other factors
influencing decision making in the EU highlighted in literature, include the
dominance of small scale farming, the presence of a strong biotechnology sector and
the share of organic production (Cooper, 2009; Kurzer and Cooper, 2007a; Kurzer
and Cooper, 2007b).
6
Chapter 1. Introduction
1.2 GM crops in the European Union
Against this general background of the impact and spread of GM crops in the world,
we turn towards the European situation and the issues at stake at this moment. Despite
the harmonization of European legislation on risk assessments, deregulation and
labeling and tracing since 2004, adoption of GM crops within the EU has been
limited. Only one Bt variety of maize (MON 810), resistant against damage from
European Corn Borer, is commercialized. Figure 2 shows the adoption through time.
The major adopter is Spain, with a total of nine Member States having cultivated GM
crops. 3 In the beginning of 2010 a GM potato with increased starch content was also
deregulated but no commercial cultivation has taken place yet.
120,000
35,000
GM maize plantings (ha)
100,000
80,000
Other EU member states, 2004-2008
30,000
25,000
PL
SK
20,000
RO
15,000
DE
PT
10,000
60,000
Other
EU
member
states
France
5,000
CZ
0
40,000
2004
2005
2006
2007
2008
Spain
20,000
0
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Years
Figure 2: Adoption of Bt maize in the EU
Evolution of plantings (ha) with Bt maize resistant against European corn borer in the EU since 1997
Source: reproduced with permission from Demont et al. (2010a)
Two issues are at the forefront of the legislative debate in Europe. A first aspect
concerns the role of socio-economic assessment in the EU regulatory process. As the
remainder of this dissertation is situated in this debate, we first focus on the second
issue, coexistence. Ceteris paribus, the regulatory problem is going to expand as the
development and commercialization of GM crops is assumed to boom in the next
3
The ninth country with experience in GM crops is Romania which deregulated HT soybean before accession to
the EU but had to abandon plantings due to this accession.
7
decade. Stein and Rodriguez-Cerezo (2010) estimate that by 2015 over 120 different
crop trait combinations will be commercialized worldwide. The introduction of
second generation and third generation GM crops may further increase regulatory
problems and economic assessments.
1.2.1 Coexistence between GM and conventional crops 4
The EU struggles with the implementation of coherent rules to assure the coexistence
between GM crop production and conventional/organic production. According to the
European Commission’s (EC) guidelines, “Coexistence refers to the ability of farmers
to make a practical choice between conventional, organic and GM crop production, in
compliance with the legal obligations for labeling and/or purity standards. The
adventitious presence of GMOs [genetically modified organisms] above the tolerance
threshold set out in Community legislation triggers the need for a crop that was
intended to be a non-GMO crop, to be labeled as containing GMOs. This could cause
a loss of income, due to a lower market price of the crop or difficulties in selling it.
[…] Coexistence is, therefore, concerned with the potential economic impact of the
admixture of GM and non-GM crops [...]” (EC, 2003). The final authority of
designing coherent coexistence policies lies on the Member State level. Since the
publication of these guidelines, some Member States have developed, and others are
still developing, a diversity of ex ante regulations and ex post liability rules on the
coexistence of GM and non-GM crops (Beckmann et al., 2006; Devos et al., 2009;
EC, 2006). The variety of coexistence rules in place and the stringency of these rules
in some Member States suggest the perverse use of coexistence rules to avoid GM
cultivation by some Member States (Devos et al., 2008b).
Research has focused on the optimal design of coexistence measures. The biggest
body of empirical work covers the technical feasibility to keep adventitious presence
of GM traces under the tolerance threshold. However, as suggested by the EC, this
could cause a loss of income, due to a lower market price of the crop or difficulties in
selling it.. Therefore some studies focus on the economic impacts of different
coexistence measures. Special focus has been on the spatial ex ante coexistence
regulations (SEACERs) such as isolation distances and buffer zones. Perry (2002)
concludes that stringent mandatory SEACERs combined with an increase in adoption
4
The next section is adapted from papers co-authored by the author of this dissertation: Devos et al. (2009),
Demont et al. (2008b; 2009a), and is therefore more elaborated than some other sections.
8
Chapter 1. Introduction
of organic cultivation restricts the cultivation of GM crops within the landscape, in
some cases even hampering the coexistence of both production systems. Dolezel et al.
(2005) confirm these findings by estimating the area lost for non-GM cultivation
under SEACERs in three Austrian regions. Moreover, this case study concludes that
in small-scale structured landscapes with a high percentage of maize cultivation,
spatial feasibility is negatively correlated to increases in the stringency of SEACERs
and GM crop adoption. Sanvido et al. (2008) examine the feasibility of SEACERs
under spatial constraints at two levels. At the aggregate level, national statistics are
used to assess whether the available arable land in Swiss communes is large enough to
enable isolating an assumed area allocated to GM maize. This approach is
complemented by a GIS (Geographic Information System) analysis at the field level.
The authors demonstrate that both the density of maize cultivation as well as the
distances between maize fields can strongly differ within the communes and hamper
coexistence, despite the feasibility to comply with SEACERs at the aggregate level.
Devos et al. (2007), using a combination of GIS data and Monte Carlo simulations,
investigate how these spatial arrangements influence the feasibility of implementing
SEACERs. The results show that clustering may be an effective strategy to facilitate
the coexistence of production systems. Therefore, Devos et al. (2008c) propose the
theoretical solution of pooling arable land to increase the feasibility of SEACERs in
the landscape. However, besides agronomical problems such as the promotion of
monoculture, clustering also creates significant transaction costs. Furtan et al. (2007)
assess the economic and institutional feasibility of such coordination through the case
of private organic clubs in Canada. Although institutional settings differ between
Canada and the EU, the results offer some insight in the dynamics. If the size of the
cluster decreases or SEACERs become more stringent, the costs of the private clubs
increase. Messéan et al. (2006) use a GIS analysis to assess the influence of
alternative SEACERs on farm level costs but do not provide any information on the
economic feasibility of the SEACERs and the impact of spatial patterns. Munro
(2008) investigates spatial feasibility of SEACERs within an economic environment.
Within the spatial restriction of a stylized rectangular agricultural landscape, he
stresses the importance of appropriate policy options as efficient outcomes will not be
achieved in an unregulated market. The feasibility of SEACERs depends mainly on
the size of the barrier which must be maintained in order to avoid cross-fertilization.
However, as the model is built on a simplified spatial economy, it does not take into
9
account the geographical influence of landscape, land fragmentation, and field
configuration on the impact of GM crops. Moving from a simplified surrogate to a
real landscape substantially increases the complexity of a model. Ceddia et al. (2009)
attempt to overcome the constraints of a stylized agricultural landscape through the
incorporation of a general aggregation index, developed by He et al. (2000), in their
model. Using this model, they assess the biological efficiency of different policy
variables including SEACERs. However some of the proposed policies would not be
feasible from an economic point of view as they would generate excessive costs to the
farmers. Moreover, the aggregation index does not account for the actual shape of
agricultural plots. From this literature review, it becomes clear that there is a need to
assess the interaction between SEACERs on the one hand and the spatial
configuration of the landscape on the other hand. Demont et al. (2008b) analyze this
interaction and illustrate that spatial feasibility of SEACERs can be significantly
reduced in densely planted areas if they trigger a domino-effect of non-GM crop
planting decisions. However, it could be useful if this interaction could be
summarized in a single measure.
Therefore Demont et al. (2009a) introduce the shadow factor as the ratio of the
total area of the shadow induced by SEACERs to the remaining total area planted
with the GM crop assuming perfect compliance with SEACERs. The shadow factor is
a measure for the opportunity costs borne by GM crop farmers per planted hectare of
the GM crop as a result of complying with SEACERs. The shadow factor is applied to
assess the proportionality condition of SEACERs. In a more recent Communication to
the Council and the European Parliament the EC states: “[…] coexistence measures
should not go beyond what is necessary in order to ensure that adventitious traces of
GMOs stay below the labeling threshold […] in order to avoid any unnecessary
burden for the operators concerned. While some Member States have taken this
advice into account, others have decided to propose or adopt measures that aim to
reduce adventitious presence of GMOs below this level. In some cases, proposed
measures, such as isolation distances between GM and non-GM fields, appear to
entail greater efforts for GM crop growers than necessary, which raises questions
about the proportionality of certain measures. […] Given that the majority of Member
States have not yet proposed technical field measures for coexistence, and that little
practical experience is available, a full evaluation of such measures has not yet been
possible. While the Commission recognizes the legitimate right to regulate the
10
Chapter 1. Introduction
cultivation of GM crops in order to achieve coexistence, it stresses that any approach
needs to be proportionate to the aim of achieving coexistence” (EC, 2006, p. 6). The
analysis suggests that flexible SEACERS based on pollen barriers are more likely to
respect the proportionality condition than rigid SEACERs based on isolation
distances. Particularly in early adoption stages, imposing rigid SEACERs may
substantially slow down GM crop adoption. The findings argue for incorporating a
certain degree of flexibility into SEACERs by advising pollen barrier agreements
between farmers rather than imposing rigid isolation distances on GM farmers.
The most recent paper in the debate, published by Beckmann et al. (2010), goes
one step further and discusses the interaction between ex ante and ex post regulations
in the presence of uncertainties and irreversibilities. Taking into account these extra
dimensions increases the effect of
mandatory SEACERs on the necessary
incremental benefits justifying GM crop production. Moreover, this inclusion
strengthens the findings by Beckmann and Wesseler (2007) that ex ante coexistence
regulations are discriminatory against small farms and suggests that a combination of
ex ante and ex post regulations is superior over precautionary ex ante measures only.
1.2.2 The role of ex ante socio-economic impact assessments in the
regulatory process
There is an ongoing active discussion on including this type of assessments in the
regulatory process of deregulation, both at the global and the European level.
At the global level, this discussion is centered around the Cartagena Protocol on
biosafety, an agreement of the Convention on Biological Diversity, was designed to
assure science-based risk assessments of the environmental effects of novel
technologies such as GM crops. The Protocol allows the inclusion of other
considerations such as socio-economics in the biosafety assessment. Article 26.1 says:
“The parties may take into account, consistent with international obligations, socioeconomic considerations […] The parties are encouraged to cooperate on research
and information exchange on any socio-economic impacts […].”. However the exact
meaning and interpretation of the consequences of this article is being discussed.
Article 26.1 does not explicitly state which socio-economic effects should be
considered. Many stakeholders believe that the scope of the economic consideration
exceeds the environmental level (Fransen et al., 2005), while some authors argue that
the scope of the Protocol is limited to factors affecting biodiversity in a strict sense
11
(Jaffe, 2005). Falck-Zepeda (2009) compares the inclusion of socio-economics in
different national biosafety regulations. It seems there is a tendency to go further than
the biodiversity related costs, including issues of food security, market opportunities
and labor division. Kikulwe (2010) argues that the inclusion of socio-economics in
the biosafety regulation should be done in a case by case approach as he discusses the
potential introduction of GM banana in Uganda.
Within Europe the issue of socio-economic assessments is present at different
fronts. First there is the safeguard clause, which gives Member States the
authorization to ban GM crop production, based on perceived novel concerns on
consumer safety and environmental risk. 5 One of the scientific questions raised in the
French report supporting the use of the safeguard clause states: “available studies only
consider the micro-economic level […]. Uncertainty and variation (climatic variation
and pest pressure) make analysis difficult. […] Price differences [between GM and
conventional] are not taken into account. […] No study assesses the impact at the
global level. […] Conclusion of the committee, not enough studies and data are
available at the micro-level, the supply chain and the international market.” 6
Following the publication of the report, the Haute conseil des biotechnologies was
inaugurated in 2009, with a specific commission assessing the ethic, social and
economic effects of biotechnology. In the Netherlands on the other hand, answering a
question of the minister of environment, the Commissie voor Genetische Modificatie
(COGEM), presented nine socio-economic criteria that are important in the
assessment of GM crops (COGEM, 2009).
At the European level, the European Commission will present a report in June
2010 that will discuss the position and possibilities for the inclusion of socioeconomic effects within the deregulation procedure of the EU, or commitology
procedure (see Chapter 5 for a description of this procedure). The European
Commission is expected to deliver a report indicating, among other things concerning
GMOs, their stance on the position of economic assessments in the deregulation
process. They will consider the opinion of the different Member States but above all
have to make sure that the solution is consistent with existing WTO rules on trade.
5
6
The use of the safeguard clause explains the disappearance of Bt maize in 2008 in France and in 2009 in
Germany. In total six EU Member States are now using the safeguard clause to ban GM cultivation on their
territory
This extract has been translated from French. The real quote can be found in: Comité de préfiguration d’une
haute autorité sur les organismes génétiquement modifies institué par le décret n°2007-1719 du 5 décembre
2007. Avis sur la dissémination du MON810 sur le territoire français
12
Chapter 1. Introduction
Without taking a stand in the discussion, including socio-economic studies in a
biosafety regulation and decision making is not straightforward, and may become an
obstacle in the regulatory process delaying or even blocking the introduction of GM
crops. First of all, there is the conceptual problem of weighing economic results with
risk assessments. On the other hand, investing resources in products with low
environmental risks but no economic value may never result in adoption and therefore
lost resources. This is certainly problematic for developing countries with limited
resources for biosafety assessments. Secondly, extra regulatory demands and its
delays will further increase the regulatory cost for innovators. These costs are already
a significant factor in the commercialization of GM crops (Kalaitzandonakes et al.,
2007) but the time value tends to be even higher than the cost of compliance itself
(Beyer et al., 2008). Third, and most important for the remainder of this dissertation,
as biosafety assessments take place before large scale introduction of the technology,
socio-economic effects have to be calculated ex ante. This setting means that no
observable adoption data exist and the researcher has to work under imperfect data,
requiring a set of tools distinct from ex post socio-economic impact studies.
Compared to ex post impact assessments, the literature on ex ante impact
assessments is limited. Partial budgeting approaches are popular to assess the effect of
a new technology at the farm level by comparing alternative farm plans (e.g. Alston et
al., 2002). An adoption model has to be used to determine estimated adoption rates
based on competing technologies. Breustedt et al. (2008) use a choice experiment to
determine the anticipated demand function while others often make exogenous
assumptions. A cost-benefit analysis goes one step further by considering all gains
and losses related to the technology that are measurable (e.g. Flannery et al., 2004).
The use of real options (e.g. Demont et al., 2004; Wesseler et al., 2007) introduces the
concept of irreversibility and uncertainty in the equation as it estimates the maximal
incremental social tolerable irreversible costs (MISTICs). However, these approaches
do not provide insight in the distributional effects of the proposed technology.
Kikulwe (2010) tackles this issue by explicitly combining a real options approach for
the consumer side and a willingness to pay analysis for the consumer side. Another
approach is to calculate consumer and producer surpluses in a partial equilibrium
model (e.g. Falck-Zepeda et al., 2000). An added property of market models is the
fact that policies, both direct and indirect related to the technology, can explicitly be
modeled (e.g. Demont and Tollens, 2004a). The relation between markets and
13
substitution effects can be captured in a general equilibrium model, simultaneously
determining prices and quantities under perfect competition (Nielsen, 2001). The
specifics of selected methodologies will be highlighted during the following chapters.
In this dissertation an attempt is undertaken to improve the available
methodologies for ex ante impact assessments of GM crops and provided additional
knowledge about the potential introduction of GM crops in Europe. However, the
proposed methodologies are not strictly limited to GM crops, but can be applied to a
variety of
innovations (e.g. an ongoing application on Jatropha production in
Tanzania). From an empirical point of view, some recent policy changes and novel
production constraints in the arena of GM crops are assessed and the consequences
evaluated. This way the dissertation provides a timely addition to the knowledge in
the policy literature, the impact assessment literature, stochastic partial equilibrium
modeling and the impact of agricultural biotechnology in Europe.
1.2.3 Dissertation’s hypotheses
The scholarly literature on the adoption of GM crops has highlights the importance of
the heterogeneous character of farmer populations on welfare creation and distribution
of IPR protected innovations (Weaver, 2004). This heterogeneity stems from
difference in farm properties such as soil, climate, managerial capacities and market
conditions. These heterogeneous characteristics lead to a downward sloping aggregate
demand curve for the technology (Falck-Zepeda et al., 2000), restricted monopoly
pricing (Weaver and Wesseler, 2004) and farmers that make a rational choice between
adopting and not adopting the technology (Lapan and Moschini, 2004). Oehmke and
Wolf (2004) develop a model to assess to which extent observed farmer rents can be
explained by adopter heterogeneity in USA’s Bt cotton market.
Although heterogeneity is implicitly captured by ex post adoption data and hence
in ex post impact assessments, Demont et al. (2008a) show that ex ante impact
assessments need to explicitly account for it in order to avoid a homogeneity bias.
This dissertation builds on their framework by focusing on corporate pricing
strategies under IPR in a heterogeneous population and the relation between
heterogeneity and the distribution of welfare effects among stakeholders. Hence the
following hypothesis is tested in Chapters 2 and 4 on the case study of HT sugar beet
introduction in the EU 27, a timely issue as the crop was introduced in the USA in
2008.
14
Chapter 1. Introduction
Hypothesis 1: Heterogeneity among farmers has no effect on corporate
pricing strategies of proprietary technologies and resulting welfare
effects.
In ex post adoption data of a technology, the valuation for the technology is
revealed by adopters through their adoption decision and technology expenditures.
From an ex ante perspective, adoption has not yet occurred and revealed preference
information of a technology is quasi-unobservable to the researcher. Stated preference
information can be collected through choice experiments such as contingent valuation
(CV) instead. However, this method requires costly survey data as surveys need to be
reproduced in different years and different regions in order to capture both the
structural and stochastic foundations of heterogeneity. These constraints severely limit
the use of CV analysis in large-scale ex ante impact assessment under time and
resource constraints. Hence, a flexible framework is proposed which explicitly models
heterogeneity of proprietary seed technology valuation among adopters under data
scarcity. The framework relies on the use of parametric methods to simulate
heterogeneity in the population under scarce data.
Hypothesis 2: Parametric approaches do not have the potential to
complement the inherent scarce data in ex ante impact assessments.
The standard framework is discussed in Chapter 2. In Chapter 7 the flexibility of the
approach is demonstrated by complementing the framework with a biological
component to upgrade the framework to a bio-economic model.
The impact assessment literature recognizes market interventions as a key
determinant of innovation because they distort the flow of benefits from R&D and,
hence, the incentives for innovation in the agricultural sector (Alston et al., 1995).
Wesseler (2005) highlights the importance of policies related to the introduction of
GM crops when describing the theoretical environment in which innovation takes
place. Consequently, different ex ante impact assessments of GM crops have
explicitly considered agricultural policies in their approach (e.g. Demont et al., 2004;
Demont and Tollens, 2004a). However, to my knowledge, no study has ever assessed
the impact of changes in policies that are situated in a different policy arena indirectly
affecting the innovation incentive. Hence the following hypothesis is assessed.
15
Hypothesis 3: EU policies not directly related with GM crops do not
affect the potential adoption and deregulation of GM crops in Europe.
This hypothesis is tested by modeling the impact of two indirect policy changes on the
potential introduction of HT sugar beet in Europe. Starting from the existing
EUWABSIM model (Demont et al., 2004; Demont, 2006; Demont and Tollens,
2004a), the effect of a recent change in the common market organization (CMO) for
sugar, primarily aimed at increasing competitiveness in the sector is assessed.
Particular attention is given to the effect of the change on the incentive for farmers to
adopt and the incentive for the innovator to develop and commercialize the
technology (Chapter 3). If there is a significant effect on these factors, it is necessary
to assess whether these induced changes coincide with the initial goal of increased
competitiveness.
In Chapter 5 the effect of a change in the European commitology procedure on
the chance of deregulating HT sugar beet is assessed. The literature review
demonstrated that from a purely economic point of view the introduction of GM crops
seems rational as farmers can gain significantly. However the possibility of
irreversible negative effects related to the introduction of these crops (Demont et al.,
2005b), affects the outcome of the approval process, eg. during the quasi moratorium
on GM crops between 1998 and 2004 in the EU (Wesseler et al., 2007) or the
safeguard clause. Through the use of a real options approach these irreversible effects
can be included in the analysis. Hence in Chapter 5, the results from the EUWABSIM
models serve as an input for this Bayesian analysis to estimate the MISTICs for HT
sugar beets in the EU. Assuming that individual voters base their behavior on these
MISTICs, the electoral outcome under the voting rules of two different EU treaties
can be compared with a recent proposal to nationalize the approval decision of GM
crops, the Barroso proposal.
A next set of more empirical hypotheses finds its origin in a project at the
Division of Agricultural and Food economics to estimate the potential impact of the
invasive maize pest
Diabrotica virgifera virgifera
Le Conte or Western corn
rootworm (WCR) and the competitiveness of different control options. WCR was
introduced in Europe in the early nineties (Baca, 1993) and continuously spread in
Central Europe since. An alien species can induce significant economic losses by
imposing an additional constraint on crop production. Therefore, the introduction and
16
Chapter 1. Introduction
establishment of an invasive pest should be understood in the socio-economic context
of crop production (Beckmann and Wesseler, 2003; Boriani et al., 2006). However,
very little information is available about the pest’s spread and damage in Europe (Dun
et al., 2010) and the uncertainty surrounding the resulting damages under different
control options is high (Rice and Oleson, 2005). Hence the research project created
the perfect environment to further expand the methodology developed in Chapter 2
and the discussion of Hypothesis 1 and 2. This resulted in two additional hypotheses
of this dissertation:
Hypothesis 4: The invasive species Western Corn Rootworm (WCR) does
not presents an economic threat for Central European maize farmers.
Hypothesis 5: EU farmers cannot gain from the introduction of Bt maize
resistant to WCR damages.
Chapter 6 introduces the nature and the actual status of the pest. Furthermore, the
results from a survey questioning Hungarian farmers about their production methods
and strategies to cope with the invasion are presented. Moreover, the survey is used to
determine the willingness to pay for a still to be commercialized pest resistant maize
variety trough a dichotomous choice model. Chapter 7 combines the lessons learned
from the survey with a stochastic crop protection model and the framework from
Chapter 2 to develop a bio-economic model to assess the impact of the pest and the
competitiveness of the different control options. The setup of the model has the added
value benefit of determining the factors influencing the appropriateness of a certain
control option. Chapter 8 extends the framework to seven other Central European
countries. The results show that the choice of control options depends to a high extent
on the local market and production situation.
The final chapter concludes by discussing the hypotheses and highlights some
issues for further research.
Figure 3 schematically presents the relation between the five hypotheses and the
chapters of this dissertation.
17
Chapter 3:
Stochastic partial
equilibrium model of HT
sugar beet introduction
under changing policies.
H1
Chapter 2:
methodologic
framework to
improve ex ante
impact
assessments.
H 1 and H2
Chapter 6:
Western Corn
Rootworm. Exploration
+ choice experiment
H4 and H5
Chapter 3: effect on
innovation incentive
H3
Chapter 4: Potential welfare
effects of HT sugar beet
H1
Chapter 5: includes
irreversibilities and assesses
EU approval decisions.
H3
Chapter 7:Bio-economic
model to assess WCR effects
in Hungary
H2, H4 and H5
Chapter 8: Bio-economic
model applied on extra
countries to deepen insights
H4 and H5
Figure 3: Schematic representation of the dissertation’s structure and the
interaction with the hypotheses (H)
18
Chapter 2. Corporate Pricing Strategies with Heterogeneous
Adopters: The Case of Herbicide Tolerant Sugar Beet
Adapted from Dillen, Demont,& Tollens,
AgBioForum, 2009, 12 (3&4): 334-345.
In ex ante impact assessment of proprietary seed technologies, the assessor operates
under scarce and imperfect data as no market has been established for the new
technology and adoption has not yet taken place. Recently the scholarly literature
focused on the importance of accounting for heterogeneity among potential adopters
to avoid homogeneity bias in the impact estimates. In this chapter we argue that
incorporation of heterogeneity in the corporate pricing strategy of the innovation is
needed to avoid a second bias in the welfare estimates, the pricing bias. Therefore a
framework is developed which explicitly incorporates heterogeneity of proprietary
seed technology valuation among adopters in both the pricing decision and the impact
assessment. The results explain the tendency of innovators to engage in third degree
price discrimination if the market structure discourages arbitrage. Finally the model
is applied on the case study of herbicide tolerant sugar beet in the EU-27.
2.1 Introduction
Since the commercial introduction of the first generation of genetically modified
(GM) crops in agriculture, the value creation and benefit sharing of these technologies
have been of great interest. In contrast to earlier, publicly funded technologies in
agriculture, most of the commercially available GM technologies are developed and
commercialized by the private sector. The laws and enforcement of intellectual
property rights (IPRs) have provided innovating firms with some monopoly power in
the market for GM seeds, affecting the value creation and benefit sharing of these
proprietary seed technologies (Acquaye et al., 2005;Falck-Zepeda et al.,
2000;Moschini et al., 1997;Moschini et al., 2000). The first generation of GM crops is
out there for more than a decade and ex post impact studies uncover the global value
creation (Brookes, 2009) and benefit sharing (Demont et al., 2007b) of these
technologies. Regardless of the variability of the impact evidence, Demont et al.
(2007b) reveal that, on average, two thirds of the global benefits of first-generation
GM technologies are shared among domestic and foreign farmers and consumers,
while only one third is extracted by the input suppliers (developers and seed
suppliers).
Previous research has shown that the heterogeneous character of farmer
populations (Weaver, 2004), results in a downward-sloping aggregate derived demand
curve for GM seed technologies (Falck-Zepeda et al., 2000). This suggests that,
despite their high value, GM seed innovations can be considered non-drastic (Arrow,
1962), because the monopolist’s pricing decision is constrained by the threat of
competition (Lemarié and Marette, 2003), leading to restricted monopoly pricing
(Weaver and Wesseler, 2004) and reduced adoption. In this chapter we argue that the
observed benefit sharing of first-generation GM technologies is a direct reflection of
heterogeneity of farmers’ technology valuation, constraining pricing strategies of
monopolistic technology providers. This is in contrast with earlier research stating
that the pricing of biotechnology innovations has a clear strategic dimension (Fulton
and Giannakas, 2001).
Oehmke and Wolf (2004) developed a model to assess to which extent observable
farmer rents can be explained by adopter heterogeneity in USA’s Bt cotton market.
Although heterogeneity is implicitly captured by ex post adoption data, Demont et al.
(2008a) show that ex ante impact assessments need to explicitly account for it in order
20
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
to avoid homogeneity bias. However, their impact estimates and the model
application by Demont et al. (2008) are based on fixed, exogenous, technology
pricing. In this article, we elaborate on their framework and relax the assumption of
exogeneity by introducing alternative corporate pricing strategies. A parametric
framework to fully incorporate farmer heterogeneity in ex ante impact assessment of
proprietary seed technologies under data scarcity is developed. Within this
framework, alternative corporate pricing strategies are simulated to assess the effect
of heterogeneity on classic adoption parameters such as profits and adoption. The
output of this framework can be aggregated through a trade model – as presented in
Dillen et al. (2008; 2009b), the next chapters of this dissertation – to assess the effect
of profit maximizing pricing strategies on total welfare distribution. The approach
offers both a tool for policy makers to implement ex ante socio-economic assessments
with limited resources and for seed technology and gene developers to assess the
value of proprietary seed technologies.
The chapter is organized as follows. In the following section, a theoretical
framework for modeling heterogeneity among potential adopters of a new technology
protected by IPRs is developed and some comparative statics are derived. In the third
and fourth sections, an empirical application on herbicide tolerant (HT) sugar beet in
the EU-27 is presented. A final section discusses the theoretical and empirical results
of our framework.
2.2 Model
We define a technology as a marketable good which allows farmers to surmount
an agricultural constraint. Moreover, we introduce the concept of technology
valuation to represent the willingness to pay (WTP) for the technology, which can
include both pecuniary and non-pecuniary attributes (e.g. Marra and Piggott, 2006).
While some papers (e.g. Demont et al., 2004; Wesseler et al., 2007) and Chapter 5
assess both the irreversible and reversible benefits and costs of the technology and
determine the maximum incremental social tolerable irreversible costs (MISTIC), this
chapter focuses on the reversible private benefits and costs as these drive adoption
among profit maximizing farmers. It is important to recognize the fact that the
innovation does not happen in a vacuum. Certain innovations alter the nature of
production decisions, i.e. from decisions over individual inputs (e.g. seeds and
pesticides) to decisions over bundles of complementary inputs (e.g. HT seed
21
complimented with a broad-spectrum herbicide) (Alexander and Goodhue, 2002).
Therefore the value of such technology should be calculated as the value of the new
production system as a whole. Previous research showed that the value of a
technology is not uniformly distributed among farmers; some realize a profit from the
technology and adopt it while others rationally choose not to adopt. In particular, GM
seed technologies will pay off differentially depending on field conditions, pest
densities, crop rotation and environmental conditions. Moreover, the technology
valuation to any particular farm will depend on managerial expertise and local market
conditions that condition the profitability of GM seed technologies relative to
alternative technologies (Weaver, 2004). Furthermore, technology valuation is
affected by the attitude towards risk of the potential adopter. Empirical evidence
shows that most farmers are risk averse (e.g. Anderson and Hardaker, 2003). Farmers
are averse of being exposed to unexpectedly low returns and this affects their
technology valuation and adoption behavior.
In ex post adoption data of a technology, technology valuation is revealed by
adopters through their adoption decisions and technology expenditures. From an ex
ante perspective, adoption has not yet occurred and revealed preference information
of a technology is quasi-unobservable to the researcher. Stated preference information
can be collected through choice experiments such as contingent valuation (CV).
However, this method requires costly survey data as surveys need to be reproduced in
different years and different regions in order to capture both the structural and
stochastic foundations of heterogeneity. Moreover, farmer preferences are elicited
based on hypothetical, rather than actual, scenarios. These constraints severely limit
the use of CV analysis in large-scale ex ante impact assessment under time and
resource constraints.
Therefore, we propose a framework which explicitly models heterogeneity of
proprietary seed technology valuation among adopters under data scarcity. Just et al.
(2004) recommend the use of a probability density function (PDF) to model
heterogeneity among producers. We adapt this modeling approach to the case of
heterogeneous technology adopters. In ex ante impact assessment, imperfect
information is endogenous to the problem and parametric approaches can be used to
complete scarce data with estimates based on assumptions, analogy and theory.
Let x represent farmers’ valuation or WTP for a new proprietary seed technology
and assume that the new technology is an innovation such that x > 0. Farmers’
22
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
technology valuation may include pecuniary benefits such as yield increases and cost
reductions as well as non-pecuniary benefits such as convenience and enhanced
flexibility of farming operations induced by the technology. Hence, it summarizes the
total value of all attributes of the technology the farmer is willing to pay for. Let f(x)
represent the PDF of individual technology valuations in a heterogeneous population
of farmers. From the ascending cumulative density function (CDF) of technology
valuations, F(x), the descending CDF, Q(x), can be derived, which can be interpreted
as a normalized demand curve, Q( x) ∈ [0,1] , for the new technology:
Q( x) = 1 − F ( x) ,
(1)
The restricted monopoly in the market for privately developed proprietary seed
technologies allows the innovator to set a monopolistic price which is higher than the
price of the conventional technology bundle. In most ex ante impact studies this
technology fee (θ) is treated exogenously (Demont et al., 2008a; Demont and Dillen,
2008; Demont and Tollens, 2004a; Flannery et al., 2004; Hareau et al., 2006; Krishna
and Qaim, 2007; May, 2003). Alston et al. (2002) endogenize θ based on first-order
statistics, i.e. they assume that the technology would be competitively priced, which
implies that the technology fee is set at the mean technology valuation. However,
neglecting higher order statistics does not account for heterogeneity and leads to
homogeneity bias in the estimation of the impact of the technology on welfare and
corporate revenue (Demont et al., 2008a). Throughout this chapter we assume that
farmers act rational, adopting a technology if valuation, net from technology price, is
positive, i.e. x − θ > 0 . Assuming constant long-run marginal costs, c, the profit
function of the monopolistic innovator is represented by:
) (θ − c)Q(θ ) .
π (θ=
(2)
The optimal price of the technology bundle, θ*, satisfies the following first-order
condition:
d π (θ ) dQ(θ )
.(θ − c) +=
Q(θ ) 0 .
=
dθ
dθ
(3)
23
Alexander and Goodhue (2002) argue that if users are heterogeneous, θ* may be
below the technology valuation of a potential adopter therefore leaving significant
rents with the adopter. Lapan and Moschini (2004) alternatively interpret this pricing
strategy as the monopolist choosing the profit-maximizing marginal adopter, m = θ*,
directly from a heterogeneous population of farmers and allowing the adoption to be
incomplete. This marginal adopter is indifferent between adopting and a status quo;
therefore all potential adopters with a higher technology valuation will adopt the new
technology leading to an adoption rate,
∞
∫ f ( x)dx .
ρ=
(4)
θ*
We define fa(x) as the adopters’ density function of technology valuation:
 f ( x)
(x > θ *)

f a ( x) =  ρ
.
*
0
(x ≤ θ )

(5)
The average net value (farmer surplus) of the new technology, α , for all adopters
can be measured by aggregating the farmer surplus, τ(x) = x – θ*, and amounts to:
∞
α = ∫ τ ( x). f a ( x) dx
θ
(6)
*
Similar to Oehmke and Wolf (2004) we derive some comparative statistics.
Assume f(x) is characterized by a mean µ, standard deviation σ and risk premium R.
In order to derive comparative statistics analytically, independence between µ and σ is
required. For R not to thwart this assumption, constant absolute risk aversion is
assumed. Constant absolute risk aversion is often used to analyze farm decisions
under risk (see Mitchell et al. (2004) for an application in GM crop literature).
Assuming an increasing von Neumann-Morgenstern utility function (Babcock et al.,
1993), we can calculate the risk premium R as,
R ( A, σ ) =
24
ln(0.5(e − Aσ + e Aσ )
,
Aσ
(7)
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
where A represents the coefficient of absolute risk aversion.
The effect of a change in average technology valuation over all farmers on θ* is
determined by
− Fµ ( x) + (c − θ ) f µ ( x)
dθ *
= −
,
dµ
− F ( x)θ − f ( x) − θ fθ ( x)
(8)
where subscripts denote partial differentiation. In order to determine the sign of
the derivative, additional assumptions are needed. The denominator is negative under
the assumption that f ( x) is unimodal and θ* lies on the lower tail of the PDF. As we
assume the innovation will be marketed, θ* > c, the nominator becomes positive.
Therefore, as the average technology valuation increases, the technology fee will
follow.
The effect of a change in variance is determined by
− F ( x)σ + (c − θ ) f ( x)σ
dθ *
= −
.
dσ
− F ( x)θ − f ( x) − dF ( x)θ
(9)
Under the same assumptions, this derivative will be negative. This means the
innovator will drop the price in order to maintain his customer base as the variance or
heterogeneity increases. This price decrease originates both from the direct effect of a
more heterogeneous population and the decrease of the risk premium which acts as a
shifter as
dR( A, σ )
< 0 for σ > 0. The effect of increased risk aversion is determined
dσ
in a similar way by
− F ( x ) A + (c − θ ) f ( x ) A
dθ *
= −
,
− F ( x)θ − f ( x) − dF ( x)θ
dA
(10)
where under the same assumptions θ* decreases if absolute risk aversion
increases as
dR( A, σ )
< 0 for A > 0. The induced changes in θ* translate to changes in
dA
the distribution of the technology’s welfare effects. Assessing this impact through a
formal model is difficult. First the resulting comparative statistics yield ambiguous
results. Secondly, in reality µ and σ are often linked which makes the analytical
25
traceability even lower. Therefore we discuss the results in the next section through an
empirical case study.
2.3 Application
The case of herbicide tolerant (HT) sugar beet is very appealing for EU agriculture as
this crop is grown in most EU countries and economic sugar production is impossible
without weed control. Moreover, the recent reform of the sugar regime (see Chapter
3) towards more market driven and a booming market for raw materials (biofuels and
bio-based chemistry) increases the need for cost reduction in the sector. We
understood that the major impediment comes from the concentrated group of refiners,
processors and manufacturers of sugar and sugar-containing products. Processors face
risks related to market acceptability of sugar and by-products (DeVuyst and
Wachenheim, 2005). However, it seems that recently the sugar processors have
opened their doors to biotechnology following the food and feed approval in
Australia, New Zeeland, Japan and the EU. The success of the commercial
introduction of HT sugar beet in the USA in 2008, reaching an unprecedented
adoption level of over 90% of the total sugar beet area in the second year (Kilburn,
2009), proves the high value of the technology for farmers 7.
The appropriate PDF of the technology valuation among potential adopters is the
centerpiece of the developed model. Oehmke and Wolf (2004) in their ex post
assessment construct the PDF through a kernel density estimator on available
adoption data. In ex ante assessment, the assessor faces the endogenous problem of
imperfect information which makes a similar approach impracticable. The amount of
data available to the researcher can be situated in an information continuum. At the
lower end of the continuum, no information is available and theoretical considerations
dictate the parametric estimation procedures. Towards the higher end of the
information continuum, in the direction of full information, empirical findings
gradually replace theory and non-parametric procedures substitute for parametric
procedures. Although the probability of selecting the correct parametric model is zero,
a parametric approach does not necessarily lead to biased results. With a false model,
parameters will converge such that the Kullback-Leibler distance between the true
7
In 2009, a district court in California ruled that the environmental assessment preceding introduction by the
United States Department of Agriculture was not adequate resulting in a suspension of the cultivation
authorization. On March 16 2010 the suspension was temporarily lifted as a cultivation ban would have severe
economic consequences. The final decision on the pending court case will follow later in 2010.
26
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
density and the best parametric estimate is minimized. Therefore it is possible that an
incorrect parametric model may have greater efficiency than the correctly modeled
density and a non-parametric model (Goodwin and Ker, 2002). In the case of data
scarcity, expert opinions are often used to construct subjective probabilities for
uncertain parameters. Although very popular, the triangular PDF might not be the best
way to model expert opinions for modeling heterogeneity of technology valuation.
Both tails are overemphasized and often experts have better knowledge on the central
tendencies than on the extremes. Therefore the PERT PDF, a special case of the Beta
distribution, is preferred for modeling expert opinions. It can range from highly
skewed till symmetrical distributions, has a close fit to normal and lognormal
distributions and attributes less weight to the extremes. The excessive reliance of the
triangular PDF on the extremes also influences the variance introduced into the
model.
0.25
0.24
Standard deviation
0.23
0.22
0.21
0.2
0.19
Triangular distribution
0.18
PERT distribution
0.17
0.16
0.15
0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 0.95
Most likely value
Figure 4: Effect of parameter specification, choice of most likely value, on the
resulting standard deviation of the triangular and the PERT PDF
In Figure 4 we observe that for the same values of the extremes, the triangular
distribution always has a higher standard deviation, therefore introducing unnecessary
uncertainty in the model if we assume that we can rely on the expert’s knowledge.
Demont et al. (2008a) use a re-parameterized version of the classic PERT
specification to construct a PDF based on their specific context of limited data
availability. In this article we are one step further in the information continuum as we
combine expert opinions with evidence from farmer surveys, which allows us to
27
select the PDF based on empirical goodness-of-fit instead of forcing a pre-defined
PDF to fit our expert data.
For economic sugar beet production, effective weed control is crucial. Yield
losses can be up to 100%, such is the poor ability of beet to compete with the large
range of weeds present in arable soils (Dewar et al., 2000). Expert opinions on the
herbicide expenditures for most of the EU countries are reported by Hermann (1996;
1997; 2006), while detailed cross-sectional survey evidence is available for the
Netherlands (IRS, 2004) and France (ITB, 2000; Lemarié et al., 2001). We use the
first data source to obtain basic first and second order moments of the underlying
data, while we use the second data source to determine higher order moments in order
to determine the appropriate functional form to be used in our parametric modeling.
We fit alternative CDFs on the available cross-sectional datasets on herbicide
expenditures in sugar beet growing. The loglogistic CDF consistently provides the
best fit based on the root mean square error (RMSE) goodness-of-fit criterion and is
therefore chosen to model heterogeneity among potential adopters (Table 1).
Table 1: Cumulative loglogistic distributions fitted on survey data on weed
expenditures
Parameter
Estimate
Model: H(z) = 1/(1+(z/γ)^(-δ)) (equation 11, assuming ϕ = 0 before adoption of HT
sugar beet)
France, 1997
France, 2000
Netherlands, 2004
Scale (γ)
Shape (δ)
Scale (γ)
Shape (δ)
Scale (γ)
Shape (δ)
103.3*
4.8* (0.2)
160.2*
4.7*
803.3*
5.0*
(1.0)
(0.4)
(0.04)
(3.7)
(0.1)
0.99852
0.99981
0.99917
0.003
0.006
0.005
R²
Goodnessof-fit
(RMSE)
Sources: IRS (2004), ITB (2000), and Lemarié (2001)
Notes: Standard deviations are shown between brackets. * = significant at the 0.01 level.
We combine the empirical results with the expert opinions through the estimation of
the three fractiles in the survey data corresponding with the minimum, maximum and
mean expert opinions. By assuming that the correspondence between both data
sources is constant, we calibrate loglogistic CDFs of herbicide expenditures on expert
opinions for major sugar producing Member States of the EU-27 in 2004 (Table 2).
With this procedure we avoid the difficulties of experts to estimate the extreme
fractiles of a distribution (Miller, 1956). This yields the following PDF, h(z), and
CDF, H(z), for our case study:
28
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
δ −1
 z.(1 − φ ) 
δ.
γ 

h( z ) =
2
  z.(1 − φ ) δ 
γ . 1 + 
 
 
  γ
H ( z) =
1
 z.(1 − φ ) 
1+ 

 γ

−δ
,
(11)
(12)
where z represents herbicide expenditures, γ the scale parameter and δ the shape
parameter of the loglogistic PDF. In order to make the concept of technology
valuation more dynamic, a shift parameter, φ > 0 , is introduced. The parameter can be
used to introduce constant absolute risk aversion premium, R(A, σ), as presented
earlier, and various exogenous price and cost effects into our static model. For the
assessment of a biotechnological innovation, the latter can include: (i) the inclusion of
coexistence measures (e.g. Devos et al., 2009), (ii) the existence of a market premium
for non-GM crops and (iii) changes in the pricing strategies of competing
technologies. In the case of HT crops, history shows that conventional herbicide
producers try to countervail the increased competition with price reductions. This
effect took place with the introduction of HT soybeans in the US and can be observed
with the introduction of generic products in the herbicide market as well (Just, 2006).
Due to data limitations, we assume the cost implications of these price reactions to be
homogeneous in the population and to shift the PDF without affecting relative
heterogeneity measured by the ratio of the mean to the standard deviation as both
statistics are multiplied by the same factor. Although spatial variation in some of these
factors certainly exists, we assume that this heterogeneity is captured by the herbicide
expenditures. In order to transform herbicide expenditures, z, to technology valuation,
x, we subtract the total cost of the HT system, excluding the technology fee, from the
total cost of the conventional system, including application and product costs. The
country-specific PDF of technology valuation can then be calculated as
fi ( x) = hi [ x + (nci − ngi )ki − pgli .gi ] ,
(13)
29
where ngi and nci represent the number of herbicide applications in the HT and
conventional system respectively, ki the cost of a single herbicide application
(assumed equal for glyphosate and conventional applications), gi the dosage and pgli
the price of glyphosate in country i. Note that (x + nciki) represents the total cost of the
conventional system while (ngiki + pgligi) represents the total cost of the HT
replacement system, excluding the technology fee. Again, due to data limitations we
assume application costs and glyphosate expenditures to be homogeneous at the
country level and to shift the PDF of technology valuation horizontally without
affecting its shape.
Based on the resulting PDF of technology valuation among potential adopters, the
innovating sector (technology developers and seed suppliers) will set its price in order
to maximize profits following the presented modeling framework. We differentiate
between two distinct corporate pricing strategies. In the first pricing strategy we use
disaggregated data to calculate the price at the member state level. This pricing
strategy is known as spatial third degree price discrimination (Schmalensee, 1981).
The second pricing strategy assumes a uniform price setting over the EU-27 sugar
beet producing countries. The resulting proprietary seed demand function for the
proprietary seed is constructed based on the area-weighted average herbicide
expenditures:
n
Q( x) = ∑ f i ( x).χ i ,
(14)
i
where n represents the number of countries taken into account and χi the share in area
over the n countries. The two different pricing strategies enable investigating the
effect of heterogeneity on the variables under research. Uniform pricing is a realistic
pricing strategy for the European sugar sector as it is common for sugar producers
(processors) to purchase seed and distribute it upstream with contracted farmers. This
practice, combined with the high concentration of the sugar sector through
multinational companies (Smith, 2007), creates a situation of oligopsony at
supranational levels which would favor a uniform pricing strategy.
With these specifications, the impact of the different pricing strategies on
adoption and value creation can be calculated following the system of equations 3-6.
30
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
2.3.1 Data
In Table 2 the data used in this chapter are presented. Similarly to Desquilbet and
Bullock (2009), we assume for simplicity that the long-run marginal costs of
supplying seed are zero, which comes down to assuming that the technology
developer maximizes revenue instead of profit. All prices are deflated to the 2008
level by using GDP deflators (World Bank, 2008). To aggregate the revenues to the
national level and for yields, we use data from F.O.Licht (2005). The shift parameter,
φ , is calibrated on the expected reduction of the price level of competing
conventional herbicides. Analogous to the effect of HT soybeans in the US we
estimate this reduction to take the value of 20% (Just, 2006). Moreover, analogously
to other studies (Demont and Dillen, 2008; Dillen et al., 2008; Dillen et al., 2009b) we
include a 5% yield increase for adopting farmers due to reduced toxicity of the
glyphosate herbicide regime in HT sugar beet cultivation. The price of glyphosate is
assumed uniform among Member States (€4.37/l) due to data limitations and to make
abstraction of those markets where glyphosate is not well established and where the
price structure is likely to change with the introduction of HT sugar beet. To construct
the demand function in the uniform pricing strategy we use the area-weighted average
of fi(x) for the countries producing three quarters of the European production
(equation 14). These regions can be considered as the main market and include the
efficient production regions where innovation incentives are positively affected by the
recent sugar reform and therefore feature a higher likelihood to adopt HT sugar beet
(see next chapters).
31
Table 2: Weed control costs under conventional and HT sugar beet technology in the EU-27
Member
State
Min
(€/ha)
Mean
(€/ha)
Max
(€/ha)
Conventional weed control
Scale
Shape
Application
parameter parameter
cost k (€/ha)
γ
δ
261
5.4
41.5
207
4.2
18.5
160
3.9
18.2
223
6.1
13.0
180
10.0
7.3a
Number of
applications
nc
2.5
3.5
3
3
3b
Total
application
costs (€/ha)
103.8
64.8
54.6
39.0
21.9
HT weed control
Number of Glyphosate
Glyphosate
applications
dose g
expenditures pgl
ng
(l/ha)
(€/ha)
2.5
6
26.2
2.5
6
26.2
2.5
6
26.2
1
3
13.1
2.5
6
26.2
Austria
156
311
467
Belgium
104
261
417
Germany
76
206
334
Spain
141
261
381
Czech
138
198
276
Republic
France
103
150
206
136
9.7
20.0
3.8
76.0
2.5
6
26.2
Finland
154
220
297
201
10.0
16.4
3.8
62.3
2.5
6
26.2
Greece
94
132
202
121
10.5
17.5
1.5
26.5
1
3
13.1
Italy
95
169
253
145
6.4
15.3
2.5
38.3
2.5
6
26.2
Ireland
64
93
122
84
9.7
13.3
3
39.9
2.5
6
26.2
Netherlands
135
176
238
164
13.5
40.4
3.5
141.4
2.5
6
26.2
Poland
121
214
332
185
6.4
7.3
3b
21.9
2.5
6
26.2
Sweden
77
186
308
149
4.3
13.3
2.9
38.6
2.5
6
26.2
UK
78
149
225
124
5.9
14.4
4.6
66.2
2.5
6
26.2
Denmark
88
212
372
166
4.4
26.0
4
104.0
2.5
6
26.2
Portugal
141c
261c
381c
223
6.1
13.0
3
41.7
1
3
13.1
d
d
d
a
b
Hungary
64
159
211
132
2.7
7.3
3
21.9
2.5
6
26.6
Sources: Hermann (2006) for most cases, complemented by Hermann (1996; 1997) if recent data is missing, Schäufele (2000) and Bückmann (2000)
a
no data available, data for Poland were used.
b
Urban et al. (2008)
c
no data for Portugal, we used data from Spain
d
no data from Hermann (1996; 1997; 2006), data from AKI (2005)
Total
application
costs (€/ha)
103.8
46.3
45.5
13.0
18.3
50.0
41.0
17.5
38.3
33.3
101.0
18.3
33.3
36.0
65.0
13.9
18.3
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
2.4 Results and Discussion
Table 3 presents the results of our pricing framework. Under third degree price
discrimination, the technology fee ranges from €50/ha to €147/ha. The area-weighted
average technology fee amounts to €98/ha. These spatially adapted technology fees
result in adoption potentials (ceilings) that range from 51% for Ireland to 97% for
Spain. The estimated adoption rates are remarkably close to potential market
penetration rates estimated by experts of national sugar beet institutes (Coyette et al.,
2002). The uniform pricing strategy yields a European-wide price of €95/ha. This fee,
less adapted to local conditions and demand, generates a dramatically different
adoption pattern from almost no adoption in Ireland to almost full adoption in Spain.
At first sight, the endogenous technology fees seem rather high compared to
commercially available GM seeds of other crops. However they are in line with the
current technology fee of HT sugar beet in the USA, i.e. €90-106/ha (KWS, 2006),
and can be explained by the high herbicide expenditures in conventional sugar beet
cultivation (Table 2). The results confirm the theoretical model as the increased
farmer heterogeneity of herbicide expenditures, generated by applying a uniform
pricing strategy over a larger area, leads to a lower technology fee at the EU-27 level.
Under price discrimination, the area-weighted average of the farmer profits
amount to €99/ha with the extremes in Spain and Ireland. Corporate revenue in the
EU-27 is €154 million. If third degree price discrimination is not possible, e.g. due to
the market structure, the farmer profits amount to €116/ha while the corporate
revenue drops to €141 million. Under increased farmer heterogeneity of herbicide
expenditures, farmers are able to capture a higher rent while the innovator looses
revenue. The lower technology fee will relocate the position of the marginal adopter
but the increased adoption rate is not enough to compensate for the reduced optimal
technology fee on a per-hectare base. Therefore the corporate goal of increasing
adoption might not be the revenue maximizing strategy. Under market structures with
strong IPRs which reduce arbitrage, third degree price discrimination is a profitable
strategy for the innovator. By dividing the population into smaller, more homogenous
segments, corporate revenue increases by progressively extracting farmer surplus
from the total value of the proprietary seed technology. Heterogeneity is minimized in
each sub-market fi(x) and revenue is maximized. Empirical evidence can be found for
Bt cotton in the US, Mexico and South Africa (Frisvold et al., 2006; Gouse et al.,
33
2004; Traxler et al., 2003). The separation is mainly introduced by different
germplasm due to the sensitive reaction of upland cotton varieties to agro-climatic
changes (Acquaye and Traxler, 2005) and preventing farmers to buy seeds in other
districts (Traxler et al., 2003). With strong IPRs, the submarkets could be introduced
through enforcement of a contract with a “no resale” clause, thereby strengthening
monopoly power. In Europe, price discrimination can be found in Bt maize in Spain
as a function of spatial pest infestation levels (Gomez-Barbero et al., 2008). These
observations indicate that the innovator is aware that patent-based uniform pricing of
a GM seed technology leaves substantial benefits with the producers, preventing full
appropriation by the innovator due to heterogeneity among farmers. In developing
countries with weak or no governance of IPRs, on the other hand, price discrimination
of HT technologies can only be efficiently implemented in the case of hybrids as
hybridization biologically strengthens IPR protection and acts as a substitute for weak
IPR enforcement (Goldsmith et al., 2003). This may explain the absence of price
discrimination in the case of HT soybeans in developing countries (Qaim and Traxler,
2005).
Due to the antagonistic response of farmer rents and corporate revenues to
changes in heterogeneity, the magnitude of total welfare depends on the shape of the
parametric model. In the case of the loglogistic PDF, the upwards effect on farmer
rents following an increase in heterogeneity more than compensates the downwards
effect on corporate revenue on a per-hectare base such that the total value increases.
The total value under uniform pricing amounts to €218/ha while it only reaches
€204/ha in the case of third degree price discrimination. Full welfare effects under
uniform pricing are calculated through the use of a partial equilibrium model,
EUWABSIM (Chapter 3).
34
Table 3: Technology fee, innovator revenue, farmer rent and total value of HT sugar beet under two alternative pricing strategies, price
discrimination and uniform pricing, in the EU-27
Member State
Austria
Belgium
Germany
Spain
Czech Republic
France
Finland
Greece
Italy
Ireland
Netherlands
Poland
Sweden
UK
Denmark
Portugal
Hungary
EU-27 area-weighted average
(total)
Price discrimination
Technology
Farmer
Adoption Corporate
fee
incremental
(%)
revenue
(€/ha)
profits
(million €)
(€/ha)
147
123
74
4.9
123
131
76
8.1
94
99
69
26.5
145
135
97
12.9
98
87
88
5.6
87
133
90
27.0
125
68
79
3.5
75
95
90
2.2
77
59
79
9.7
50
32
51
0.7
121
92
92
11.2
102
79
80
22.4
86
74
65
2.5
78
117
86
8.8
106
122
80
4.3
145
135
97
1.2
108
63
84
2.6
98
99
(154)
Total
value
(€/ha)
270
254
193
280
185
220
193
170
136
82
239
181
160
195
228
280
171
204
Uniform pricing
Technology
Farmer
Adoption Corporate
fee
incremental
(%)
revenue
(€/ha)
profits
(million €)
(€/ha)
95
188
84
3.8
95
174
98
7.2
95
108
63
26.3
95
192
99
9.0
95
101
91
5.6
95
131
78
26.9
95
102
97
2.7
95
63
43
1.8
95
45
43
8.6
95
1
5
0.003
95
151
97
8.6
95
96
85
22.0
95
71
47
2.5
95
99
59
8.5
95
145
82
4.0
95
192
99
0.7
95
78
92
2.4
95
116
(141)
Total
value
(€/ha)
283
269
203
287
196
226
197
158
140
96
246
191
166
194
240
287
173
218
Note: For comparison, the market potentials estimated earlier by experts of national sugar beet institutes in six selected countries are 75% for Belgium, 71% for Germany,
100% for Spain, 90% for France, 93% for the Netherlands and 89% for the UK (Coyette et al., 2002)
Demont et al. (2008a) demonstrate the homogeneity bias that emerges from not
incorporating second order statistics in ex ante impact assessment. They conclude that
the bias is a decreasing function of the mean and an increasing function of the
variance of f(x). This result is in contrast with earlier statements in the literature that
the distribution is of minor importance as the lower and higher tails of the distribution
will compensate each other (Breustedt et al., 2008). However, by using an exogenous
technology fee, θ , they potentially introduce a second source of bias, i.e. pricing bias.
In the literature on HT sugar beet, different assumptions and estimates have been
reported for the technology fee, varying from €25/ha (Flannery et al., 2004), €3040/ha (Märländer, 2005), €40/ha (Demont, 2006), €32-48/ha (May, 2003), €38/ha
(Gianessi et al., 2003) and €77/ha (Lemarié et al., 2001) in Europe to €128/ha
(Gianessi et al., 2002), €133/ha (Burgener et al., 2000), €157/ha (Rice et al., 2001),
and €164/ha (Kniss et al., 2004) in the USA. We illustrate the magnitude of the
exogenous pricing bias for France by comparing farmer surplus generated under the
assumption of an exogenous technology fee of €40/ha with surplus generated under
the assumption of third-degree price discrimination. The assumption of exogeneity
would lead the researcher to overestimate the potential adoption ceiling by 11%
(100% instead of 89%) and the value captured by farmers by 39% (€185/ha instead of
€133/ha). If the technology developer commercializes HT sugar beet seed into the
European sugar industry through uniform pricing, endogeneizing the technology fee
but assuming third-degree price discrimination instead of uniform pricing would
generate a second source of bias, i.e. farmer benefits would still be overestimated by
1.5% (€133/ha instead of €131/ha). The small bias is explained by the fact that
France, the largest sugar producer, would play a determinant role in EU uniform
pricing. The total pricing bias on farmer surplus would then be the sum of both biases
and would amount to 41% (€185/ha instead of €131/ha).
Our results underline the importance of incorporating heterogeneity in ex ante
impact assessment of proprietary seed innovations, especially if the impact assessor is
operating under data scarcity and imperfect expert information. Not accounting for
heterogeneity introduces both homogeneity bias and pricing bias into the impact
estimates. Moreover, the presence of heterogeneity explains the strategy of the
innovating seed sector to engage in third degree price discrimination as the latter
maximizes corporate revenue but possibly reduces total welfare.
36
Chapter 2. Corporate Pricing Strategies with Heterogeneous Adopters
Finally, our proposed parametric modeling framework provides a practical tool for
breeders, technology developers, agricultural economists, crop protectionists and
biosafety regulators to estimate the value of future proprietary seed technologies
under data scarcity. In particular it can be useful for large-scale ex ante impact
assessment under time and resource constraints, such as the socio-economic
assessment of new technologies under the Cartagena protocol. Falck-Zepeda (2009)
argues that inclusion of socio-economic considerations may become an obstacle to
potentially valuable technologies, particularly for developing countries facing higher
barriers in terms of biosafety regulatory compliance due to resource constraints.
37
38
Chapter 3. European Sugar Policy Reform and Agricultural
Innovation
Adapted from Dillen, Demont, & Tollens,
Canadian Journal of Agricultural Economics, 2008, 56:533-553
In July 2006, the European Union’s Common Market Organization (CMO) for sugar
underwent the first radical reform since its establishment in 1968. In this study we
model the incentives for adoption of a new technology, herbicide tolerant sugar beet,
before and after the policy reform. We take the observed reform-induced changes in
production incentives as given. A stochastic partial equilibrium model is built to assess
the innovation incentives under the old and the new CMO. We use the model to analyze
the effect of the policy reform on the adoption of new technologies. Our findings show
that the new CMO does not jeopardize the goal of increased competitiveness in the
European sugar market as a whole. The adoption incentive of high cost farmers is
significantly reduced under the new CMO, decreasing their ability to cope with future
changes in the sugar market. Medium cost producers on the other hand have grater
incentives under the new CMO to adopt new technologies while low cost producers are
left unaffected. These results show that the new CMO disfavors adoption by high cost
producers of a new technology. The reduced adoption leads to lower flexibility and
competitiveness of these farmers and therefore coincides with the goals of the reform to
crowd out high cost producers.
3.1 Introduction
In 2006, the European Union (EU) implemented the first radical reform of the Common
Market Organization (CMO) for sugar since its establishment in 1968. Two key factors
are said to have initiated this reform. First, there is the conviction of the EU sugar
policy on the 28th of April 2005, when after a complaint from Brazil, Thailand and
Australia, it was decided that the export of out-of-quota sugar did not comply with
World Trade Organization (WTO) rules because of a cross-subsidy effect. This socalled C-sugar is supposed to be traded on the world market at low prices due to the
high support for in-quota sugar and is therefore indirectly supported by the policy. The
second constraint is the Everything But Arms (EBA) agreement which would offer free
access on the European market to sugar from the least developed countries (LDCs)
starting in 2009. To cope with this anticipated increased competition from LDCs, a
more competitive European sugar sector was a necessity.
Several studies have attempted to estimate the effects of alternative reform
scenarios of the European sugar sector and the CMO itself, both at the farm level and at
the aggregate level (Bogetoft et al., 2007; Buysse et al., 2007; de Bont C.J.A.M. et al.,
2006; Frandsen et al., 2003; Gohin and Bureau, 2006; Nolte, 2008; Schmitz, 2002).
These studies have focused on the demand and supply side of the sugar market. In
contrast with previous studies which model the impact of the sugar policy reform on the
incentives for production and consumption assuming unaltered incentives for
innovation, we explicitly model the incentives for innovation under changing sugar
policies, while taking the observed reform-induced changes in incentives for production
and consumption as given.
Market interventions are a key determinant of innovation because they distort the
flow of benefits from R&D and, hence, the incentives for innovation in the agricultural
sector (Alston et al., 1995). However, although agricultural innovation could assist the
European sugar sector to become more competitive, so far no studies have modeled to
which extent this new policy alters the incentives for the adoption of new technologies. 8
In this chapter we assess whether the secondary effect of a policy reform, i.e. its impact
on innovation incentives, is consistent with the objectives of the reform to increase the
8
It is important to make a distinction between the incentive to innovate for innovators (gene developers and seed
suppliers) and the incentive to adopt innovations for farmers. The surplus of an innovation is the total value to
the innovator and farmers of a cost-reducing innovation, while the incentive to adopt innovations is purely based
on the share of the surplus accruing to farmers. We are dealing with proprietary innovations in this paper and
assume the farmer pays a technology premium for a license to the developer of the technology.
40
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
competiveness of the European sugar sector. We test whether the null hypothesis of the
chapter, the policy reform has no influence on the stream of benefits from innovation,
holds or not. This statistical test complements Demont (2006) who provided a first and
rough attempt to predict the impact of the new CMO.
Our model is a stochastic partial equilibrium model and the specific innovative crop
analyzed is genetically modified (GM) herbicide tolerant (HT) sugar beet. The case of
herbicide tolerance in sugar beet growing is very appealing for EU agriculture as sugar
beet is grown in most of its Member States and weed control is crucial to economic beet
production. Previous research has forecast that EU beet growers could substantially
gain from HT technologies (Demont et al., 2004; Demont, 2006; Demont and Tollens,
2004a). Most European countries have sufficient research experience in GM sugar beet,
but for the time being their processing industries remain largely reluctant to accept this
technology in fear of losing domestic and export markets (DeVuyst and Wachenheim,
2005). In the USA, in contrast, many processors have cleared their growers to plant GM
HT sugar beet in the spring of 2008, developed by the gene developer Monsanto and
seed company Betaseed (part of the KWS group), following approval for feed en food
use in Mexico, Japan, Australia, New Zealand the EU and others. Secondly, the demand
for this technology by farmers (groups) is important in the USA, where farmers have
extensive experience with GM crops since 1996.
A key factor in modeling the European CMO for sugar is the specification of
responsiveness of out-of-quota supply to world prices in the distorted world market.
Beet growers may overshoot their quotas for two reasons, namely (i) as a precautionary
measure to ensure quota fulfillment, or (ii) to speculate on the world market. In the
literature, alternative argumentations on the incentives behind observed out-of-quota
production have been presented (see Buysse et al., 2007 for a review). In this article we
propose a balanced multi-criteria system to identify Member States that consistently
supply out-of-quota sugar in response to world prices.
The chapter is organized as follows. In the next section, we introduce the stochastic
partial equilibrium model which explicitly models the old and the new EU sugar
regimes. The model simulates the introduction of GM HT sugar beet in a disaggregated
European Union with spillovers to the sugar beet sector in the rest of the world. We use
the model to study the incentive to innovate under the two regimes. In the model
section, we calibrate the model with secondary data. In the results section, we report the
Monte Carlo simulation outcomes. The final section discusses the results.
41
3.2 Model
The focus of the framework lies at the level of the farmers, who are the main actors in
the adoption process of new varieties of crops. The incentive to innovate and adopt is
assessed in a rational choice model (see Scot, 2000, for a review and common
critiques), that is, a farmer makes his decision purely based on the per-hectare profit of
adoption. For suppliers of the innovative input, HT sugar beet in this example, only the
effect on revenue is estimated.
To analyze the potential benefits of the introduction of HT sugar beet in the sugar
sector, we set up a spatial model that differentiates between sugar cane and sugar beet.
The sugar beet region is further divided into the EU-27 and the ROW, including mainly
the rest of Europe, followed by the USA. We believe that a three-region model,
covering the EU-27 (disaggregated later on in the chapter in order to obtain a more
realistic supply response for the EU) and the beet and cane growing regions in the rest
of the world (ROW) captures well the essence of production and trade in the global
sugar market.
Moschini and Lapan (1997) argue that impact assessments of new technologies
should account for the relevant institutional and industry structures responsible for the
actual development of these technologies. Most of the recent agricultural biotechnology
innovations have been developed by private firms protected by intellectual property
rights (IPR), such as patents, which confer temporary restricted monopoly rights to the
discoverer. Prices for IPR protected inputs are higher than they would be in a perfectly
competitive market. Our model draws on the impact assessment framework of
Moschini et al. (2000), which exogenously incorporates monopolistic power in the
input market. They develop a profit function for farmers operating in the output market
which explicitly accommodates the characteristics of the HT technology. By
introducing the simplifying assumption of exogenous monopolistic pricing of the
technology in the profit function, they separate the incentives for innovation from the
incentives for the adoption of the proprietary technology:
π i , j ( p, ρ ) = Ai , j + ρα i +
(1 + ρβi )Gi , j
1 + ηi
p1+ηi − µi ρ ,
(1)
where the HT technology is parameterized through four parameters, i.e. the adoption
rate ρ, the technology-induced per-hectare profit gain αi, the boost βi in percentage and
the per-hectare licensing premium or ‘technology fee’ μi. Furthermore, Ai,j and Gi,j
42
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
represent scale parameters (i represents the region and j the year), p the sugar price, and
ηi the elasticity of yield with respect to sugar prices (as defined by Moschini et al.,
2000). Applying Hotteling’s lemma, the yield function can be derived as
(1 + ρβ )Gi , j pηi . Next, the scarce factor to the industry, i.e. land, is represented, as being
available with an upward sloping non-linear constant-elasticity supply function as a
function of the average per-hectare profit, that is,
θ
Li , j = λi , jπ i , j ( p, ρ ) i , j .
(2)
Multiplying the land supply function with the yield function results in a region- and
year-specific supply function:
θ
=
Qi , j ( p, ρ ) λi , j [π i , j ( p, ρ )] i , j (1 + ρβi )Gi , j pηi
(3)
where λi,j is a scale parameter and θi,j the land supply elasticity (area elasticity) with
respect to sugar beet profit per hectare. Our model includes 18 regions, i.e. the ROW
cane region (i = 0), the ROW beet region (i = 1), 16 EU-27 Member States (i = 2, 3, …,
17), and covers nine agricultural seasons (2006-2014) (see Table 4). 9
Aggregation of 18 supply functions allows modeling the effect on world sugar
prices of the interaction between two aggregate regions, the EU-27 and the ROW, as a
consequence of the introduction of the HT technology. However, the structure of these
functions implies that all 18 regions in the model are able to participate in the aggregate
supply response to prices. While all regions certainly respond to a certain regionspecific incentive price, in reality only the most competitive producers respond to world
prices. Due to domestic price regulations and high production costs, uncompetitive
Member States are isolated from world market incentives. Their supply is inelastic and,
instead of exporting on the world market, any surplus generated through enhanced
yields will be absorbed by their domestic quotas and will free up land allocated to sugar
beet. Hence, we distinguish between world price responsive (WPR) and world price
nonresponsive (WPN) Member States, which will be identified through a multi-criteria
framework at a later stage in this chapter. The aggregate EU sugar supply function in
year j is given by,
9
In the remainder of the text, the term ‘region’ will be used for aggregates (ROW beet, ROW cane, EU-27), but also
for individual EU-27 member states. We merged Belgium and Luxemburg in a single region ‘Belgium’ and
omitted nine remaining New Member States, who supply less than 8% of sugar in the EU-27. We also omitted
Ireland which gave up production since the reform of the CMO, making comparison between the two regimes
impossible.
43
QEU, j ( =
p, ρ EU )
17
∑Q
=i 2
=
( p, ρi )
i, j
∑Q
+
∑Q
i, j
=i WPI
=i WPR
i, j
( p, ρi )
(4)
In Equation 4, ρEU represents the 16x1 adoption vector of the new technology in the
EU with elements ρi (i = 2, 3, …, 17). This aggregate sugar supply function is very
detailed in that it contains 10 parameters per member state totaling 160 parameters, of
which 64 are related to the new technology.
Table 4: Regional specification of incentive prices and elasticities under the new
and the old sugar EU sugar policy for the Member States included in the model
ϕi,j*
Region
Incentive price
Former
6
5
New
6
5
Former
world price
world price
New
world price
world price
Belgium
4
8
world price (C)
Denmark
Germany
3
4
7
8
B sugar price
world price (C)
Greece
Spain
France
0
3
4
7
7
8
A sugar price
B sugar price
world price (C)
Italy
Netherlands
1
2
7
7
Austria
4
8
A sugar price
mixed price (A, B and
a fixed quantity of C
sugar)
world price (C)
world price
(industrial)
instit. price
world price
(industrial)
instit. price
instit. price
world price
(industrial)
instit. price
instit. price
ROW cane
ROW beet
Area
elasticity
Elasticity
of yield
0.290
0.202
0
0
0.055
0.08
0.034
0.074
0.08
0.08
0.228
0.226
0.172
0
0.08
0.08
0.712
0.041
0.08
0.08
world price
0.154
0.08
(industrial)
Portugal
0
7
A sugar price
instit. price
0.228
0
Finland
1
7
A sugar price
instit. price
0.064
0.08
Sweden
2
7
B sugar price
instit. price
0.030
0.08
United
4
8
world price (C)
world price
0.176
0.08
Kingdom
(industrial)
Czech Rep.
4
7
world price (C)
instit. price
0.569
0.08
Hungary
3
7
B sugar price
instit. price
0.5686
0.08
Poland
3
7
B sugar price
instit. price
0.5667
0.08
Sources: Devadoss and Kropff (1996), Poonyth (1998), Frandsen et al. (2003), Banse et al. (2005), and
Confédération des Betteraviers Belges (2002).
*
Legend for ϕi,j:
0 = A-quota not fulfilled
1 = A-quota fulfilled
2 = A-quota fulfilled and significant part of B-quota
3 = A-quota and B-quota fulfilled but world price nonresponsive (WPN) sugar
production
4 = A-quota and B-quota fulfilled but and world price responsive (WPR)
sugar production
5 = ROW but world price nonresponsive (WPN) sugar production
6 = ROW and world price responsive (WPR) sugar production
7 = post-reform quota fulfilled
44
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
Next, we model innovation adoption. Innovation adoption occurs in a large, open
economy with technology spillovers and shape the two-region framework of Alston et
al. (1995) to the specific features of the EU’s CMO for sugar. For each beet growing
region i, the four technology-specific parameters ( ρi , α i , βi , µi ) engender a pivotal,
divergent shift of the supply curve (Equations 1-3). At the centre of the analysis is the
calculation of a counterfactual world price pj in year j to isolate the effect of the
technology-induced supply shift from other exogenous changes in supply and demand.
It represents what the world price would be if all supply and demand conditions are
identical except for the introduction of the new technology. Hence, in our analysis we
represent the world price as a function of the worldwide adoption vector: pj(ρW).
The EU’s export supply curve in year j can be modeled as:
ES j ( p, ρ EU ) = QEU , j ( p, ρ EU ) − C j
(5)
where C j is EU sugar demand in year j, assumed exogenous and independent of the
price due the inelastic response to sugar prices in the EU. The EU’s export supply
expansion, generated by a technology-induced pivotal shift of the EU’s aggregate
supply function, would cause the world price to decline from pj(0) to pj(ρEU). This price
decrease is determined using a reduced-form equation, extracted from the University of
Missouri’s FAPRI world sugar model, which calculates the world sugar price as a
function of actual and lagged EU net sugar exports (Poonyth et al., 2000). 10 For each
year j the reduced-form equation transforms the observed world price into the price that
would result from the EU’s technology-induced export expansion in year j and j – 1:

ES j ( p j (0), ρ EU, j ) − ES j ( p j (0),0)
ES j −1 ( p j −1 (0), ρ EU, j −1 ) − ES j −1 ( p j −1 (0),0) 
p j ( ρ EU, j ) =
p j (0) 1 + σ 1
+σ 2

ES j ( p j (0),0)
ES j −1 ( p j −1 (0),0)


(6)
with σ1 = -1.0 and σ2 = 0.46. The variables σ1 and σ2 are the short- and long-run
flexibility, respectively, and they reflect sugar export demand elasticities that are twice
as large in the long-run as in the short-run. The positive value for the coefficient of the
lagged export supply expansion term reflects the output contraction of the ROW as a
reaction on the world price decline from pj(0) to pj(ρEU). The reaction stems mainly
10
If we assume EU imports to be exogenous, we can calculate the world price as a function of the EU’s export
supply expansion. Under the old CMO the import was fixed, due to fixed ACP (African, Caribbean, and Pacific)
import agreements. Under the new CMO the EBA agreement will offer free access to the European sugar market
for the LDCs in 2009. In a recent simulation, Nolte (2008) observes that the extra import under the EBA only
has a marginal effect on the import and on the price level (-1%) in Europe due to the lower internal prices
following the reform. Therefore, as the EBA will marginally affect our results we assume it to be exogenous
under the new CMO.
45
from the ROW cane sector. The ROW beet region supplies less than 2% of globally
traded sugar (F.O.Licht, 2005) and can therefore be considered ‘small’, i.e. facing an
infinitely elastic export demand function and not able to influence world prices
significantly through innovation in the beet sector solely. ROW sugar policies further
weaken the link between world prices and domestic prices and supply. Therefore, we
model the ROW beet region as WPN and, as a result, in our model world prices are not
affected by technology spillovers to this region. In addition, if we assume no spillovers
of biotechnology in sugar beet breeding to the ROW cane sector, the following equality
holds: pj(ρW) = pj(ρ0, ρ1, ρEU), independent of ρ0 and ρ1.
3.2.1 The Old Market Organization for Sugar
The overall world price change can be transmitted to EU domestic prices using the
principles of the EU’s former CMO for sugar, which came into full effect in 1968 and
ended on the 30th of June 2006. At the start of each year j, the Council of the European
i
Union fixed an intervention price ( p EU
, j ) for sugar and minimum prices for beet. The
quotas, consisting of A- and B-sugar ( Qa , j and Qb , j ), filled under the high price support
system, were set at a higher level than internal consumption Cj, at the intervention price
i
p EU
, j . This overproduction Qd , j (= Qa , j + Qb , j – Cj) was exported on the world market
and hence subsidized. This export subsidy system was completely auto-financed by
levies on A- and B-quota production. Consumers, who paid a high internal intervention
price, subsidized the internal within-quota production. A levy τ aj of maximum 2% of
the intervention price applied to the entire quota. Moreover, B-quota production
received an additional, more variable, levy τ bj of maximum 37.5% of the intervention
price. Both levies were chosen to satisfy the auto-financing constraint, AFCj, which was
a function of the world price (Combette et al., 1997):
i
a
i
b
i
pEU,
0 (7)
jτ j ( p j (ρ W ))(Qa , j + Qb , j ) + pEU, jτ j ( p j (ρW ))Qb , j − (Qa , j + Qb , j − C j )( pEU, j − p j (ρ W )) =
The levies had to fill the gap between the world price and the high internal price for
quota production which was in excess of consumption and exported on the world
market. If AFCj did not solve by combining Equations 7 and 8, the system of Equations
7 and 9 was solved. Finally, when the latter neither yielded a solution, a multiplicator νj
was defined solving the system 7 and 10:
46
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
τ aj ( p j [ρ W ]) ∈ [0,0.02[
 b
τ j ( p j [ρ W ]) = 0
(8)
τ aj ( p j [ρ W ]) = 0.02
 b
τ j ( p j [ρ W ]) ∈ [0,0.375[
(9)
τ aj ( p j [ρ W ]) = 0.02(1 + υ j )
 b
τ j ( p j [ρ W ]) = 0.375(1 + υ j )
(10)
By feeding the technology-induced world price pj(ρW) into Equation 7, the system
of Equations 7 to 10 yielded an estimate of the levies that had to be imposed on quotaproduction to satisfy AFCj. This specification clearly visualizes how the levies were a
function of the world price, while the world price on its turn was a function of
worldwide adoption. World prices were transmitted to levy changes through solving
AFCj. For each member state, A- and B-quota prices could be deducted from the State’s
intervention price pii, j and the levies:
pia, j ( p j (ρ W )) = pii, j [1 − τ aj ( p j (ρ W ))]
(11)
pib, j ( p j (ρ W )) = pii, j [1 − τ aj ( p j (ρ W )) − τ bj ( p j (ρ W ))]
(12)
By substituting pj(ρW) into equations 11 and 12, the model allows us to transform
world price changes into domestic quota price changes. Thus, under the CMO the farmgate price was endogenous since it depended on sugar production, internal demand and
the gap between the intervention and the world price. All out-of-quota production could
either be: (i) stocked to be carried over to the following marketing year, enabling to
smooth out annual production variations, or (ii) exported on the world market at the
world price.
The opposite effects of technology-induced cost-reduction and depression of world
and domestic prices are transmitted to the average profit functions by imputing the
corresponding prices and adoption rates (Equation 1). Note that the profit functions are
a function of (i) region-specific and (ii) worldwide adoption rates, the latter through the
[
world price: π i , j pia, j ( p j (ρ W )), ρ i
[
]
[
for A quotas, π i , j pib, j ( p j (ρ W )), ρ i
]
for the B
]
quotas, and π i , j p j (ρ W ), ρ i for C-sugar beet. If Li , j (π ) denotes the optimal allocation
of land to sugar beet in member state i in year j, the variation in producer surplus
47
(relative to the benchmark without adoption) due to the innovation can be measured in
the market of land supplied to the industry (Moschini et al., 2000). The producer
surplus change strongly depends on the position of the supply curve and, hence, the
member state’s competitiveness in sugar production. Therefore, we introduce the
categorical parameter ϕi,j (Table 5). Depending on the value we attribute to this
parameter, the model automatically selects the appropriate formula for the calculation
of the welfare effects. Detailed information about the formulas can be found in
(Demont, 2006). For a high-cost EU member not fulfilling its A quota we set ϕi,j = 0.
Portugal and Greece are the only examples. A high-cost EU member state, fulfilling its
A- but not its B-quota is assigned φi,j = 1. The beet growers in these countries aim at
fulfilling their A quota and in order to ensure this objective they choose to accept a
minimal precautionary overproduction, even in low-yield years. For medium-cost EU
countries fulfilling their A quota and a significant part of its B quota, we set ϕi,j = 2.
Table 5: Multi-criteria framework for identifying WPR sugar producers in the EU
Belgium
Denmark
Germany
Greece
Spain
France
Italy
Netherlands
Austria
Portugal
Finland
Sweden
UK
Frandsen
(2003)
Frandsen (2003)
revisiteda
WPR
WPR
WPR
WPR
Gohin
(2006)
WPR
Buy-out in
first two yearc
No
WPR
WPR
No
No
WPR
Yes
Yes
No
WPR
WPR
WPR
Poonyth
(1998)
WPR
WPR
WPR
WPR
wrongb
WPR
WPR
Yes
No
No
Yes
Yes
Yes
No
Score
WPR
(3/5)
WPR
(5/5)
WPR
(4/5)
WPR
(3/5)
WPR
(4/5)
Czech republic
WPR
Yes
Hungary
WPR
Yes
Poland
WPR
No
Sources: Frandsen et al. (2003), Gohin and Bureau (2006), and Poonyth (1998).
a
We extend the criterion of Frandsen et al. (2003) to the period 1996-2006 (F.O.Licht, 2005).
b
The authors did not use the correct mixed price in their econometric model.
c
The fact that countries decide not to sell any quota despite this incentive means they are able to produce
at the lower internal price, hence being competitive.
48
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
As mentioned previously, the estimation of world price effects crucially hinges on
the specification of world price responsiveness of sugar supply. Particular attention
needs to be given to the distinction between WPR and WPN Member States. 11 Buysse
et al. (2007) summarize the literature on the specification of the sugar supply curve and
the incentive to produce C-sugar. The first approach they discuss is developed by Gohin
and Bureau (2006) who show that, in contrast to the WTO decision on C-sugar, crosssubsidization from the high price quota to the fixed costs of C-sugar production cannot
be sustained in the long run. The second motivation for the production of C-sugar
discussed is an insurance strategy against losses due to not filling the quota. However,
Adenäuer and Heckelei (2005) show that this insurance strategy is only one factor in
the decision, temporal variability being at least as important. Therefore, Buysse et al.
(2007) define a precautionary sugar beet supply as a function of the quota endowment
and rent, taking into account farm-specific characteristics that play a role in this
variability. Unfortunately, the analysis of Buysse et al. is based on farm level modeling
and cannot be easily extrapolated to the aggregate level. Frandsen et al. (2003) define
WPN Member States as those that overshoot their quota by less than twice the standard
deviation of total production. They argue that WPN Member States would normally not
supply C-sugar, since the rents of the latter are not sufficient to cover production costs.
Since quota fulfillment is the primary objective of WPN Member States, we assume
that risk premiums and stock decisions are exogenous.
In order to get a more balanced distinction between WPN and WPR Member States
we extend the single-criterion framework of Frandsen et al. (2003) and present a multicriteria framework in Table 5, by adding four additional criteria based on the literature
and the success of the buy-out scheme. In our multi-criteria framework, we define WPR
Member States as those that satisfy more than half, i.e. at least three out of five, of the
criteria in Table 5. We revisit the criterion proposed by Frandsen et al. (2003) and apply
it on a longer time series of data, i.e. 1996-2006. Further, we base our third and fourth
criteria on econometric evidence provided by Poonyth (1998) and Gohin and Bureau
11
A reviewer interestingly pointed out that our exogeneity assumption imposed on the categorical parameter φi,j is
strong. We agree that the parameter φi,j should be a function of prices (and adoption). This would imply that
regions could move between categories due to the net effect of innovation and reducing prices. However,
introducing this endogeneity would make our model extremely complex. In this paper, we decided to develop a
static multi-criteria framework to identify WPN and WPR regions, based on historical data (Table 4), due to the
important influence of world price responsiveness on world prices. Ideally, we would need to apply this
framework on all categories and make it dynamic, i.e. by endogenizing prices. These upgrades would
unnecessarily complicate the present framework, because we would need to make strong assumptions on ‘when’
regions upgrade from one category to a higher one, i.e. when regions become quota-fillers or when
precautionary overproduction of quota-fillers becomes structural.
49
(2006). Finally, we use observed post-reform behavior of Member States as a fifth
indicator. Application of the buy-out scheme in the post-reform period indicates the
inability of a member state to produce at reduced post-reform internal prices and, hence,
the inability to respond to world prices, which are even lower than internal prices. Our
enhanced framework suggests adding Belgium to Frandsen et al.’s (2003) original list
of WPR Member States, i.e. Austria, France, Germany, and the UK.
3.2.2 New Common Market Organization for Sugar
On the first of July 2006 a new CMO for sugar was introduced. The key features of the
reform were (i) a progressive cut of the EU institutional price up to 36% over four
marketing years, (ii) direct compensatory payments of 64.2% of the estimated revenue
loss over three marketing years and (iii) a single quota arrangement for the term
2006/07-2014/15. The new institutional sugar price is not fixed in the short-run but in
the long run it is. In order to facilitate the desired reduction in production, a buy-out
scheme was established. In the four year period following the reform, sugar producers
can sell their quota to the EU for an annually declining compensatory payment, which
evolves from €730/ton in the first two years to respectively €625/ton and €520/ton in
the third and fourth year. This should stimulate less competitive producers to reduce or
abandon production. If the reduction were to be insufficient in 2010 to reach the goals
set by the EU (12.5 million tons), the EU could decide on a linear quota cut for all
European producers (European Parliament, 2006).
For our model, the introduction of the new CMO has several structural implications.
Supply and quota decisions depend to a large extent on the prices for sugar.
Furthermore, this decision is influenced by a restructuring amount to be paid for each
quota held which creates an incentive to sell excess quota as soon as possible.
Therefore, we introduce two new post-reform categories of competitiveness (for an
overview of all categories, see the legend of the categorical parameter ϕi,j below (Table
5). Producers not filling their quota under the old regime (ϕi,j = 0, 2), will sell their
excess quota under the new institutional price and join the group of post-reform quotafillers (ϕi,j = 7), together with pre-reform quota-fillers (ϕi,j = 1, 3) although the latter
might sell some of the initially allocated quota due to the reduced sugar prices. WPR
Member States (ϕi,j = 4) are affected the most, since they have to give up all sugar
50
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
production supplied to the world market but we assume that they remain WPR after the
reform (ϕi,j = 8).
Due to WTO obligations, export of sugar is limited to 1.4 million tons white sugar
per year. As a rule, this amount is filled with excess quota sugar. Only if the budget is
not sufficient to fill the gap between the institutional price and the world price or the
excess supply of quota sugar is smaller than 1.4 million there is a possibility to export
out of quota sugar. Therefore, EU sugar production meant for the world market is
marginal. Low cost producers can supply their excess out-of-quota sugar only to the
small European industrial market. On this market European sugar producers are price
takers since they have to compete with imported sugar and the limited size of this
market. They get some rents through reduced costs but only marginally influence world
prices. The ROW cane industry is assumed to respond to the world price, but since no
technology spillovers are assumed between beet and cane industries, cane producer
surplus is only affected through world prices.
The innovation rents of A- and B-quota producers are estimated taking into account
member state-specific pricing systems, quota and price responsiveness (see Demont,
2006 for detailed formulas). The innovation rents of the European sugar beet producers
post reform are calculated following the same formulas depending on the categorical
variable ϕi,j. The innovation rents of WPR Member States under the old regime are
measured along the supply curve of land to the sugar beet industry (Moschini et al.,
2000):
∆PS i , j ( p j (ρ W ), ρ i ) =
π i , j ( p j ( ρ W ), ρ i )
∫L
i, j
π i , j ( p j ( 0 ), 0 )
(π )dπ
(13)
For WPN Member States we assume inelastic land supply; instead of allocating more
land, they respond to yield-enhancing technologies by freeing up land allocated to sugar
beet. Hence, their innovation rents are measured as follows:
∆PS i , j ( p j (ρ W ), ρ i ) =
Qi , j
y ( p j (ρ W ), ρ i )
[π ( p (ρ
i, j
j
W
]
), ρ i ) − π i , j ( p j (0),0 )
(14)
Finally, to calculate the profit of the input suppliers in the case of GM sugar beet, we
can reasonably assume that they will apply uniform monopolistic pricing at the EU-27
level (see Chapter 2, updated to correct year). Therefore the gross revenue captured by
the input suppliers can be modeled as the accumulation of monopolistic technology fees
51
(μi) on the entire area of land supplied to the sugar beet industry in equilibrium and
planted with GM crops (Moschini et al., 2000):
18
Π j ( p j (ρ W ), ρ W ) = ∑ ρ i , j Li , j ( p j (ρ W ), ρ i ) µ i ,
(15)
i =0
3.3 Data and model calibration
Our model is calibrated on the observed production data for the period 2004-2006.
Observed yields, incentive prices (see above), world sugar prices on the London n°5
exchange, quantities ( Qi , j ) and quota ( Qi ,aj and Qi b, j ) are taken from F.O.Licht
(F.O.Licht, 2005), USDA (2006b), and FAO (2007). Our counterfactual scenario
projects the old regime up to 2014, assuming that production remains fixed at the prereform 2005 level. The factual scenario is based on observed post-reform 2006 data.
Future price and consumption levels come from the World Sugar Outlook (F.O.Licht,
2005) and future yields are derived through linear extrapolation of historical trends
(F.O.Licht, 2000; F.O.Licht, 2005). The sugar prices, as forecast by FAPRI, account
for the new CMO and could therefore be influencing our outcome. However, the new
regime, although eliminating a lot of the European export, is not the major factor
influencing world sugar prices. For example right after the introduction of the new
CMO, world prices went down due to record yields in Brazil and Thailand in 2007.
Other major influences are the demand for bio-ethanol and decreasing stocks (as in
India during 2007). We assume that only WPR producers supply industrial sugar after
the reform and up to an amount of 1.5 million ton (SUBEL, 2007), shared among
producers and weighted according to their quota. The other Member States are
assumed to just fulfill their quota.
52
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
Table 6: Global production of sugar beet and profits from HT technologies in
2006
Area
Production
(ha)
(t white sugar)
Belgium
86,078
960,248
Denmark
48,595
475,047
Germany
405,136
4,090,902
Greece
37,490
294,897
Spain
114,334
1,207,097
France
349,906
4,328,332
Italy
26,375
185,281
Netherlands
92,373
991,380
Austria
48,348
499,314
Portugal
7,091
65,467
Finland
32,316
158,930
Sweden
53,278
420,344
United Kingdom
118,737
1,257,945
Czech Republic
69,621
568,496
Hungary
63,519
497,047
Poland
292,579
2,047,234
EU-27
1,901,149
19,845,900
ROW beet
3,088,900
14,826,720
ROW cane
21,634,040
98,867,570
World
26,624,089
133,540,190
Sources: Chapter 4, F.O.Licht (2005),
Note: n.a.: not applicable, a EU area-weighted average
Yield
(t/ha)
11.2
9.8
10.1
7.9
10.6
12.4
7.0
10.7
10.3
9.2
4.9
7.9
10.6
8.2
7.8
7.0
10.4
4.8
4.6
5.0
Profit from HT sugar beet
(€/ha)
174
145
108
63
192
131
45
151
188
223
102
71
99
101
78
96
117a
117a
n.a.
n.a.
To calibrate the average profit function, we need an approximate estimate of the profit
in all regions. Thelen (2004) compares per-hectare profits among four beet producers
(Poland, Ukraine, USA and Germany) and six cane producers (Brazil, Australia,
Thailand, South-Africa, India and USA). Since after an extensive sensitivity analysis
it appears that this is just an inconsequential scaling parameter, we use the estimate of
Germany for the EU-27 and calculate the area-weighted averages for the ROW cane
and beet regions. All cost and price data are first deflated and actualized to the
agricultural season 2006/07 using the GDP country deflators from the World Bank’s
World Development Indicators (World Bank, 2008), and then converted to Euro using
the Euro/USD exchange rate for 2006. The technology fee and adoption ceilings is
estimated according to Chapter 2. We allow for technology spillovers to the ROW
beet region but, as mentioned before, we assume no changes in the ROW cane region.
As we carry out the analysis before adoption has been observed, the relevant
adoption data are not yet available. Moreover, the estimation of certain parameters,
such as elasticities, is surrounded by uncertainty. Therefore we construct subjective
distributions for these parameters, using all prior information available. , We generate
stochastic distributions of the outcomes of the model using Monte Carlo simulations.
53
For sugar beet, it is widely accepted that conventional herbicides harm the crop
because of the low selectivity of the used herbicides (Märländer, 2005). Field trials
comparing HT sugar beet with conventional control practices suggest that yield boosts
vary between 0% and 8% (Bückmann et al., 2000; May, 2003; Wevers, 1998), due to
greater weed control and reduced crop injury. We construct a triangular distribution
for of this parameter (minimum = 0%, most likely value = 4%, maximum = 8%) to
capture this uncertainty. The per-hectare profit gains of HT sugar beet in all EU
Member States are calculated according to Chapter 2 (Table 6). The profit estimates
are based on parametric modeling of heterogeneity of herbicide expenditures of beet
growers through loglogistic probability distribution functions which are extrapolated
to the entire period 2006-2014 in our model.
To calibrate the model, we need to define incentive prices pˆ i , j for all regions, i.
The incentive price for the ROW, pˆ 0, j corresponds to the world price (Table 4). For
EU Member States, the incentive price depends on the state’s competitiveness and the
national pricing system applied to pay beet growers and processors. The incentive
prices under the former CMO for sugar are modeled in a dynamic way and depend on
the world price, which, in turn, depends on world-wide adoption rates. Incentive
prices can be A-sugar prices pia, j ( p j (ρ W )) , B-sugar prices pib, j ( p j (ρ W )) , a member
[
]
state-specific mixed price pim, j pia, j ( p j (ρ W )), pib, j ( p j (ρ W )), p j (ρ W ) , or the world
price pj(ρW). For the new CMO, the incentive price for in-quota sugar ( pio, j ) is set
equal to the reference price level (which is decreasing in time) and the out-of-quota
incentive price is the world price pj(ρW). The model is calibrated on the preinnovation equilibrium, i.e. we set ρW = 0.
In Table 5 we combine different literature sources and our categorical system
mentioned above to calibrate supply curves with member state-specific incentive
prices and area elasticities. To calibrate θi,j it is useful to relate this parameter to the
more standard notion of elasticity of land supply with respect to sugar prices. If we
define ri,j as the farmer’s share (rent) of unit revenue, the parameter θi,j can be
calibrated as (Sobolevsky et al., 2005):
 πˆ i 

 pˆ i , j y i , j 
θ i , j = ψ i ri , j = ψ i 
54
(16)
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
Since our model features disaggregated area response (ψi) and yield response (ηi)
to prices we need to find elasticities that correctly represent farmers’ behavior and
incentives in the global sugar beet sector. In a quota system with fixed prices, the
annual within-quota price variation is too small to obtain reliable estimates of the
elasticities of supply. Quota rents of WPN Member States are not significantly
affected by supply response and therefore their supply elasticities do not influence the
results. WPR Member States, in contrast, significantly affect world prices and global
welfare through technological innovation. Therefore for them, precise estimates of
area response to world prices are needed. Poonyth et al. (2000) report short- and longrun area elasticity estimates for 13 EU Member States 12 As they do not include any
standard deviations for ψi, we construct symmetric triangular distributions with the
short-run estimate as the minimum value, the long-run estimate as the maximum
value, while we assume that the most likely value is the average of the SR and LR
estimates. Devadoss and Kropff (1996) report supply elasticities for all major sugar
producers in the world. For the ROW cane and ROW beet regions, we calculate a
production-weighted average supply elasticity of 0.269 and 0.207, respectively. For
Greece and Portugal we use Devadoss and Kropff (1996) supply elasticity estimate of
0.228 for A-quota sugar. As supply elasticities already incorporate yield response to
prices, we set ηi = 0 for these Member States. We borrow area elasticities for the new
Member States and yield elasticities from the ESIM-model of Banse et al. (2005) and
use the aggregate EU-27 elasticity of yield for all Member States, ηi = 0.08. Unless
otherwise stated, for all structural elasticities we construct symmetric triangular
distributions centered on the average value and ranging from zero to twice the average
value, based on strict positivity from theory.
3.4 Results
We conduct a Monte Carlo simulation of 3,000 iterations to generate stochastic
distributions for the results, using the software program @Risk from Palisade
Corporation. Given the assumed, estimated and retrieved parameters, in each random
12
Only the supply elasticities of WPR regions influence our results as in our model only these regions are able to
influence world prices (see Demont 2006, p. 118 for more details). For the five WPR countries, identified
through our multi-criteria system (Table 5), Poonyth (1998) econometrically estimated area equations as
functions of world price, exchange rates, wage index, allocated quota and sugar yield. He used two stage least
square regressions and obtained highly significant t-values. Moreover, in none of the equations the Durbin Htest’s null hypothesis of no autocorrelation was rejected. Therefore, we concluded that the supply elasticity
estimates stemmed from well-behaved structural equations.
55
draw the model is calibrated on the pre-innovation structural parameters so as to
retrieve pre-innovation acreage, quantity, yield, and price data for each year j. Two
alternative scenarios have been assessed: first the observed factual introduction of the
new regime in 2006, and secondly the counterfactual projection of the former CMO
up to 2014. The basic question is whether the differences in innovation incentives
among the two regimes are robust under a set of reasonable assumptions.
To analyze hypotheses of interest on transformed prior distributions of stochastic
simulations, Davis and Espinoza (1998) use the distribution-free Chebychev
inequality on the pairwise differences between alternative outcomes of the model.
However, Griffiths and Zhao (2000) argue that this inequality is unnecessarily
imprecise and recommend the use of ordinary probability intervals. If the nullhypothesis of this chapter, H0: the policy reform has no influence on the stream of
benefits from innovation, is true, a probability interval of the pairwise differences of
the alternative outcomes under both CMOs should contain the value zero. If it is
rejected, the probability intervals show a significant deviation from zero, leading to
higher or lower incentives after the reform.
Our results reveal there are significant differences between the old and the new
sugar regime for both farmers and input suppliers. Table 7 shows the 95% probability
intervals of the simulation model’s outcome for the different Member States welfare
gains of the new technology. The reduction of institutional sugar prices in the EU
means lower revenue for sugar producers. However not all producers are affected the
same way as there supply reacts to different incentive prices. High cost sugar
producers who responded to the high A-quota price face the steepest price drop. The
latter means a significant reduction of adoption incentives under the new regime. This
effect is observed for Portugal, Italy and Finland, where farmers gain on average
25€/ha less from the innovation under the new CMO. For middle cost producers,
before adjusting supply according to B-quota prices, the situation is different.
Although the average revenue (over the produced A- and B- quota) reduces, their
incentive price in fact increases. The new reference price for sugar is set between the
old A- and B-quota prices. Therefore B-quota producers could gain from the reform.
This gain is higher during the restructuring period due to the gradual reduction of the
institutional price. Once the restructuring period is over in 2010, the institutional price
is close to the old B-sugar price, decreasing the additional incentives for innovation.
This theoretical explanation is confirmed by the outcomes of the model. For medium56
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
cost B-sugar producers such as Spain, the Czech Republic, Hungary, Poland,
Denmark and Sweden, we observe significant positive effects on the incentives. The
Netherlands, although also a B-sugar producer, has no significant outcome due to the
use of the mixed pricing system in the former regime, which sets the incentive price a
little above the B-sugar price, close to the new institutional price. Intuitively, we
would have expected C-producers to have higher innovation incentives after the
reform due to a smaller depressing effect of EU sugar exports on world prices.
However, C-producers’ innovation incentives seem to be largely unaffected by the
reform. This holds for the highly competitive WPR Member States Austria, Belgium,
Germany, France, and the UK. These results show we can reject the null hypothesis of
the chapter, as benefits stream of the innovation are indeed affected by the reform.
The issue is whether this result interferes with the goals of the new CMO of increased
competitiveness. Since the innovation incentive for high cost producers is reduced,
innovation is discouraged compared to the former CMO. On the other hand, producers
with lower costs are encouraged to innovate or remain unaffected by the reform.
Therefore low cost producers are favored to reduce production costs even further and
are better prepared to cope with future changes in the market caused by biofuels or
price changes. This increases the competitiveness of low cost producers and could
eventually even lead to a further crowding out of high cost producers. These results
are in line with the predictions of Demont (2006).
57
Table 7: 95% probability intervals of farmers’ profit differences (€/adopted hectare) and input suppliers’ revenue (m€)
Belgium
Denmark
Germany
Greece
Spain
France
Italy
Netherlands
Austria
Portugal
Finland
Sweden
United Kingdom
Czech Republic
Hungary
Poland
Input suppliers revenue
Belgium
Denmark
Germany
Greece
Spain
France
Italy
Netherlands
Austria
Portugal
Finland
Sweden
United Kingdom
Czech Republic
Hungary
Poland
Input suppliers revenue
2006
(-21.9; 88.5)
(10.6; 137.8)
(-23.7; 97.0)
(-44.6; 3.9)
(15.4; 94.1)
(-27.7; 110.0)
(-35.0; 2.5)
(-19.0; 94.5)
(-20.7; 87.0)
(-119.4; -5.9)
(-41.6; -1.1)
(13.5; 81.1)
(-14.0; 54.6)
(10.0; 71.2)
(2.8; 35.2)
(9.8; 61.6)
(-4.3; -8.1)
2011
(-23.7; 94.7)
(-15.8; 96.1)
(-25.8; 103.8)
(-86.3; 0.40)
(-0.24; 29.3)
(-30.1; 117.1)
(-65.7; -0.73)
(-76.5; 63.8)
(-22.6; 93.3)
(-190.2; -14.9)
(-80.4; -6.1)
(5.5; 50.2)
(-15.3; 58.7)
(1.5; 23.2)
(-3.9; 11.2)
(0.38; 24.8)
(-10.2; -16.6)
2007
(-20.2; 83.1)
(2.2; 111.9)
(-21.9; 91.7)
(-64.4; 1.4)
(9.6; 59.3)
(-25.6; 103.6)
(-48.5; 0.41)
(-40.4; 70.8)
(-19.1; 81.9)
(-151.0; -10.8)
(-58.1; -3.7)
(10.5; 64.2)
(-12.8; 51.0)
(6.3; 46.7)
(0.70; 22.7)
(6.4; 41.9)
(-10.9; -17.6)
2012
(-24.3; 96.5)
(-16.3; 97.7)
(-26.5; 105.8)
(-86.2; 0.51)
(-0.32; 29.8)
(-30.8; 119.2)
(-66.3; -0.65)
(-77.4; 65.0)
(-23.1; 95.1)
(-191.3; -15.0)
(-81.4; -6.1)
(5.5; 50.8)
(-15.7; 59.9)
(1.5; 23.6)
(-4.0; 11.4)
(0.34; 25.2)
(-10.1; -16.4)
2008
(-22.1; 88.8)
(-7.2; 102.0)
(-24.0; 97.5)
(-76.7; 0.93)
(4.7; 41.4)
(-28.1; 110.2)
(-57.3; -0.24)
(-59.4; 65.4)
(-21.0; 87.3)
(-171.0; -13.0)
(-69.1; -5.0)
(7.8; 55.2)
(-14.2; 54.8)
(3.9; 32.7)
(-1.4; 16.0)
(3.6; 31.6)
(-10.6; -17.2)
2013
(-24.7; 97.9)
(-16.7; 99.0)
(-26.9; 107.3)
(-86.2; 0.59)
(-0.42; 30.1)
(-31.3; 120.8)
(-67.0; -0.60)
(-78.2; 66.0)
(-23.5; 96.6)
(-192.6; -14.9)
(-82.5; -6.1)
(5.5; 51.4)
(-16.0; 60.8)
(1.5; 23.9)
(-4.1; 11.5)
(0.32; 25.6)
(-10.1; -16.3)
2009
(-23.1; 92.3)
(-15.3; 93.8)
(-25.1; 101.2)
(-86.4; 0.31)
(-0.19; 28.7)
(-29.3; 114.3)
(-64.4; -0.79)
(-75.1; 62.3)
(-21.9; 90.8)
(-187.4; -14.5)
(-78.3; -6.0)
(5.4; 49.2)
(-14.9; 57.0)
(1.4; 22.7)
(-3.8; 10.9)
(0.42; 24.2)
(-10.4; -16.9)
2014
(-25.3; 99.8)
(-17.2; 100.6)
(-27.5; 109.3)
(-86.1; 0.70)
(-0.56; 30.6)
(-32.0; 123.0)
(-67.7; -0.49)
(-79.4; 67.2)
(-24.0; 98.5)
(-193.8; -14.9)
(-83.5; -6.1)
(5.4; 52.1)
(-16.4; 62.1)
(1.5; 24.3)
(-4.2; 11.7)
(0.29; 26.0)
(-10.0; -16.2)
2010
(-23.3; 93.2)
(-15.4; 94.8)
(-25.4; 102.3)
(-86.4; 0.34)
(-0.19; 29.0)
(-29.6; 115.5)
(-65.0; -0.77)
(-75.7; 62.9)
(-22.1; 91.9)
(-189.0; -14.9)
(-79.3; -6.0)
(5.5; 49.6)
(-15.1; 57.7)
(1.5; 22.9)
(-3.8; 11.1)
(0.40; 24.5)
(-10.6; -17.4)
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
The situation for the input suppliers is different. Under the new sugar regime, total
production decreases following the abolishment of quota by high cost producers. As a
result, the total annual technology revenue for the supplier declines (Table 7). This
might explain the observed low priority that seed producers are currently attributing to
the regulatory approval of HT sugar beet for cultivation in Europe (Monsanto, 2007).
Ceteris paribus, the decrease in revenue implies a decrease in profit and the incentive
for innovation for the technology provider. The latter, however, might be partly offset
by the fact that higher farmer profits accelerate adoption and enable innovators to
capture revenue in a shorter time interval. Moreover, other policy measures can
significantly influence the profit of innovating firms. Kalaitzandonakes et al. (2007)
have shown that the regulatory costs for registering new GM varieties are very high.
Creating an environment which favors innovation and research could lower the cost to
introduce new technologies in the market. In this case lower revenue is not necessary
correlated with lower profits.
The results also affect US and Canadian sugar industries and consumers. There is
of course a GM rent accruing to farmers due to reduced production costs. However
this GM rent does not change under the EU sugar reform because of the high
domestic protection of the sugar market. The fact that innovation has a marginal effect
on world prices – the new CMO completely eliminates the risk of immiserizing
growth (Bhagwati, 1958) – represents perhaps the most dramatic change induced by
the reform. For the North American government, this implies a reduction in tax
expenses to protect beet and cane growers from world price declines. These savings
accrue directly to taxpayers in these countries. In contrast with the former regime,
world prices are not affected by the innovation, keeping sugar prices constant
minimizing the welfare losses for sugar cane producers, mostly in developing
countries. The increase of competitiveness in the European Union and North America
also implies that trade barriers become less necessary to maintain domestic
production. Therefore, endogenizing the innovation in trade policies would allow
trade partners to reduce high trade barriers and the latter would perfectly fit in the
ongoing WTO trade negotiations which spur free trade. On the other hand, in a world
with rising agricultural commodity prices, a competitive European sugar sector could
start exporting on the world market without subsidies and as such decrease the world
price significantly after all.
59
3.5 Discussion
The first historical radical reform of the European sugar regime has greatly affected
the sector. Besides the direct goals of the reform to reduce production through
crowding out of high cost producers, our analysis has shown that the stream of
benefits from a innovative technology are significantly altered. An alternative way of
increasing competitiveness in the sugar sector could result from spurring innovation.
Therefore, a desirable side effect from the reform would be that it increases the
incentives for innovation, or at least, does not decrease the incentives of already
competitive producers. It is important here to distinguish between ‘motives’ and
‘incentives’ for innovation. Their uncompetitive character provides them motives for
innovation, and even more so after the reform if they are not able to cope with lower
prices. The incentives for innovation, on the other hand, do not directly stem from
their competitiveness (although indirectly through technical efficiency), but mainly
from the marginal revenue they obtain from genetic yield improvements and, hence,
institutional prices. Due to lower institutional prices, farmers in uncompetitive
Member States have weaker incentives to adopt new technologies under the new
sugar regime. During the beginning of the restructuring period the losses in
technology rents are small due to the gradual decrease of prices, but they increase
with time. This means farmer gain less from adopting HT sugar beet then before the
reform. Former B-sugar producers, on the other hand, capture higher gains under the
new regime because the new institutional price is slightly higher than the old B-sugar
price. Hence, the new regime spurs innovation and adoption by leaving competitive
producers unaffected, providing incentives for innovation for medium-competitive
producers, while eroding incentives for uncompetitive producers and stimulating
crowding out of this group. This decrease in adoption incentive makes them less
likely to adopt new technology. Therefore future changes in the sugar market will be
difficult to bear and could lead to a targeted crowding out of high cost producers. This
result coincides with the goals of the European sugar CMO.
However, the reform could result in loss of incentive for gene developers and seed
companies regarding R&D investments due to the contracted market for innovated
sugar beet seed in Europe. This is consistent with the currently observed lack of
interest of the European biotechnology sector in continuing GM sugar beet research.
60
Chapter 3. European Sugar Policy Reform and Agricultural Innovation
Other policy measures favoring innovation could anticipate and offset this decrease in
incentive.
Finally, one effect of the introduction of GM beet has not been considered in this
study. The decrease in production costs in the European sugar industry could open
doors to cost-effective bio-ethanol production, where consumer attitudes are not
expected to refrain biotechnology research. Further research is needed to assess the
incentives for innovation under these production conditions.
61
62
Chapter 4. Global Welfare Effects of GM Sugar Beets under
Changing Sugar Policies
Adapted from Dillen, Demont & Tollens,
AgBioForum 12(1):119-129
Ex post impact studies of genetically modified (GM) crops indicate that society is
capturing sizeable gains of agricultural biotechnology. In Europe, in contrast, due to
limited adoption, research has been largely restricted to ex ante technology and
policy impact assessment of GM crop cultivation. In this study we assess the impact of
a hypothetical introduction of herbicide tolerant (HT) sugar beet in the global sugar
sector under both the former and the actual European Common Market Organization
(CMO) for sugar. The model starts from a farm level analysis, introducing a perfect
corporate pricing strategy under restricted monopoly power, which is expanded to a
partial equilibrium model of the world sugar trade. We show that even under the
given condition of private market power, significant gains accrue to farmers and
consumers while a smaller part goes to the seed sector (gene developers and seed
suppliers). The global value of HT sugar beet for society in the period 1996-2014 is
estimated at €15.4 billion of which 29% is captured by EU farmers, 31% by farmers
and consumers in the rest of the world and 39% by the seed sector. However, the
global sugar sector is foregoing most of this value as the technology is currently only
accepted by the USA sugar industry.
4.1 Introduction
In the scholarly literature, the debate on the economic impact of genetically modified
(GM) crops has been characterized by polar viewpoints. 13 In contrast with public
research, in the case of proprietary GM seed innovations societal interest is typically
not focused on the rate of return of biotechnology research, but on its welfare
distribution among the private upstream seed sector (gene developers and seed
suppliers) and downstream stakeholders (farmers, processors, distributors and
consumers) in the supply chain. The published ex post impact assessments indicate
that farmers clearly capture sizable gains despite the proprietary nature of the
innovation (Demont et al., 2007b; Qaim, 2009). According to Chapter 2,
heterogeneity among farmers drives value sharing of proprietary seed technologies by
restricting monopoly.
In the European Union (EU), only a limited number of Member States have been
growing GM crops so far and only a few ex post impact assessments have been
published, i.e. on Bt maize in Spain (Demont and Tollens, 2004b; Gomez-Barbero et
al., 2008) and the Czech Republic (Demont et al., 2008a), and HT soybeans in
Romania (Brookes, 2005). Some ex ante EU distributional impact studies on GM
sugar beet are documented as well (Demont, 2006; Demont and Tollens, 2004a),
reporting a global welfare increase of €1.1 billion during the five-year period 19962000, shared among EU producers (26%), the global seed industry (24%) and farmers
and consumers in the rest of the world (50%). However, these studies did not take into
account heterogeneity among farmers and relied on an exogenous set technology fee,
opening the door for homogeneity and pricing bias. Chapter 2 addresses the question
but as argued by Frisvold et al. (2003), assessing distributional effects without trade
effects is of little use.
Despite the official end of the moratorium and new approvals of GM crops,
adoption of national guidelines on coexistence has been relatively slow in the EU and
due to regulatory uncertainty and consumer hostility, the adoption of GM crops is still
limited. This means that the EU is still in a state of quasi-moratorium regarding the
introduction of GM crops Devos et al. (2008b), foregoing important benefits of these
technologies. However, with upwards price pressure in world food markets, shortages
13
For example, despite the vast literature on the positive welfare effects of GM crops, in public debate and some
recent papers the relation between adoption of GM cotton in India and increased farmer suicides is
investigated (e.g. Herring, 2008).
64
Chapter 4. Global Welfare Effects of GM Sugar Beets under Changing Sugar Policies
in the non-GM feed market and development of a significant bio-energy market, it
seems the tide could be changing and the demand for the introduction of GM crops
might increase in the EU in the near future. In this chapter we therefore assess the
potential global welfare implications of HT sugar beet for upstream and downstream
stakeholders in the global sugar beet sector under perfect corporate pricing strategies.
Since our analysis covers both the future and the past, results both include the welfare
forgone and the potential future benefits of the technology. Furthermore, this study
also takes into account the change in the Common Market Organization (CMO) for
sugar that Europe underwent in 2006.
4.2 Model
The general structure of the EUWABSIM model is described in Chapter 3. However,
the model now covers 19 agricultural seasons, from 1996-2014. Hence, it captures the
accession of new Member States in the EU in 2004 and 2007. This broadened scope
allows us to introduce two model novelties. First, because HT sugar beet is not yet
adopted, the characteristics of the adoption pattern are not known yet. Therefore, for
each member state we construct a logistic adoption curve (Griliches, 1957) calibrated
on the adoption ceilings estimated in Chapter 2, which take into account heterogeneity
of weed control expenditures in the different Member States.
ρi ( j ) =
ρ max,i
1 + exp(−a − bj )
(1)
We further complete the parameterization of the adoption curve through analogy.
More specifically, we impute adoption pattern parameters, a and b, of a comparable
technology in the USA, i.e. HT soybeans (USDA, 2006a). 14 The inclusion of regional
adoption ceilings allows for a more realistic representation of the adoption of the
technology and, hence, consists in a major upgrade of the former EUWABSIM
model.
Second, we apply the framework presented in Chapter 2 for two values of j, 1996
and 2004. This means we allow the monopolistic innovator to set a new technology
fee in 2004 to adapt to the expansion of the EU and the changed practices in
14
We believe that the US case of HT Roundup Ready® soybeans is comparable with the EU case of HT sugar
beet, because of (i) the common embedded technology of herbicide tolerance, (ii) the ubiquitous importance
of each crop on both continents, and (iii) the comparable importance of exports of the refined products.
65
conventional herbicides. Hence, ρmax,I and the profit per hectare are altered in 2004 for
all Member States.
4.3 Data and Model Calibration
In the simulation model, our counterfactual scenario assumes hypothetically that both
the EU’s beet sugar industry and the ROW beet region embraced the technology since
the marketing year 1996/97, and progressively adopted it up to 2014/15. Our model is
calibrated on the observed production data from the past period 1996-2006. Observed
yields, incentive prices, London n°5 world sugar prices, quantities and quota are taken
from various sources (European Commission, 1999; F.O.Licht, 2000; F.O.Licht,
2005; FAO, 2006; USDA, 2006b). Forecasted data are borrowed from the FAPRI
(2006) model, linear extrapolations of historical yield trends and from decision
290/2007 from the EU (European Parliament, 2007). We assume that only low cost
producers will supply industrial sugar and this up to an amount of 1.5 million ton
(European Parliament, 2007) (SUBEL, 2007), shared according to their quota. The
other Member States are assumed to just fill their new quota. All cost and price data
are first deflated and actualized to the agricultural season 2006/07 using the GDP
country deflators form the world development indicators, and then converted to Euro
using the exchange rate of 2006. Institutional prices are deflated using both
agricultural and financial exchange rates.
We assume a uniform pricing strategy in which the innovating firm
monopolistically prices the technology in two stages, i.e. in 1996 upon introduction
and in 2004 upon the introduction of new Member States. The adoption ceilings then
represent the maximal adoption of the technology under the restricted monopoly held
by the seed sector. Heterogeneity of weed control expenditures is based on estimated
herbicide and application costs (Table 8). Since we are only focusing on a single
technology in a single sector, in our model the technology cannot ‘spillover’ to the
ROW cane region. Therefore, we allow technology spillovers to the ROW beet
region, subject to a similar adoption pattern, but assume a ceteris paribus in the ROW
cane region. As a result, our estimated welfare effects have to be interpreted as
functions which are fed with an assumed adoption pattern.
66
Chapter 4. Global Welfare Effects of GM Sugar Beets under Changing Sugar Policies
Table 8: Heterogeneity of herbicide expenditures and estimated technology fee
and adoption ceilings for herbicide tolerant sugar beet adoption in the EU-27
Mean herbicide expenditures
(€/ha)a
1996
2004
167.6 (37.3)
226.9 (110.2)
180.5 (83.4)
180.8 (84.8)
216.0 (85.4)
178.7 (95.3)
228.1 (46.8)
122.9 (21.6)
280.2 (98.7)
233.1 (73.5)
134.1 (55.5)
138.2 (26.4)
199.8 (37.1)
85.9 (16.5)
194.25 (64.9)
151.4 (45.4)
144.8 (101.2)
165.8 (22.6)
Technology fee
(€/ha)
1996
2004
98
88
98
88
98
88
98
88
98
88
98
88
98
88
98
88
98
88
Adoption
ceiling
1996
2004
89%
91%
88%
92%
90%
69%
99%
63%
100%
100%
43%
89%
93%
1%
74%
53%
69%
100%
Belgium
Denmark
Germany
Greece
Spain
France
Ireland
Italy
The
Netherlands
Austria
246.9 (104.3)
275.9 (99.1)
98
88
87%
96%
Portugal
280.26 (98.7)
280.4 (98.8)
98
88
99%
100%
Finland
276.8 (81.0)
2040 (37.6)
98
88
99%
100%
Sweden
159.6 (100.6)
162.7 (77.7)
98
88
47%
60%
United
130.1 (42.2)
130.1 (42.2)
98
88
66%
73%
Kingdom
Czech Republic
n.a.
183.1 (34.0)
n.a.
88
n.a.
92%
Hungary
n.a.
166.7 (162.1)
n.a.
88
n.a.
46%
Poland
n.a.
192.6 (57.4)
n.a.
88
n.a.
87%
Notes: a standard deviation between brackets; n.a. = not applicable as these countries were not part of
the EU-27 in 1996.
Source: Estimated and calculated through the framework Chapter 2 with data from Hermann, yield
boost not included (1996, 1997, 2006).
As we carry out the analysis from an ex ante perspective, i.e. before adoption has
taken place, the relevant adoption data (yield increases and cost reductions) are not
yet available. Moreover, the estimation of certain parameters, such as elasticities, is
surrounded by uncertainty. Therefore, using the computer program @Risk from
Palisade Corporation, we incorporate subjective distributions for these parameters into
the model, using all prior information available. Through Monte Carlo simulations,
stochastic distributions are generated for the outcomes of the model.
Technology-induced cost reduction estimates are crucial to economic surplus
calculations. We reproduced the 2004 farm level profit estimates of Chapter 2 for the
agricultural season 1996 upon the hypothetical introduction of HT sugar beet.
Furthermore we assume that the ROW beet area is able to achieve a cost reduction
similar to the EU-27 and use the area-weighted average for this region.
To calibrate the model, we need to define regional ‘incentive prices’ for all
regions depending on the category introduced earlier. For the ROW the world price is
used. For EU regions, the incentive price depends on the region’s production
efficiency and the national pricing system applied to pay beet growers and processors.
67
The incentive prices for the former CMO for sugar are modeled in a dynamic way and
depend on the world price, which, on its turn, depends on world-wide adoption rates.
Incentive prices can be either A or B sugar prices, a region-specific mixed price, or
the world price. For the new CMO for sugar the incentive price for in quota sugar is
fixed (although decreasing in time) and the out-of-quota incentive price is the world
price. Chapter 3 introduces a multicriteria decision tool to assign the right incentive
price to different Member States (Table 4). Elasticity calibration remains unchanged
from Chapter 3.
4.4 Results and Discussion
We conduct a Monte Carlo simulation of 6,000 simulations to generate stochastic
distributions for our welfare estimates, using the software @Risk from Palisade
Corporation. Table 9 reports the mean values. The downstream sector (global
producers and consumers) captures the largest share (61%), while the seed sector
extracts 39% of the total welfare created despite the perfect corporate pricing strategy.
This result is in line with ex post impact studies on first-generation GM crops where,
on average, a value sharing of two thirds downstream and one third upstream is
observed (Demont et al., 2007b). 31% of the benefits accrue to the ROW if we
assume that beet producers in these countries are able to achieve similar cost
reductions from the technology as in the EU-27, and are not able to export the
technology-induced surplus on the world market which would further erode the world
market price (i.e. they are assumed to be WPN). The results presented further include
welfare effects foregone in the past and potential benefits in the future. Biennial price
and welfare effects are reported in Table 9. Worldwide, sugar beet growers gain €8.0
billion almost equally shared between EU-27 producers (57%) and ROW producers
(43%). The seed sector extracts €6.1 billion of the global welfare gains. If we do not
take into account any market effects, 57% of the benefits would flow to the beet
growers, while 43% would accrue to the seed sector.
68
Chapter 4. Global Welfare Effects of GM Sugar Beets under Changing Sugar Policies
Table 9: Biannual price and welfare effects of the global adoption of herbicide
tolerant sugar beet during the adoption scenario 1996-2015
Year
Price effects (%)
World sugar price
A sugar price
B sugar price
96/98
98/00 00/02
02/04
04/06
06/08
08/10
10/12
12/14 14/15
NPV
LSR
99.6
99.9
99.8
99.4
99.9
99.7
98.9
99.9
99.2
98.7
99.9
99.2
98.6
99.9
99.6
98.3
n.a.
n.a.
98.4
n.a.
n.a.
98.4
n.a.
n.a.
98.4
n.a.
n.a.
98.4
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
Welfare effects (million €)
Belgium
6
Denmark
5
Germany
31
Greece
5
Spain
25
France
13
Ireland
3
Italy
15
The Netherlands
10
Austria
4
Portugal
1
Finland
3
Sweden
2
United Kingdom
7
Czech Republic
0
Hungary
0
Poland
0
EU-27 producers
128
EU-27 consumers
0
ROW cane
-232
ROW beet
107
Net ROW producers -125
ROW consumers
296
Net ROW
171
Seed sector
189
Total
489
10
11
56
9
52
21
8
32
24
6
3
6
4
11
0
0
0
253
0
-313
205
-108
387
279
403
935
15
17
88
17
83
31
11
50
36
9
4
9
6
16
0
0
0
392
0
-529
316
-213
637
424
585
1,401
18
21
97
18
96
33
14
63
43
10
5
11
8
18
0
0
0
454
0
-596
369
-227
685
458
716
1,628
34
25
98
11
90
106
1
44
68
19
6
10
10
30
30
18
114
712
0
-730
383
-347
865
518
715
1,945
31
20
89
8
49
100
0
21
41
17
3
8
10
27
18
10
73
522
0
-999
391
-609
1,171
562
628
1,712
30
16
92
4
42
94
0
18
34
17
2
5
8
26
13
7
60
468
0
-1,046
398
-649
1,226
578
601
1,646
31
16
94
4
41
97
0
17
33
17
2
5
8
27
13
7
59
468
0
-1,110
399
-711
1,306
595
588
1,651
31
16
95
4
41
99
0
17
33
17
2
5
8
27
13
7
59
472
0
-1,167
405
-762
1,384
622
585
1,680
16
8
48
2
20
50
0
9
16
9
1
2
4
14
6
3
29
238
0
-606
205
-402
723
322
292
851
223
178
888
115
691
603
64
390
389
127
36
79
74
210
80
45
331
4,523
0
-7,222
3,461
-3,761
8,610
4,848
6,069
15,440
0.2
-3.1
0.2
-2.8
-3.3
0.3
-2.0
-2.2
-3.2
0.6
-3.6
-3.6
-2.0
0.3
-4.3
-2.2
-4.1
-1.2
n.a.
-0.4
-2.8
-0.7
n.a.
n.a.
n.a.
-0.7
Welfare distribution (%)
EU-27 producers
26
EU-27 consumers
0
Net ROW
35
Seed sector
39
Total
100
27
0
30
43
100
28
0
30
42
100
28
30
0
33
37
100
28
0
35
36
100
28
0
36
36
100
28
0
37
35
100
28
0
38
34
100
29
0
31
39
100
n.a.
n.a.
n.a.
n.a.
n.a.
37
0 0
28
27
44
37
100
100
Note: n.a. = not applicable; NPV = net present value in the agricultural season 2006/07 of the
accumulated welfare effects; LSR = land supply response to the technology
The depressing effect on world prices engendered by innovating WPR regions causes
ROW consumers to gain €8.6 billion, but this is for a large part offset by the ROW
cane growers’ loss of €7.2 billion. Since we assume that the technology spillovers to
the ROW beet sector do not depress the world price (WPN assumption), the EU is not
affected by HT sugar beet adoption in the ROW. Instead, through the inclusion of
WPR Member States, our model implicitly allows for the EU eroding its own
profitability through technological innovation, an ambiguity called immiserizing
growth (Bhagwati, 1958). However, our results show that the CMO for sugar largely
protects domestic producers against this perverse side effect of innovation. The model
69
suggests a world price decrease of 1.6% over a period of 19 years due to the
progressive adoption of the innovation. This estimate is relatively small compared
with the estimated annual price declines of 0.64% in the case of Bt cotton adoption in
the USA (Falck-Zepeda et al., 2000) and 0.88-0.97% (Moschini et al., 2000; Qaim
and Traxler, 2005) in the case of Roundup Ready® soybean adoption in the USA and
South America.
Under the former CMO, EU institutional prices were exogenously fixed, i.e. no
important price declines were possible. As a result, the benefits essentially accrued to
farmers without affecting EU processors and consumers. However, if weed control
based on GM HT technology increases the sugar beet’s sucrose content (Kniss et al.,
2004), processors will be expected to gain from the technology as the processing costs
are approximately the same per ton of beets regardless of sugar content (DeVuyst and
Wachenheim, 2005). Moreover, if the EU government endogenized public and private
agricultural research expenditures (see e.g. Swinnen and De Gorter, 1998) in the
CMO for sugar, benefits would be shared among farmers and consumers. Under the
new CMO, where no institutional prices for beets exist, created benefits can be shared
between farmers and sugar processors through lower beet prices. The welfare increase
for sugar processors could increase the pull by the sugar lobby to accept GM sugar
beet in the EU. The global welfare gain over the entire 19-year period considered,
finally, would accumulate to a 2006/07 net present value (NPV) of €15.4 billion. As
we assume no supply response for the majority of beet producers, the enhanced yields
of the technology engender important land contractions in the beet industry. Table 9
reports the average land supply response (LSR) to the technology. Our model predicts
that due to the adoption of HT sugar beet, the EU-27 beet area will shrink with 1.2%
on average. WPN Member States’ beet areas are expected to decline between 2.0%
and 4.3%, whereas WPR regions are expected to allocate more land to sugar beet, i.e.
between 0.2% and 0.6%, in response to increased profits. The ROW beet region will
remove 2.8% of sugar beet area from cultivation, while the ROW cane area will
shrink with about 0.4%. On a global scale, the sugar industry is expected to contract
its area allocated to sugar beet and cane with 0.7%.
In Table 10, we present some descriptive statistics of the global 2006/07 NPV of
HT sugar beet during the period 1996-2015. Given the assumed subjective
distributions, EU-27 producer surplus ranges from €3.8 billion to €5.3 billion in 95%
70
Chapter 4. Global Welfare Effects of GM Sugar Beets under Changing Sugar Policies
of the cases. Total welfare increase is less robust, ranging with the same probability
from €12.5 billion to €18.5 billion.
Table 10: Descriptive statistics of the global 2006/07 net present value of
herbicide tolerant sugar beet during 1996-2015
EU-27 producers
EU-27 consumers
Net ROW
Seed sector
Total
Minimum
3,245
0
2,563
4,310
10,999
2.5% confidence limit
3,751
0
3,366
4,777
12,512
Mean
4,523
0
4,848
6,069
15,440
97.5% confidence limit
5,347
0
6,333
7,355
18,462
Maximum
5,998
0
7,415
7,838
20,161
Table 11: Normalized regression coefficients of the estimated welfare effects in
the agricultural season 2006/07 in function of the model parameters
World
price
EU-27
producers
ROW
cane
ROW
beet
ROW
consumers
Net
ROW
Model parameter
Short-run flexibility
0.905
0.094
0.905
0.709
-0.905
-0.088
Long-run flexibility
0.375
0.039
0.375
0.293
-0.375
-0.037
Area elasticity ROW 0.000
0.000
0.000
0.000
0.000
0.016
cane
Area elasticity ROW -0.004 0.000
-0.004 -0.003 0.004
0.000
beet
Yield change ROW
0.000
0.000
0.000
0.534
0.000
0.861
Yield change EUa
-0.018 0.133
-0.018 -0.014 0.018
0.002
R2
0.982
0.997
0.982
0.989
0.982
0.998
Note: a The normalized regression coefficients are averaged over all EU regions.
Seed
sector
Total
0.003
0.000
0.000
-0.020
-0.008
0.009
0.000
0.000
-0.052
-0.001
1.000
0.444
0.038
0.999
Normalized regression coefficients in Table 11 reflect the robustness of the model to
individual parameter values. The coefficient of determination R2 is high in all
regressions, which suggests that the linear response surface sufficiently explains the
variation in the iterations. We investigate the coefficients for the most recent
agricultural season, 2006/07, the sensitivity estimates for the other seasons being
similar. The short-run flexibility (≤ 0), which can be interpreted as the inverse of the
ROW export demand elasticity, is the main driver of technology-induced world price
movements. A higher short-run flexibility implies a more elastic export demand
curve, engendering (i) a smaller technology-induced world price decline, (ii) a smaller
loss for all farmers (positive coefficient, columns 3, 4 and 5) and (iii) a smaller gain
for ROW consumers (negative coefficient, column 6). For the global welfare gains,
the opposing effects are largely cancelling each other out. Sensitivities to the lagged
sugar export supply expansion coefficient (long-run flexibility) are smaller because of
two reasons. First, we assumed a more narrow distribution for this parameter.
71
Secondly, as we assumed a monotonically increasing adoption curve, lagged
technology-induced EU sugar export supply expansions are smaller than actual
expansions such that it has a smaller effect on welfare gains, regardless of its
stochastic distribution. All yield increases have an important effect on global welfare.
As the EU model is disaggregated, each region features a separate stochastic yield
boost and the aggregate effect is partly cancelled out. However, for individual EU
regions the coefficients are larger, ranging from 0.011 for Hungary to 0.216 for
Germany. The ROW cane area benefits from all factors that prevent the EU (i) to
achieve large cost reductions in adopting HT sugar beet, e.g. a small yield boost, and
(ii) to export its surplus on the world market, e.g. an elastic export demand and/or
inelastic supply. As the ROW cane region does not innovate in our model, its welfare
is essentially a function of the world sugar price. Therefore, the world price and the
ROW cane region equations share the same regression coefficients. Table 11 reports a
small but significantly negative effect of a yield increase on the seed sector’s profits.
In highly protected sectors, such as quota systems, yield-enhancing technologies
negatively affect their own demand, as farmers who are non-responsive to world
prices (WPN) will decrease their land allocated to the crop, lowering the derived
demand for enhanced seed. This phenomenon has long been observed in the EU
market for sugar beet seed, which is gradually decreasing due to increasing
productivity and to decreasing acreage (Bijman, 2001). Including the market for
biofuels or modeling the introduction of GM technologies in sugar cane production
could also be included in further updates of the EUWABSIM model.
72
Chapter 5. The Barroso Proposal of Nationalizing GM
Approval: A Look at HT-Sugar Beets under Changed
European Sugar Policy
The president of the European Commission José Manuel Barroso proposed in 2009 to
nationalize the approval of genetically modified crops. We use a stochastic partial
equilibrium model with irreversibility to analyze the effect of the Common Market
Reform for sugar on the maximum incremental social tolerable irreversible costs
(MISTICs) of genetically modified herbicide tolerant sugar beet. Voting for approval
based on the MISTICs is assessed at the national and the EU level. Results show that
the Barroso proposal is more proportional to the economic incentives of technology
than the different EU treaties.
5.1 Introduction
Despite the first commercial introduction of genetically modified (GM) crops almost
15 years ago, the discussion on their benefits and risks is still at the forefront of the
debate in the European Union (EU). Over the years, the focus of the debate has
shifted over diverse frames of reference as freedom of research, environmental risk,
food safety, consumer protection, bioethics, economic policy and international trade
(Scholderer, 2005). Since the end of the de facto moratorium in 2004, harmonized EU
regulations are in place for deregulation for cultivation and imports, and for labeling
and tracing. Despite this harmonization, only two GM crops are deregulated for
cultivation and only one Bt maize variety is commercialized, with adoption limited to
eight EU Member States (Devos et al., 2009).
Two major regulatory reasons for this standstill can be determined. First, several
Member States use a safeguard clause to ban GM crop production on their territory,
based on their perception that novel concerns on consumer safety and environmental
risk exist (Protection, 2009; Sheingate, 2009). Secondly, the EU struggles with the
implementation of coherent national regulations to ensure the coexistence of GM and
non-GM production systems (Beckmann et al., 2010; Demont et al., 2009a; Devos et
al., 2009). Moreover, it has been suggested that national authorities exploit their
authority on coexistence regulations to design measures that hamper cultivation on
their territory (Devos et al., 2008b).
In his visionary talk at the end of his first term as the president of the European
Commision, José Manuel Barroso suggested a solution for this stalemate, “… it
should be possible to combine a Community authorization system, based on science,
with freedom for Member States to decide whether or not they wish to cultivate GM
crops on their territory” (Barroso, 2009). The idea, first proposed by Austria at the
Council of Environment Ministers in Luxembourg in June 2009, would confer the
decision of GM deregulation to the subsidiarity principle and allow the individual
Member States some degree of sovereignty in the field. The Barroso proposal would
converge the authority on deregulation with the authority on cultivation practices
under the coexistence regulations. Consequently, decision making can be determined
solely
at
the
national
level
instead
of
the
supranationalism
or
the
intergovernmentalism level, potentially increasing regulatory efficiency (Lieberman,
2006). Although the implementation of the Barroso proposal has to be verified
74
Chapter 5. The Barroso Proposal of Nationalizing GM approval
against the regulation of internal markets as specified in the Lisbon Treaty and the
signed WTO treaties, it has the potential to end the standstill in the deregulation of
new GM crops and their cultivation, as it is backed by a substantial amount of
Member States (GMO safety, 2009).
In this study, we assess how the Barroso Proposal could affect the deregulation of
herbicide tolerant (HT) sugar beet in the EU. This is done through a comparison of the
proposal with the EU Treaty it would replace. For such a comparison the driving
forces behind the political process have to be understood. The established literature on
the decision making and voting behavior in the EU
has highlighted different
determinants in the position of a Member State towards GM crop cultivation,
including the dominance of small scale farming, the presence of a strong
biotechnology sector and the share of organic production (e.g. Cooper, 2009; Kurzer
and Cooper, 2007b). In this chapter, the assumption is made that the decision is made
based on economic rationale, i.e. the EU tries to maximize the welfare of EU citizens.
However, as highlighted Chapter 1, irreversible effects and uncertainty surround the
introduction of GM crops and influence the political outcome (e.g. the safeguard
clause, precautionary principle) (Wesseler et al., 2007). Hence these concepts need to
be explicitly incorporated in the political economic framework. A Bayesian decision
analysis, i.e. real options, does this by estimating the maximum incremental social
tolerable irreversible costs (MISTICs) that justify the immediate release of the
technology (Ansink and Wesseler, 2009; Batie, 2003; Demont et al., 2004; Gollier
and Treich, 2003; Hennessy and Moschini, 2006; Mooney and Klein, 1999; Morel et
al., 2003; Wesseler et al., 2007). In his dissertation, Demont (2006) takes a similar
approach. However, his setup assumes that the EU functions as a single decision
maker weighing (with different weights depending on the scenario) the benefits and
risk of individual Member States in order to reach a societal optimal outcome. in
reality, the decision at the EU level is reached through a political process in which
representatives support national stands on the introduction. Hence, in this chapter we
take one step back and start from the voting behavior of individual citizens and follow
their vote through the regulatory process. The underlying assumption is that the
magnitude of a citizen’s MISTIC determines the individual voting behavior. The
higher a voter’s MISTIC, the higher his incentive to support the introduction of the
technology.
75
The earlier described stochastic partial equilibrium model EUWABSIM (Dillen et
al., 2009a; Dillen et al., 2008) provides the necessary data for the Bayesian analysis.
Demont et al. (2004) apply a real options approach on the same model but without the
inclusion of the changes described in Chapters 2 and 3. Moreover in this chapter, the
model output is used to calibrate the assumed stochastic process, which assures a
technology specific calibration, different from earlier approaches.
Applying the framework on the case of HT sugar beet deregulation is particulary
interesting as two policies affecting the process recently changed. In 2006, the
Common Market Organization (CMO) for sugar underwent its first drastic reform
since the establishment in 1968 (Chapter 3). Hence, we can estimate the MISTICs
under both the old and the new CMO and its differential effect on deregulation.
Secondly, in 2009 the EU adopted a new treaty, anticipating the enlargement of the
EU, which rewrites the procedures of decision making in the EU. Comparing both
treaties allows the assessment of different voting rules within the EU on the likelihood
of deregulating HT sugar beet and its relation to the economic incentives.
The chapter is structured as follows. In the next section, a comparison is made
between the conventional sugar beet production and the potential HT sugar beet
system, particularly with regard to the irreversible benefits and costs associated. In the
third section, both the Bayesian decision analysis and the stochastic partial
equilibrium model, yielding social reversible benefits, are introduced. Special
attention is given to the calibration procedure of the stochastic process and the
differences with earlier approaches. The fourth section presents the MISTIC values
and describes the impact of the CMO reform on these values. The fifth section
discusses the individual voting behavior and assesses the likely outcome under the
different voting rules considered. The final section discusses and concludes.
5.2 Herbicide Tolerant Sugar Beet
Herbicide tolerant sugar beet is very appealing for EU agriculture as it is grown in
most EU countries and economic sugar production is impossible without weed
control. HT sugar beet allows the farmer to use a single broad-spectrum herbicide
instead of mixtures of different active ingredients (a.i.). Hence, it allows easier and
more flexible control of weeds within a crop than by conventional herbicides or
mechanical means. Switching from conventional sugar beet cultivation to HT sugar
beet cultivation implies a shift in production system, not just using a single different
76
Chapter 5. The Barroso Proposal of Nationalizing GM approval
input (Alexander and Goodhue, 2002; Dillen et al., 2009a). This change in production
system has different agronomic and economic consequences. In this section we focus
on the quantifiable social irreversible costs and benefits accompanying the change in
production system based on Chapter 2, Bückmann et al. (2000) and Schäufele (2000)
(displayed in Table 12).
The conventional herbicide mix is replaced by the application of a broad spectrum
herbicide, hence both the type of a.i. and the amount of a.i applied per hectare is
altered. Similar to Demont et al. (2004) we estimate the external cost of releasing a.i.
in the environment based on Pretty et al. (2001). Accounting for the annual human
health cost and the loss of biodiversity, each kilogram of a.i. released in the
environment has an external cost to society of €1.13. 15 The monetarization by Pretty
et al. (2001) does not account for the difference in toxicity level of a.i.. For instance
for HT maize, Devos et al. (2008a) show that, due to the lower acute toxicity of
broad-spectrum herbicides and the lower potential to contaminate ground water, the
pesticide occupational and environmental risk (POCER) indicator was reduced
significantly under the HT regime. Therefore the aforementioned value can be
considered as a conservative estimation. Interestingly Table 12 indicates that in some
countries the shift to HT sugar beet would increase the amount of a.i. released. This is
in contrast with some earlier studies (e.g. Coyette et al., 2002) predicting a decrease in
a.i. following the introduction of HT sugar beet. This can be explained by a recent
reduction of the number of a.i. in conventional sugar beets (Eurostat, 2007).
According to industry sources, this is a result of a tendency to increased efficiency in
conventional cultivation in view of a revision of the EU regulations on herbicide
application.
The HT production system also alters the number of herbicide applications as
described in Chapter 2. Diesel use per application per hectare is estimated by
Rasmusson (1998) at 1.43l/ha which translates to 3.56kg CO2/ha following Phipps
and Park (2002). Using the price for CO2 estimated by the Intergovernmental Panel on
Climate Change (IPCC, 2001) the monetary value can be calculated.
Finally, farmers may be more inclined to adopt reduced or zero tillage systems as
the control of perennial grass weeds comes at no extra cost under a broad-spectrum
15
Pretty et al. (2001) consider USA, Germany and the UK. We extrapolate the results for the UK to the rest of
Europe. All monetary data were discounted to 2006 using the World Development Report Indicators
published by The Worldbank (www.worldbank.org)
77
herbicide (May, 2003). Results from the adoption of HT canola in Canada show a
strong correlation with the adoption of reduced tillage. Similar results have been
reported for the adoption of HT soybeans in the United States (Kalaitzandonakes,
2003). The adoption of reduced or zero tillage systems provides a number of
environmental benefits. This includes an increase in biodiversity and reduced nutrient
run-off. Carbon sequestration has also been mentioned but still remains controversial
(Baker et al., 2007). The IPCC recommends to use a 10% increase in soil carbon
sequestration for a change from conventional tillage to zero tillage and a 5% increase
for a change from conventional tillage to reduced tillage in temperate climates (West
and Post, 2002). The effect of adoption of reduced tillage systems induced by HT
sugar beets remains questionable. Romaneckas et al. (2009) report no yield effect of a
shift from conventional to reduced tillage for Lithuania while Koch (2009) report a
yield decrease for Germany. Given the controversy assessing soil carbon
sequestration and the possibility of a negative yield effects on sugar beets we do not
consider potential benefits of reduced or zero tillage adoption with the introduction of
HT sugar beets, but note that this may result in an underrepresentation of irreversible
environmental benefits.
5.3 The Economic Model
The introduction of a novel technology is surrounded by uncertain benefits and costs,
both affecting private stakeholders and society as a whole. The decision maker has to
decide whether to release the technology immediately, or wait until further
information becomes available. In the context of ex ante impact assessments, different
methodologies are developed to account for a part of these uncertainties (e.g. Demont
et al., 2008a; Dillen et al., 2010a). However, these methodologies do not take into
account the presence of irreversible and time effects. In the context of HT sugar beet,
these could include e.g. pollen drift, health issues, herbicide resistance or effects on
biodiversity. The Bayesian decision analysis of real options, suggested by Morel et al.
(2003) in the context of GM crops, offers a tool to account for these irreversible
effects. Some papers have followed this approach to increase the understanding of the
socio-economic impact of the potential introduction of GM crops (e.g. Demont et al.,
2004; Wesseler et al., 2007).
In order to optimize the welfare of EU citizens, the decision making unit has to
weigh the expected social reversible net benefits against the social irreversible net
78
Chapter 5. The Barroso Proposal of Nationalizing GM approval
costs. Through the explicit inclusion of the possibility of postponing the introduction
hysteresis is introduced. The technology should only be released if the reversible net
benefits are greater than the irreversible net costs multiplied by a factor higher than
one, the hurdle rate. The real option approach does allow the quantification of this the
hurdle rate, h, through contingent claim analysis and standard real option pricing
models (Demont et al., 2004; Dixit and Pindyck, 1994).
𝛽𝛽
(1)
ℎ = 𝛽𝛽 −1
with
1
𝛽𝛽 = 2 −
𝑟𝑟−𝛿𝛿
𝜎𝜎 2
𝑟𝑟−𝛿𝛿
1
2𝑟𝑟
+ �� 𝜎𝜎 2 − 2� + 𝜎𝜎 2 > 1
(2)
where r is the riskless rate of return, δ the convenience yield and σ the drift rate of a
geometric Brownian motion.
The full extent of the social irreversible costs and benefits of GM crops are highly
uncertain and central in the precautionary principle followed by the EU. However, the
net social reversible benefits can be determined as presented in Chapter 4. Hence,
reformulating the aforementioned decision criteria, the maximum incremental social
tolerable irreversible costs, MISTICs, can be calculated, justifying the immediate
release of the technology,
𝐼𝐼 ∗ = 𝑅𝑅 + 𝑊𝑊/ℎ
(3)
where R represents the quantifiable social irreversible benefits and costs and W the
social reversible benefits. The threshold value I* defines a space where immediate
introduction of the GM crop would be rational.
79
Table 12: The irreversible benefits and costs as a result from a switch from conventional sugar beet to a HT sugar beet production
system
Dosage of
herbicide
use 2003
(kga.i./ha)
Glyph.
dose
(l/ha)
Glyph.
dosage
(kg
a.i./ha)
Difference
dosage
(kg
a.i./ha)
External
irreversible
herbicide
reduction
benefits
(€/ha)
#
Conv.
app.
#
Glyph.
app.
Difference
# app.
1.6
6
2.16
-0.6
-0.63
2.5
2.5
0.0
Austria
3.0
6
2.16
0.8
0.95
3.5
2.5
1.0
Belgium
1.8
6
2.16
-0.4
-0.41
4
2.5
1.5
Denmark
2.8
6
2.16
0.6
0.72
3.8
2.5
1.3
Finland
3.4
6
2.16
1.2
1.40
3.8
2.5
1.3
France
2.4
6
2.16
0.2
0.27
3
2.5
0.5
Germany
3.7
3
1.08
2.6
2.95
1.5
1
0.5
Greece
0.2
6
2.16
-2.0
-2.21
3
2.5
0.5
Ireland
1.5
6
2.16
-0.7
-0.74
2.5
2.5
0.0
Italy
3.6
6
2.16
1.4
1.62
3.5
2.5
1.0
Netherlands
0.1
3
1.08
-1.0
-1.10
3
1
0.1
Portugal
6.9
3
1.08
5.8
6.55
3
1
0.1
Spain
2.1
6
2.16
-0.1
-0.07
2.9
2.5
0.4
Sweden
3.1
6
2.16
0.9
1.06
4.6
2.5
2.1
UK
6
2.16
1.6
1.85
3
2.5
0.5
Czech Republic 3.8
3.8
6
2.16
1.6
1.85
3
2.5
0.5
Hungary
2.8
6
2.16
0.6
0.72
3
2.5
0.5
Poland
Sources: Eurostat (2007), Bückmann et al. (2000),Pretty et al. (2001),Schäufele (2000),Phipps and Park (2002)
Diesel use
(l/Appl.ha)
Saving
in
diesel
use
(l/ha)
Avoided
carbon
dioxide
emission
(kg/ha)
External
irreversible
benefits
(€/ha)
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
0.0
1.4
2.1
1.8
1.8
0.7
0.7
0.7
0.0
1.4
0.1
0.1
0.6
2.9
0.7
0.7
0.7
0.00
4.99
7.48
6.48
6.48
2.49
2.49
2.49
0.00
4.99
0.35
0.35
1.99
10.47
2.49
2.49
2.49
-0.63
1.44
0.34
1.37
2.04
0.52
3.20
-1.96
-0.74
2.12
-0.11
7.55
0.13
2.10
2.10
2.10
0.97
Chapter 5. The Barroso Proposal of Nationalizing GM approval
5.3.1 Social reversible effects
To assess the social reversible benefits of HT sugar beet, an adapted version of the
EUWABSIM model is used. The model, described in earlier chapters is a stochastic
partial equilibrium model calculating the technology induced worldwide welfare
effects of a hypothetical introduction of HT sugar beet. Starting from a corporate
profit maximizing framework with heterogeneous adopters, developed by Dillen et al.
(2009a), it simulates the introduction of a technology protected by intellectual
property rights in an open economy (Alston et al., 1995; Moschini and Lapan, 1997).
19 regions are included, each of them modeled by a non linear constant elasticity
supply function: 17 EU regions 16, the rest of the world sugar beet region and a sugar
cane region. These differentiated supply functions are aggregated into an EU and
ROW supply function. This specification allows for heterogeneity among Member
States and for technology spillovers in the ROW beet region. The social reversible
benefits are calculated as welfare changes, both consumer and producer surpluses
(Demont, 2006). Both the old and the new CMO are explicitly modeled for the time
span of 2006-2014, to incorporate the policy specific price and supply effects. The
main differences lie in the decreased institutional price for sugar and the reduced
export possibilities for the EU. This translates in a diminished production under the
new CMO, shifting production to low cost producers (Bogetoft et al., 2007; Buysse et
al., 2007; Gohin and Bureau, 2006). Dillen et al. (2008) show that, as a secondary
effect, the incentive to adopt HT sugar beet decreases significantly for the high cost
producers, stimulating the crowding out effect of these producers in time.
The technology adoption process follows a logistic pattern following Griliches
(1957) as described in Chapter 4. The long term adoption ceilings, 𝜌𝜌𝑚𝑚𝑚𝑚𝑚𝑚 ,𝑖𝑖 , are
extracted from the study by Dillen et al. (2009a) that estimates these ceilings based on
a uniform European wide technology premium of €95/ha. The other parameters in the
adoption function are calibrated on the diffusion of HT soybean in the USA (Dillen et
al. 2009b). This innovation relies on the same technology and reaches adoption rates
similar to 𝜌𝜌𝑚𝑚𝑚𝑚𝑚𝑚 ,𝑖𝑖 (NASS, 2009). Producer surpluses are calculated as the region
annual welfare per hectare multiplied with the adopted acreage, which forces the
welfare function (𝑊𝑊𝑖𝑖 ,𝑐𝑐 (𝑡𝑡)) to follow a similar logistic pattern.
16
The remaining EU member states either do not produce sugar beets or very low amounts. Our 17 regions
represent 92% of the EU’s sugar production (Eurostat, 2009). Results are only presented for 16 EU regions as
Ireland abolished sugar beet production under the new CMO, making a comparison impossible.
81
Wi ,c (t ) =
Wmax,i ,c
(1)
1 + exp(−aW − bW t )
where 𝑎𝑎𝑊𝑊 and 𝑏𝑏𝑊𝑊 are a constant of integration and the diffusion rate respectively that
are assume constant for the different Member States (Chapter 4), t is time, i
differentiates the Member States and c is a dummy variable specifying the old or the
new CMO. 𝑊𝑊𝑚𝑚𝑚𝑚𝑚𝑚 ,𝑖𝑖,𝑐𝑐 is the highest annual welfare effect during the time period
considered from 2006 to 2014. The 2006 net present value of the social reversible
benefits can be calculated as
∞
𝑊𝑊06,𝑐𝑐 = ∫0 𝑊𝑊𝑖𝑖 ,𝑐𝑐 (𝑡𝑡) 𝑒𝑒𝑒𝑒𝑒𝑒−𝜇𝜇𝜇𝜇 𝑑𝑑𝑑𝑑
(2)
where µ is the risk-adjusted rate of return derived from the capital asset pricing
model. 17 Hence, our results refer to the year 2006, the year in which we assume the
start of adoption and when the CMO reform for sugar took place.
5.3.2 Social irreversible effects
The known social irreversible effects, ri, of the HT sugar beet innovation were
described in section 2 and presented in Table 12. They are approximated by,
(3)
𝑟𝑟𝑖𝑖 = ∑(𝐴𝐴𝑖𝑖 , 𝑁𝑁𝑖𝑖 )
with Ai the social effect of a reduced number of a.i. and, Ni the social irreversible
benefit of reduced carbon dioxide. We assume that ri is not affected by the change in
CMO. 18 Assuming that the per hectare social irreversible benefits and costs are
proportional to the adoption function, the 2006 present value, R06,i can be calculated
by
∞
𝑅𝑅06,𝑖𝑖 = ∫0 𝑅𝑅𝑖𝑖 (𝑡𝑡)𝑒𝑒𝑒𝑒𝑒𝑒−𝜇𝜇𝜇𝜇 𝑑𝑑𝑑𝑑
where Ri (t ) = ri
ρ max,i
1 + exp(−aW − bW t )
(4)
.
5.3.3 Calibration of the MISTICs
To calculate the hurdle rate, h, some additional calibration parameters are needed. As
a solution to the real options was reached through contingent claim analysis, a riskfree rate of return has to be decided. Following earlier papers on GM crops we set r
17
18
Following Demont et al. (2004) and Wesseler et al. (2007) we set µ at 10.5%.
This is a simplification. As the reform of the CMO reduced production in the EU and crowed out a group of
farmers, the average cultivation properties may have changed. However, data to assess this change in practices
is not available at this time.
82
Chapter 5. The Barroso Proposal of Nationalizing GM approval
at 4.5% to complement the risk adjusted rate, µ, of 10.5% (Demont et al., 2004;
Wesseler et al., 2007).
Central to the real options approach is the stochastic process followed by the value
of the technology. The standard assumption is that the time series of technology payoffs follows a geometric Brownian motion (Dixit and Pindyck, 1994). Different
approaches haven been used in the literature to determine the drift rate, α, and the
variance, σ, of the Brownian motion. Previous studies calculated hurdle rates using
time series data based on past performance of economic variables such as prices,
revenues or gross margins (Demont et al., 2004; Pietola and Wang, 2000; Rahim et
al., 2007; Wesseler et al., 2007), while other studies use simulation results projecting
returns of a technology (Ndeffo Mbah et al., 2010; Purvis et al., 1995; Winter-Nelson
and Amegbeto, 1998)
or investments (Hinrichs et al., 2008; Musshoff and
Hirschauer, 2008; Odening et al., 2005). In this study we calculate the maximum
likelihood estimators for α and σ using the results of the EUWABSIM model. Wi,j,c is
the net welfare effect of the technology per adopted hectare in country i, in year j
under policy c given the logistic adoption pattern. The EUWABSIM output can be
transformed into a differential time series for the period 2006-2014. This data output
represents the technology specific growth under the policy’s and the technology’s
particularities and can be used to estimate α and σ, needed to calculate β and the
hurdle rate, 19
𝑊𝑊 𝑖𝑖,𝑗𝑗 ,𝑐𝑐
𝜎𝜎 = 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 �ln 𝑊𝑊
𝑖𝑖,𝑗𝑗 −1,𝑐𝑐
𝑊𝑊 𝑖𝑖,𝑗𝑗 ,𝑐𝑐
𝛼𝛼 = 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 �ln 𝑊𝑊
𝑖𝑖,𝑗𝑗 −1,𝑐𝑐
�
�.
(5)
(6)
However, as the EUWABSIM model has a stochastic nature, this differential time
series is not deterministic due to uncertain effects and inputs. Table 11 presented the
stochastic variables influencing the outcome of the EUWABSIM model such as yield
boost and elasticities. How these are transposed to the technology pay-off depends on
the applied sugar price system and the volume traded on the world market. Therefore,
the parameters and the hurdle rate, h, are calculated for each single iteration of the
Monte Carlo simulation representing a certain possible time series of the technology.
Aggregating the simulation results leads to a stochastic hurdle rate instead of a
19
There is no reason the technology pay-off would not follow a geometric Brownian motion. Seasonality plays no
role as annual averages are used for the calibration nor does time dependency. The assumption of
lognormality does not cause problems as long as adopters can temporarily replace HT sugar beet with
conventional without additional costs. For HT sugar beet this assumption seems not problematic.
83
deterministic value in earlier studies. The hurdle rate’s simulated distribution is
approximated by a triangular probability density function to incorporate this
stochastic nature in the calculation of the MISTICs. 20
5.4 Model Results
Table 13 presents the results of the modeling exercise i.e. ρmax,i, Wmax,i, the mean
hurdle rate and the values of W, R and I* as annuities (subscript a). Wa are the highest
in Spain, the Netherlands, Austria and Portugal under both the old and the new CMO.
For some countries Wa decreases under the policy reform while for others the value
increases. For, a detailed discussion on the effect of the policy change on the welfare
creation, the reader is referred to Chapter 3 as several factors play a key role; the
changing price structure, the old quota structure and changes through time. ρmax,I
ranges from 43% in Greece and Italy to 99% in Portugal. The range can be explained
through the use of a uniform technology fee while herbicide expenditures are not
uniform among Member States. Hence, countries with previously high herbicide costs
or low heterogeneity among farmers tend to have higher adoption ceilings. The social
irreversible net effects vary among countries but are generally small ranging from €0.36/ha to €5.10/ha.
The hurdle rates differ significantly between the old and the new CMO. Under the
new CMO, where the export sugar is limited and there is limited effect of world
market prices on domestic sugar prices, the variation though time, and hence the drift
of the Brownian motion, is very low, leading to hurdle rates close to 1. If we look
very closely we can see that world price responsive countries (i.e. Austria, Belgium,
Germany, France, UK) have the highest hurdle rates as they are marginally exposed to
price changes on the world market. Under the old CMO hurdle rates are generally
higher. The reason is country specific and is a result of the higher variance in
technology pay-off due to different systems of sugar pricing and the volume exported
on the volatile world sugar market. A higher hurdle rate under the old CMO indicates
that social irreversible costs have to be compensated by a higher amount of social
reversible net benefits in order to justify an immediate deregulation of HT sugar beet.
20
As no theoretical correct distribution for this rate is known, a triangular PDF was chosen for transparency of the
model.
84
Chapter 5. The Barroso Proposal of Nationalizing GM approval
However, as Wa also changes (Chapter 3), the outcome is not straightforward. 21
Under the old CMO the highest annual MISTIC values are found in Portugal
(€254/ha) and Spain (€169/ha) and the lowest in Germany (€14/ha) and Denmark
(€21/ha). This is a combined effect of the fact that high cost producers have a higher
share of production under the A-quota, and thus benefit from higher sugar prices, and
the higher hurdle rate of sugar exporting countries. The total annual MISTICs within a
Member State these amount to €19 million and €15 million in Poland and France
respectively. This is mainly driven through the magnitude of adopted hectares in these
countries. Spain also has a high MISTIC value driven by the high potential adoption
rate and high benefits per hectare. Portugal, despite high Wa values and high adoption,
has low MISTICs due to the negative value of Ra and low acreage.
Under the assumption that the negative externalities from HT sugar beet
introduction stay at the sugar beet farm, we can assess the MISTIC farmers are
willing to accept to justify a release of the technology. Portugal and Czech Republic
have present the highest MISTIC at the level of the sugar beet holding. For Portugal
the reason can be found in the high value per adopted hectare combined with the high
adoption. The high value in Czech can be explained by the size of typical sugar beet
farms in the country. The lowest MISTICs can be found in Germany with €131/farm.
Under the new CMO, the MISTICs range from €48,1/ha in Italy to €197.6/ha in
Portugal. This range is smaller due to the smaller value of α and thus hurdle rate.
Comparing the individual outcomes under both regimes reveals that the new MISTICs
are higher for all countries except Greece, Italy, Finland and Portugal. As these
countries are generally considered high cost sugar producers, the conclusion can be
drawn that competitive countries have a higher incentive to adopt the technology
under the new CMO through increased Wa and lower hurdle rates. At the national
level the MISTICS are high in Germany and France but low for Portugal and Finland.
At the farm level Czech has by far the highest MISTIC with €3838/ha followed by
Denmark, Germany and France, a group of competitive sugar producers. These farm
level values indicate that if the decision was left to them whether or not to introduce
the technology, they would allow a significant amount of irreversible costs as the
value created by the technology is large.
21
The hurdle rates for the old CMO for sugar differ significantly from Demont et al. (2004) and are generally
lower. By calibrating the Brownian motion on a time series of gross margins several factors, Demont et al.
introduce sources of variability, e.g. weather effects, world market prices and other technologies in sugar beet
production, that do not directly relate to the technology of HT sugar beet leading to higher hurdle rates.
85
Table 13: EUWABSIM results, adoption ceilings (ρmax), mean hurdle rates,
annual social reversible benefits (Wa), social irreversible benefits (Ra) and
maximum incremental tolerable social irreversible costs (I*a) per hectare of HT
sugar beet and per sugar beet growing farmer.
Member State
Wmax
Wa
(€/ha)
Ra
(€/ha)
Hurdle
Rate
Ia*
(€/ha)
Ia*
(€)
98%
161
103
0.97
1.077613
97
8 107 383
4.79E-02
605
Denmark
82%
158
102
0.19
4.841197
21.3
857 356
6.76E-01
200
Germany
63%
84
51
0.22
3.836102
13.6
5 346 466
5.85E-01
131
Greece
43%
122
83
0.94
1.05967
79.4
2 933 103
3.82E-02
204
Spain
99%
245
164
5.10
1.000278
169.1
15 088 785
1.83E-04
978
France
78%
104
66
1.09
1.448304
46.5
15 859 971
1.97E-01
530
Italy
43%
97
64
-0.22
1.232448
51.9
9 860 526
1.28E-01
274
Netherland
97%
218
142
1.4
1.363038
105.9
9 283 404
1.75E-01
705
Austria
84%
176
113
-0.36
1.135094
99.6
4 680 685
7.97E-02
522
Portugal
99%
382
254
-0.07
1.000081
253.7
1 698 012
5.27E-05
2534
Finland
97%
182
121
0.91
1.000105
122.1
3 000 929
6.75E-05
1409
Sweden
47%
75
49
0.04
1.047497
47
2 375 252
3.11E-02
673
United Kingdom
59%
91
59
0.85
2.747777
22.2
2 557 346
4.54E-01
337
Czech Republic
91%
135
91
1.3
1.000665
92.6
4 872 339
4.41E-04
4640
Hungary
92%
90
60
1.32
1.000254
61.5
3 003 420
1.65E-04
1206
Poland
85%
123
82
0.56
1.004259
82.4
18 504 185
2.95E-03
229
98%
198
132
0.97
1.009099
131.9
11 031 382
3.44E-03
823
Denmark
82%
229
145
0.19
1.001713
144.8
5 076 405
6.40E-04
1186
Germany
63%
124
83
0.22
1.010634
82.2
32 244 302
3.79E-03
789
Greece
43%
105
65
0.94
1.002209
66.2
1 315 412
5.06E-04
91
Spain
99%
292
185
5.10
1.001611
189.7
13 470 372
6.05E-04
874
France
78%
150
102
1.09
1.014367
101.3
34 613 129
4.69E-03
1158
Italy
43%
77
48
-0.22
1.002026
48.1
5 509 687
6.77E-04
153
Netherland
97%
248
155
1.4
1.001956
156.4
10 583 648
7.08E-04
804
Austria
84%
212
141
-0.36
1.008735
139.7
6 572 343
3.62E-03
733
Portugal
99%
309
199
-0.07
1.00111
198.3
506 231
4.26E-04
756
Finland
97%
148
95
0.91
1.001445
96.2
1 617 853
5.80E-04
760
Sweden
47%
115
73
0.04
1.001841
72.8
2 538 996
6.37E-04
719
United Kingdom
59%
114
77
0.85
1.012847
76.6
8 838 270
4.57E-03
1164
Czech Republic
91%
167
106
1.3
1.00217
107.5
4 029 889
7.12E-04
3838
Hungary
92%
105
67
1.32
1.001456
67.9
2 255 003
5.58E-04
906
Poland
85%
152
97
0.56
1.001884
97.2
18 719 120
7.01E-04
231
Old CMO sugar
Belgium
New CMO sugar
Belgium
86
ρmax
Coefficient
of variation
Ia*/farm
(€)
Chapter 5. The Barroso Proposal of Nationalizing GM approval
5.5 Voting Assessment
Following Directive 2001/18/EC of the EU, the decision whether or not to release a
GM crop for commercial use typically begins at the level of the national competent
authorities. 22 They notify the EU and the European Food Safety authority starts with
the preparation of a scientific risk assessment. Based on this opinion, the European
Commission (EC) formulates a draft decision and presents this to the regulatory
committee, a group of Member State experts. Within this group a qualified majority is
needed to reach consensus and for the Commission to adopt the regulation. If no
qualified majority is reached, the EC’s draft is forwarded to the Consilium, better
known as the Council of Ministers. Here, a qualified majority is again needed to adopt
the regulation. If no decision is reached within three months, the Commission itself
may deregulate the GM crop for EU-wide use (Sheingate, 2009). Kurzer and Cooper
(2007a) argue that agricultural biotechnology is one of the rare occasions where
consumers have overruled the EU’s policymaking process. Through their organized
action, consumer movements have transformed a pro-GM approval system to a
market situation where GM food is labeled and its availability low. This study follows
another approach and assumes that consumers solely influence the EU decision
making through their voting behavior. Moreover, the assumption is made that the
individual’s MISTIC value determines the tendency to vote for or against approval of
a GM crop. If an individual has a high MISTIC value, he/she is more inclined to vote
for approval. I*vote,c, the MISTIC per voter per year from HT sugar beet under policy
c, is calculated by dividing Ia* through the number of people older than 18 (Eurostat,
2009) and shown in Table 14. The individual voting behavior can be represented by
∗
1 𝑖𝑖𝑖𝑖 𝐼𝐼 < 𝐼𝐼𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣
,𝑐𝑐
𝑉𝑉𝐼𝐼𝐼𝐼𝐼𝐼 = �
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
(7)
where I is the actual incremental social tolerable irreversible costs. Although I cannot
be determined, it is clear that if I*1vote < I*2vote than P (Vind=1| I*1vote) < P (Vind=1|
I*2vote). Under the assumption that all citizens have a similar exposure to irreversible
social net costs, I*vote,c is uniformily distributed within each Member State. Therefore,
Vind directly translates to a decision at the national level and consequently the
decision of their representatives in the Council of Ministers.
22
The website http://www.gmo-compass.org provides an overview about the steps for gaining approval for
planting GM crops in the EU.
87
At the Council of Ministers, a qualified majority is essential to proceed with the
deregulation of GM crops in Europe. The criteria to reach a qualified majority are
described in the EU’s treaty. The aim of a qualified majority is to introduce a
correction for the differences population between Member States in the voting
procedure. The voting weights are presented in Table 14. Under the Nice Treaty,
which entered into force on 1 February 2003, a qualified majority is reached if the
countries pro regulation represent more than half of the Members States and 74% of
the voting weights and 62% of the EU population (Felsenthal, 2001). Hence, VQMnic,
has the following form:
𝑉𝑉𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄 = �
1 𝑖𝑖𝑖𝑖 ∑𝑗𝑗𝑗𝑗𝑗𝑗 𝑣𝑣𝑗𝑗 ≥ 0.74 ∑𝑁𝑁 𝑣𝑣𝑛𝑛 ∧ 𝑗𝑗 ≥ 0.5𝑛𝑛 ∧ ∑𝑗𝑗𝑗𝑗𝑗𝑗 𝑝𝑝𝑗𝑗 ≥ 0.62 ∑𝑁𝑁 𝑝𝑝𝑛𝑛
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
(8)
with N the set of n Member States, j the countries voting in favor of the proposal, v
the weights assigned in the voting process to each Member State according to the
Nice treaty and p the population within country j or n.
On 1 December 2009, the Lisbon Treaty replaced the Nice treaty. The Lisbon
Treaty includes, among other things ensuring efficient governance in an expanding
EU, new criteria to reach a qualified majority. However, in a transitional phase until
31 October 2014 the voting rules of the Nice Treaty stay in place. From 2014 on, the
criteria to pass are a positive vote by a majority of countries (55%) representing 65%
of the population or every situation where the criteria to block are not met. To block a
proposal, at least 4 Member States have to vote against the proposal or in cases where
not all members participate, the minimum number of members representing more than
35% of the population of the participating Member States, plus one member,
𝑉𝑉𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄 =
⎧
⎪
⎪∑
⎨
⎪
⎪
⎩
𝑘𝑘𝑘𝑘𝑘𝑘
0 𝑖𝑖𝑖𝑖 ⋕ 𝑘𝑘 = 4 ∧ ∑𝑘𝑘𝑘𝑘𝑘𝑘 𝑝𝑝𝑘𝑘 ≥ 0.35 ∑𝑁𝑁 𝑝𝑝𝑛𝑛 𝑜𝑜𝑜𝑜 𝑖𝑖𝑖𝑖
𝑝𝑝𝑘𝑘 ≤ 0.35 ∑𝑁𝑁 𝑝𝑝𝑛𝑛 : 𝑡𝑡ℎ𝑒𝑒 ⋕ 𝑜𝑜𝑜𝑜 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑡𝑡𝑡𝑡 (∑𝑘𝑘𝑘𝑘𝑘𝑘 𝑝𝑝𝑘𝑘 ≥ 0.35 ∑𝑁𝑁 𝑝𝑝𝑛𝑛 ) + 1
1 𝑖𝑖𝑖𝑖 𝑗𝑗 ≥ 0.55𝑛𝑛 ∧ ∑𝑗𝑗𝑗𝑗𝑗𝑗 𝑝𝑝𝑗𝑗 ≥ 0.65 ∑𝑁𝑁 𝑝𝑝𝑛𝑛 𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎𝑎𝑎𝑎 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
with k the countries voting against the proposal (European Union, 2007).
The proposal introduced by Barroso at the end of 2009 has the aim to circumvent
the rules of the qualified majority by shifting the authority on deregulation to the
national level. Consequently in our model the decision whether or not to allow
cultivation of GM crops on their territory is transformed to an autonomous
88
Chapter 5. The Barroso Proposal of Nationalizing GM approval
dichotomous process directly reflecting the individual voting behavior within the
Member State. Hence, the outcome under the Barroso proposal can be represented by,
𝑉𝑉𝐵𝐵𝐵𝐵𝐵𝐵 ,𝑖𝑖 = 𝑉𝑉𝑖𝑖𝑖𝑖𝑖𝑖 = �
∗
1 𝑖𝑖𝑖𝑖 𝐼𝐼 < 𝐼𝐼𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣
,𝑖𝑖,𝑐𝑐
.
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
(10)
With the voting rules laid out, the results from Table 14 can be used transfer the
individual vote decision to a EU deregulation. Each value of I*vote,c can be considered
as a threshold value for a positive vote in a particular country following equation 7. In
order to evaluate the set of equations 8-10 we start with the Member State with the
highest I*vote,c and add the next Member State in the row until a qualified majority is
reached. Hence, a threshold for I can be determined which would justify the
immediate release of the technology under the differ. To allow for the fact that not all
Member States are present in our analysis, the relative shares were used representing
the situation that only these countries would have voting rights on the approval of HT
sugar beet.
First consider the situation as present in the EU, the voting rules of the Treaty of
Nice combined with the new CMO for sugar. The results show that
𝑉𝑉𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄 ,𝑛𝑛𝑛𝑛𝑛𝑛 �
€0.19
1 𝑖𝑖𝑖𝑖 𝐼𝐼 < 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 ∗𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
(11)
As long as the irreversible social incremental costs are below €0.19/ voting citizen
deregulation will happen. Replacing the Nice Treaty with the Lisbon Treaty results in
a higher threshold value,
𝑉𝑉𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄 ,𝑛𝑛𝑛𝑛𝑛𝑛 �
€0.28
1 𝑖𝑖𝑖𝑖 𝐼𝐼 < 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 ∗𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
.
(12)
For the Lisbon Treaty the opportunity to block a proposal at the Council of Ministers
was also explored in a similar way. However, the requirement to block would only be
reached for I equal to €0.37 making this an impossibility given the fact that a qualified
majority would be reached at €0.28. Similar to previously, we assume that I*1 < I*2
means that P (V=1| I*1) < P (V=1| I*2) or a higher threshold means a higher likelihood
of deregulation. Comparing VQMnic,new and VQMlis,new shows that the introduction of the
Lisbon Treaty would increase the likelihood of deregulation significantly due to a
47% increase in the threshold value. This is in line with the aim of the Lisbon Treaty
to facilitate decision making in a larger EU and reach a consensus more easily.
The same reasoning and sequence of equations can be followed for the old CMO
yielding the following result,
89
𝑉𝑉𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄 ,𝑜𝑜𝑜𝑜𝑜𝑜 = 𝑉𝑉𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄 ,𝑜𝑜𝑜𝑜𝑜𝑜 �
€0.20
1 𝑖𝑖𝑖𝑖 𝐼𝐼 < 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 ∗𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦
(13)
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
This result suggests that the under the old CMO for sugar the change of treaties does
not affect the likelihood of deregulating HT sugar beet.
Table 14: The MISTIC per voting citizen and the voting weights under different
EU treaties
Member State
Belgium
Denmark
Germany
Greece
Spain
France
Italy
The Netherlands
Austria
Portugal
Finland
Sweden
United Kingdom
Czech Republic
Hungary
Poland
Total
Population
over 18
years
8331936
4216893
67725607
9178742
36017564
49006703
48608776
12752453
6647030
8567879
4151882
7113513
46799543
8359568
8108480
30293256
Ia*/voting
person new
CMO,
Ia*/voting
person old
CMO,
( €/year)
(€/year)
I*vote,i ,new
1.32
1.20
0.48
0.14
0.37
0.71
0.11
0.83
0.99
0.06
0.39
0.36
0.19
0.48
0.28
0.62
I*vote,I,old
0.97
0.20
0.08
0.32
0.42
0.32
0.20
0.73
0.70
0.20
0.72
0.33
0.05
0.58
0.37
0.61
Voting
weight
under Nice
Treaty
12
7
29
12
27
29
29
13
10
12
7
10
29
12
12
27
277
Voting
weight
under
Lisbon
Treaty
11
5.5
82
11
46
64
60
17
8.3
11
5.3
9.2
62
10
10
38
450.3
Sources: Eurostat (2009), European Union (2007)
These results lead to an important conclusion. From a Chapter 3 we know that the
new CMO disfavors the adoption of HT sugar beet for high cost producers. Moreover
the results in Table 13 show a higher value of I*a is higher for lows cost producers
under the new CMO. Hence, the incentives for deregulation in low cost producers
have increased under the new CMO. Nevertheless, the results show that IQMnic,new <
IQMnic,old indicating that the likelihood for deregulation is lower if voted under the
Nice Treaty. However, under the Lisbon Treaty, IQMlis,new > IQMlis,old translating in an
increased likelihood of deregulation coinciding with the economic incentives. This
90
Chapter 5. The Barroso Proposal of Nationalizing GM approval
confirms the analysis by Leech (2002) that the rules to reach a qualified majority can
be too stringent to be an (economic) effective decision making rule.
Let us now turn towards a situation in which I is marginally higher than the
highest threshold to reach deregulation under any of the four scenarios. Applying the
rules of the Barroso proposal, equation 10, results in the fact that all countries for
which I*vote,c > €0.29 would deregulate while neither under the Nice Treaty nor under
the Lisbon Treaty deregulation would take place. Comparing the threshold value of
€0.29 with the results in Table 14 highlights the countries that would deregulate HT
sugar beet under the Barroso proposal. Under the old CMO all Member States except
Denmark, Germany, Italy, Portugal and the United Kingdom would deregulate the
cultivation of HT sugar beet on their territory. Under the new CMO, all countries
except Greece, Italy, Portugal, the United Kingdom and Hungary would deregulate.
Hence, the shift to the new sugar policy would increase the chance of deregulation in
two competitive sugar beet countries, Germany and Denmark, while a high cost
producer as Greece now has a lower chance of deregulating. Overall, the more
competitive Member States have a higher voting threshold under the new CMO,
increasing the potential of deregulation under the Barroso proposal. Hence, the
Barroso proposal, as expected, provides a better response to the invididual incentives
of the Member States and the effects of the sugar policy reform.
Finally, when the absolute values of I*vote,c are considered, it seems a rational
decision by the EU not to engage in an immediate release of HT sugar beet. Despite
the high potential value of the technology for sugar beet farmers (Dillen et al. 2009b),
and their resulting high MISTICs (Table 13), I*vote,c values are very low. As long as
individuals are prepared to pay as little as €0.08 for German citizens under the old
CMO, or €1.32 for Belgian citizens under the new CMO, it is better to postpone the
introduction of HT sugar beet. These results are in the same magnitude as the analysis
by Demont et al. (2004).
5.6 Discussion
This chapter presents a Bayesian decision analysis of the introduction of herbicide
tolerant sugar beet in the EU. The model is built on a stochastic partial equilibrium
model complemented with a real options approach calibrated on the same partial
equilibrium model. The aim of the model is twofold. First, the effect of a change in
the Common Market Organization for sugar on the MISTICs that justify immediate
91
release of the technology is assessed. Secondly, the decision making process at the
EU level is evaluated based on these calculated MISTIC values. In particular the
effect of the Barroso proposal compared to the conventional European treaties is
estimated.
The results show that the MISTIC value per hectare under the new CMO increases
for all Member States except for some high cost producers, Finland, Greece and
Portugal. This observation is a combined effect of the change in value added per
hectare through the adoption of HT sugar beet, as presented earlier in Dillen et al.
(2008) and the effect on the hurdle rate. The new CMO, through reduced export
opportunities and changed price effects makes the return from the technology more
predictable, hence decreasing the uncertainty and the hurdle rate, resulting in
increased MISTICs. From a regulators point of view this means that deregulating HT
sugar beet under the new CMO would be justified under higher irreversible social
costs, hence increasing the economic incentive for approval.
Under the assumption that voters base their position towards GM crops on the
magnitude of their personal MISTIC value, the outcomes of different EU decision
making protocols are assessed. The results show that increased economic rationale of
adopting HT sugar beet under the new CMO is only transformed in a higher chance of
deregulation under the Lisbon Treaty while the Nice Treaty even slightly decreases
the chance. The Barroso proposal to nationalize deregulation decisions has the closest
link to Member States’ economic incentives and hence the individual voter. Under
irreversible social costs that would never qualify for a qualified majority under the EU
treaties, a large majority of the Member States would engage in an immediate release
of the technology. Besides the closer link to economic incentives, the Barroso
proposal brings the legislative framework of deregulation at the same level as that of
coexistence, possibly increasing the efficiency of the resulting decisions and
regulations.
Finally, looking at the absolute values of the MISTICs the analysis shows that if
citizens are prepared to pay, depending on the Member State and the CMO, between
€0.06 and €1.32 annually to avoid the introduction of HT sugar beet, release should
be postponed. This result indicates that despite high economic benefits from HT sugar
beet and high MISTICs for sugar beet farms the benefits are not enough on a per
capita level to justify an immediate release. If the benefits of HT sugar beet would
increase in the future through a more liberalized sugar CMO or quality effects for the
92
Chapter 5. The Barroso Proposal of Nationalizing GM approval
consumer this could change. These are the same deterministic mechanisms that
influence the strength of consumer movements highlighted in the analysis by Kurzer
et al. (2007a).
93
94
Chapter 6. Are Farmers Willing to Pay for a Maize Variety
Resistant to Diabrotica virgifera virgifera Damage?
Western corn rootworm (WCR) (Diabrotica virgifera virgifera Le Conte), an invasive
species in Europe, presents a novel threat to European maize production. The
species’ larvae feed on maize roots, causing yield losses and lodging. Designing
socio-economic optimal control strategies is an important but difficult task for
European farmers and policy makers. The reason is twofold. First of all the species’
damage is highly variable and information in European settings limited. Secondly,
some of the control options are still under development or not deregulated. This study
attempts to shed some light on the issues at stake through a survey among Hungarian
maize farmers. European farmer’s perceptions about WCR have never been
documented before. Therefore we present data on the actual behavior of farmers
under this new constraint and their perceptions about the species. Farmers recognize
the threat of the species as they list it as the most important maize pest in Hungary.
Moreover, they estimate the average potential yield loss caused by the species be
taken at 22%. This explains why farmers list it as the most important maize pest in
Hungary. The introduction of damage resistant varieties, being conventionally bred
or through biotechnology, could increase the farmer’s toolbox and decrease the
dependence on chemical control options. The willingness-to-pay analysis shows that
farmers are willing to pay a significant premium, €70/ha, for the pecuniary and nonpecuniary benefits of such a variety. This information is valuable for seed suppliers
and policy makers, currently developing or deregulating these varieties.
6.1 Introduction
The threats of invasive species to agriculture are well documented and studied
(Pimentel et al., 2001). An introduced exotic species can induce significant economic
losses by imposing an additional constraint on crop production. Therefore, the
introduction and establishment of an invasive pest should be understood in the socioeconomic context of crop production (Beckmann and Wesseler, 2003; Boriani et al.,
2006). This chapter focuses on the Western corn rootworm (Diabrotica virgifera
virgifera Le Conte: WCR), a recently introduced invasive species threatening
European maize production. The species probably originated in Central America but
has spread over the United States Corn Belt, reaching the east coast of North America
(Gray et al., 2009). The costs arising from controlling WCR were estimated at around
$1 billion annually in the USA (Metcalf, 1986). Moreover, WCR is the insect pest
which causes the highest insecticide use in the world (Baufeld and Enzian, 2005c).
Over a decade ago the species was introduced into Europe, where it was first observed
near Belgrade, Serbia in 1992 (Baca, 1993). Since, at least four other introductions
from the USA into Europe took place, presumably via airplanes (Ciosi et al., 2008).
As a response to the introduction, the European Commission designed a series of
eradication and containment strategies detailed in Commission Decision 2003/766/EC
and later amendments (European Commission, 2003). Despite these regulations, the
pest was successful in spreading through Central and Eastern Europe. Two types of
infested areas have been identified in Europe (Figure 5): (i) areas of continuous
spread (in Central and Southeastern Europe and northwestern Italy), and (ii) several
disconnected outbreaks that did not persist over time or did not spread due to enforced
eradication measures in place. The Central European outbreak now extends over 11
countries, from Austria to the Ukraine and from southern Poland to Serbia. Studies
indicate that the spread of WCR to the rest of climatic suitable Europe is only a matter
of time (Baufeld and Enzian, 2005b). The steady spread of this invasive species
causes a threat to maize production in times with generally good prospects for
European maize production. Among the structural causes of this expected
appreciation are the steady rise in global commodity demand driven by urbanization
and changes in dietary patterns (notably for meat) in many parts of the world (in
particular India, China and Latin America) and the emergence of new market outlets
such as the biofuels market (European Commission, 2008). Furthermore, the
96
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
abolishment of intervention prices should increase fluidity in Eastern European
markets and therefore offer a larger marketing potential for European maize.
Figure 5: The presence of Western Corn Rootworm in Europe in 2008
(permission granted by C.R. Edwards, Purdue University)
Most of the damage to the maize crop is caused by larvae of the univoltine beetle
feeding on the root system, although some economic damage may occur through adult
beetles feeding on silk. The latter is especially important in high value crops such as
sweet and seed maize. Predicting and assessing the damage caused by the species is
very difficult as exogenous factors such as drought and soil conditions affect the
resulting yields to a high extent (e.g. Rice and Oleson, 2005). Larval damage may be
offset by the regrowth of the root system if water is available during the appropriate
time window of the growing process (Simic et al., 2008). This leads to difficulties, or
even impossibilites, in determining the correlation between population pressure and
damages (Urias-Lopez and Meinke, 2001). This uncertainty inflicts problems in
determining the optimal crop protection (CP) strategy in the field. In particular
conditions the pest can lead to a total loss of production due to lodging or goose
necking, while in other conditions no economic damage can be confirmed despite a
high pest population pressure.
97
In Chapters 7 and 8, a bio-economic model is developed to evaluate the
competitiveness of different CP options and shed light on the optimal design of these
strategies. However, before we turn to a economic assessment, two complementary
research aims are addressed in this chapter. Different papers (e.g. Dillen et al., 2010a;
Dillen et al., 2010b; Dun et al., 2010; Wesseler and Fall, 2010) highlight the need for
European data generation on WCR. Although data is available from the USA, very
limited knowledge exist about damages and practices in European crop production.
This will also proof key during the calibration of the bio-economic model later. Hence
in this chapter we present the results from a survey among Hungarian maize farmers
regarding their perceptions of the pest and the CP options. Hungary offers a good case
study as the species already appeared on its territory in 1995, via the southern county
Csongrad. In 2000, the first economic damage to maize roots by larvae was detected
and no uninfested plots were left by the end of 2003 (Boriani et al., 2006). Moreover,
maize is the most important crop in Hungary with around 1.2 million hectares sown
each year and 3.2 million tons exported in 2008 (Eurostat, 2009). On average around
40% of the maize area is sown in continuous cultivation (no rotation). This has
important consequences as continuous cultivation increases the potential for WCR
population build-up and damage. This share differs in each county and depends on
farm size, the degree of specialization of the farmer and the profitability of competing
crops. In absence of detailed data one can assume a high percentage of continuous
maize in smaller fields of individual farmers (Toth, 2005). Land allocation constraints
combined with a high private demand for maize (e.g. own livestock) reduce the crop
rotation options of smaller farmers resulting in reduced flexibility.
Secondly, the same survey is used to assess whether Hungarian farmers are
willing to pay for WCR resistant maize. For the moment only a small toolbox of CP
strategies is available to Hungarian farmers. However, different companies are
working on the development and deregulation of a maize variety resistant to WCR
damage. Hence assessing the valuation for such a technology is valuable for
technology providers and developers to assess the potential impact of their innovation.
Furthermore, a high willingness to pay (WTP) may also indicate shortcomings in the
available toolbox. Compared to the calculation of the monetary benefits of such a
technology in Chapter 7, the WTP approach also recognizes the presence of
technology properties that are not directly valuated in the market: increased farmer
health from reduced handling of chemicals, safety for the environment, reduced
98
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
management and scouting activity, certainty of protection (risk reduction), easier
application of the protection, … (Alston et al., 2002). Most of these non-pecuniary
benefits are either credence or experience based (Marra and Piggott, 2006). Therefore,
the willingness-to-pay (WTP) is influenced by the knowledge and beliefs of the
farmer about the technology and his perceptions of the pest problem. Earlier papers
(e.g. by Krishna and Qaim (2007) and Marra et al. (2001)) demonstrate the significant
importance of information provision in the valuation and adoption of Bt technology.
However, at the time of introduction, not all non-pecuniary benefits may be known to
the farmer. Piggott and Marra (2008) show that the effects of non-pecuniary benefits
on farmer’s WTP do indeed fluctuate over time. In an initial phase the demand for a
new technology is determined by its monetary gains, while in a later phase demand
becomes less dependent on price changes as the non-pecuniary benefits are revealed.
The WTP for a pest resistant maize variety has never been assessed in Europe
before.. This information may provide guidance to policymakers in order to develop
appropriate guidelines or deregulate certain control options. The remainder of this
chapter is structured as followed. In the next section the contingent valuation model
used to estimate the WTP for WCR damage resistant maize varieties is presented. In
the data section, the survey structure is documented and both some general
characteristics of maize cultivation in Hungary and specific farmer behavior under
WCR are presented. Finally, the results of the WTP model are presented and
discussed.
6.2 Methodology
Contingent valuation is a common methodology to measure a respondent’s WTP for a
good. It allows the researcher to extract a stated valuation where the revealed
valuation does not exist or non-pecuniary properties have a high value. In our case as
we assess a not yet commercialized maize variety with specific properties both of
these constraints are present. In the context of resistant crop varieties, contingent
valuation has mainly been applied to genetically modified (GM) crops (e.g. Alston et
al., 2002; Marra et al., 2004; Marra and Piggott, 2006). In this study this
differentiation between different breeding techniques is not made. The reason is
threefold. First of all GM crops have not been commercialized in Hungary and
therefore farmers never cultivated any GM crops. This means they have to rely on
secondary information and beliefs which may bias their WTP. This leads to the
99
second reason. The institutional environment in Hungary is strongly opposed to GM
cultivation. (Protection, 2009). Hence, not specifying the technology avoids political
statements during the evaluation. Finally, the resulting WTP now bears valuable
information to a broader range of agricultural stakeholders, including GM crop
producers and conventional maize breeders, as it indicates the technology price to be
used in marketing strategies and business development plans. On the other hand the
WTP may be optimistic for GM crops as costs for coexistence and segregation (Devos
et al., 2009) are not included in the analysis.
The setup of a contingent valuation study is as follows. The respondent is
introduced to the technology under research through a thorough story about both the
advantages and the disadvantages of its nature and choice. Afterwards a statement of
the respondent is extracted. In this chapter, a bounded choice model is chosen as this
form is generally superior to an open-ended estimation due to its more market-like
situation (Bateman et al., 2002). Bishop and Heberlein (1979) introduce a single
bounded model where the respondent is presented a single monetary amount, varied
among respondents, and has to decide whether he is willing to pay this amount for the
presented situation or not. Hanemann et al. (1991) in turn introduce a variant, the
dichotomous bounded choice model. Similarly to the earlier approach, a respondent is
first presented a certain monetary amount and has to state his opinion. Conditional on
this answer a follow-up question is chosen. If the respondent was willing to pay the
initially proposed amount, a second, but higher, bid is presented. If the respondent
rejects the first amount, a lower follow-up amount is proposed. The authors show,
both theoretical and empirical, that such a dichotomous approach is more efficient and
cost effective than the previous methods. In this study, a bivariate probit model
initially developed by Cameron and Quiggin (1994) is used. The probit functions is
estimated via a maximum likelihood estimation and has the advantage over a logit
model of reaching convergence easier in a bivariate model. 23 We define y1j=1 if the
response to the first question was “yes” and 0 if “no”, and y2j similarly coding for the
second answer, with d1j= 2 y1j -1, d2j= 2 y2j -1, ti the value of the bid, µi the mean, σi
the standard deviation, ρ the correlation and εi the error terms, the jth contribution to
the bivariate probit likelihood function becomes
23
The underlying assumption of a probit regression is that the error term has a normal distribution. This was
checked through the use of a residual plot and no systematic deviation was observed.
100
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
L j ( µ t ) = Φ ε 1ε 2 (d1 j (
t1 − µ1
σ1
), d 2 j (
t 2 − µ2
σ2
), d1 j d 2 j ρ )
(1)
This model is very general and flexible by allowing for the possibility of different
means, dispersion, and non perfect correlation across the two responses. We used a
Wald test to assess whether µ1= µ2 and σ1= σ2. Independent from which parameters
are introduced in the model we can reject the hypothesis. This means respondents
reveal a WTP significantly different in the two rounds of questions. This phenomenon
is well documented in literature (i.e. Cameron and Quiggin, 1994; Haab and
McConnel, 2002; Hanemann et al., 1991). Several explanations have been proposed.
Respondents may feel exploited if the price goes up in the second question, may
perceive the value of the product as decreased if the price is reduced or answer
strategically on the expectation of a third question creating a bargaining situation.
Despite these different preference sets, Haab and McConnel (2002) argue that the
bivariate probit model still offers higher efficiency than the single bounded approach
if the correlation between the two independent probit functions significantly differs
from zero. As this is the case in our data, the bivariate probit model is selected and the
parameters of the first equation estimated, as the preference set in the first bid can be
assumed to be the unaltered set. To avoid a starting point bias, we randomize the first
amount presented within a predetermined range (Herriges and Shogren, 1996). The
mean WTP is estimated by evaluating at variable mean values.
6.3 Data
A computer assisted telephone survey of maize farmers was undertaken in October
2008 by Kleffmann &Partner Kft Hungary. A stratified sample was randomly taken
from their database of maize producers. The stratification was based on the cultivation
of maize in specific regions of Hungary; the Danube valley (n=200), North (n=30)
and the Great Plain (n=220).
The surveyed farmers grew maize on 34% (standard deviation (σ) =19%) of their
cultivated area or 253ha (σ=394ha). If farmers were cultivating maize in continuous
cultivation in 2008, it covered on average 36% (σ= 28%) of the area sown with silage
maize and 44% (σ=31%) in the case of grain maize. Farmers in the Great Plain have a
significant higher percentage of continuous maize, which coincides with the figures of
AKI (2007) stating that gross margins are the largest in this region. If maize is
101
cultivated in a rotation scheme, complementary crops are winter wheat, barley,
sunflower and rape seed. Most farmers rely on a limited rotation system, with 62%
stating winter wheat as their primary pre crop and 44% stating it as the primary post
crop. In a secondary rotation scheme, one has winter barley and sunflower and to less
extent rape seed.
The presence of livestock on the farm influences the decision to cultivate maize.
Almost half of the farmers (43%) cultivating maize have some kind of livestock on
the farm, 51% of these have pigs and 49% have cattle, which are exclusive. This herd
can be complemented with some other animals (poultry 12%, sheep 10%). Farmers
engaging in mixed production cultivate significant higher acreage with silage maize
and in the case of cattle also the total maize area is significantly higher. Silage maize
is mainly produced for the spot market (26%) or for farm use (35%), only a limited
amount (11%) is planted on a contract base for which the expected price is lower than
on the spot market. The supply of own fodder is almost sufficient since only 9% of
the mixed farms buys extra fodder. Grain maize on the other hand is mainly used on
farm (47%) or produced on a contract basis (34%) with limited sales on the spot
market (19%).
The official infestation data state a full coverage of the Hungarian territory by
2003. However initial interviews with farmers seemed to contradict this statement. As
WCR is an invasive species, information on its spread is essential. Therefore farmers
were explicitly asked the data of the first detection of WCR in their own field. Figure
6 shows the percentage of maize fields infested through time according to official
sources and our survey. Boriani et al. (2006) indicate a full coverage of the territory
by the end of 2003 while at that time only 40% of the respondents detected WCR in
their field. Even in 2008 only 72% of the farmers detected the species in their maize
field. This discrepancy between the official figures and local knowledge points
towards an information lag and a lag between infestation and damage which may lead
to problems in designing appropriate control options based on scouting techniques.
102
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
120.00%
Cummulative presence
100.00%
80.00%
60.00%
in farmer perception
40.00%
Official (Borriani et al.
2006)
20.00%
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
0.00%
Figure 6: Presence of WCR in Hungarian maize fields according to official data
and the survey results
The respondents were also asked about the importance of WCR compared to other
maize pests. The results show that WCR is by far the most important pest 24, for
which 82% of the selected respondents applied some control measure. 11% of the
farmers controlled for Aphids, 10% for European Corn Borer, followed by the other
European pests such as cut and wireworms. This ranking is in line with previous
research on pest pressure in Hungarian agriculture from before the introduction of
WCR and highlight the importance of WCR control in Hungarian agriculture (Nagy et
al., 1999; Szõke et al., 2002).
As the pest is such is an important constraint to agricultural production, farmers
were asked to estimate the potential yield loss of WCR within their field. As expected,
the highly variable impact of the pest, due to exogenous effects, on yield damages is
reflected in the farmers’ perceptions. Therefore an Epanechnikov kernel function is
used to estimate the non parametric probability density function (PDF) to get an
insight in higher order statistics of the yield impact (Figure 7).
24
Limited amount of respondents (56%) had knowledge about all the pests within their field. We only used these
farmers to determine importance of different pests and control options.
103
Kernel density estimate of western corn rootwom in Hungary
0
.005
Density
.01
.015
.02
.025
(kernel=Epanechikov, bandwith=7.6)
0
10
20
30
40
50
60
70
Percentage yield loss
80
90
100
Figure 7: Kernel density estimate of WCR damage in Hungary
The resulting PDF is positively skewed with about 50% of the farmers stating they do
not fear any damage from WCR. On the right hand side of the distribution there are
also farmers stating their whole maize production may be lost under high pest
pressure. A possible explanation for this heterogeneity can be found in the different
production systems. Indeed, as highlighted before, continuous cultivation of maize
increases the population density and hence the potential for damages. As a
consequence, a uniform crop protection strategy for all Hungarian farmers may be
difficult to achieve. Averaging the perceived potential yield loss results in an
estimated yield loss of 22% if no control option would be taken.
Fortunately, farmers have access to different crop protection options to moderate
the effect of WCR in maize. The survey probed to the actual use of these tools in
Hungary. The spectrum and application procedures of these options are discussed in
detail in Chapter 7 or in the reviews by Levine and Oloumi-Sadeghi (1992) and Gray
et al. (2009). The results show that he most widely used protection against WCR in
Hungary is soil decontamination or soil insecticides with 59% of the maize farmers
applying it on an average area of 203 ha. At the time of the survey only one active
ingredient (tefluthrin) was allowed for use in maize (Ripka, 2008). Crop rotation and
seed treatment follow closely with respectively 47% on 228 ha and 44% on 196 ha.
104
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
Foliar application of chemicals is only used by 20% of the farmers on 91 ha on
average, with application rates between 1 and 4 (average 1.28). Foliar treatment is
only economical for the production of high value maize such as seed production as
they are under threat of silk feeding by adult beetles. All the control options indicate a
very high coefficient of variation (~1) which is endogenous to the heterogeneous
nature of maize production systems in Hungary. Nevertheless some interesting
properties of the control options can be extracted from the data. Crop rotation has a
significant negative dependence with soil insecticides but positive with seed and foliar
treatment. This can be explained by the fact that despite the use of crop rotation, and
seed treatments are applied as an insurance strategy and their mode of action against
wireworms (Muška, 2008). Seed treatment and foliar application are also a combined
strategy. In continuous cultivation of maize soil insecticides are the preferred sole
option as they have a higher efficiency under high population pressure (Ma et al.,
2009). Farmers using soil insecticides perceive the potential damage significantly
higher than farmers using any of the other techniques, once again confirming its status
as the best available solution under high population pressure. Respondents were also
asked about the practice of crop rotation in their field after detecting WCR. 52% of
the farmers stated a decrease in acreage of continuous maize with on average 59% of
their previous maize area(σ=30%). If this figure is to be compared with the share of
farmers indicating crop rotation as a control option (47%), it seems that 5% of the
farmers introduced it as a precautionary measure.
Finally farmers were asked about their actual loss in the last two cropping season
applying the aforementioned control options. In 2007, 20% of the farmers had
economic damage with an average yield loss of 20% (σ=19%). In 2008 only 12% of
the farmers had damage with an average yield loss of 15% (σ=14%). Interesting to
note is that 8.5% of the farmers had an economic loss in both years. This either means
their pest protection was not adequate, they miscalculated the severity of the pest or
they valued the damage from WCR lower than the price of the control measure.
To calibrate the dichotomous bivariate probit in the WTP analysis the bid prices
were randomly varied from 1000 to 10000 HUF/ha. This range was based on the
actual price of maize and the findings in previous valuation studies of Bt maize.
Conditional on whether the respondent was willing to pay the proposed value, a
second value was given being 30% higher if the initial answer was “yes”, 30% lower
if “No”. The variables used in the regression model are presented and elaborated in
105
Table 15. For the likert scale respondents were asked the importance of the factor in
their decision about the chosen control option against pests.
6.4 Results
The results from the dichotomous bounded choice model (equation 1) can be seen in
Table 15. The different variables included in the analysis are described in the same
table. A first group of variables assess the importance of a control option’s specific
characteristics on the WTP value for the technology of WCR resistant maize. It seems
that for Hungarian farmers factors such as personal health, then environment,
insurance and ease of management is not important. What is important to them is the
compatibility of a control strategy with existing machinery. As soil insecticides and
foliar application need specialized machinery (e.g. high clearance tractors and
airplanes) or contracted services which are not needed with the systemic protection of
resistant plants, this results in a higher WTP for WCR resistant maize.
Another group of variables consists of the farm’s demographic characteristics. An
important factor is the location of the farm. If a maize farm is located in Great Plain,
they are willing to pay significantly more than farmers in the rest of the country. The
gross margin for maize production is consistently higher in this region, making the
switch to crop rotation costly (AKI, 2007). Therefore they are willing to pay for a new
technology allowing them to produce maize in the future. A similar reasoning can be
followed for the significant positive coefficient accompanying the percentage of
maize in continuous cultivation. Farmers specialized in maize production have a need
for an efficient technology to guarantee the production of maize. The other
demographic variables (education, lease, livestock and age)
have no significant
effect. This could be induced by two reasons. All farmers may have the same
information on WCR and control options, or information is scattered and based on
personal experiences not captured by the demographic variables. The latter option is
not convincing in the setting of Hungary with highly educated farmers (70% of the
sample has a higher education training of which 70% a university degree in
agriculture) and information sourcing. The survey shows farmers stating own
experience as their major information source regarding insect control is limited to
9.1% of the population. With 49%, almost half of the farmers state an independent
source (advisors, agricultural magazines, lectures and professional literature) as the
determining source of information. Private companies are only the major source in
106
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
16% of the cases. Therefore we can assume that the extension services are efficient
and well used and information is uniform among farmers (although not necessarily
correct). The coefficients for both grain and silage prices indicate that high maize
prices spur the investment in WCR damage resistant varieties. This is rational as
under high prices there is more value in the production to protect. The significantly
negative effect of 2008’s yield is challenging but may be explained by the fact that
farmers with high maize yields feel they already possess the necessary tools to
manage the presence of WCR. Farmers with lower yields, possibly partially explained
by the WCR damage, are willing to pay more for a protecting technology.
Table 15: Results and parameters of the bivariate probit analysis to assess the
WTP for a WCR resistant maize variety
Variable
Description
Coefficient
Mangament
Likert scale 1-5 on the importance in control option decision
-0.01
Yield value
Likert scale 1-5 on the importance in control option decision
0.17
Insurance
Likert scale 1-5 on the importance in control option decision
-0.06
Personal health
Likert scale 1-5 on the importance in control option decision
-0.23
Equipment
Likert scale 1-5 on the importance in control option decision
0.25**
Environment
Likert scale 1-5 on the importance in control option decision
-0.2
IPM
Experience Integrated Pest management (1=yes,0=no)
Danube
1 if the farmer is located in the Danube valley, 0 otherwise
Plain
1 if the farmer is located in the Great Plain, 0 otherwise
Lease
The percentage of the cultivated area leased
-0.05
Maize
The percentage of the cultivated area planted with maize
0.64
Livestock
1 if the farmer has livestock, 0 otherwise
0.21
Price grain
Anticipated farm gate price for 2008 season for grain maize (HUF/ton)
0.0002***
Price silage
Anticipated farm gate price for 2008 season for silage maize (HUF/ton)
0.0003***
Grain continuous
Share of grain maize area under continuous cultivation
Yield grain
Expected yield in 2008 (ton/ha) for grain maize
-0.11*
Yield Silage
Expected yield in 2008 (ton/ha) for silage maize
-0.0004***
Education
1 if university education, 0 otherwise
-0.13
Age
Age of respondent
0.01
Infestation
Years since first detection in field (2009)
0.04
Future damage
1 if farmer expected damage in the future, 0 otherwise
-0.18
Off farm income
percentage of income from off farm employment
-0.03
0.63***
0.19
1.15**
0.01***
-2.74**
Intercept
-180.3
Log-likelihood
1
96.9**
chi squared
Estimated mean WTP
Average WTP for the proposed variety (€/ha)
70.3
Notes:1presents a goodness of fit test based on the log-likelihood. The regression parameters where
chosen based on this criteria. *=significant at the 0.1 level, **=0.05, ***0.01
107
A final significant and interesting result of the analysis is the effect of the farmer’s
experience with Integrated Pest Management (IPM). IPM is an environmentally
sensitive approach to pest management. IPM programs use information on the life
cycles of pests and their interaction with the environment. This information, in
combination with available pest control methods, is used to manage pest damage by
the most economical means, and with the least possible hazard to people, property,
and the environment. The positive effect means that farmers having an interest in
sustainable cultivation methods such as IPM value the resistant variety higher. This
could be a direct effect of the high toxicity of the allowed soil insecticides tefluthrin.
Finally, the specified bivariate probit model leads to an average WTP of €70.3/ha.
We tried to use the bootstrap methodology introduced by Krinsky et a. (1986) in order
to construct a confidence interval for the average WTP but failed to reach
convergence. A reason for this may be found in the high variability among farmers
and the different preference sets used by the respondents
6.5 Discussion
The results of our extensive survey seem to indicate that WCR presents a serious
threat to farmers. According to Hungarian farmers, on average 22% of the maize
yields may be lost if no action is undertaken but this may amount to a full yield loss in
specific cases. Hence, Hungarian farmers are using crop protection strategies such as
crop rotation, soil insecticides and seed treatment to moderate the effect of the
invasive species on their maize. Especially in continuous cultivation of maize, which
is common practice, the dependence on chemical control options is high with
potential environmental side effects. Maize resistant to WCR damage could expand
the spectrum of available control options and lessen the dependence on chemical
solutions. It seems that farmers are aware of the potential of such technologies as they
are willing to pay €70/ha as a premium to conventional seeds. Especially the reduced
need for specialized machinery increases the WTP for the resistant variety and is
important to farmers. Farmers that are inflexible due to specific production
constraints, such as land, reveal a higher WTP for resistant maize. Therefore the
marketing potential is mainly located in the Great Plains and among continuous maize
farmers.
108
Chapter 6. Are Farmers Willing to Pay for a WCR Resistant Maize Variety?
This kind of data helps to provide some insights in the practices of farmers to cope
with WCR, which has been limited up to now and is essential for the economic
assessment in the next chapters.
109
110
Chapter 7. On the Competitiveness of Diabrotica virgifera
virgifera Crop Protection Strategies in Hungary: a Bioeconomic Approach
Adapted from Dillen, Mitchell, & Tollens,
Journal of Applied Entomology, 134(5):395-408
Diabrotica virgifera virgifera Le Conte or Western Corn Rootworm (WCR), is a
major pest of cultivated maize, Zea mays L., introduced into Europe in 1992. Since
then, the beetle spread through Central Europe, leading to a continuous pest
population in 11 European countries. This chapter presents an economic assessment
of different crop protection strategies against this invasive species in the Hungarian
maize sector. A bio-economic model, using Monte Carlo sampling, estimates the
potential damage from WCR under no control and the value created by Bt maize, seed
treatment, soil insecticides and cultural control. At the same time potential market
shares for the different damage abatement options under perfect information are
deducted. The potential damages under a no control scenario are estimated at
€176/ha for grain maize farmers on average, which points out the need for well
designed damage abatement strategies. For land-constrained farmers cultural control
is a valuable damage abatement strategy, being the optimal choice in 69% of the
cases. In monoculture Bt maize is the best option as it creates the highest value in
78% of the cases. However, as Bt maize active against WCR is not deregulated in the
European Union, soil insecticides in 54% of the cases and seed treatment in 46% of
the cases are the rational choices. As the value created by Bt maize is positive, not
deregulating Bt maize in Hungary leads to benefits foregone ranging from €16/ha for
land-constrained grain farmers to €49/ha in the case of silage maize under
monoculture. Finally, the results of the sensitivity analysis can be used to develop a
multi-criteria tool to aid farmers in applying the appropriate damage abatement
strategy. This could decrease the dependency of farmers on scouting techniques and
economic thresholds of WCR presence.
7.1 Introduction
As discussed in the introduction of Chapter 6, the introduction of an invasive species
in agriculture hast to be understood in its socio-economic situation of crop production.
Earlier socio-economic literature on WCR in the USA mainly estimated economic
thresholds of insecticide applications (e.g. Foster et al., 1986), while recent literature
focused on integrated practices (Van Mellor et al., 2006), the value of introducing Bt
maize as a crop protection (CP) strategy (Alston et al., 2002) or on dynamic control
strategies (Crowder et al., 2005). In Europe, where the invasive species is still
spreading, economic damages are not yet stabilized, uncertainty about the efficiency
of different control measures is not yet resolved and future trends are difficult to
predict, economic assessments have been limited. Wesseler and Fall (2010) assess the
efficiency of containment strategies while others estimate the potential value of
damage and crop protection options (e.g. Baufeld and Enzian, 2005b; Fall and
Wesseler, 2008; Hatala Zsellér et al., 2006; Schaafsma et al., 1999).
In this chapter, a bio-economic simulation model is developed to assess the
absolute and relative competitiveness of alternative CP strategies and their potential
market share in a European setting, which is accepted to be different from the USA
situation (Kiss et al., 2005). In the first section different CP strategies are presented.
In a second part, the bio-economic model is described. As available data is scarce
and not all CP strategies commercialized, we opt for a stochastic ex ante simulation
model, explicitly incorporating heterogeneity among maize farmers. The impact of
different CP strategies is compared to the counterfactual of unrestricted damages.
However, instead of using a proxy to determine the main source of heterogeneity as in
Chapter 2, the impact of the production constraint, WCR, is explicitly modeled
through a stochastic biological component. Via a hierarchical modeling procedure this
component is then used as an input for the economic model. This approach
ameliorates the earlier economic studies as it reduces the need for expert opinions and
decouples the explicit link between CP options and yield losses. In the third section
data are presented. The fourth section presents the results of the model and the
sensitivity analysis. These results can be used to develop a multi-criteria decision tool
aiding the farmer in choosing the optimal control option given his constraints. The last
section concludes and discusses.
112
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
7.2 Control Options
7.2.1 Crop rotation
Crop rotation is tailored to the behavioral characteristics of the univoltine beetle and
has been successfully applied in the USA for decades. Egg deposition takes place
during late summer, predominantly in maize fields, where the eggs overwinter and
hatch the next spring. The larvae have relatively limited mobility, and therefore feed
on the roots in their vicinity. However, the species depends on monocotyledons to
terminate their lifecycle. Moreover, the potential of crops besides maize to act as a
feeding source is low (Branson and Ortman, 1967). Therefore, rotating maize with a
non-host crop offers a practical solution to limit pest population growth and overcome
damage. However, since the early nineties, damage in first year maize occurs in some
parts of the USA despite rotation (Levine and Oloumi-Sadeghi, 1996). In their review
Gray et al. (2009) point towards a lack of fidelity to maize as an oviposition site for
adult females as a trigger for the development of a rotation resistant pest population.
However, this variant has not evolved in Europe and the relatively higher
agrobiodiversity of European agriculture may not exert as strong of a selection
pressure on this trait. European maize production differs regionally but is generally
different from the USA due to a more diverse array of cropping patterns (Kiss et al.,
2005). This diminishes the pressure towards rotation resistant pest populations.
However, the same high agro-biodiversity increases the potential presence of other
host crops in the agricultural landscape. Moeser and Vidal (2004; 2005), studying
European populations of WCR, found that feeding on some grasses was significant
and adults emerged. Gloyna and Thieme (2007) tested WCR development on barley,
oats, spelt, triticale and wheat. On all of these crops, possible to use as a rotational
crops except oats, adult WCR development took place. Unfortunately none of these
studies assessed whether these adults were fecund and if so, where egg deposition
took place. In this study we assume crop rotation is a feasible CP strategy for infested
European maize fields in the present situation.
Crop rotation comes at a cost to farmers, the magnitude of which we use to
classify farmers in this chapter. Farmers with excess land and low specialization in
maize production encounter minimal costs to use new crop rotations, as they can
produce a crop mix equivalent to the situation before crop rotation was introduced as
a CP strategy by simply reallocating land. However, farmers facing land constraints
113
and specialize in maize production will have to adapt their output mix by growing a
second best crop, reducing farm profits. Within this group of land constrained
farmers, a special group can be differentiated who practice continuous (non-rotated)
maize production. This group mainly consists of small-scale farmers engaging in
livestock production or farmers producing under long term contracts and will not
readily adopt crop rotation as a CP strategy, but focus on other available solutions.
7.2.2 Chemical control
Both in the USA and in selected European countries, chemical CP strategies are
applied to both reduce the WCR population and prevent damages. Soil insecticides
and seed treatments target the larvae to protect maize roots from feeding damage. Soil
insecticide applications are placed in furrow, over the row or in a T-band during
planting.
The efficacy and consistency of soil insecticides depends upon a number of
factors: the applied active ingredient, timing of the application because of limited
persistence in the soil (about 6 weeks), leaching, physical and chemical composition
of soil, mechanical and operational aspects, …(see Gerber, 2003 for a detailed
review). Seed treatments on the other hand are placed directly onto the seeds via a
coating, thereby optimizing both spatial and temporal application while reducing
management requirements. Although these products provide adequate protection
under low pest population pressure, they tend to be more variable in protection than
traditional soil insecticides under high pest population pressure (Cox et al., 2007;
Horak et al., 2008; Ma et al., 2009). On the other hand, some of the active ingredients
used in seed treatment offer protection against wireworms, thereby increasing the
spectrum of application (Muška, 2008).
Adult control can be achieved through foliar application of insecticides. Although
adult control may be rational to reduce silk feeding in seed production and for sweet
maize, it does not directly protect roots from larval feeding. However, by reducing the
population, egg laying is reduced, thus reducing larvae damage in the next season.
Foliar spraying is either done by high clearance tractors or aerial application. As this
chapter only focuses on grain and silage maize, and the availability of the required
mechanical equipment in the countries under research is low, adult control is not
incorporated in the analysis.
114
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
Based on the experiences gained in the USA where insecticide resistance for some
active ingredients developed in WCR populations (Meinke et al., 1998), cautious
resistant management is a key factor in maintaining the efficiency of the different
control options. In this respect, the limited and reducing availability of active
ingredients allowed for use in the EU due to environmental concerns could create a
genetic bottleneck. In Hungary, only one active ingredient was available as a soil
insecticide in the 2008 planting season (Ripka, 2008). Combined with the findings of
Ciosi et al. (2008), demonstrating multiple introductions of WCR in Europe
increasing the probability of adaptations to management strategies, this points towards
the need for a diversified approach against WCR in Europe. Moreover, neither soil
insecticides nor seed treatment is an efficient way to control pest population
dynamics. Both offer farmers the ability to reduce damage but not necessarily to
reduce the presence of WCR in the field (Furlan et al., 2006), introducing a potential
for lagged damages.
7.2.3 Bt maize
Agricultural biotechnology is a recent addendum to the toolbox to prevent WCR
damage. The genome of the maize, resistant to WCR damage, contains a coleopteran
specific insecticidal toxin from the soil bacterium Bacillus thuringiensis (Bt). This socalled Bt maize was commercially introduced in 2003 in the USA. Adoption of the
technology by US farmers was rapid so that by 2005, about 1.8 million hectares were
planted in the USA with WCR resistant Bt maize. In 2009, 63% 25 of the US maize
area was planted with some type of Bt maize. The adoption of was successful at the
expense of soil insecticides. The Bt technology gives a systemic protection to WCR
and, as it is incorporated in the root, its performance is less likely to be affected by
environmental conditions, planting time, soil conditions or calibration of machinery
(Mitchell, 2002). The results in the USA indeed show a high efficiency of the trait to
avoid economic loss and high efficacy (Ward et al., 2005). The threat of resistance
against the active ingredient Cry3bb1, is reduced through the implementation of a
high-dose/refuge area in the USA. In Europe, the cultivation of this Bt maize
(MON88017) is not deregulated, but allowed for import as food and feed crop
(AGbios, 2008).
25
This figure also includes the area of Bt maize that only carries resistance against the European corn borer
(Ostrinia nubilalis) because disaggregated data are not available.
115
7.3 Bio-economic Model
Unfortunately, very little data is available on the extent of WCR damage and the
efficiency of different CP strategies in Hungary. Moreover, one of the potential CP
strategies, Bt maize, has not been deregulated. This imperfect and scarce data justifies
the setup of the model as an ex ante impact assessment. We compare the situation in
which every single farmer tries to optimize his farm income with a counterfactual
situation where no CP strategy would be applied, potentially resulting in high yield
losses. The structure of the model is similar to the approach in Chapter 2 which to
cope with the particularities of imperfect data in an ex ante impact assessment through
parametric modeling. Of particular interest in that chapter was the role of
heterogeneity among farmers. This heterogeneity, stemming from differences in soil,
climate, managerial capacities, etc, creates a different value for the available CP
strategies between farmers which has to be accounted for to avoid the homogeneity
bias. In the case of WCR another important source of uncertainty influences the
valuation of the technology in the population of potential adopters. As presented
earlier, the damage caused by WCR depends on several exogenous factors which
results in a low correlation between pest population, CP strategies and resulting
yields. To explicitly incorporate the unpredictable nature of the yield effect, this effect
is introduced through a hierarchical model starting from a biological component
which’s output feeds the economic farm model. Hence, compared to earlier
approaches, we take one step back and explicitly model the source of variance instead
of relying on observable proxy variables.
The biological component of the model follows Mitchell et al. (2004) predicting
the root rating under different CP strategies conditional on the counterfactual
untreated situation 26. For notation, let RRno denote the root rating without CP and
RRseed, RRsoil and RRBt respectively denote the root rating with a seed treatment, with
soil insecticide and with Bt maize active against WCR larvae. The model assumes
that the root rating with a treatment follows a beta probability density function (PDF)
with parameters conditional on the root rating without CP.
Specifically, the
distribution with control treatment is a beta PDF with a mean µ and standard
deviation σ as follows:
26
The root rating, sometimes referred to as the Iowa scale, is a measure to quantify the damage on a maize root
(Hills and Peters, 1971). The scale categorizes maize roots from no visible damage (1) to the equivalent of
three or more nodes of roots pruned off (6).
116
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
µ=
1 + m1 ( RRno − 1)
σ = e( s + s ( RR
0
1
no −1))
(2)
(3)
where m1, s0 and s1 are parameters to estimate via a maximum likelihood estimation.27
These functions normalize the root rating without control to the 0 to 5 range, then,
because the root rating must be between 1 and 6, the function for the mean has a
minimum of 1 when the observed root rating without control is also 1 (i.e., no
rootworm damage without control implies on average no root damage with control).
The standard deviation uses an exponential function to ensure that it is always
positive, with the intercept term e s0 defining the standard deviation when the root
rating without control is 1. The conditional beta PDF constructed needs a lower and
an upper limit. The lower limit is 1.0 as this denotes no visible damage and the upper
limit is assumed to be the maximum of RRno and 2.0. Therefore the upper limit is no
higher than the observed root rating without control, except when the observed root
rating without control is very low. Given the values or functions for the mean,
standard deviation, minimum, and maximum, the standard equation for the beta
function is used to derive functional forms for the parameters of the conditional beta
density.
A specific procedure using Monte Carlo simulation can now be used to generate
pseudo observations of the impact of WCR and the CP options in Hungary. In each
iteration, a value is drawn from a hypothetical PDF representing the counterfactual
situation. This value is then inserted in the biological model and initiates the draw of a
root rating for each single CP option conditional on this counterfactual value. Hence,
the difference in root rating between, ∆RRi =RRno - RRi where the subscript i denotes
the different abatement strategies, can be determined for each iteration. This pseudo
observation of ΔRRi is then translate in a yield effect according to
=
Yi Ybase (1 − ∆RRi .α )
(4)
where Ybase is the unaffected yield in Hungary, and α the percentage yield loss per
root rating extracted from the composed-error model by Mitchell et al. (2004). We
27
Other estimation techniques would have been possible. However, the maximum likelihood is not biased and
allows for the specification of the beta distribution with strict boundaries. Incorporating these constraints in
other techniques would create estimation problems that are out of scope of this chapter.
117
make abstraction from the standard deviation around α and assume variability is
captured by the conditional PDF. The specification of the model contains the implicit
assumption that the yield losses in the USA and Europe are equal despite the various
differences in production systems. However, due to the limited availability of data on
European damage and the deregulated status of Bt maize this is a necessary
assumption. In Chapter 8 this assumption is further discussed.
To transform the biological component resulting in a yield effect into an economic
decision model, the model is grounded in a rational choice framework assuming
farmers are utility maximizers. In this chapter we differentiate two utility functions.
The first utility function is a linear function of the profit derived from different CP
strategies. The second utility function is a negative exponential Von NeumannMorgenstern utility function. This specification allows introducing constant absolute
risk aversion (CARA). The assumption of CARA preferences is often used to analyze
farm decisions under risk (Just and Pope, 2002). As the absolute risk aversion level
does not convey sufficient information to indicate whether the implied level of risk
aversion is reasonable, Babcock et al. (1993) calculate the risk premium as a
percentage of the wealth at risk. The resulting certainty equivalent (CE) may lead to
an ordering of control options different from the risk neutral case (see Mitchell and
Hutchison, 2008 for an overview of economic methods to incorporate risk into
entomological contexts).
A partial budgeting approach is followed to assess the marginal effect of WCR on
farm profits. Partial budgeting is a commonly used method to estimate the net
increase in farmer returns due to a small change in management relative to some base
case (Olson, 2004).
π q = P(Yi − Yno ) − cq
where q = soil or seed treatment
(5)
π Bt = (1 − r )( P(YBt − Yno ) − YBt ∆P − cBt ) + r ( P(1 − ∆P)Yno )
(6)
π rot =
•
2( wother − wmaize )
3
P the price for identity preserved (IP) non –GM maize and ΔP the price
discount for Bt maize.
•
(7)
ci the cost of the different control measures on a per hectare base.
118
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
•
r the percentage area not planted with Bt maize while applying the Bt CP
strategy, including both the refuge area under the integrated resistance
management plans and the area planted with ex ante coexistence measures
under the European regulations (Devos et al., 2009 offers a comprehensible
overview). As refuge areas have to be planted in the direct vicinity of the Bt
maize field, we assume they are located in the buffer compulsory under the
coexistence guidelines and will therefore have to be sold at the discounted
price for Bt maize.
•
Yi is the average yield under the application of the alternative control
measures.
•
wmaize is the gross margin of maize and wother of the complementary crops in a
3-year rotational plan.
As highlighted in Chapter 2, the segment of adopting farmers in the population is
a function of the price/cost of the different CP options. However, as Bt maize active
against WCR is not deregulated in Europe, the technology fee is not yet established.
Chapter 2 presents a framework to calculate the technology fee assuming a revenue
maximizing strategy by the IPR owner. If the technology fee is not endogenized, i.e.
through competitive pricing or expert opinions, a pricing bias could emerge. Central
in the framework is the concept of technology valuation, the appraisal a potential
adopter has for the new technology compared to the other solutions available. The
technology valuation therefore depends on the threat of competition from other CP
strategies. The bio-economic model allows the calculation of the added value of Bt
maize compared to the next best alternative. Under this specification, the technology
valuation, θ , is represented by
=
θ π Bt0 − max [π soil , π seed , π rot ]
(8)
where π Bt0 is the profit made from Bt maize assuming a zero price level for the
technology. The heterogeneity and uncertainty introduced in the model through the
pseudo observations, implies that θ has a PDF, f(θ), representing the distribution of
technology valuation among farmers. Farmers on the right/upper tail will have a high
valuation and will likely adopt the technology while farmers at the lower/left tail
would not adopt even at a zero price. From this PDF a normalized demand curve for
119
the technology can be constructed via the descending cumulative distribution Q(θ) =
1 - F(θ) with θ the price of the technology. As Bt maize is protected by IPRs, we
assume that the innovator has some market power which allows him to set θ at a rate
maximizing his revenue function:
π (θ ) = θ Q (θ ).
(9)
The optimal price of the technology bundle, θ*, satisfies the following condition:
θ * = arg max π (θ ).
θ
(10)
With the assumption of rational farmers, a farmer will adopt if his technology
valuation exceeds θ*. Moreover, the technology fee, θ*, equals the marginal adopter,
indifferent between accepting and refusal (Lapan and Moschini, 2004). With the
marginal adopter known, the potential adoption rate, ρ, and the average value ( ω )
accruing to adopters from introducing Bt maize can be calculated,
∞
ρ = ∫ f (θ )dθ
(11)
θ*
=
ω
∞
∫ (θ − θ
*
) f a (θ ) dθ
(12)
θ*
where
 f (θ )

f a (θ ) =  ρ
0

(θ > θ * )
(13)
(θ ≤ θ * )
7.3.1 Data
Data are gathered from various sources and transformed in order to serve as an input
for the stochastic bio-economic model and presented in Table 16. In this section we
discuss some of the most important data. Further implications of the data use is
discussed in Chapter 8 if we compare the model in different countries.
120
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
Table 16: Input data for the bio-economic model of Chapter 7
Parameter
Value
Source
Yield grain maize (t/ha)
PERT (0; 7.70; 10.26)
Based on AKI (2008)
Price grain maize(Ft/t)
Lognormal(29278;11140)
“
Yield silage (t/ha)
PERT(0;31.85;42.46)
“
Price silage (Ft/t)
Lognormal(4874;775)
“
Cost seed treatment (Ft/ha)
Uniform(10130;12836)
Hatala Zselléret al. (2006)
Cost soil insecticide (Ft/ha)
17472
“
RR in counterfactual
PERT(1;2;6)
Assumption
Yield increase / RR (α)
0.114
Mitchell et al. (2004)
Price discount (%)
Triangular(0;2;3)
Assumption
Technology fee (€/ha)
Triangular(18;23;28)
Endogenized
Refuge (%)
20
AGBIOS (2008)
Gross Margin Maize (Ft/ha)
PERT(68627.6;87959.5;94444.6)
Based on AKI (2008)
Gross Margin Barley (Ft/ha)
PERT(41350.6;58762.0;73847.9)
“
Gross Margin Winter wheat (Ft/ha)
PERT(48337.0;67416.8;78270.4)
“
Exchange rate (Ft/€)
252
Oanda (2008)
-0.2
Assumption based
Goodwin (2009)
Correlations
Price grain-yield grain
on
The lognormal PDF on price data is constructed based on the mean and standard
deviation of a time series from 2000 to 2008 (AKI, 2008). Following Goodwin (2009)
we introduce a negative correlation between yields and prices to account for the effect
that under a high supply of maize, market prices may decrease.
Yield itself is an output from germplasm, management and environmental
influences. Due to the correlation between space and time, following from the
systemic nature of crop production, the central limit theorem is not applicable.
Moreover, crop yield distributions are assumed to be skewed due to the biological
constraints at the tail of the distribution (Goodwin and Ker, 2002). We assume crop
yields follow a linear trend in time and deviate from this trend due to external
constraints. Assuming that the deviation is proportional to the trend we calibrate our
uninfested yields, Ybase on trend adjusted yields. Regressing yields in time we get
expected yields in year t of Yˆt and the error, eˆt , which are used to recenter the latest
yield data for year t = 2008 as the base,
ˆ
∧
Ybase= (1 + Yeˆ2008 ) Y 2008 .
2008
(14)
121
Under the constraints of scarce data we opt for the PERT distribution to model the
base yield as it is well suited to represent subjective estimates with lower emphasis on
the extremes when data availability for these fractiles is low (see Chapter 2 Lau et al.,
1998). The distribution is parameterized on the centered yield with zero and 150% of
Ybase as the extremes. 28
Data for estimation of equations 1 and 2 were assembled from field trials
conducted by entomology faculties at universities in Illinois, Nebraska and Ohio. This
data was combined data provided by Monsanto for early comparisons of Bt maize for
rootworm control with conventional control by soil insecticides. In many cases, trap
crops were planted the previous year to ensure a high amount of larval pressure.
These field trials are randomized plots comparing the various rootworm control
options with an untreated control plot. Their original purpose was often to provide
third-party evaluations of the control efficiency of the various WCR larval control
options available to farmers. The majority of these data were for soil insecticides (949
observations), with a few for Bt maize (75 observation) and seed treatments (38
observations) because soil insecticides were the most popular form of larval control in
use at the time and Bt maize was relatively new. Later evaluations of Bt maize and
soil insecticides switched to using the node injury scale of Oleson et al.(2005), so few
observations were available for this currently popular form of larval control. The
results of the estimation are shown in Table 17 which serves as an input to construct
the conditional PDFs of root ratings under different CP strategies. To apply equation 3
an estimation of RRno, the counterfactual situation is needed. However, due to the
containment regulations in place and the use of CP, this counterfactual situation is not
observable in Hungary. Therefore we have to make an assumption based on expert
opinions and compare the resulting yield effects with the survey data from Chapter 6
and other secondary data sources to check consistency before continuing with the
calculations (a detailed set of assumptions is given in Dillen et al., 2009c).
28
The specification of the base yield is conservative as it assumes that the used historic yield data are not yet
affected by WCR. This is debatable in those countries already suffering from a high pest population pressure
such as Hungary.
122
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
Table 17: Estimation results of the parameters determining the conditional beta
distribution of root ratings under different damage abatement strategies
Damage abatement
strategy
Seed Treatmenta
Soil Insecticideb
Bt maizec
a
Parameter
m1
s0
s1
m1
s0
s1
m1
s0
s1
Estimate
0.674
-2.145
0.440
0.427
-1.500
0.231
0.262
-1.687
0.317
Standard Error
0.032
0.552
0.139
0.005
0.059
0.018
0.036
0.121
0.057
t Statistic
20.99
-3.89
3.15
77.91
-25.41
13.04
7.31
-13.94
5.54
p Value
< 0.001
< 0.001
0.002
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
38 observations, RMSE = 0.479 using the mean as the prediction.
949 observations, RMSE = 0.492 using the mean as the prediction.
c
75 observations, RMSE =0.416 using the mean as the prediction.
Source: based on: Gray and Steffey (1998), Meinke et al.(2000a; 2000b; 2001; 2002; 2003; 2004a;
2004b), Eisley (2002) and Mitchell (2002).
b
Costs for seed treatment and soil insecticide are taken from the Hungarian State
Phytosanitary Institute. Prices are in line with the €40-60/ha for insecticides according
to Takacs et al. (2007) and data we received from wholesalers in Hungary. To reflect
variation in the price for seed treatments, we use a uniform PDF with the appropriate
ranges while we assume a fixed price for soil insecticides as only teflutrin was
allowed for use in the 2008 season. In the case of Bt maize, costs arise from different
sources as reflected in equation 5. The refuge area needed to ensure effective insect
resistance management is assumed to be equal to the requirements in the USA, i.e.
20% of the planted area. We assume that this refuge area at the same time allows
farmers to comply with the European legislation on ex ante coexistences. The farmer
does not control for WCR in the refuge area and therefore encounters yield losses
while offering it in the market as non IP maize. In theory there may be a price
discount for Bt maize as consumers perceive it to be an inferior product. European
markets in GM crops are premature, but for a price discount for Bt maize did not
develop in the maize market (Gomez-Barbero et al., 2008; Skevas et al., 2009).
However, to be conservative, a subjective triangular PDF ranging from no discount to
3% is introduced.
Cultural control options are evaluated based on alternative cultivation options. As
the gross margin of complementary crops is highly uncertain at the time of decision
making, the associated cost has a high variability. AKI (2008) documents gross
margins of commodities with their spatial and temporal variation throughout
Hungary. Using the standard methodology to calibrate a PERT PDF based on the
123
most likely value and the extremes, the appropriate distributions are constructed. As
winter wheat and barley are the main crops rotated with maize (see Chapter 6), we use
this rotation scheme in the model. We believe that the broad variability in gross
margins covers the local deviations from the assumed rotation scheme. However, not
all farmers switching from continuous maize to a rotation scheme encounter a
reduction in revenue. As introduced before, we distinguish three groups of farmers. A
first group of land-unconstrained farmers has the possibility to engage in crop rotation
as a CP strategy without altering their output mix and hence there revenues. These
farmers are only marginally affected by the WCR infestation and we assume that their
damage of WCR is zero. Land-constrained farmers that have to reduce maize
production under crop rotation bear a cost due to the lower gross margins on the the
rotation crop. Within this group of land-constrained farmers, there is a sub-population
that is characterized by the compulsory cultivation of continuous maize. The
underlying reason is the high degree of specialization, often related to the presence of
livestock on the farm or long term contracts (i.e. biofuels production). As land
resources are essential in this differentiation, smaller farmers have a higher chance of
engaging in continuous maize production.
Finally, for CARA farmers we follow Mitchell et al. (2004) and set the risk
premium at 20% of the standard deviation on profits of the different CP options, to
account for moderately risk averse farmers.
7.3.2 Simulations and Results
The calculations are carried out in Excel, using the @RISK add-in by Palisade
Corporation allowing Monte Carlo sampling in Excel. Each simulation was run for
10 000 iterations to ensure convergence of the results. Before turning the attention to
the competitiveness of the different control options, the results of the specification of
the biological component are evaluated (Figure 8). The average yield loss in
Hungarian maize production reaches on 28%. This is reasonably close the results of
the survey in Chapter 6 where farmers estimated this value to be 22%. Earlier
literature also presents a range for the potential yield loss. Schaafsma et al. (1999)
estimate the effect for Germany at around 10%, Baufeld and Enzian (2005a) use 1013%, while Macleod et al. (2007) assume 25-30% for the UK. Our result is on the
higher edge of this range. However, as Hungary is characterized by a high percentage
124
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
of continuous maize (Toth, 2005) a high population pressure in the counterfactual
may be expected.
With this specification, the efficiency of the different CP strategies can be
evaluated. Crop rotation offers the highest yield protection as we assume no damages
under this strategy. Bt maize has an average efficiency of 96% while soil insecticides
and seed treatment offer a lower efficiency, respectively 93% and 88%.The estimated
yield loss from WCR leads to a potential average loss of €176/ha in grain maize
cultivation. Wesseler and Fall (2010) estimate these costs in the same magnitude at
€137/ha.
1
Cummlative percentage
0.8
0.6
Yield crop rotation (mean 6.8 t/ha)
0.4
Yield seed treatment (mean 6.0 t/ha)
Yield soil insecticide (mean 6.3 t/ha)
0.2
Yield Bt maize (mean 6.5 t/ha)
11
10
9
8
7
6
5
4
3
2
1
0
0
Yield grain maize (t/ha)
Figure 8: Yield under different damage abatement strategies in Hungarian field
conditions
In a next step, the model is setup to estimated the distribution of the technology
valuation, f(θ). Based on the Kolmogorov-Smirnov test, the Weibull PDF was chosen
as a parametric representation of f(θ). This PDF allows the calculation of the
endogenized technology fee, θ*. Chapter 2 shows that pricing the technology
differently in separate submarkets, e.g. grain and silage maize, increases the revenue
of the innovator. However, grain and silage maize markets are not clearly distinct at
the time of planting. The farmer’s choice between silage and grain can be reviewed
during the growth process if exogenous factors favor the one or the other system (e.g.
125
ear diseases, price changes, etc.). Moreover, the area under silage maize is small in
Hungary, making it a niche market. Therefore we assume that the price is set in the
grain maize market, leading to a technology fee of €23/ha. This value lies within the
range of magnitude found in other papers. Alston et al. (2002) price the technology
competitive to soil insecticides, at about $30/ha in the USA. Fall et al. (2008) use
different scenarios with a technology fee ranging from €27.65/ha to €102.7/ha for
different European Member States, but abstracting from costs arising from refuge and
price discount.
Table 18: Value created by different damage abatement strategies against WCR
in Hungary for risk neutral and risk averse farmers (€/ha)
Risk Neutral
Mean
Coefficient of
variation
Certainty
Equivalent
Rent Bt maize
148.2
0.6
132.2
Rent seed treatment
100.1
0.8
87.8
Rent soil insecticide
113.0
0.9
96.4
Rent crop rotation
169.0
0.8
152.6
Land-constrained
16.1
0.6
17.6
Continuous maize
37.0
0.7
35.7
Rent Bt maize
46.8
0.9
38.4
Rent seed treatment
-12.0
2.4
-17.5
Rent soil insecticide
-10.0
4.5
-18.8
Rent crop rotation
40.7
1.7
27.6
Land-constrained
25.6
0.5
26.5
Continuous maize
48.5
0.5
46.3
Grain
Added value created by Bt maize
CARA
Silage
Added value created by Bt maize
With all the necessary variables defined, the initial research question can be
addressed. The results presented in Table 18 show that in grain maize production all
different CP strategies on average create a positive rent. For land-constrained farmers,
crop rotation offers the highest value, €169/ha, followed by Bt maize, soil insecticides
and seed treatment. The heterogeneity among farmers and the particular properties of
each CP strategy suggest that the highest average value does not mean the strategy is
optimal in all situations. Therefore Table 19 indicates the percentage of pseudo
observations in which a certain strategy is optimal. For land-constrained farmers, crop
rotation is the optimal solution in 69% of the cases, while Bt maize is the best strategy
in 28% of the cases, leaving the chemical options only optimal for a marginal number
126
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
of farmers. For farmers committed to continuous maize, Bt maize is optimal in 78%
of the cases followed by soil insecticides in 15% and seed treatment in the remaining
7%.
In silage maize production, the average benefits are lower due to the lower value
of the crop. Bt maize now offers the highest average protection and creates a value of
€47/ha, followed by crop rotation and the chemical options. Interesting is that both
soil insecticides and seed treatment on average seem not to generate benefits. Looking
into detail in the pay-off, the model shows that seed treatment would be preferred to
no control in 26% of the cases, while soil insecticides do slightly better with 34% of
the cases. For land-constrained silage farmers, Bt maize is optimal in 55% of the
cases, crop rotation in 36% of the cases, while for 8% of the farmers it is optimal not
to engage in any CP measure as the value of the damage is too small. In continuous
maize, the dominance of Bt maize increases to 87%.
Table 19: Distribution of optimal crop protection strategies against WCR in the
heterogeneous farmer population of Hungary
Risk neutral farmers
Grain
Continuous
78%
Landconstrained
36%
1%
7%
2%
7%
Soil insecticide
1%
15%
3%
12%
Crop rotation
69%
n.a.
60%
n.a.
No control
0%
0%
0%
0%
Bt maize
55%
87%
61%
90%
Seed treatment
0%
1%
0%
1%
Soil insecticide
0%
3%
0%
2%
Crop rotation
36%
n.a.
29%
n.a.
No control
Bt maize
Seed treatment
Landconstrained
28%
Continuous
CARA farmers
81%
Silage
8%
9%
11%
7%
Grain
(Bt maize not commercialized)
Seed treatment
8%
46%
7%
47%
Soil insecticide
3%
54%
4%
53%
Crop rotation
89%
n.a.
89%
n.a.
0%
13%
0%
11%
Soil insecticide
0%
28%
0%
25%
Crop rotation
67%
n.a.
62%
n.a.
No control
33%
59%
38%
64%
Silage
(Bt maize not commercialized)
Seed treatment
n.a.: not applicable
127
As Bt maize is not yet deregulated in the European Union, the results presented above
do not reflect the actual situation. Therefore in Table 19, the distribution of CP
strategies among the heterogeneous population of farmers is shown in the absence of
the Bt technology for grain maize. In that scenario, land-constrained farmers would
opt for crop rotation in 89% of the cases, followed by seed treatment in 8% and soil
insecticides in 3% of the cases. In monoculture, soil insecticides become the dominant
strategy with 54% followed by seed treatment with 46%. If we compare these values
with the results of the survey in Chapter 6 on the actual practices in Hungary, these
figures seem to be close to the actual situation, as 59% of the farmers stated they used
soil insecticides and 44% used a seed treatment. Land-constrained silage maize
farmers will either apply crop rotation, 67%, or no control at all. In a situation of
silage under monoculture, 13% of the farmers would opt for seed treatment, 28% for
soil insecticides while the remainder would not apply any CP strategies. The
discrepancy between the outcome with Bt maize deregulated and the actual situation
suggests that the deregulated status of Bt maize comes at a cost (benefits foregone) to
those farmers having to use the less efficient technologies available. These benefits
foregone can be calculated as the added value from adopting Bt maize and shown in
Table 18. Land-constrained farmers on average forego €16/ha for grain maize and
€26/ha for silage maize. The latter is higher as the technology fee is tailored to the
particularities of the grain maize market, creating an added value which is higher in
the silage maize market. For farmers engaging in monoculture, this value increases to
€37/ha and €49/ha respectively.
To farmers the absolute value may not be the final parameter in the decision
whether or not a CP strategy is optimal. For risk averse farmers, the variability in the
value created plays an important role. This variability is shown in Table 18 by the
coefficient of variation and used to both calculate the risk premium and the CE. The
low variability in the value created by Bt maize makes it attractive for the moderately
risk averse farmer. For grain maize this translates into increased rationale for Bt
maize, now the optimal CP strategy for land-constrained farmers in 36% of the cases
(Table 19). The application range of crop rotation is affected the most, only optimal in
60% of the cases. The choice for soil insecticides and seed treatments reduces to 2%
and 3% respectively. In continuous grain maize, Bt maize would be the optimal
choice in 81% of the cases for risk averse farmers. Similar results for other situations
can be found in Table 19.
128
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
As the results indicate, there is no single dominant CP strategy. Therefore farmers
need a set of criteria to determine the optimal CP strategy given their own constraints.
Up to this point no insight is provided in these decision criteria. Literature has not
addressed this issue up to now. But also in this study we either implicitly assumed to
have only knowledge about the average value in Hungary, or full information in order
to determine the potential market share for each optimal CP strategy. The reality for
the farmer is somewhere in between. Farmers have to operate under imperfect
information as yield, prices and damages are not known at the time of decision
making, but farmers possess private information which differentiates them from the
other farmers in the population. This private information may aid farmers in their
decision if the factors determining the value of a certain CP strategy are known. To
determine these factors and their influence on the value creation, a sensitivity analysis
is run through a normalized stepwise linear regression (R2 always higher than 0.9).
The results can be seen in Figure 9 through Figure 11. As the regression is
normalized, a coefficient of 1 means an increase of one standard deviation increases
the dependent variable equally with 1 standard deviation. Figure 9 shows the
significant coefficients for the value created by Bt maize. As expected the higher the
root rating in the no control scenario, the higher the value of the protection by Bt
maize. A higher yield and higher maize price also increase the value of the Bt maize.
On the other hand an increase in the technology fee or a decrease in the efficiency of
Bt maize decreases the value. The other individual values follow the same reasoning.
In the farmer’s adoption decision the change in the mutual relationship of CP
strategies is of higher interest. Figure 10 shows the determinants of the added value
created by Bt maize. The negative influence of the root rating under no control is
interesting. Often Bt maize is believed to be the best option under high pest
population pressure. However as the efficiency of Bt maize is not 100% and the
refuge area suffers from WCR losses under high pressure, a high population pressure
leads to an increase in the competitiveness of crop rotation. The same reasoning
explains the negative coefficients on the grain price and yield. However, farmers with
a high gross margin in maize production prefer Bt maize. This can be explained by the
fact that gross margins are determined by the price of both inputs and outputs. The
framework allows for this as the gross margin of maize production is introduced
exogenously. Finally consider the tradeoff between seed treatments and soil
insecticides as this reflects the actual situation in Hungarian continuous maize fields
129
(Figure 11). The major determinant is the efficiency in protecting maize from WCR
damage. Furthermore, high yields and high maize prices increase the value of soil
insecticides as the potential monetary loss increases and the higher price of soil
insecticides is justified.
Figure 9: Normalized regression coefficient of Bt maize value
Figure 10: Normalized regression coefficients of added value from Bt maize with
respect to the next best option
130
Chapter 7. On the Competitiveness of WCR Control Options in Hungary
Figure 11: Normalized regression coefficients of the difference in value between
seed treatment over soil insecticides
The results in the above paragraphs can be used to develop a multi-criteria decision
framework to aid the adoption decision of CP strategies while decreasing the
dependence on inefficient scouting techniques and uniform economic thresholds. At
the same time the results can help to anticipate in the changes in the market for CP
technologies under changing prices and policies.
7.4 Discussion
The results of the bio-economic model demonstrate that the value of the potential
WCR damage in the Hungarian maize sector is high, with average potential losses for
grain maize of €176/ha under no control. Therefore applying a CP strategy would be
rational for all farmers cultivating grain maize. With the profit maximizing corporate
pricing strategy resulting in a tech fee of €23/ha, lower value silage maize, 8% should
not apply any CP strategies as they are too expensive. Crop rotation offers the highest
average value of protection in grain maize while in silage maize Bt maize offers the
highest value. However due to the variability in Hungarian maize production they are
not optimal for all farmers. In the actual situation where Bt maize is not deregulated
crop rotation is optimal in 89% of the cases for land-constrained grain maize farmers,
followed by seed treatment and soil insecticides. For those farmers engaging in
monoculture, excluding the option of crop rotation, soil insecticides are optimal in
54% of the cases.
131
With the hypothetical introduction of Bt maize resistant against WCR damage, the
situation changes significantly. Bt maize offers a higher average value than chemical
options and in certain situations outperforms crop rotation as a CP strategy in grain
maize production. In silage maize, the average value from Bt maize is higher than any
other CP strategy. In monoculture Bt maize would be rational for 78% of the grain
maize farmers and 87% in silage maize. This comes mainly at the expense of soil
insecticides and seed treatment as their potential market share decreases significantly.
The non-commercialization of Bt maize in Hungary comes at a cost as farmers and
innovators forego the possible benefits from adopting, ranging from €16/ha for landconstrained grain farmers to €49/ha for silage maize under monoculture and €23/ha
for the innovator
The model results determine the factors influencing the performance of the
different CP strategies. This knowledge can be exploited by the farmer’s knowledge
of his farming practices in order to determine the optimal CP strategy for his
particular situation. The same results could be used in a multi-criteria decision support
tool for farmers. This would yield more appropriate CP strategies than simple
scouting techniques, threshold values or prophylactic applications.
Comparing the results of this chapter with the WTP for WCR resistant maize
varieties in Chapter 6 gives some insight in the non-pecuniary value of the
technology. While the pure bio-economic assessment yields an added value of the
technology in grain maize between €16/ha and €49/ha the WTP analysis resulted in a
technology valuation of €70/ha. Although calculating an average of the monetary
value in Hungary is impossible without detailed information on the share of the
different type of production systems, it is fair to assess the non-pecuniary benefits
introduced earlier in the range of €21-€54/ha. Alston et al. (2002) in their assessment
get a lower estimation of 12$/ha in the USA setting. The only study we know of
indicating a measure of WTP for non-pecuniary benefits of insect control options in
Hungary (or neighboring countries) reports values €59-99/ha (Demont et al., 2005a)
which is in line with our results.
A more general discussion of the results of the bio-economic model can be found
in Chapter 8 where we apply the model to seven other Central European countries
giving a deeper insight in the results.
132
Chapter 8. The Western Corn Rootworm, a New Threat to
European Agriculture: Opportunities for Biotechnology?
Adapted from Dillen, Mitchell, Van Looy &Tollens,
Pest Management Science, in press
During the early 1990s, the Western Corn Rootworm, a maize pest, invaded the
European continent. The continuous spread of the pest has introduced a new constraint
into European maize production. As the damage caused by the invasive species is highly
variable and different crop protection (CP) strategies are available, farmers’ optimal
strategies are not obvious. This study uses a simulation model to assess the
competitiveness of different CP strategies in seven Central European countries.
Results indicate a high degree of heterogeneity in the profitability of different CP
strategies depending on the production parameters in each country. In general, crop
rotation and Bt maize offer the best solution to farmers, but in continuous (non-rotated)
maize cultivation, chemical CP options may capture part of the market. For Austrian
continuous maize production, we find that not deregulating Bt maize implies that farmers
forego revenues of up to 59€/ha.
In the presence of WCR, producing maize in an economically sound method requires
incorporating country and farm specific characteristics into the decision framework.
Also, not deregulating Bt maize has direct monetary consequences for many farmers that
could influence total maize output and resistance management.
8.1 Introduction
The threats of invasive species to agriculture are well documented and studied
(Pimentel et al., 2001) An introduced exotic species can induce significant economic
losses by imposing an additional constraint on crop production, so that
the
introduction and establishment of an invasive pest should also be understood in the
socio-economic context of crop production (Beckmann and Wesseler, 2003; Boriani
et al., 2006). This chapter focuses on a recently introduced invasive species, WCR,
threatening European maize production. The species, first introduced into Europe in
the early 1990s from the USA, now has a continuous population in 11 European
countries (Figure 5). Predicting and assessing the yield loss caused by the pest
damage is difficult, as exogenous factors such as water availability and soil conditions
are important (see Chapter 7). Therefore farmers face difficulties in designing
economically optimal crop protection (CP) strategies. Some studies assess part of the
economic puzzle surrounding WCR in Europe, mainly estimating potential damages
and the effect of different CP and containment strategies. Chapter 7 highlights these
studies and evaluates the on-farm competitiveness of different CP strategies in
Hungary based on a stochastic model explicitly incorporating the uncertainty and
heterogeneity of damages and losses among maize farmers. Results show that the
competitiveness and suitability of a CP option depend on several specific criteria, not
only on a single economic threshold value as proposed in earlier literature (Foster et
al., 1986). This chapter applies the WCR bio-economic model to seven infested
countries not previously examined for WCR economics: Austria, Czech Republic,
Poland, Romania, Slovakia, Serbia and Ukraine. It also expands the discussion of the
modeling approach in a more qualitative way and the problem of WCR in Europe.
The insights gained from the analysis can be used by both farmers and policy makers
to design appropriate practices and to minimize the impact of the invasive species on
European agriculture.
The remainder of the chapter is structured as follows. The next section provides an
overview of maize production in each nation and documents the status of WCR.
Section 8.3 introduces the important concepts of the stochastic bio-economic model
and the data used, while in the last two sections, model results are presented and
discussed.
134
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology?
8.2 Maize Production and WCR in each Country 29
8.2.1 Austria
The area cultivated with grain maize has been increasing steadily over the last 5 years
in Austria, leading to an all time high of 194 000 ha in 2008 (AGES, 2008). In
addition, Austrian farmers annually cultivate 80 000 ha of silage maize. Combined,
maize covers about 22% of the available arable land in Austria and has consistently
higher yields than in the other countries discussed in this study, indicating intensive
production. Indeed, because of spatial constraints such as elevation, the cultivation of
maize is predominantly concentrated in the southeastern part of the country bordering
Slovenia and Hungary. According to the Austrian Agency for Health and Food Safety,
the area under continuous maize ranges from 0% in the low density areas to 25% in
high density areas (AGES, 2008). Two types of farms are likely to have continuous
maize—small farms with animal husbandry and farms with specialized irrigation
systems.
WCR was initially detected near the borders with Hungary and Slovakia during
the summer of 2002. By 2008, the pest was present in 65% of the Austrian maize
area. Despite the presence of WCR in most maize fields, no economic damage has
been officially reported (AGES, 2008).
8.2.2 Czech Republic
The Czech Republic produces both grain and silage maize. Although the area sown
with silage maize has been decreasing in recent years, it still dominates, constituting
63% of the cultivated maize area in 2008 (Eurostat, 2009; Trnka et al., 2007). With
287 900 ha sown, maize ranks fourth in terms of cultivated area in the country, behind
soft wheat, barley and rape seed, with half of this production concentrated in Central
Czech and South Moravia. WCR entered into the Czech Republic in 2002 via the
southeastern border. Pheromone trap data reveal an annual increase of the WCR
29
Very little peer reviewed data on maize production and WCR damage exist for the countries under research,
so we must rely on government data, unpublished information and personal communication. Part of our goal is to
summarize this information, as such a summary does not exist, and to keep the sources and process as transparent
as possible so that interested readers can consult the same sources. Some further data is detailed in (Dillen et al.,
2009c).
135
population and by 2008, the beetle had spread to all major maize producing areas. In
the Czech Republic, South Moravia is the area with the highest density of maize
cultivation and the longest established WCR populations, and so this area is under the
most acute danger of economic loss, but none had been reported before 2009 (SPA,
2009).
8.2.3 Poland
Grain maize is a relatively minor commodity in Polish agriculture, the dominant crops
being soft wheat, barley, rye and triticale. In 2008, 317 000 ha of grain maize and
418 000 ha of silage maize were harvested in Poland, or about 5% of the total arable
land (Eurostat, 2009). Despite the low importance of grain maize in Poland, it is still
the 6th largest producer of grain maize in Europe, with 3.5% of the total sown area.
The main maize producing areas within Poland are found in the central provinces of
Dolnoslaskie, Wielkopolski and Podlaski, representing 38% of the sown maize area
and almost half of the grain maize production (Polish Statistics, 2008). Within this
area, continuous maize is a common practice and nationally about 23% of all maize is
grown in continuous cultivation (European Commission, 2006b). In Poland, 56% of
livestock are located on farms with less than 20 ha of cultivated land, while 59% of
cultivated grain maize is on farms with livestock. Therefore, continuous maize is
predominantly found on small farms that need maize as a source of feed and have low
flexibility due to a land constraint.
WCR was first detected in 2005 near an international road leading to Slovakia,
with the nearest maize field situated 6 km away (EPPO, 2008). In 2006 and 2007, the
beetle spread throughout the whole southern part of Poland, with findings in 9
provinces. Before 2007, beetles where only found in trapping devices and neither
larvae nor economic damage were detected. However, because of the high percentage
of continuous maize in the area of infestation, economic damage is an impending
threat. As the beetle spreads towards the central regions of Poland, the most important
maize growing regions will likely be infested within the next few years.
8.2.4 Romania
The political transition process in Romania focused on privatizing land ownership and
downsizing farms to change collective agriculture to individual agriculture (Lerman et
al., 2007). The farmer response to this process was to shift cultivation to maize, as the
136
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology?
crop is generally easy to cultivate and uses relatively few inputs (Balint and Sauer,
2006). As a result, 2.5 million hectares were sown with maize in 2008, which ranks
second in the European Union. Within Romania, maize occupies on average about
30% of total arable land, but in the eastern half of the country, which represents about
60% of the area sown with maize, this reaches 42% (Eurostat, 2009; National Institute
for Statistics Romania, 2009). In 2006, about 50% of the sown area was in continuous
maize (Rosca, 2006).
WCR was first reported in Romania in 1996 near the Hungarian border. Since that
time, WCR has spread rapidly due to the high percentage of land planted to maize in
(EPPO, 2008). In 2006, about half of the country was infested and preventing spread
to the rest of Romania was difficult, with a front line over 400 km wide and the most
densely planted areas in the east not yet infested. Despite these favorable settings,
official data show that economic damage can only be found on a few square
kilometers in south west Romania (Rosca, 2006).
8.2.5 Serbia
Grain maize is by far the most important crop in Serbia. With an average area of 1.2
million hectares sown annually, it occupies 38% of the total arable land in Serbia
(Statistical Office of the Republic of Serbia, 2007). The main maize growing areas are
located in river valleys and in the northern part of the country, with continuous maize
production common. A recent survey found that maize was planted for more than one
year on the same plot for 48% to 53% of the fields studied (Sivcev et al., 2009).
However, in some communities the share of continuous maize was 80% before the
invasion of WCR (Sivcev, 2008).
Serbia was the location of the first successful invasion by WCR in Europe (Baca,
1993). The first economic damage occurred in 1992, though the actual introduction
likely took place earlier but was not documented. The high percentage of maize under
continuous cultivation created ideal conditions for rapid population growth and the
pest soon covered the whole region. In 2003 economic damage was recorded on 3000
ha, but since that time, damage has been sporadic and spatially fragmented, averaging
1000 ha annually due to changing production systems and crop protection practices.
In 2009, Sivcev et al. (2009) reported 12% of the fields under research had more than
6 beetles/trap/day, which is considered the economic threshold value. As sampling
was not representative, we cannot extrapolate this value to the rest of the country.
137
Serbian farms are small, averaging 2.49 ha of cultivated land per holding, which
allows for little flexibility in the production system (Statistical Office of the Republic
of Serbia, 2007). Maize production is especially prevalent on the smaller farms, as
maize only accounts for 10% of arable land on larger farms. Few farmers use
insecticides to manage WCR damage. The preferred CP option of Serbian farmers
seems to be crop rotation, as it is promoted by most extension services in the region.
This shift away from continuous cultivation due to WCR is suggested as one of the
causes for the 236 000 ha decrease in the sown area between 1991 and 2008
(Statistical Office of the Republic of Serbia, 2007).
8.2.6 Slovakia
In Slovakia, maize is the third most extensively planted crop grown, with 141 000 ha
of grain maize and 74 000 ha of silage maize sown in 2008, only exceeded by winter
wheat and barley (Eurostat, 2009). Silage maize is predominantly used for on-farm
purposes—on average, 88% of silage maize was used for fodder production in 2006
and 2007. The main grain maize producing region is located in the southern parts of
Slovakia, bordering Hungary. Within this region, maize cultivation is intensive, often
irrigated, with a significant portion under continuous cultivation. The area under
continuous maize cultivation fluctuates annually, but over the period 2004-2007, 16%
of the maize was on average cultivated under continuous cultivation (Cagan, 2008).
WCR was first detected in Slovakia in 2000, reported in three districts bordering
Hungary. The first economic damage occurred in 2004 on about 340 ha and by 2005,
7% of the maize area had a beetle population exceeding the economic threshold and
2.66% (6419 ha) of the area had economic larval damage. The area with economic
damage has continued to increase despite the use of CP options and by 2007,
economic larval damage occurred on 4.86% of the sown area, with yield losses in
continuous maize reaching 60% in some cases (Cagan, 2008).
8.2.7 Ukraine
As the potential bread basket of Europe, maize is one of the most important grain and
forage crops in Ukraine, with about 2 million hectares planted to grain maize
annually, mainly in eastern and southern Ukraine (FAO, 2007). Due to the ease of
cultivation and its high feeding value, maize is disproportionally popular among
household farms, as they have less flexibility in the crop rotation. Nationally, 26% of
138
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology?
private farmers have difficulties using crop rotation, but in some areas this constraint
applies to almost half of the farmers (Lerman et al., 2007). It is important to note that
Ukrainian farmers are characterized by a high degree of heterogeneity as a result of
the differences in resources during the recent transition period (Zorya and von
Cramon-Taubadel, 2006). Continuous maize growing averages 10-12% of the sown
area nationally, but is significantly higher in the intensive production regions.
In August 2001, WCR was detected west of the Carpathian Mountains near the
Hungarian and Romanian borders. Since that time, the beetle has spread throughout
the southeastern part of Ukraine bordering the infested zones in Romania, Hungary,
Slovakia and Poland. No economic damage has been reported yet, but the WCR
population is steadily increasing and, because continuous maize is a common practice
in the infested zone (25% of the sown maize area), damage is expected in years to
come. In 2008, 15 000 ha of arable land (30% of it planted with maize) were officially
under quarantine.
8.3 Bio-economic Model
WCR is not yet fully established in the countries examined here, nor are all CP
options deregulated and commercially available. Therefore, we use a bio-economic
model as an ex ante impact assessment. For this analysis, we use partial budgeting to
estimate the net increase in farmer returns per hectare due to adoption of a specific CP
option relative to the hypothetical base case of an established WCR population and no
CP options used. We use a stochastic model to account for some of the uncertainties
faced by farmers making their CP choices.
The basic model structure for farmer returns per hectare with each CP option is
the product of the maize price and maize yield, minus the cost of production and the
cost of the CP option, with the price, yield and costs potentially varying for each CP
option. For example, for Bt maize, the maize price includes a discount because the
maize is GM and costs include required refuge/co-existence measures. As discussed
in Chapter 7, a recent stream of literature has focused on developing partial budgeting
methods to avoid biases arising from earlier methodologies. An important aspect of
this literature is the use of probability distributions for various economic and
biological parameters and Monte Carlo simulations to account for the uncertainty
inherent in these parameters and for the heterogeneity among farmers and across
seasons. As explained by Demont et al. (2008a), explicitly accounting for
139
heterogeneity in this manner avoids the homogeneity bias introduced by using only
parameter means. The bio-economic model used here is described in detail in Chapter
7, but we highlight some of the specific properties more qualitatively here to deepen
the understanding of the model and its consequences.
Root ratings and yield loss. Uncertainty exists for both the damage caused by the
WCR and the consistency and effectiveness of each CP option, so the model uses
available information to construct prior probability density functions (PDFs) used for
Monte Carlo simulations. To estimate WCR damage to maize and the effectiveness of
different CP options, the model follows a specific procedure. For each iteration, the
root rating under the hypothetical no control scenario is drawn from a beta PDF with a
minimum of 1 and a maximum of 6. A root rating for each CP option is drawn from a
beta PDF with parameters that depend on the drawn root rating without control (see
Chapter 7 for the full set of equations). Finally, the difference in the root ratings
between the no control case and each CP option is used to generate the percentage
yield difference between the no control case and each CP option. Following Mitchell
et al. (2004), the percentage yield difference is 11.4% of the difference in the root
rating between the no control case and each CP option. This calculated percentage
yield difference for each iteration is then used in the economic model.
WCR damage and pest population data are limited in Europe, so the conditional
PDF of the root rating is based on USA data. Results from field trials by land grant
universities in Illinois, Nebraska and Ohio are complemented by data from Monsanto
(Meinke et al., 1998; Mitchell et al., 2004). This assumption implies that the effect of
WCR is similar in Europe and the USA, although production systems differ
significantly (Kiss et al., 2005). It is important that European data be generated to
improve future assessments (Dun et al., 2010).
Prices and yields. Farmers never have certainty about prices and yields, even
though the time at which decisions are made differ for the CP options (soil insecticide
at sowing, seed treatment at buying and Bt maize even earlier as notification to
regulatory agencies may be required). To capture this uncertainty and to reflect the
heterogeneity in farming practices within each country, PDFs are also used for many
of the prices and yields as part of the Monte Carlo simulation. This uncertainty and
heterogeneity stems from differences in soil, climatic and managerial conditions and
has important consequences for an ex ante impact assessment. A few price parameters
are fairly similar across the countries examined here (e.g., the cost of seed treatments,
140
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology?
soil insecticides, and Bt maize), and so have values that do not vary across countries,
though the values can be random for each iteration. Table 20 summarizes the
distributions used for each parameter, or constant values for those parameters not
drawn from distributions, and the source of these distributions or values. However,
many price and yield parameters vary among the countries and so require countryspecific parameters, which are reported in Table 21. For example, grain and silage
yields in each country are drawn from beta PDFs and prices from lognormal PDFs,
but with different parameters for each country (Table 21). PDFs for gross margins for
maize and key rotational crops in each nation are the net return (revenue minus costs)
under the no control case.
Table 20: Probability density distributions and deterministic values for price and
yield parameters constant that are considered constant across regions
Parameter
Value
Source
Cost seed treatment (€/ha)
Uniform(37.3;47.4)
Cost soil insecticide (€/ha)
62
a
Hatala Zsellér et al. (2006)
“
b
Bt maize price discount (%)
Triangular(0;2;3)
Assumption
Additional cost Bt maize (€/ha)
Triangular(18;23;28)b
Chapter 7
Refuge (% sown area)
20%
Maize price grain-yield correlation
-0.2
AGBIOS (2008)
Assumption based on Goodwin
(2009)
a
b
Numbers are the minimum and maximum of the distribution.
Numbers are the minimum, mode, and maximum of the distribution.
Price and yield data for this study were gathered from several sources. For
transparency and consistency, we kept the number of sources limited, but verified
them with national experts. We use prices for the different CP options from the
Hungarian market because Hungary is an important maize producer in the region, is
fully infested with WCR and has a commercial market for most CP options.
Moreover, some countries examined here do not allow active ingredients such as
tefluthrin for use in maize production (PPDB (http://www.eu-footprint.org)). These
assumptions are in line with earlier work (Takacs et al., 2007; Wesseler and Fall,
2010). Based on Chapter 7, we assume a technology fee for Bt maize between €18/ha
and €28/ha, with an average equal to the optimal technology fee calculated of €23/ha.
This assumption implies that there is no scope for price discrimination between
countries, which coincides with the assumption of equal pest pressure in our
counterfactual. Also, we introduce a negative correlation between the yield and maize
prices as a higher supply likely decreases local maize prices (Table 20). Lastly, as
141
explained in the model section, we use Monte Carlo simulations to obtain robust
results, to capture second order statistics and to facilitate sensitivity analysis, using
@Risk from the Palisade Corporation with 10 000 iterations.
Quality of information. As previously discussed, many of the parameters are not
deterministically known. We try to incorporate this uncertainty using observable
historical data, expert opinions, and similar information, but nevertheless, must
operate under imperfect data. However, farmers may possess private information
about these parameters, not observable to researchers. This private information will
allow the farmer to extract more information than researchers from the posterior
distribution constructed through the Monte Carlo simulation. Therefore, we present
three different types of results. First, we report the average and coefficient of variation
of the net benefit for each CP option for farmers in each country. Second, potential
adoption patterns for each CP option in each country are presented under the
assumption that farmers have perfect information. Finally, regression-based
sensitivity analysis shows the relative contribution of the various factors underlying
our results and the effect of private farmer information on the competitiveness of the
different CP options.
142
Table 21: Country-specific distributions* and parameters for bio-economic model
Austria
source
Czech Republic
a
source
Poland
source
Eurostat (2009)
PERT(0; 6.8; 9.1)
Eurostat (2009)
PERT(0; 6.205; 8.27)
Eurostat (2009)
Silage yield (t/ha)
PERT(0;
10.7566;
14.342)a
PERT(0; 46.2; 65.3)a
Eurostat (2009)
PERT(0; 38.3; 51.1)a
Eurostat (2009)
Eurostat (2009)
Price grain (€/t)
Lognorm(127.2;49.76)
Lognorm(150.7;43.1)b
VUZE
Price silage (€/t)
Lognorm(25.1;5)
Landwirtschaftskamm
er
Fall et al. (2008)
PERT(0;
47.91;
63.88)
Lognorm(102;7.7)
Lognorm(18.0;2.9)b
Fall et al. (2008)
Lognorm(12.5;2.862)
Fall et al. (2008)
Grain yield (t/ha)
c
Eurostat (2009)
Gross margin maize (€/ha)
PERT(-83;69;216)
Brookes (2008)
Uniform(15;108)
Brookes (2008)
PERT(363;489;626)
Brookes (2008)
Gross margin wheat (€/ha)
PERT(-171;-165;-147)
Brookes (2008)
Uniform (10;48)c
Brookes (2008)
PERT(186;229;240)
Brookes (2008)
Brookes (2008)
PERT(118;131;172)
Brookes (2008)
PERT(268;277;316)
Brookes (2008)
Gross margin barley (€/ha)
PERT(-271;-269;-260)
Gross margin oilseed rape
(€/ha)
Exchange rate (€1)
n.a.
Serbia
PERT(0; 5.9; 8.85)
Brookes (2008)
Uniform(16;51)
n.a.
n.a.
source
c
28.32
Slovakia
a
source
Romania
source
Ukraine
source
Statistics
Office of
Serbia
PERT(0;8.3;11.1)
Eurostat (2009)
PERT(0; 3.9; 5.85
Eurostat
(2009)
PERT(0; 3.7; 5.55)
FAO (2009)
PERT(0;27.6;36.8)a
Eurostat (2009)
PERT(0; 19.2; 28.8)
PERT(0; 12; 18)
FAO (2009)
Statistics
Office of
Serbia
Lognorm(89;4.8)
Statistical office
Slovakia + Eurostat
(2009)
Fall et al. (2008)
Lognorm(149.7;49.2)
Eurostat
(2009)
Eurostat
(2009)
Lognorm(55;8.08)
FAO (2009)
Lognorm(9.1;3)
PERT(170;340;510)
FAO(2004)
PERT(260.25;347;433.75) Brookes (2008)
PERT(156;303;450)
PERT(104;208;312)
Schaafsma(1
999)
FAO(2005)
Soya= PERT(110.5;221;331.5)
FAO(2004)
PERT(81;108;135)
Brookes (2008)
PERT(58;116;174)
PERT(87.5;175;262.5)
FAO(2005)
Sunflower=PERT(148;296;444)
FAO(2004)
PERT(60.75;81;101.25)
Brookes (2008)
PERT(32;64;96)
Schaafsma(1
999)
Brookes
(2008)
Brookes
(2008)
Brookes
(2008)
PERT(63.25;126;189.25
)
n.a.
FAO(2005)
n.a.
Lognorm(87.76;30.99
n.a.
Lognorm(28.4;4.8)
Lognorm(9.1;4.5)
n.a.
n.a.
n.a.
82
38.9
3.38
Unspecified sources are unpublished data and personal communications from 2008.
*
PERT is the beta-PERT distribution and Lognorm is the lognormal distribution. a Numbers are the minimum, most likely, and maximum of the distribution.
standard deviation of the distribution c Numbers are the minimum and maximum of the distribution.
7.79
b
Numbers are the mean and
8.4 Results
For each CP option examined, Table 22 reports the average net benefit (€/ha) and
coefficient of variation (CV), where the net benefit is the increase in farmer returns
from applying the CP method compared to the hypothetical situation of no WCR
control. Results vary significantly depending on the local situation. In most countries,
Bt maize offers the highest average benefit, ranging from 0.4 €/ha in Ukraine to 100.7
€/ha in Austria. In two countries, Czech Republic and Serbia, crop rotation performs
best, with an average net benefit of 136.4 €/ha and 25.9 €/ha, respectively. Chemical
CP options on average generate benefits with substantially lower means and higher
variability as the coefficients of variation show. This relatively higher variability
would decrease the value further for risk averse farmers. For silage maize, the results
are similar. Czech Republic generates the highest average benefits from crop rotation,
95 €/ha, while Bt maize dominates in the other countries, reaching an average net
benefit of 104 €/ha in Austria. In Ukraine, none of the CP options generates an
average net benefit higher than the no control case, implying that farmers on average
would earn higher returns using no control and accepting lower yields rather than
using any of the CP options. Again, the chemical CP options generate significantly
lower average benefits in all countries.
Note that the average net benefit does not necessarily equal the actual benefit
earned by an individual farmer. Because of heterogeneity among farmers, net returns
and the relationships between the different CP options differ, so that in the same
country, some farmers will earn higher net benefits with one CP option, while other
farmers will earn higher net benefits with a different CP option. Thus, for each CP
option, Table 23 reports the percentage of the iterations for which the benefit was the
largest relative to all other CP options. These percentages can be interpreted as the
potential adoption rate of each CP option, assuming farmers have perfect information
regarding their prices, yields and losses from WCR. Thus, in the Czech Republic,
where average benefits of crop rotation are high, 91% of the land constrained grain
maize farmers would likely apply crop rotation, while only 7% would likely opt for Bt
maize, while the remaining farmers would be rational in not applying any control.
However, for farmers in the Czech Republic obliged to cultivate maize in a
continuous system, results in Table 23 suggest that 81% of the grain farmers would
choose Bt maize, 10% would choose soil insecticides and 2% would choose seed
144
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology?
treatments and the group of farmers not using a CP option increases to 7%. On the
opposite end of the spectrum, Table 4 suggests that 81% of Slovakian grain farmers
would choose Bt maize, only 2% would engage in crop rotation, and 11% would not
apply any CP strategy. As crop rotation has a limited scope in Slovakia, few
continuous maize producers would likely adopt crop rotation, rather results imply that
83% would likely turn to Bt maize. In Austria, results suggest that soil insecticides are
preferred by 10% of the land constrained farmers and 13% for continuous maize
producers, the highest share in any of the countries and as high as crop rotation. Seed
treatments will likely have the highest potential in Romania, where results show that
16% of the farmers, both land constrained and continuous maize producers, would
likely apply seed treatments. In Serbia crop rotation and Bt maize will likely be
preferred by a similar share of the farmers, with 46% and 38% respectively.
Interestingly, in Ukraine 41% of the land constrained farmers and 58% of the farmers
with continuous maize cultivation would likely choose not to adopt any CP option.
For the remaining countries and the analysis of adoption in silage maize, the results in
Table 23 are similar.
145
Table 22: Mean and coefficient of variation (CV) of net benefit and surplus created by Bt maize (€/ha) in each country for each
Western Corn Rootworm crop protection option
Austria
Grain
Czech Republic
Poland
Serbia
Slovakia
Romania
Ukraine
Net Benefit
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Bt maize
100.7
1.0
71.9
1.0
37.8
1.1
28.2
1.3
47.2
1.0
99.8
0.9
0.4
33.8
Seed treatment
22.0
3.0
5.2
8.6
-13.9
-1.9
-19.4
-1.3
-8.8
-3.5
20.9
3.0
-35.1
-0.3
Soil insecticide
48.3
2.1
20.4
3.5
-13.7
-3.0
-23.3
-1.7
-4.7
-10.2
48.2
2.0
-50.9
-0.3
Crop rotation
16.4
9.6
136.4
0.8
-89.3
-0.8
25.9
3.0
-55.4
-1.3
62.9
2.4
-6.2
-5.9
Land constrained
48.3
0.6
10.5
0.7
44.4
0.5
30.6
0.6
45.0
0.5
33.1
0.6
23.4
0.6
Continuous
58.5
0.7
52.0
0.6
45.5
0.5
44.8
0.4
47.2
0.5
36.5
0.7
40.5
0.3
Bt maize
103.9
0.8
54.9
0.9
46.9
0.9
----
----
34.1
1.1
103.7
0.8
-5.9
-1.5
Seed treatment
19.1
-3.1
-8.2
3.9
-13.3
2.1
----
----
-18.5
1.3
18.0
-3.1
-39.5
0.2
Soil insecticide
43.2
-2.0
-3.2
15.6
-12.3
3.6
----
----
-21.9
1.7
42.8
-2.0
-58.7
0.1
Crop rotation
7.3
18.8
95.2
0.8
-86.7
-0.9
----
----
-85.6
-0.7
54.3
2.5
-19.8
-1.6
Surplus created by Bt maize
Silage
Net Benefit
Surplus created by Bt maize
Land constrained
52.4
0.5
12.1
0.5
49.7
0.5
----
----
46.5
0.4
30.8
0.4
24.2
0.5
Continuous
61.7
0.6
51.3
0.5
51.0
0.6
----
----
47.3
0.4
31.4
0.4
44.6
0.4
Table 23: Adoption pattern for Western Corn Rootworm crop protection options implied by Monte Carlo simulations.
Crop Protection
Austria
Czech Republic
Poland
Serbia
Slovakia
Romania
Ukraine
LC
C
LC
C
LC
C
LC
C
LC
C
LC
C
LC
C
Bt maize
72%
79%
7%
81%
81%
81%
38%
74%
81%
83%
71%
72%
22%
42%
Seed treatment
2%
3%
0%
2%
1%
1%
0%
1%
1%
2%
16%
16%
0%
0%
Soil insecticide
10%
13%
0%
10%
3%
3%
1%
3%
5%
5%
0%
0%
0%
0%
Crop rotation
10%
n.a.
91%
n.a.
0%
n.a.
46%
n.a.
2%
n.a.
0%
n.a.
37%
n.a.
No control
5%
5%
2%
7%
14%
14%
15%
22%
11%
11%
12%
12%
41%
58%
Bt maize
79%
84%
13%
88%
89%
89%
n.a.
n.a.
83%
88%
18%
18%
12%
20%
Seed treatment
2%
2%
0%
1%
0%
2%
n.a.
n.a.
1%
1%
1%
1%
0%
0%
Soil insecticide
8%
10%
0%
4%
2%
0%
n.a.
n.a.
1%
4%
0%
0%
0%
0%
Crop rotation
8%
n.a.
85%
n.a.
1%
n.a.
n.a.
n.a.
1%
n.a.
0%
n.a.
26%
n.a.
No control
3%
3%
2%
7%
8%
8%
n.a.
n.a.
14%
7%
82%
82%
62%
80%
Seed treatment
13%
20%
1%
38%
11%
12%
1%
9%
12%
14%
46%
51%
42%
1%
Soil insecticide
36%
52%
0%
52%
23%
26%
3%
18%
27%
32%
0%
0%
0%
1%
Crop rotation
23%
n.a.
96%
n.a.
4%
n.a.
59%
n.a.
9%
n.a.
8%
n.a.
0%
n.a.
No control
28%
29%
3%
10%
62%
62%
38%
73%
52%
53%
46%
49%
58%
99%
Seed treatment
14%
20%
1%
46%
11%
12%
2%
14%
8%
9%
2%
2%
0%
0%
Soil insecticide
37%
51%
0%
48%
23%
26%
5%
34%
17%
19%
0%
0%
0%
0%
Crop rotation
23%
n.a.
96%
n.a.
5%
n.a.
71%
n.a.
3%
n.a.
0%
n.a.
28%
n.a.
No control
27%
28%
3%
6%
61%
62%
23%
52%
71%
72%
98%
98%
72%
100%
Grain
Silage
Grain (Bt maize not commercialized)
Silage (Bt maize not commercialized)
LC: Land Constrained, C: Continuous maize, n.a.: not applicable
Note that these results do not resemble the actual situation, since Bt maize has not
been deregulated in any of the countries examined here. Because Bt maize creates
large average benefits for farmers relative to the counterfactual no control case and is
the optimal CP option for many farmers, non-commercialization implies substantial
benefits not being captured by farmers. Hence Table 22 also reports the added benefit
of Bt maize relative to the next best option. Alternatively, this added benefit can be
interpreted as the benefits lost due to the non-commercialization of Bt maize. For land
constrained grain maize farmers in the Czech Republic, this lost benefit amounts to 11
€/ha, while in continuous maize, in the absence of crop rotation, the lost benefit
increases to 52 €/ha. For the other countries, the difference between continuous
cultivation and land constrained farmers is smaller. Also, these results show that the
value per adopted hectare is the highest in Austria’s intensive maize growing areas, at
almost 60 €/ha. As can be seen in Table 22, the results for silage maize are similar,
but slightly higher on average.
The unavailability of Bt maize also affects potential adoption shares of the other
CP options. As Table 23 shows, depending on the competitiveness of these CP
options, the share previously captured by Bt maize is divided differently. If crop
rotation is favorable, as in Czech Republic, almost all land constrained farmers (96%)
will likely adopt crop rotation. In other countries, the gap is mainly filled by the
chemical CP options. In Austria, 36% of the land constrained grain maize farmers
would probably adopt soil insecticides and 13% would likely adopt seed treatments,
while for specialized maize farmers with continuous maize, these adoption rates
would be 52% and 20% respectively. In Romania on the other hand, seed treatments
are the dominant chemical CP option, with adoption rates likely around 46% and 51%
in land constrained and continuous production systems respectively. In the absence of
Bt maize, results in Table 23 show that many farmers in some of these countries
would probably not apply any CP option. Under continuous maize cultivation, 53% of
famers would likely not use any control in Slovakia, while 62% would likely do so in
Poland, 73% in Serbia and 99% in the Ukraine. Again, results are very similar for
silage maize. This result occurs because the costs of chemical CP options are too high
compared to their yield benefit.
The Monte Carlo simulation model allows determination of the relative
contribution of the various factors underlying these results, since each iteration can be
treated as a “pseudo observation.” A regression-based sensitivity analysis, Figure 1
148
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology
shows the normalized regression coefficients for the factors influencing the average
net benefit of Bt grain maize. These normalized regression coefficients as calculated
by the @RISK software show the change in the net benefit of Bt maize if the
respective parameter is increased by one standard deviation. Results depend on the
specific situation in each country, but some general tendencies are apparent. For
example, if the effectiveness of Bt maize is low, the net benefit of Bt maize is reduced
significantly, as the relatively large negative results for “Damage Bt maize” indicate.
On the other hand, if chemical treatments suffer from poor performance, the
competitiveness of Bt maize increases, as the relatively large positive results for
“Damage soil insecticide’ indicate. A surprising result is the negative influence of
damage under the hypothetical no control scenario (“Damage counterfactual”)—the
greater the WCR damage under no control, the lower the net benefit of Bt maize. This
result occurs because of the policies we assumed if/when Bt maize is deregulated. We
anticipate that the area assigned to refuge and ex ante spatial coexistence measures
will be untreated for WCR, implying higher losses in these areas under higher
population pressure. Though losses on the Bt maize are hardly changed, famers
adopting Bt maize will suffer greater losses on refuge/coexistence areas, thus reducing
the overall value of Bt maize. Finally, the higher the gross margin for maize
production, the more farmers gain from Bt maize as they can use Bt maize to protect
their more valuable crop.
Damage Bt maize
1
Damage soil
Price seed
insecticide
treatment
0.5
Gross margin
winter wheat
0
Price maize
-0.5
Gross margin
barley
Czech
Slovakia
Austria
-1
Damage
counterfactual
Poland
Romania
Serbia
Technology fee
Gross margin
maize
Yield
Ukraine
Damage seed
treatment
Figure 12: Normalized regression coefficients of the factors influencing the value
added by Bt maize
149
8.5 Discussion
In Ukraine and Serbia, chemical CP options have very limited scope as their cost
outweighs the benefits in a low yielding environment, so that in practice farmer
choices are limited to Bt maize or crop rotation. Farmers earning high revenue from
maize will protect this revenue by using Bt maize, while other farmers will choose
crop rotation if feasible, and no CP option in most cases. In Ukraine, this effect is
augmented by the strong negative effect on the benefit of Bt maize from a slight
increase in the competitiveness of rotational crops, which is especially the case for
small mixed farms with flexibility constraints. This modeling result is consistent with
observations that Serbian farmers have not adopted chemical CP strategies for WCR
control, but typically use crop rotation to manage the pest (Sivcev, 2008).
In the maize producing areas of Poland and Slovakia, the cost of engaging in crop
rotation as a CP method is high because the revenue from producing maize is high
compared to other crops and is highly specialized. As a result, the chemical CP
options directly compete with Bt maize in the market. In Figure 12, Poland and
Slovakia exhibit high coefficient values on the axes for damage in soil insecticides
and seed treatments. If a farmer assesses a low efficacy for chemical CP due to his
farm characteristics, he will be inclined to protect his high revenues through Bt maize
and vice versa. For Austria, the highest yielding country in this study, similar
conclusions can be drawn. However, because the value of the maize is high, and
therefore the value of potential damage, it is mainly the more effective soil
insecticides that compete with Bt maize, and to a lesser extent seed treatments. For
Romania, a low yielding country with little scope for crop rotation due to the typical
farm structure, seed treatment is the best solution for many farmers as it offers some
protection at a lower price.
Results in Table 23 show that the scope for chemical CP options is limited in
markets where Bt maize is deregulated. In the actual situation with Bt maize not
commercially available in these countries, chemical CP options are economically
viable in most of these countries. The more expensive and more effective soil
insecticides will likely be popular in high yielding countries, while seed treatments
will be preferred by lower yielding operations. Not deregulating Bt maize in Europe
has important consequences besides the direct benefits foregone by producers
discussed earlier. First, the success of crop rotation as a CP option may reduce the
150
Chapter 8. Western Corn Rootworm: Opportunities for Biotechnology
total area sown with maize in a time when returns for coarse grains are expected to
remain high. This shift has been observed in Serbia since 1992. Second, the
availability of soil insecticides in the European Union is still under discussion due to
environmental considerations. In 2008, only one active ingredient, tefluthrin, was
deregulated for use in maize. This over-reliance on a single compound could lead to
the development of insecticide resistance among the WCR as has occurred in the USA
for some compounds (Meinke et al., 1998). Finally, a large number of farmers would
choose to not use any CP option and thus be exposed to increased risk and income
variability due to the pest. As crop rotation is mainly favored in low yielding, low
revenue maize production, this implies that small farmers risk losing substantial parts
of their harvest if WCR damage is unexpectedly high in a season. Bt maize would
offer a practical and economic sound tool to minimize the impact of WCR in
European maize production.
The information extracted from the sensitivity analysis can potentially aid farmers
in their decision-making process. A farmer can use his private information to position
himself relative to the population of farmers in his country. Comparing his own farmspecific characteristics with the overall farm population in his country and relating
them to the normalized regression coefficients in Figure 12 gives an indication on the
relative competitiveness of specific control options within his constraints. For
example, the large positive coefficient for “Gross margin maize” in Figure 1 for
Serbia indicates that a Serbian farmer with a gross margin for maize production
relatively higher than average in Serbia can conclude he is likely to gain from the
adoption of Bt maize. Similarly, the relatively high coefficients for “Price maize,”
“Yield” and “Damage soil insecticide” for Poland in Figure 12 imply that a Polish
farmer may want focus on his managerial capacities and application constraints for
soil insecticides, as these are the most important factors influencing his decisionmaking. In the future, a more detailed multi-dimensional framework could assist
farmers in designing optimal CP options given their specific constraints, instead of
relying on arbitrary economic population thresholds.
Finally, we wish to emphasize the importance of the input data used for
determining parameter values and our results. Although input data were selected to
reflect reality, assumptions always influence the outcome in ex ante assessments. Two
key assumptions are important in this respect—prices and yield effects of the WCR.
Prices influence the relative competitiveness of the different CP strategies, so that our
151
results can alternatively be explained by the fact that the prices used for chemical
control options were too high for the degree of crop protection provided relative to the
cost and protection offered by Bt maize. However, these insecticide and Bt maize
prices are not static. Insecticide prices used in this analysis were observed in a market
where Bt maize is not deregulated and therefore the insecticides were priced only to
compete with crop rotation. If Bt maize is deregulated and commercialized in these
countries, the prices of chemical CP options may decrease, changing the results of this
study. A similar phenomenon occurred for herbicide resistant crops in the USA and
the introduction of generic herbicides when the patent on glyphosate expired (Just,
2006). Similar logic applies to the cost of Bt maize—if Bt maize were
commercialized at prices substantially different from those used here, results would
change.
Another crucial source of data is the impact of the WCR on maize yields in
Europe. Our bio-economic model used data for three key relationships—the damage
caused by WCR under the no control scenario, the damage caused by WCR with each
CP option (or equivalently, the efficacy of each CP option), and the yield effect of
WCR damage. Only fragmentary data to describe these relationships are available in
Europe, particularly for many of these counties, which make an economic assessment
difficult. Therefore this study relies on data from the USA, which has disadvantages,
as it assumes that the damages and yield losses are similar in US and European
systems although there is a significant difference in cultivation practices and
circumstances (Dun et al., 2010). Our results suggest that European maize producers
will face substantial yield and income losses from WCR and will likely adopt various
crop protection measures to reduce these losses, including Bt maize if it is available.
However, our results rely on non-European data for some key relationships, indicating
the tremendous need to collect, analyze and disseminate field data and economic
analyses based on European data.
152
Chapter 9. Conclusions and Further Considerations
In this chapter the dissertation’s hypotheses, introduced in Chapter 1, are critically
assessed using the results from the various chapters. A full statistical proof is
impossible due the nature of ex ante impact assessments. However, the different
chapters provide substantial information to discuss the hypotheses. Furthermore, some
recommendations for further research are formulated based on the lessons learned
from the dissertation.
Hypothesis 1: Heterogeneity among farmers has no effect on corporate pricing
strategies of proprietary technologies and resulting welfare effects.
The importance of heterogeneity is recognized in the scholarly literature about
adoption and technological innovation (e.g. literature review in Chapters 1 and 2).
The value of a technology is not uniformly distributed among farmers; some realize a
profit from the technology and adopt it while others rationally choose not to adopt. In
particular, GM seed technologies will pay off differentially depending on managerial
capacities, market and field conditions, pest densities, crop rotation and
environmental conditions. However in the literature of ex ante impact assessments
little attention is given to this important concept, especially under imperfect and
scarce data. Previous studies generally focused on the value of the new technology for
the average farmer. Implicitly, these studies assume that the group of farmers with a
high potential value is compensated by a group of farmers with low benefits centering
the welfare estimation around the average farmer. However, some stakeholders in the
debate surrounding GM crops, may actively seek out a limited subsample of the
heterogeneous population to support their claims about GM crops. Therefore, Chapter
2 presented a framework explicitly accounting for heterogeneity in ex ante impact
assessments and assesses the implicit assumption in earlier studies.
Building upon the work started by Demont et al. (2008a), Chapter 2 explicitly
takes into account the presence of heterogeneity through the use of probability density
functions (PDF) and Monte Carlo simulation. Instead of focusing on some available
data points, the use of PDFs allows for the evaluation of the technology for each
single farmer in the population of potential adopters. This approach eliminates two
potential sources of biases in ex ante impact assessments that do not take into account
heterogeneity. The first bias, homogeneity bias, may arise in the conventional
approach through insufficient knowledge about the distribution of technology
154
Chapter 9. Conclusions and Further Considerations
valuation in the population. The segments of adopting and non-adopting farmers are
not directly observable. Hence the implicit assumption that both segments will cancel
each other out around the average farmer is not straightforward and may lead to a
bias. Particularly for innovations that possess some degree of market power such as
proprietary seed varieties. This constraint is overcome by explicitly determining a
marginal adopter, indifferent between adopting and not-adopting, and calculating the
benefits explicitly for the adopting segment of the population. Similar to Demont et
al. (2008a), it is
argued that taking into account second order statistics avoids
homogeneity bias.
Before these individual determinants could be assessed, the solution for a second
possible bias, the pricing bias, is to be introduced. Under the assumption that the
innovator operates in a market with market power due to his intellectual property
rights and that he is a profit maximizer, the decision on the profit maximizing
technology fee can be made endogenous. This specification makes the marginal
adopter dependent on the shape of the chosen PDF. The framework can be used to
trace the effect of higher order statistics on the magnitude of the technology fee,
potential adoption rates and the resulting welfare effects. The higher the average
valuation for the technology in the population, the higher the technology fee profit
maximizing technology fee, which follows the intuition. Perhaps of higher interest is
the formal conclusion that a more heterogeneous population of potential adopters
decreases the value per hectare for the innovator. The price is dropped in order to
maintain a profit maximizing customer base. Unfortunately, the influence of these
variables on the created farmer benefits are not analytically tractable. However,
compared to the conclusions reached by Demont et al. (2008a), the calculations show
that higher order statistics such as skewness also play a key role, and the arising
pricing and homogeneity bias is determined by the full shape of the probability
density curve. It is apparent that lower heterogeneity leads to higher profit shares for
the innovator, extracting more of the generated value.
Exploration of the consequences was completed through a case study with a
specific PDF. Applying the framework on HT sugar beet yields interesting data on
adoption and technology fees under different corporate pricing strategies. Moreover,
introducing the micro level analysis into the partial equilibrium model EUWABSIM
yields insight in the distribution of benefits among downstream and upstream sectors.
Surprisingly, assuming profit maximizing innovators in the heterogeneous population
155
of sugar beet farmers, results in a distribution according to the rule of thumb defined
by Demont et al. (2007a), stating that 2/3 of the benefits accrue to farmers and
consumers. Hence it is suggested that the observed rule of thumb is a direct result of
the heterogeneity among farmers and not a result of a strategic choice by the
innovator.
The importance of heterogeneity on the value and its distribution can help to
explain why innovators engage in spatial third degree price discrimination. By setting
a different technology price in different regions, the innovator tries to reach a more
homogeneous population of potential adopters. This differential price strategy has
been observed in Spain with pest pressure as a homogenizing variable (GomezBarbero et al., 2008). Heterogeneity may also play a key role in the benefit sharing of
future introductions of IPR protected technology. For the innovator, a more
homogenized, standardized population of potential adopters increases the share of the
value that can be extracted. Hence, it is a profitable strategy for the innovator to
homogenize the population through packaging of the technology with compulsory
goods. For example, credit can be offered to farmers in order to increase the liquidity
of farmers and their possibility to pay for a certain technology. Another strategy is to
provide insurance policies with the GM technology, as decreased risk decreases the
heterogeneity in the population. Interestingly, the very nature of some first generation
GM crops available in the market seems to result in a homogenized population of
potential adopters. HT crops for instance, shift the production system from based on
individual scouting decisions and individual decision on mixing appropriate
herbicides, to a standardized herbicide-seed decision. This potentially increases the
share accruing to innovators in the next wave of GM crop introductions aimed at other
production constraints.
The discussion suggests that Hypothesis 1 does not hold as heterogeneity has been
proven to affect corporate pricing strategies and welfare effects to a high extent and
may even be an important determinant of welfare distribution. Hence, not explicitly
accounting for it in ex ante impact assessments leads to biases in the estimation
results.
The aforementioned possibility of an homogenizing effect of GM crop
introduction on the farm triggered an idea for further research in the dissertation of
Demont (2006). This idea has been further elaborated during the period of this PhD
and will be shortly introduced here.
156
Chapter 9. Conclusions and Further Considerations
Agricultural production differs from other economic sectors due to endogenous
structural constraints in the production process (Nerlove, 1996; Schmitt, 1991).
Among other properties, a farm’s output depends on nature’s influences, mainly
weather, pests and diseases, and on the optimal timing of application of production
inputs depending on farm-specific weather variations which are hard to anticipate in
advance. Hence, timeliness costs are high as yields may decline significantly in the
case of deviation from optimal timing (Ndeffo Mbah et al., 2010; Short and Gitu,
1991; Swanton et al., 1999). The effects of these structural constraints in agriculture
on labor or production organization have been debated in literature. Stiglitz (1974)
started a stream of literature on share tenancy based on a principal-agent framework.
Another stream of research is focused on the transaction costs involved in agriculture
(Allen and Lueck, 1996; Allen and Lueck, 1998; Allen and Lueck, 2002; Beckmann,
2000; Roumasset, 1995). Especially labor specific moral hazard problems and their
related transaction costs burden agricultural production and are often listed as a reason
for the persistence of small scale farming despite a long term tendency towards
consolidation and industrial agriculture (Allen and Lueck, 2002; Beckmann, 2000;
Johnson and Ruttan, 1994; Nerlove, 1996; Rizov, 2003).
However, the direct effect of timeliness costs on contracting within a farm and the
boundaries of the farm are not well studied. Allen and Lueck (2002) present some
predictions based on a stylized framework arguing that decreased time sensitivity
would lead to increased contracting. This is in line with the seminal paper in New
Institutional Economics by Masten et al. (1991) highlighting the importance of
temporal specificity in governance structures promoting internal integration. GM
crops may have the opportunity to decrease this temporal specificity through the
increased production flexibility. Therefore, research on this interaction and its effect
on contracting and even the boundaries of the agricultural firm is needed to assess
these indirect effects of GM crops. Some empirical papers have touched upon some
related structural effects of adoption (Fernandez-Cornejo et al., 2005; Useche et al.,
2009) such as off farm income. However, no structured framework exists to address
these issues in a more formal way.
157
Hypothesis 2: Parametric approaches do not have the potential to complement
the inherent scarce data in ex ante impact assessments.
Chapter 2 and the discussion on Hypothesis 1 highlighted the need for higher order
statistics to incorporate the important factor of farmer heterogeneity in ex ante impact
assessments. But the nature of ex ante impact assessments creates an endogenous
problem of imperfect information for the assessor. The amount and type of data
available to the researcher can be situated in an information continuum (Figure 13). At
the lower end of the continuum, no information is available and theoretical
considerations dictate parametric theoretic estimation procedures. Towards the higher
end of the information continuum, in the direction of full information, empirical
findings gradually replace theory and non-parametric procedures substitute for
parametric procedures.
Little information
Intermediate
Full information
• Expert opinions
• Theory and
parametric models
• Resource extensive
• Combination data
with theory
• Data sets
• Non-parametric
models
• Resource intensive
Figure 13: The information continuum in ex ante impact assessments
Although the probability of selecting the correct parametric model is zero, a
parametric approach does not necessarily lead to biased results. With a false model,
parameters will converge such that the Kullback-Leibler distance between the true
158
Chapter 9. Conclusions and Further Considerations
density and the best parametric estimate is minimized. Therefore it is possible that an
incorrect parametric model may have greater efficiency than the correctly modeled
density and a non-parametric model (Goodwin and Ker, 2002). The position of the
assessor in the information continuum is to a high extent dependent on the resources
available to gather data. As argued in the introduction, the inclusion of socioeconomic impact assessments in biosafety regulations may lead to a need for reliable
results with limited resources. Hence, in this study two different parametric
approaches based on limited available data have been developed and tested.
The approach in Chapter 2 matches an extensive dataset for a limited geographical
region with expert opinions on that dataset. The approach yields a structural equation
for the PDF and fractals determined by the experts. Consequently the appropriate PDF
for the remaining regions can be constructed by overlaying the region specific expert
opinions with the structural shape of the PDF. The resulting PDF can then be used as
a measure of heterogeneity in the pricing strategy and to discover the marginal
adopter to identify the segments of adopters and non-adopters in the population.
Although the aforementioned method is one step up from previous approaches
constructing a PDF solely on expert opinions (Demont et al., 2008a), heterogeneity is
still simplified through the use of a proxy variable instead of the underlying
determinants. The bio-economic model described in Chapter 7 and applied in Chapter
8 tries to trace back the heterogeneity to the driving forces of pest damages, control
options and market situations through the use of a hierarchical model. Based on
experimental field data, a yield response to pest damage under different control
options is estimated. This response is modeled as a beta PDF conditional on a
counterfactual pest damage distribution for the situation and location considered.
Sampling from these two hierarchical distributions results in a structural
representation of the yield effects which can be used in the partial budgeting approach
of Chapter 2 and further limit the reliance on assumptions and experts.
Hence, in contradiction with Hypothesis 2, parametric procedures have the
potential to combine scarce data and theoretical consideration in such way that the
homogeneity biases can be avoided.
The developed approaches do not differentiate between do not explicitly
differentiate different sources of heterogeneity. However, in reality, the sources of
heterogeneity among farmers can often be subdivided into a spatial and a temporal
dimension. The application of the developed framework on a panel data set
in
159
Demont et al. (2009b) recognizes these two dimension but when fitting the resulting
PDF they aggregate both dimension is a single spatio-temporal PDF. However, for
some technologies, disaggregating these two dimensions may be essential to capture
the consequences introducing the technology. A technology may have a different
moderation effect on the two dimension. An example at the center of attention is
drought tolerance. A drought tolerant maize variety will reach the US market in 2012
and different research consortia such as the Water Efficient Maize for Africa
(WEMA) are developing crops for the near future in other continents. Drought has a
spatial component but it is mainly the temporal variability that causes problems in the
agricultural production process. Theoretically there is no problem to transform the
frameworks developed in Chapter 2 and Chapter 7 to a three dimensional model
explicitly accounting for the two sources of heterogeneity. Let us consider the
continuous PDFs s(x) and t(y) representing the spatial and the temporal heterogeneity
respectively. A joint distribution can be constructed,
𝑓𝑓𝑥𝑥𝑥𝑥 (𝑥𝑥, 𝑦𝑦) = 𝑠𝑠𝑥𝑥⃓𝑦𝑦 (𝑥𝑥
⃓𝑦𝑦)𝑡𝑡𝑦𝑦 (𝑦𝑦)
where 𝑠𝑠𝑥𝑥⃓𝑦𝑦 (𝑥𝑥
⃓𝑦𝑦) is a conditional distribution and 𝑡𝑡𝑦𝑦 (𝑦𝑦) a marginal distribution. This
three dimensional distribution can be projected in a plain to fit equation 1 in Chapter
2,
𝑟𝑟
𝑟𝑟
𝐹𝐹(𝑟𝑟) = ∫0 ∫0 𝑓𝑓𝑥𝑥𝑥𝑥 (𝑥𝑥, 𝑦𝑦)𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑.
For some PDFs a closed form of this multivariate distribution does exist. However, if
the distribution in the spatial dimension and temporal dimension have different
functional forms and the analytical traceability will reduce. Two possible solution can
be explored. First a numerical procedure can be used to approximate the function or
calculate the results. If longitudinal data is available a joint distribution could be
constructed through the use of a Copula function (Greene, 2008).
Hypothesis 3: EU policies not directly related with GM crops do not affect the
potential adoption and deregulation of GM crops in Europe.
It is well know that policy interventions alter the stream of benefits from innovations
(Alston et al., 1995). For a highly regulated sector such as agriculture this has
160
Chapter 9. Conclusions and Further Considerations
important consequences. Hence, when describing the institutional environment
affecting the introduction of GM crops, Wesseler (2005) pinpoints policy as an
important factor. Therefore, studies assessing the impact of innovations typically
account for these market distortions (e.g. Chapter 4). However, very little studies have
assessed the effect of changes in policies on the incentives for innovation and
adoption. To my knowledge, no study has ever assessed the impact on innovation of
policies that are situated in a certain policy arena but may have secondary effects on
innovation. In this dissertation the effect of two EU policies originally serving other
goals on innovation is assessed.
In Chapter 3 the effect of the 2006 change in the Common Market Organization
(CMO) for sugar on the innovation incentive of farmers and innovators is assessed.
The pairwise differences between the technology’s added value per adopted hectare
under the old and the new CMO for sugar are evaluated. Hence the chapter specific
hypothesis, the policy reform has no influence on the stream of benefits from
innovation, can be tested. Focusing on the farmers, the results show that for high cost
producers there is a significant reduction of the incentive to innovate. Reduced
innovation leads to decreased competitiveness in the longer run. Therefore, this result
is in line with the initial aim of the policy reform to shift production to lower cost
regions and crowd out the high cost producers. Hence, the secondary effect of the
sugar reform does not interfere with the initial aim of the policy.
To understand the impact of the CMO reform on the innovator, the seed
developer, data from both Chapter 3 and Chapter 4 can provide some insights. In
Chapter 4 one can see that the CMO for sugar does drastically decrease the area
planted with sugar beet due to lower internal prices. As the source of revenue from the
innovator is the technology fee, a premium per hectare, a decreased hectarage means
decreased revenue. In Chapter 3 this hypothesis is analyzed by the pairwise
comparison of the EU revenue before and after the reform. As expected, the results
show a significant decrease in revenue from the technology introduction. A decreased
anticipated revenue reduces the incentive for R&D in the sugar sector. From a policy
point of view, this is an unwanted side effect of a reform aiming at increased
competitiveness in the sugar sector.
The discussion in Chapter 3 only considers reversible effects of the technology
introduction. However, the introduction of GM crops may have properties that cannot
be quantified or are even unknown at the time of the assessment. For GM crops these
161
potential effects may include gene drift, biodiversity effects, human health etc. As
these sources of uncertainty are not easily predicted, an approach as described earlier
is not feasible. Moreover, some of these potential effects have an irreversible
character, once they arise, a ban on the technology does not restore the situation to its
state before introduction. These properties of a novel technology results in a waiting
value or option. In time, the uncertainty about a technology may reduce and provide
more insights in whether or not to allow introduction at that time. Therefore, Chapter
5 introduces a Bayesian decision model to integrate these sources of uncertainty and
irreversibility. This model transforms the neo-classical rule, “introduce the technology
if the benefits are at least as high as the costs” to “the technology should only be
released if the reversible net benefits are greater than the irreversible net costs
multiplied by a factor higher than one, the hurdle rate. The hurdle rate can be
quantified by the standard procedures of the real options approach. However, with the
introduction of GM crops the net reversible benefits can be assessed (Chapter 4) but
the irreversible net costs are uncertain. Hence the decision criteria is reformulated in
order to quantify the maximal incremental social tolerable irreversible cost (MISTIC)
that would justify an immediate release of the technology.
Comparing the analysis including irreversibility and uncertainty with the neoclassical approach deepens the insights in the effects of the technology and introduces
a second approach to assess the effect of policy reforms. In Table 13 we see that the
hurdle rates, both under the old and the new CMO, are higher than one, leading to a
more conservative decision rule as in neo-classical economics. Interestingly, the
hurdle rates are higher under the old regime, especially for those countries exporting
sugar to the world market. Indeed the exposure to world market prices increases the
uncertainty of the stream of benefits. Denmark presents a clear example of the impact
of including uncertainty and irreversibility in the economic rationale. Under the old
CMO reversible net benefits are €103/ha, but the presence of high uncertainty leads to
a MISTIC of only €21.3/ha. Under the new CMO this average benefit increases to
€145/ha but the MISTICS increase to €144.8/ha. From this perspective, an
introduction of the technology in Denmark is more likely to be justified than under the
old regime. A similar but less pronounced effect is present for most Member States.
Only for Greece, Italy and Portugal do the MISTICs decrease with the CMO reform.
Hence the increased stability under the new CMO provides an environment in which
162
Chapter 9. Conclusions and Further Considerations
introduction of HT sugar beets is more likely from an economic optimizing strategy
point of view.
In conclusion to Hypothesis 3, it seems the reform of the sugar policy did affect
the potential adoption of HT sugar beet in the EU. The change favors low cost
producers through two avenues, i.e. by increasing the incentives for innovation and
hence the MISTICs, but also by lowering uncertainty by limiting world market
influence and, hence again increasing MISTICs
Irreversibilities have a prominent role in the approval process of GM crops in the
EU (Wesseler et al., 2007). Hence an assessment has been done of the effect of
changes in this approval process on the likelihood of deregulation of HT sugar beet
based on the MISTICs value. Under the assumption that the magnitude of individual
MISTICs determines the voting behavior of an individual, the outcome under two EU
treaties and the recent Barroso proposal are evaluated. The two treaties modeled, the
Nice and Lisbon Treaty, differ in the way a qualified majority is to be reached. For the
old CMO for sugar, the shift from the Nice Treaty to the Lisbon Treaty only has a
marginal effect on the MISTIC threshold value needed to reach a qualified majority
and hence the likelihood of deregulation. For the new CMO however, there is an
important difference in this likelihood. The threshold value increases with 47% under
the Lisbon Treaty. Hence the likelihood of approval increases significantly under the
newer Lisbon Treaty. This leads to the conclusion that the increased economic
incentive to innovate under the new CMO only translates to a higher chance of
deregulation under the reformed Lisbon Treaty. This is in line with the aim of the
alteration of the rules to reach a qualified majority proportional to incentives, despite
the expansion of the EU.
The Barroso proposal suggest to shift the deregulation decision to the national
level. The results under this proposal show that a high degree of inertia is present in
the EU decision making if the position of different Member States differs. A threshold
value that would not lead to deregulation at the European level, would transfer to
deregulation in all but five Member States under the Barroso proposal. Interestingly,
the countries restrained to deregulate are countries that are considered to have high
production costs. Consequently, the proposal would allow the countries with low
costs to approve the technology. Thus, as a secondary effect the Barroso proposal may
increase long term competitiveness of the sector as discussed earlier. Shifting the
authority on deregulation to the national level has the added benefit of converging the
163
authority on approval with the authority on coexistence between GM crops and
conventional systems. This may increase the efficiency of policy making and avoid
perverse effects of the legislative framework (e.g. Demont et al., 2010b; Devos et al.,
2008b).
Both the case studies presented in this dissertation prove that policies do affect the
likelihood of adoption and deregulation, even if they were initially designed for other
purposes. Hence, Hypothesis 3 does not hold. Therefore, long term competitiveness
studies or studies assessing the impact of novel technologies should try to include
these policies explicitly in the analysis.
Hypothesis 4: The invasive species Western Corn Rootworm does not presents
an economic threat for Central European maize farmers.
In the nineties, the invasive species WCR started spreading through Europe. The
species presents a potential economic threat to European maize production, mainly
because its larvae feeds on the roots of the maize plants. The specific nature of the
species makes the prediction and assessment of damage caused by the species very
difficult, as exogenous factors such as climatic and soil conditions affect the resulting
yields to a high extent (Rice and Oleson, 2005). Larval damage may be offset by the
regrowth of the root system if water is available during the appropriate time window
of the growing process. This leads to difficulties in determining the correlation
between population pressure and damages. Due to this uncertainty, the perception
about the actual threat to European maize is diverse and the available economic
studies scattered (e.g. Baufeld and Enzian, 2005b; Fall and Wesseler, 2008; Hatala
Zsellér et al., 2006; Schaafsma et al., 1999). In Chapter 6 through Chapter 8 different
approaches are used to evaluate whether the species actually presents an economic
risk to farmers.
In Chapter 6, the results of a farmer survey in Hungary are presented. As Hungary
was fully infested by the species in 2003, farmers should have enough experience
with the species to answer the questions. The results show that farmers consider WCR
as the major maize pest in Hungary. Interestingly, despite the perceived danger, only
70% of the farmers stated they detected WCR in their field. This indicates that despite
the official infestation, not all farmers actively observe the beetle within their
plantings. This is also reflected by the anticipated damage by WCR in their field as
164
Chapter 9. Conclusions and Further Considerations
presented in Figure 7. 50% of the farmers indicate that even without treatment WCR
would not cause any yield loss on their field. This is in contrast with the other half of
the population expecting high damages from the species, up to 100%, resulting in an
average perceived damage of 22%. The contradiction can be explained by two factors.
First, farmers with extensive crop rotation may never reach a population causing
economic damages as rotation terminates the life cycle of the species. Secondly,
farmers may have difficulties in disentangling the damages by the species from yield
reduction due to the reasons discussed above. This was demonstrated in the USA
where before introduction of a resistant GM variety a large group of farmers indicated
not having problems with the species but where adoption led to yield increases
nevertheless.
For Hungary, with a high percentage of maize in continuous cultivation, favoring
a population build-up, the perceived threat can also be deducted from the high rate of
insecticide application according to the survey. Moreover, some farmers indicated
they suffered damages in the last cropping seasons despite the application of crop
protection (CP) strategies. Hence, the results from the Hungarian survey suggest that a
threat exists but that it is highly dependent on the production system and the
marketing situation of maize.
Chapter 7 looks at the potential threat from a modeling point of view. Starting
from data gathered in the USA, combined with expert data and the survey from the
Chapter 6, a stochastic hierarchical model is developed as presented earlier. The
model results indicate the range of potential yield losses under a hypothetical no
control situation. As expected, the resulting distribution of potential effects ranges
from no yield loss to 60% yield loss in special circumstances. Fortunately, farmers
have access to different CP options moderating the impact of WCR on their maize
yield. The model results show that none of the available strategies; crop rotation, soil
insecticides and seed treatments, reduce the chance of a yield loss to a zero percentage
level. Therefore, a stochastic partial budgeting model is built to explore all possible
outcomes under different crop protection strategies, and more importantly, the
determinants of the relative economic competitiveness of different strategies.
Unraveling the determinants allows the specification of optimal CP strategies in
different maize production systems.
In Chapter 8, the framework is applied to seven other Central European countries
to get a better insight in the species’ threat. The combined analysis shows that
165
,although there exists a high variability, crop rotation is generally a rational decision
for the majority of farmers. Only in those countries where yields are low and prices
are low (such as Ukraine) a majority of farmers will not control for the species as the
price for the different CP are not compensated by the provided protection. If feasible,
crop rotation is often a viable CP strategy. However, for farmers with compulsory
continuous maize cultivation due to specialization or land constraints, chemical
control options are essential. For specific results a reference is made to Table 23.
Combining the results from the modeling approach with the results from the
survey rejects Hypothesis 4 stating there is no economic threat for maize production
by the species. Especially in a time when prospects for maize production are good this
may have serious consequences. Therefore there exists a high need to design CP
strategies consistent with the individual situation of the farmer to protect production.
Further work along the track started in this dissertation could eventually lead to a
multi-criteria framework to be used as an aid to develop economic rational control
strategies instead of uniform tools such as economic threshold values. For this
framework to evolve to a decision tool, there is an urgent need for data gathering in
the EU. A first effort should be made to gather data on the pest-yield relationships in
European field conditions. In the present framework, the implicit assumption is made
that the yield effects in Europe are similar to the effect in the USA. This link is not
proven nor completely rational as agricultural practices differ significantly between
the two continents. However, it is the best available information we have so far.
Secondly, as the maize cultivation practice is such an important determinant, detailed
information on each maize production system is essential. Ideally, geographic explicit
land use and land tenure data would allow the development of an improved biological
model including population build-up in specific regions dependent on the
environmental parameters. Moreover, such data would allow aggregation to the
national level or the incorporation in a trade model to understand the threat of the
species in more depth.
Hypothesis 5: EU farmers cannot gain from the introduction of Bt maize
resistant to WCR damages.
As indicated in the discussion of Hypothesis 4, besides crop rotation, two different CP
strategies are commercially available on the European market. At the same time it is
166
Chapter 9. Conclusions and Further Considerations
clear that WCR presents a threat to Central European maize production in the present
market context. In the USA, another CP option has been commercialized in 2003, a Bt
maize variety resistant to WCR damage. The adoption of this technology has been
swift, highlighting the potential value of the technology for farmers (Alston et al.,
2002). In Europe this variety has not been deregulated and commercialized for
cultivation, but the trait MON88017 is allowed for feed and food use. Hypothesis 5
states that deregulating this Bt maize trait in Europe would not create economic
benefits for European farmers. This hypothesis is tested via two approaches. In
Chapters 7 and 8 the bio-economic model includes Bt maize as an addition to the
toolbox of CP options and estimates the monetary value added by the technology or
alternatively the value foregone by not deregulating the technology. The survey
described in Chapter 6 uses a choice experiment to elicit the valuation of Hungarian
farmers for a WCR resistant variety, hence including non-pecuniary properties in the
estimation.
The approach in Chapter 7 addresses the strictly monetary gains from deregulating
the technology. The results suggest that despite an average anticipated technology fee
of €23/ha, profits range between €10/ha and €62/ha, depending on the production
system, for farmers rationally adopting the technology. The largest benefits accrue to
farmers that engage in continuous maize cultivation. Consequently, predicted market
shares of the different CP options change drastically with the introduction of Bt
maize. Before the introduction chemical CP would be optimal on a substantial area (as
confirmed through our survey), e.g. 49% in Austria, mainly by farmers having either
high yields or high pest pressure. These is exactly the market segment where Bt maize
is highly valuable, hence the area with chemical application in Austria reduces to 10%
after the introduction of the technology. This effect shows that besides an added value
for Bt maize, it would move farmers away from chemical control options creating a
positive externality from the technology.
In Chapter 6 the willingness to pay (WTP) for a WCR resistant maize variety in
Hungary is estimated. Due to the discussed constraints, this analysis does not only
contain Bt maize but also resistance through other breeding techniques. Estimating the
WTP has the benefit of accounting for both the monetary benefits, similar to those
calculated by the bio-economic model, and the non-pecuniary benefits present in the
technology (e.g. Marra and Piggott, 2006). The dichotomous choice model reveals
that farmers are willing to pay €70/ha for the technology on average. This is
167
substantially more than the results from the modeling approach. Hence, the nonpecuniary benefits such as reduced scouting etc. do have an important value to
farmers. For Hungarian farmers the compatibility from the technology with their
existing material is an important plus of the resistant technology. Moreover, the high
WTP for a novel CP options signals that farmers are not satisfied with the control
options available at the time of the survey.
The results from both approaches show that Hypothesis 5 does not hold and Bt
maize resistant against WCR damage would create significant pecuniary and nonpecuniary benefits. Moreover, introducing the technology would drastically change
the use of chemical pesticides in maize production.
168
List of References
Acquaye, A. and Traxler, G. (2005). Monopoly Power, Price Discrimination, and
Access to Biotechnology Innovations. AgBioForum 8((2&3)):127-133.
Adenäuer, M. and Heckelei, T. (2005). Economic incentives to supply sugar beets in
Europe. Modelling Agricultural Policies: State of the Art and New Challenges.
Parma: Monte Universita Parma: 588-609.
AGbios. GM crop database. http://www.agbios.com. 19-12-2008.
AGES. 2008. Maize production in Austria. Personal Communication.
AKI (2005). Hungarian Food and Agricultural Statistics: 2004. Budapest: Research
and Information Institute for Agricultural Economics (AKI).
AKI (2007). Results of Hungarian FADN Farms: 2005. Budapest: AKI.
AKI. 2008. Statistical data on Hungarian agriculture. Personal Communication.
Alexander, C. and Goodhue, R. (2002). The Pricing of Innovations: An Application to
Specialized Corn Traits. Agribusiness 18(3):333-348.
Allen, D. W. and Lueck, D. (1996). The Transaction Cost Approach To Agricultural
Contracts. In D. Martimort (ed.), Agricultural Markets: Mechanisms, Failures,
and Regulations. Amsterdam: Elsevier, 31-64.
Allen, D. W. and Lueck, D. (1998). The nature of the farm. Journal of Law and
Economics 41(2):343-386.
Allen, D. W. and Lueck, D. (2002). The Nature of the Farm. Cambridge,
Massachusetts: The MIT press.
Alston, J. M., Hyde, J., Marra, M. C., and Mitchell, P. D. (2002). An ex ante analysis
of the benefits from the adoption of corn rootworm resistant transgenic corn
technology. AgBioForum 5(3):71-84.
Alston, J. M., Norton, G. W., and Pardey, P. G. (1995). Science Under Scarcity:
Principles and Practice of Agricultural Research Evaluation and Priority
Setting. Ithaca, NY: Cornell University Press.
Anderson, J. R. and Hardaker, B. J. (2003). Risk aversion in economic decision
making: Pragmatic guides for consistent choice by natural resourece managers.
In J. Wesseler, H. P. Weikard, and R. D. Weaver (eds.), Risk and Uncertainty in
Environmental Economics. Cheltenham, U.K.: Edward Elgar Publishing, 171187.
Anderson, K., Valenzuela, E., and Jackson, L. A. (2008). Recent and Prospective
Adoption of Genetically Modified Cotton: A Global Computable General
Equilibrium Analysis of Economic Impacts. Economic Development & Cultural
Change 56(2):265-296.
Ansink, E. and Wesseler, J. (2009). Quantifying type I and typ II errors in decision
making under uncertainty: The case of GM crops. Letters in Spatial and
Resource Science 2(1):61-66.
Arrow, K. J. (1962). Economic welfare and the allocation of resources for invention.
In R. R. Nelson (ed.), The Rate and Direction of Inventive Activity: Economic
and Social Factors. Princeton, NJ: Princeton University Press.
Babcock, B., Choi, K., and Feinerman, E. (1993). Risk and probability premiums for
CARA Utility functions. Journal of Agricultural and Resource Economics
18(1):17-24.
Baca, F. (1993). New member of the harmful entomofauna of Yugoslavia Diabrotica
virgifera virgifera LeConte (Coleoptera,Chrisomelidae). XXI skup entomologa
Jugoslavije. IWGO Newsletter 13:21-22.
Baker, J. M., Ochsner, T. E., Venterea, R. T., and Griffis, T. J. (2007). Tillage and soil
carbon sequestration- What do we really know? Agriculture, Ecosystems &
Environment 118:1-5.
170
Balint,B. and Sauer,J. Distorted prices and producer efficiency: Romanian maize.
46th annual meeting of the German association of agricultural economists 2006.
Giessen.
Banse,M., Grethe,H., and Nolte,S. Documentation of the ESIM model structure, base
data and parameters. 2005. Göttingen and Berlin, University of Göttingen and
Humboldt University.
Barfoot, P. and Brookes, G. (2007). Global impact of biotech crops: Socio-economic
and environmental effects, 1996-2006. AgBioForum 11(1):21-38.
Barroso,J.M. Political guidelines for the next Commission. 2009. Brussels, European
Commission.
Bateman, I. A., Carson, R. T., Day, B., Hanemann, M., Hanley, H., Hett, T., JonesLee, M., Loomes, G., Mourato, S., Özdemiroglu, E., Pearce, D. W., Sugden, R.,
and Swanson, J. (2002). Economic Valuation With Stated Preference
Techniques: A Manual. Cheltenham, UK: Edgar Elgar.
Batie, S. (2003). The Environmental Impacts of Genetically Modified Plants:
Challenges to Decision Making. American Journal of Agricultural Economics
85(5):1107-1111.
Baufeld,P. and Enzian,S. Diabrotica virgifera virgifera, its potential spread and
economic and ecological consequences in Germany. 2005a. Kleinmachnow,
Germany, Federal Biological Research Centre for Agriculture and Forestry.
Baufeld, P. and Enzian, S. (2005b). Maize growing, maize high risk areas and
potential yield losses due to Western corn rootworm (Diabrotica virgifera
virgifera) damage in selected Euroopean countries. In S. Vidal, U. Kuhlmann,
and C. R. Edwards (eds.), Western Corn Rootworm: Ecology and Managment.
CABI publishing, 285-302.
Baufeld, P. and Enzian, S. (2005c). Western corn rootworm (Diabrotica virgifera
virgifera), its potential spread and economic and ecological consequences in
Germany. Symposium Proceedings No.81: Introduction and Spread of Invasive
Species:149-154.
Beckmann, V. (2000). Transaktionkosten Und Institutionelle Wahl in Der
Landwirtschaft. Zwischen Markt, Hierarchie Und Kooperation. Berlin: Editon
Sigma.
Beckmann, V., Soregaroli, C., and Wesseler, J. (2006). Coexistence rules and
regulations in the European Union. American Journal of Agricultural
Economics 88(5):1193-1199.
Beckmann, V., Soregaroli, C., and Wesseler, J. (2010). Ex-Ante Regulation and ExPost Liability under Uncertainty and Irreversibility: Governing the Coexistence
of GM Crops. Economics: The Open-Access, Open Assessment E-Journal
4(2010-9).
Beckmann, V. and Wesseler, J. (2003). How Labour Organization May Affect
Technology Adoption: An Analytical Framework Analysing the Case of
Integrated Pest Management. Environment and Development Economics 8:437450.
Beckmann, V. and Wesseler, J. (2007). Spatial Dimension of Externalities and the
Coase Theorem: Implications for Co-existence of Transgenic Crops. In W.
Heijman (ed.), Regional Externalities. Berlin Heidelberg New York: Springer,
223-242.
Beyer,J.C., Norton,G.W., and Falck-zepeda,J. The cost of biotechnology regulations
in the Phillipines. 2008 Annual meeting of the American Association of
Agricultural Economists. 2008. Orlando, Florida.
171
Bhagwati, J. N. (1958). Immiserizing growth: A geometrical note. Review of
Economic Studies 25:201-205.
Bijman, J. (2001). KWS: going beyond sugar beet. AgBioForum 4(1):46-51.
Bishop, R. C. and Heberlein, T. A. (1979). Measuring Values of Extramarket Goods:
Are Indirect Measures Biased? American Journal of Agricultural Economics
61(5):926-930.
Bogetoft, P., Boye, K., Neergaard-Petersen, H., and Nielsen, K. (2007). Reallocating
sugar beet contracts: can sugar production survive in Denmark? European
Review of Agricultural Economics 34(1):1-20.
Boriani, M., Agosti, M., Kiss, J., and Edwards, C. R. (2006). Sustainable management
of the western corn rootworm , Diabrotica virgifera virgifera LeConte
(Coleoptera: Chrysomelidae), in infested areas: experiences in Italy, Hungary
and the USA. EPPO Bulletin 36(3):531-537.
Branson, T. F. and Ortman, E. E. (1967). Host rang of larvae of the western corn
rootworm. Journal of Economic Entomology 60:201-203.
Breustedt, G., Muller-Scheessel, J., and Latacz-Lohmann, U. (2008). Forecasting the
Adoption of GM Oilseed Rape: Evidence from a Discrete Choice Experiment in
Germany. Journal of Agricultural Economics 59(2):237-257.
Brookes, G. (2008). European Arable Crop Profit Margins 2006/2007.
Brookes,G. (2009). Global impact of biotech crops: Income and production effects
1996-2007. AgBioForum, in press.
Brookes, G., Tun-Hsiang, Y., Tokgoz, S., and Elobeid, A. (2010). The production and
price impact of biotech corn, canola, and soybean crops. AgBioForum 13(1):2552.
Brookes, G. (2005). The farm-level impact of herbicide-tolerant soybeans in
Romania. AgBioForum 8(4):235-241.
Bückmann, H., Petersen, J., Schlinker, G., and Märländer, B. (2000). Weed control in
genetically modified sugar beet - Two years experiences of a field trial series in
Germany. Zeitschrift Für PflanzenKrankheit Und PflanzenSchutzung 17:353362.
Burgener,P.A., Feuz,D.M., and Wilson,R.G. The economics of transgenic sugarbeet
production. 2000. Western Agricultural Economics Association Annual
Meetings, Vancouver, Canada.
Buysse, J., Fernagut, B., Harmignie, O., de Frahan, B. H., Lauwers, L., Polome, P.,
Van Huylenbroeck, G., and Van Meensel, J. (2007). Farm-based modelling of
the EU sugar reform: impact on Belgian sugar beet suppliers. European Review
of Agricultural Economics 34(1):21-52.
Cagan,P. 2008. Information on WCR in Slovakia. Personal Communication.
Cameron, T. A. and Quiggin, J. (1994). Estimation Using Contingent Valuation Data
from a "Dichotomous Choice with Follow-Up" Questionnaire. Journal of
Environmental Economics and Management 27(3):218-234.
Ceddia, M. G., Bartlett, M., and Perrings, C. (2009). Quantifying the effect of buffer
zones, crop areas and spatial aggregation on the externalities of genetically
modified crops at landscape level. Agriculture, Ecosystems and Environment
129(1-3):65-72.
Ciosi, M., Miller, N. J., Kim, K. S., Giordano, R., Estoup, A., and Guillemaud, T.
(2008). Invasion of Europe by the western corn rootworm, Diabrotica virgifera
virgifera: multiple transatlantic introductions with various reductions of genetic
diversity. Molecular Ecology 17(16):3614-3627.
172
COGEM. Socio-economic aspects of GMOs: Building blocks for an EU sustainability
assessment of genetically modified crops. 2009. Commissie Genetische
Modificatie.
Combette, P., Giraud-Heraud, E., and Réquillart, V. (1997). La politique sucrière
européenne après les accords du GATT: Une analyse de quelques scénarios
d'évolution. Economie Et Prévision 127(1):1-13.
Confédération des Betteraviers Belges. 2002. Personal Communication.
Cooper, A. (2009). Political Indigestion: Germany Confronts Genetically Modified
Foods. German Politics 18(4):536-558.
Cox, W. J., Shields, E., and Cherney, D. J. R. (2007). Seed-applied insecticides
inconsistently affect corn forage in continuous corn. Agronomy Journal
99:1640-1644.
Coyette, B., Tencalla, F., Brants, I., and Fichet, Y. (2002). Effect of introducing
glyphosate-tolerant sugar beet on pesticide usage in Europe. Pesticide Outlook
13(5):219-223.
Crowder, D. W., Onstad, D. W., Gray, M. E., Mitchell, P. D., Spencer, J. L., and
Brazee, R. J. (2005). Economic Analysis of Dynamic Management Strategies
Utilizing Transgenic Corn for Control of Western Corn Rootworm (Coleoptera:
Chrysomelidae). Journal of Economic Entomology 98(3):961-975.
Davis, G. C. and Espinoza, M. C. (1998). A unified approach to sensitivity analysis in
equilibrium displacement models. American Journal of Agricultural Economics
80(4):868-879.
de Bont C.J.A.M., van Berkum S., Jager J.H., and Helming J.F.M. Suikerbeleid;
gevolgen van de Europese besluiten voor de Nederlandse akkerbouw en de
Europese suikermarkt. 2006. Den Haag, LEI.
Demont,M., Devos, Y., and Sanvido, O. (2010a). Towards flexible coexistence
regulations for GM crops in the EU. EuroChoices, in press.
Demont, M., Dillen, K., Daems, W., Sausse, C., Tollens, E., and Mathijs, E. (2010b).
On the proportionality of EU spatial ex ante coexistence regulations: Reply.
Food Policy 35(2):183-184.
Demont,M. Economic impact of agricultural biotechnology in the European Union:
Transgenic sugar beet and maize. 2006. Leuven, Katholieke Universiteit
Leuven. Dissertationes de Agricultura.
Demont,M., Cerovska,M., Daems,W., Dillen,K., Fogarasi,J., Mathijs,E., Muška,F.,
Soukup,J., and Tollens,E. Genetically modified crops in the New Member
States: How much value and for whom? 2007a. Leuven, Katholieke Universiteit
Leuven. Working Paper.
Demont, M., Cerovska, M., Daems, W., Dillen, K., Fogarasi, J., Mathijs, E., Muška,
F., Soukup, J., and Tollens, E. (2008a). Ex ante impact assessment under
imperfect information: Biotechnology in New Member States of the EU.
Journal of Agricultural Economics 59(3):463-486.
Demont, M., Daems, W., Dillen, K., Mathijs, E., Sausse, C., and Tollens, E. (2008b).
Regulating coexistence in Europe: Beware of the domino-effect! Ecological
Economics 64(4):683-689.
Demont, M. and Dillen, K. (2008). Herbicide tolerant sugar beet: The most promising
first-generation GM crop? International Sugar Journal 110(1318):613-617.
Demont, M., Dillen, K., Daems, W., Sausse, C., Tollens, E., and Mathijs, E. (2009a).
On the proportionality of EU spatial ex ante coexistence regulations. Food
Policy 34(6):508-518.
173
Demont, M., Dillen, K., Mathijs, E., and Tollens, E. (2007b). GM Crops in Europe:
How much value and for whom? EuroChoices 6(3):46-53.
Demont, M., Rodenburg, J., Diagne, M., and Diallo, S. (2009b). Ex ante impact
assessment of herbicide resistant rice in the Sahel. Crop Protection 28(9):728736.
Demont, M. and Tollens, E. (2004a). Ex ante welfare effects of agricultural
biotechnology in the European Union: The case of transgenic herbicide tolerant
sugarbeet. In R. E. Evenson and V. Santaniello (eds.), The Regulation of
Agricultural Biotechnology. Wallingford, UK: CAB International, 239-255.
Demont, M. and Tollens, E. (2004b). First impact of biotechnology in the EU: Bt
maize adoption in Spain. Annals of Applied Biology 145(3):197-207.
Demont,M., Tollens,E., and Fogarasi,J. Potential impact of biotechnology in Eastern
Europe: Transgenic maize, sugar beet and oilseed rape in Hungary. 2005a.
Leuven, Katholieke Universiteit Leuven. Working Paper.
Demont, M., Wesseler, J., and Tollens, E. (2004). Biodiversity versus transgenic
sugar beet: The one Euro question. European Review of Agricultural Economics
31(1):1-18.
Demont, M., Wesseler, J., and Tollens, E. (2005b). Irreversible costs and benefits of
transgenic crops: What are they? In J. Wesseler (ed.), Environmental Costs and
Benefits of Transgenic Crops. Dordrecht, NL: Springer, 113-122.
Desquilbet, M. and Bullock, D. S. (2009). Who pays the costs of non-GMO
segregation and identity preservation? American Journal of Agricultural
Economics 91(3):656-672.
Devadoss, S. and Kropf, J. (1996). Impacts of trade liberalizations under the Uruguay
Round on the world sugar market. Agricultural Economics 15(2):83-96.
Devos, Y., Cougnon, M., Vegucht, S., Bulcke, R., Haesaert, G., Steurbaut, W., and
Reheul, D. (2008a). Environmental impact of herbicide regimes used with
genetically modified herbicide-resistant maize. Transgenic Research 17:10591077.
Devos, Y., Demont, M., Dillen, K., Reheul, D., Kaiser, M., and Sanvido, O. (2009).
Coexistence of genetically modified (GM) and non-GM crops in the European
Union. A review. Agronomy for Sustainable Development 29:11-30.
Devos, Y., Demont, M., and Sanvido, O. (2008b). Coexistence in the EU - return of
the moratorium on GM crops? Nature Biotechnology 26(11):1223-1225.
Devos, Y., Reheul, D., Thas, O., De Clercq, E., Cougnon, M., and Cordemand, K.
(2007). Implementing isolation perimeters around genetically modified maize
fields. Agronomy for Sustainable Development 27:1-11.
Devos, Y., Thas, O., Cougnon, M., De Clercq, E., Cordemans, K., and Reheul, D.
(2008c). Feasibility of isolation perimeters for genetically modified maize on an
intra-regional scale in Flanders. Agronomy for Sustainable Development 28.
DeVuyst, C. S. and Wachenheim, C. J. (2005). American Crystal Sugar: Genetically
enhanced sugar beets? Review of Agricultural Economics 27(1):105-116.
Dewar, A. M., May, M. J., and Pidgeon, J. (2000). GM sugar beet - The present
situation. British Sugar Beet Review 68(2):22-27.
Dillen, K., Demont, M., and Tollens, E. (2009a). Corporate pricing strategies with
heterogeneous adopters: The case of HR sugar beet. AgBioForum 12(3&4):334345.
Dillen, K., Demont, M., and Tollens, E. (2008). European sugar policy reform and
agricultural innovation. Canadian Journal of Agricultural Economics 56:533553.
174
Dillen, K., Demont, M., and Tollens, E. (2009b). Global Welfare Effects of GM Sugar
Beet under Changing EU Sugar Policies. AgBioForum 12(1):119-129.
Dillen, K., Mitchell, P. D., and Tollens, E. (2010a). On the competitiveness of
different damage abatement strategies against Diabrotica virgifera virgifera: a
bio-economic approach. Journal of Applied Entomology 134(5):395-408.
Dillen,K., Mitchell, P. D., Van Looy, T., and Tollens, E. (2010b). The Western Corn
Rootworm, a new threat to European agriculture: Opportunities for
biotechnology? Pest Management Science, in press.
Dillen,K., Van Looy,T., and Tollens,E. Socio-economic assessment of controlling the
invasive species Diabrotica virgifera virgifera in Central Europe. 2009c.
Leuven, Division of Agricultural and Food Economics, Katholieke Universiteit
Leuven. Working paper.
Dixit, A. and Pindyck, R. S. (1994). Investment Under Uncertainty. Princeton:
Princeton University Press.
Dolezel, M., Pascher, K., and Grabherr, G. (2005). Regionality as a key parameter for
co-existence of genetically modified maize with conventional and organic
maize. In A. Messéan (ed.), Proceedings of the 2nd International Conference on
Co-Existence Between GM and Non-GM Based Agricultural Supply Chains.
Montpellier: Agropolis Productions, 203-206.
Dun, Z., Mitchell, P. D., and Agosti, M. (2010). Estimating Diabrotica virgifera
vrigifera damage functions with field data: Applying an unbalanced nested error
component model. Journal of Applied Entomology 134(5):409-419.
EC (2003). Commission Recommendation of 23 July 2003 on guidelines for the
development of national strategies and best practices to ensure the coexistence
of genetically modified crops with conventional and organic farming. Official
Journal of the European Communities L189:36-47.
EC. Report on the implementation of national measures on the coexistence of
genetically modified crops with conventional and organic farming. 2006.
Brussels, Commission of the European Communities. Communication from the
Commission to the Council and the European Parliament.
Eisley,B. Evaluation of soil applied and seed treatment insecticides for corn for
control of rootworm larvae, 2002. 2002. Department of Entomology, The Ohio
State University, Columbus.
Elbehri, A. and Macdonald, S. (2004). Estimating the impact of transgenic Bt cotton
on West and Central Africa: A general equilibrium approach. World
Development 32(12):2049-2064.
EPPO. Diabrotica virgifera. http://www.eppo.org. 15-11-2008.
European Commission. Sugar-Beet and Alternative Crops: Comparison of Costs and
Margins, by Region. 1999. European Commission. EC-DGVI.A3 - FADN
Results.
European Commission. Commission Decision of 24 October 2003 on emergency
measures to prevent the spread within the Community of Diabrotica virgifera Le
Conte (notified under document number C(2003) 3880) (2003/766/EC). 2003.
European Commission. Europeans and biotechnology in 2005: Pattern and trends.
2006a. Directorate-General for research. Eurobarometer.
European Commission. Final report of a mission in Poland to evaluate the measures
taken against Diabrotica virgifera virgifera. 2006b. Brussels.
European Commission. Prospect for agricultural markets and income in the European
Union 2007-2014. 2008. Brussels.
175
European Parliament (2006). Council Regulation (EC) No 318/2006 on the common
organisation of the markets in the sugar sector. Official Journal of the European
Communities.
European Parliament (2007). Council regulation EC 290/2007. Official Journal of the
European Communities.
European Union (2007). Treaty of Lisbon. Official Journal of the European Union
50(C306/1).
Eurostat. The use of plant protection products in the European Union (1992-2003).
2007. Luxemburg, Office of Offical Publications of the European Communities.
Eurostat. Agricultural production database. http://ec.europa.eu/eurostat. 26-11-2009.
F.O.Licht (2000). World Sugar Statistics 2001. Kent: F.O.Licht.
F.O.Licht (2005). World Sugar Yearbook 2006. Kent: F.O.Licht.
Faber, D. C., Molina, J. A., Ohlrichs, C. L., Vander Zwaag, D. F., and Ferro, L. B.
(2003). Commercialization of animal biotechnology. Theriogenology 59(1):125138.
Falck-zepeda, J. (2009). Socio-Economic Considerations, Article 26.1 of the
Cartagena Protocol on Biosafety: What are the Issues and What is at Stake?
AgBioForum 12(1):90-107.
Falck-Zepeda, J. B., Traxler, G., and Nelson, R. G. (2000). Surplus distribution from
the introduction of a biotechnology innovation. American Journal of
Agricultural Economics 82(2):360-369.
Fall,E.H. and Wesseler,J. Report on Environmental and Socio-Economic Analysis.
2008. Diabr-Act -Harmonise the strategies for fighting Diabrotica virgifera
virgifera.
FAO. Key statistics of food and agriculture external trade. FAO. 2006.
FAO (2007). FAOSTAT Agricultural Data. Rome: FAO.
FAO (2009). FAOSTAT Agricultural Data. Rome: FAO.
FAO investment centre/ EBRD cooperation programme. Serbia and Montenegro:
review of the Sugar Sector. 2004. Rome, European Bank for Reconstruction and
Development.
FAO investment centre/ EBRD cooperation programme. Ukraine: Review of the
sunflower oil sector. 2005. Rome, European Bank for Reconstruction and
Development.
Felsenthal, D. S. (2001). The Treaty of Nice and qualified majority voting. Social
Choice and Welfare 18(3):431.
Fernandez-Cornejo, J., Hendricks, C., and Mishra, A. (2005). Technology adoption
and off-farm household income: The case of herbicide-tolerant soybeans.
Journal of Agricultural and Applied Economics 37(3):549-563.
Fernandez-Cornejo,J. and McBride,W.D. Adoption of bioengineered crops. 2002.
Washington, Economic Research Service, USDA. Agricultural Economic
Report.
Flannery, M.-L., Thorne, F. S., Kelly, P. W., and Mullins, E. (2004). An economic
cost-benefit analysis of GM crop cultivation: An Irish case study. AgBioForum
7(4):149-157.
Foster, R. E., Tollefson, J. J., Nyrop, J., and Hein, G. L. (1986). Value of adult corn
rootworm population estimates in pest managment decision making. Journal of
Economic Entomology 79:303-310.
Frandsen, S. E., Jensen, H. G., Yu, W., and Walter-Jørgensen, A. (2003). Reform of
EU sugar policy: Price cuts versus quota reductions. European Review of
Agricultural Economics 30(1):1-26.
176
Fransen,L., La Vina,A., Dayrit,F., Gatlabayan,L., Santosa,D.A., and Adiwibowo,S.
Integrating socio-economic considerations into biosafety decisions: The role of
public participation. 2005. Washington DC, World Resource Institute. WRI
white paper.
Frisvold, G. B., Reeves, J. M., and Tronstad, R. (2006). Bt cotton adoption in the
United States and China: International trade and welfare effects. AgBioForum
9(2):69-78.
Frisvold, G. B., Sullivan, J., and Raneses, A. (2003). Genetic improvements in major
US crops: The size and distribution of benefits. Agricultural Economics 28:109119.
Fulton, M. and Giannakas, K. (2001). Agricultural biotechnology and industry
structure. AgBioForum 4(2):137-151.
Fulton,M. and Keyowski,L. The producer benefits of herbicide-resistant canola.
AgBioForum 1999.
Furlan, L., Canzi, S., Di Bernardo, A., and Edwards, C. R. (2006). The ineffectiveness
of insecticide seed coatings and planting-time soil insecticides as Diabrotica
virgifera virgifera LeConte population suppressors. Journal of Applied
Entomology 130(9/10):485-490.
Furtan, W. H., Guzel, A., and Weseen, A. S. (2007). Landscape clubs: Co-existence
of genetically modified and organic crops. Canadian Journal of Agricultural
Economics 55(2):185-195.
Gaskell, G., Allum, N., Wagner, W., Kronberger, N., Torgersen, H., Hampel, J., and
Bardes, J. (2004). GM foods and the misperception of risk perception. Risk
Analysis 24:185-194.
Gerber, C. K. (2003). Testing the areawide pest management concept on the western
corn rootworm (Diabrotica virgifera virgifera LeConte) in northwestern Indiana
and east-central Illinois. Purdue University.
Gianessi,L.P., Sankula,S., and Reigner,N. Plant biotechnology: Potential impact for
improving pest management in European agriculture, sugarbeet case study.
2003. Washington, DC, National Center for Food and Agricultural Policy.
Gianessi,L.P., Silvers,C.S., Sankula,S., and Carpenter,J.E. Plant biotechnology
current and potential impact for improving pest management in U.S. agriculture:
An analysis of 40 case studies. 2002. Washington, National Center for Food and
Agricultural Policy (NCFAP).
Gloyna, K. and Thieme, T. (2007). Cereals and weeds as potential host plants for
Diabrotica virgifera. IWGO Newsletter 28(1):44.
GMO safety. National cultivation bans on GM plants: An end to the political deadlock
in the EU? http://www.gmo-safety.eu/en/news/698.docu.html. 2009.
Gohin, A. and Bureau, J.-C. (2006). Modelling the EU sugar supply to assess sectoral
policy reforms. European Review of Agricultural Economics 33(2):223-247.
Goldsmith, P. D., Ramos, G., and Steiger, C. (2003). A tale of two businesses:
Intellectual property rights and the marketing of agricultural biotechnology.
Choices Third Quarter:25-30.
Gollier, C. and Treich, N. (2003). Decision-making under scientific uncertainty: the
economics of the precautionary principle. Journal of Risk and Uncertainty
27(1):77-103.
Gomez-Barbero, M., Berbel, J., and Rodriguez-Cerezo, E. (2008). Bt corn in Spain:
the performance of the EU's first GM crop. Nature Biotechnology 26(4):384386.
177
Goodwin, B. (2009). Payment limitations and acreage decisions under risk aversion:
A simulation approach. American Journal of Agricultural Economics 91(1):1941.
Goodwin, B. and Ker, A. P. (2002). Modeling price and yield risk. In R. Just and R.
Pope (eds.), A Comprehensive Assessment of the Role of Risk in U.S.
Agriculture. Norwell: Springer, 290-323.
Gouse, M., Pray, C. E., and Schimmelpfennig, D. E. (2004). The distribution of
benefits from Bt cotton adoption in South Africa. AgBioForum 7(4):187-194.
Gray, M. E., Sappington, T. W., Miller, N. J., Moeser, J., and Bohn, M. O. (2009).
Adaptation and Invasiveness of Western Corn Rootworm: Intensifying Research
on a Worsening Pest. Annual Review of Entomology 54(1):303-321.
Gray, M. E. and Steffey, K. L. (1998). Corn rootworm (Coleoptera: Chrysomelidae)
larval injury and root compensation of 12 maize hybrids: An assessment of the
economic injury index. Journal of Economic Entomology 91(3):723-740.
Greene, W. (2008). Econometric Analysis. Upper Sadle River, New Jersey: Pearson
Prentice Hall.
Griffiths, W. and Zhao, X. Y. (2000). A unified approach to sensitivity analysis in
equilibrium displacement models: Comment. American Journal of Agricultural
Economics 82(1):236-240.
Griliches, Z. (1957). Hybrid corn: An exploration in the economics of technological
change. Econometrica 25(4):501-522.
Gruère,G. and Sengupta,D. Biosafety decisions and perceived commercial risks. The
role of GM-free private standards. 2009. Washington DC, International food
policy research institute. IFPRI Discussion Paper.
Haab, T. C. and McConnel, K. E. (2002). Valuing Environmental and Natural
Resources: The Econometrics of Non-Market Valuation. Cheltenham UK:
Edwar Elgar.
Hanemann, M., Loomis, J., and Kanninen, B. (1991). Statistical Efficiency of DoubleBounded Dichotomous Choice Contingent Valuation. American Journal of
Agricultural Economics 73(4):1255-1263.
Hareau, G. G., Mills, B. F., and Norton, G. W. (2006). The potential benefits of
herbicide-resistant transgenic rice in Uruguay: Lessons for small developing
countries. Food Policy 31(2):162-179.
Hatala Zsellér,I., Ripka,G., and Vörös,G. Some economic aspects of Western Corn
Rootworm for the Hungarian maize production. IWGO 22nd conference 2006.
Vienna.
He, H., Dezonia, B., and Mladenoff, D. (2000). An aggregation index (AI) to quantify
spatial patterns of landscapes. Landscape Ecology 15(7):591-601.
Hennessy, D. A. and Moschini, G. (2006). Regulatory actions under adjustment costs
and the resolution of scientific uncertainty. American Journal of Agricultural
Economics 88(2):308-323.
Hermann, O. (1996). Enquête sur les parasites et sur le coût des traitements
phytosanitaires dans la culture de la betterave sucrière en Europe. C.I.B.E.
XXXVIth Congress Proceedings. Paris: C.I.B.E., 1-8.
Hermann, O. (1997). Enquête sur le coût de la lutte phytosanitaire dans la culture de
la betterave sucrière en Europe. IIRB 60th Congress Proceedings. Brussels:
IIRB, 469-475.
Hermann, O. (2006). Survey on pests and diseases in sugar beet growing and the cost
of plant protection products in Europe. CIBE Congress 2006. Cracow.
178
Herriges, J. A. and Shogren, J. F. (1996). Starting Point Bias in Dichotomous Choice
Valuation with Follow-Up Questioning. Journal of Environmental Economics
and Management 30(1):112-131.
Herring, R. (2008). Resolving Conflicting Evidence on 'GMOs' in Rural India: The
Warangal Method. International Journal of Multiple Research Approaches 2(2).
Hinrichs, J., Musshoff, O., and Odening, M. (2008). Economic hysteresis in hog
production. Applied Economics 40(3):333-340.
Horak,A., Ayoub,S., Nowakowski,R., Peczze,R., and Skillman,S. Cruiser seed
treatment, Force granular insecticide and Karate with Zeon technology foliar
spray as components of an integrated program against Corn Rootworm. Closing
EU-project meeting DIABRACT. 2008. Göttingen.
Hu,R., Huang,J., Lin,H., Rozelle,S., and Pray,C.E. Bt cotton in China: Are secondary
insect infestations offsetting the benefits in farmer fields? 2006. 10th ICABR
International Conference, Ravello (Italy), June 29-July 2.
Huang, J., Hu, R., Rozelle, S., and Pray, C. E. (2005). Insect-Resistant GM Rice in
Farmers' Fields: Assessing Productivity and Health Effects in China. Science
308:688-690.
IPCC (2001). Climate Change 2001: Impacts, Adaptation and Vulnerability.
Contribution of Working Group II to the Third Assessment Report of the
Intergovernamental Panel on Climate Change. Cambridge: Cambridge
university press.
IRS. Bietenstatistiek 2004. 2004. Bergen Op Zoom, Netherlands.
ITB. 2000. Données extraites des enquêtes itinéraires culturales réalisées par l'ITB.
Personal Communication.
Jaffe, G. (2005). Implementing the Cartagena Biosafety Protocol through national
biosafety regulatory systems: an analysis of key unresolved issues. Journal of
Public Affairs 5(3&4):299-311.
James, C. (2009). Global status of commercialized GM crops:2008. ISAA Brief 39.
James,C. Global status of commercialized biotech/GM crops: 2009. 2010. ISAAA.
Johnson, N. L. and Ruttan, V. W. (1994). Why Are Farms So Small? World
Development 22:691-706.
Just, R. (2006). Anticompetitive impacts of laws that regulate commercial use of
agricultural biotechnologies in the United States. In R. Just, J. M. Alston, and D.
Zilberman (eds.), Regulating Agricultural Biotechnololy: Economics and
Policy. New York: Springer, 353-396.
Just, R., Hueth, D., and Smith, A. (2004). The Welfare Economics of Public Policy.
Cheltenham,UK: Edward Elgar.
Just, R. E. and Pope, R. (2002). A Comprehensive Assessment of the Role of Risk in
U.S. Agriculture. Boston: Kluwer.
Kalaitzandonakes, N., Alston, J. M., and Bradford, K. J. (2007). Compliance costs for
regulatory approval of new biotech crops. Nat Biotech 25(5):509-511.
Kalaitzandonakes, N. G. (2003). The Economic and Environmental Impacts of
Agbiotech: A Global Perspective. New York: Kluwer Academic/Plenum
Publishers.
Kikulwe, E. (2010). On the introduction of Genetically Modified Bananas in Uganda.
Social benefits, Costs, and consumer Preferences. Wageningen University.
Kiss, J., Komaromi, J., Bayar, K., Edwards, C. R., and Hatala Zsellér, I. (2005).
Western corn rootworm and the crop rotation systems in Europe. In S. Vidal, U.
Kuhlmann, and C. R. Edwards (eds.), Western Corn Rootworm Ecology and
Managment. Wallingford: CABI Publishing, 189-220.
179
Kniss, A. R., Wilson, R. G., Martin, A. R., Burgener, P. A., and Feuz, D. M. (2004).
Economic evaluation of glyphosate-resistant and conventional sugar beet. Weed
Technology 18(2):388-396.
Koch, H. J. (2009). Yield decrease in sugar beet caused by reduced tillage and direct
drilling. European Journal of Agronomy 30(2):101.
Krinsky, I. and Robb, A. L. (1986). On Approximating the Statistical Properties of
Elasticities. The Review of Economics and Statistics 68(4):715-719.
Krishna, V. V. and Qaim, M. (2007). Estimating the adoption of Bt eggplant in India:
Who Benefits from public-private partnership? Food Policy 32(5-6):523-543.
Kurzer, P. and Cooper, A. (2007a). Consumer Activism, EU Institutions and Global
Markets: The Struggle over Biotech Foods. Journal of Public Policy
27(02):103-128.
Kurzer, P. and Cooper, A. (2007b). What's for Dinner? European Farming and Food
Traditions Confront American Biotechnology. Comparative Political Studies
40(9):1035-1058.
KWS. 2006. Price prognosis of RR sugar beet in the USA in 2008. Personal
Communication.
Lapan, H. E. and Moschini, G. (2004). Innovation and trade with endogenous market
failure: The case of genetically modified products. American Journal of
Agricultural Economics 86(3):634-648.
Lau, H. S., Lau, A. H.-L., and Ho, C. J. (1998). Improved Moment-Estimation
Formulas Using More Than Three Subjective Fractiles. Management Science
44(3):346-351.
Leech, D. (2002). Designing the voting system for the Council of the European
Union. Public Choice 113:437-464.
Lemarié,S., Desquilbet,M., Diemer,A., Marette,S., Levert,F., Carrère,M., and
Bullock,D.S. Les répartitions possibles entre les acteurs de la filière agroalimentaire des gains éventuels tirés des plantes transgéniques en France. 2001.
Rennes, INRA. Rapport au commissariat Général du Plan.
Lemarié, S. and Marette, S. (2003). Substitution and complementarities in the
Biotechnology and pesticide markets: A theoretical framework. In N. G.
Kalaitzandonakes (ed.), The Economic and Environmental Impacts of
Agbiotech: A Global Perspective. New York: Kluwer Academic/Plenum
Publishers, 287-305.
Lerman,Z., Sedik,D., Pugachov,N., and Goncharuk,R. Rethinking agricultural reform
in Ukraine. 2007. Halle,Germany, IAMO. Studies on the agricultural and and
food section in Central and Eastern Europe.
Levine, E. and Oloumi-Sadeghi, H. (1992). Managment of diabrotici rootworms in
corn. Annual Review of Entomology 36:229-255.
Levine, E. and Oloumi-Sadeghi, H. (1996). Western corn rootworm larval injury to
corn grown for seed production following soybeans for seed production.
Journal of Economic Entomology 89:1010-1016.
Lieberman, S. (2006). The so-called 'moratorium' on the licensing of new genetically
modified(GM) products by the European Union 1998-2004: a study in
ambiguity. Environmental Politics 15(4):592.
Ma, B. L., Meloche, F., and Wei, L. (2009). Agronomic assessment of Bt trait and
seed or soil-applied insecticides on the control of corn rootworm and yield.
Field Crops Research 111(3):189-196.
Macleod,A., Baker,R., Cheek,S., Eyere,D., and Cannon,R. Pest risk analysis for
western corn rootworm. 2007. York,UK, Central Science Laboratory.
180
Marks, L. A., Kalaitzandonakes, N. G., Allison, K., and Zakharova, L. (2002). Time
Series Analysis of Risk Frames in Media Communication of
Agrobiotechnology. In V. Santaniello, R. E. Evenson, and D. Zilberman (eds.),
Market Development for Genetically Modified Foods. Wallingford: CAB
International, 217-225.
Marks, L. A., Kalaitzandonakes, N. G., Allison, K., and Zakharova, L. (2003). Media
Coverage of Agrobiotechnology: Did the Butterfly Have an Effect? Journal of
Agribusiness 21(1):1-20.
Märländer, B. (2005). Weed control in sugar beet using genetically modified
herbicide-tolerant varieties - A review of the economics for cultivation in
Europe. Journal of Agronomy & Crop Science 191:64-74.
Marra, M., Hubbel, B. J., and Carlson, G. A. (2001). Information quality, technology
depreciation and Bt cotton adoption in the Southeast. Journal of Agricultural
and Resource Economics 26(1):158-175.
Marra, M. C. and Piggott, N. E. (2006). The value of non-pecuniary characteristics of
crop biotechnologies: A new look at the evidence. In R. E. Just, J. M. Alston,
and D. Zilberman (eds.), Regulating Agricultural Biotechnology: Economics
and Policy. New York: Springer, 145-177.
Marra,M.C., Piggott,N.E., and Carlson,G.A. The net benefits, including convenience,
of Roundup Ready® soybeans: Results from a national survey. 2004. Raleigh,
NC, NSF Center for Integrated Pest Management. Technical Bulletin.
Masten, S. E., Meehan, J. W., Jr., and Snyder, E. A. (1991). The Costs of
Organization. Journal of Law, Economics, & Organization 7(1):1-25.
May, M. J. (2003). Economic consequences for UK farmers of growing GM herbicide
tolerant sugar beet. Annals of Applied Biology 142(1):41-48.
Meinke, Lance J., Brown, J., Campbell, B., and McCormick, B. 2003 corn rootworm
soil
insecticide
evaluation.
http://entomology.unl.edu/fldcrops/trials/crwsoilmead03.pdf. 2003.
Meinke, Lance J., Brown, J., Campbell, B., and McCormick, B. 2004 corn rootworm
soil
insecticide
evaluation.
http://entomology.unl.edu/fldcrops/trials/crwsoilmead04.pdf. 2004a.
Meinke, Lance J., Brown, J., Campbell, B., and McCormick, B. 2004 corn rootworm
soil
insecticide/seed
treatment
efficacy
and
yield
experiment.
http://entomology.unl.edu/fldcrops/trials/crwmeadseed04.pdf. 2004b.
Meinke, Lance J., Brown, J., Stebbing, J., Campbell, B., and McCormick, B. 2002
corn
rootworm
soil
insecticide
evaluation.
http://entomology.unl.edu/fldcrops/trials/crwsoilmead02.pdf. 2002.
Meinke, Lance J., Brown, J., Stebbing, J., and McCormick, B. 2000 corn rootworm
seed
treatment
evaluations.
http://entomology.unl.edu/fldcrops/trials/crwseedmead00.pdf. 2000a.
Meinke, Lance J., Brown, J., Stebbing, J., and McCormick, B. 2000 corn rootworm
soil
insecticide
evaluations.
http://entomology.unl.edu/fldcrops/trials/crwsoilmead2000.pdf. 2000b.
Meinke, Lance J., Brown, J., Stebbing, J., and McCormick, B. 2001 corn rootworm
soil
insecticide
evaluations.
http://entomology.unl.edu/fldcrops/trials/crwmead01.pdf. 2001.
Meinke, L. J., Siegfried, B. D., Wright, R. J., and Chandler, L. D. (1998). Adult
susceptibility of western corn rootworm (colepotera: Chrysomelida) populations
to selected insecticides. Journal of Economic Entomology 91:594-600.
181
Messéan,A., Angevin,F., Gómez-Barbero,M., Menrad,K., and Rodríguez-Cerezo,E.
New case studies on the coexistence of GM and non-GM crops in European
agriculture. 2006. European Communities. Technical Report.
Metcalf, R. (1986). Foreword. In J. Krysan and T. Miller (eds.), Methods for the Study
of Pest Diabrotica. New York: Springer Verlag, vii-xv.
Miller, G. (1956). The Magical Number Seven, Plus or Minus Two: Some Limits on
our Capacity for Processing Information. Psychological Review 63:81-97.
Mitchell,P.D. Yield Benefit of Corn Event MON863. Faculty paper series 2002.
Department of Agricultural economics Texas A&M University.
Mitchell, P. D. and Hutchison, W. D. (2008). Assessing and communicating the value
and risk of IPM. In E. B. Radcliffe and W. D. Hutchison (eds.), Integrated Pest
Managment. Cambridge, UK: Camebridge University Press.
Mitchell, P. D., Gray, M. E., and Steffey, K. L. (2004). A Composed-error Model for
Estimating Pest-damage Functions and the Impact of the Western Corn
Rootworm Soybean Variant in Illinois. American Journal of Agricultural
Economics 86(2):332-344.
Moeser, J. and Vidal, S. (2004). Do Alternative Host Plants Enhance the Invasion of
the Maize Pest Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae,
Galerucinae) in Europe? Environmental Entomology 33:1169-1177.
Moeser, J. and Vidal, S. (2005). Nutritional resources used by the invasive maize pest
Diabrotica virgifera virgifera in its new South-east-European distribution range.
Entomologia Experimentalis Et Applicata 114(1):55-63.
Monsanto. 2007. Personal Communication.
Moon, W. and Balasubramanian, S. (2004). Public Attitudes toward
Agrobiotechnology: The Mediating Role of Risk Perceptions on the Impact of
Trust, Awareness, and Outrage. Review of Agricultural Economics 26(2):186208.
Mooney, S. and Klein, K. K. (1999). Environmental Concerns and Risk of Genetically
Modified Crops: Economic Contributions to the Debate. Canadian Journal of
Agricultural Economics 47:437-444.
Morel, B., Farrow, S., Wu, F., and Casman, E. (2003). Pesticide resistance, the
precautionary principle, and the regulation of Bt corn: real option and rational
option approaches to decision-making. In R. Laxminarayan (ed.), Battling
Resistance to Antibiotics. An Economic Approach. Washington, DC: Resources
for the Future, 184-213.
Moschini, G. (2006). Pharmaceutical and Industrial Traits in Genetically Modified
Crops: Coexistence with Conventional Agriculture. American Journal of
Agricultural Economics 88(5):1184-1192.
Moschini, G. and Lapan, H. E. (1997). Intellectual property rights and the welfare
effects of agricultural R&D. American Journal of Agricultural Economics
79(4):1229-1242.
Moschini, G., Lapan, H. E., and Sobolevsky, A. (2000). Roundup Ready soybeans
and welfare effects in the soybean complex. Agribusiness 16(1):33-55.
Munro, A. (2008). The spatial impact of genetically modified crops. Ecological
Economics 67(4):658-666.
Muška,F. WCR in the Czech Republic - possibilities protection maize. Closing EU
project meeting DIABRACT. 2008. Göttingen.
Musshoff, O. and Hirschauer, N. (2008). Investment planning under uncertainty and
flexibility: the case of a purchasable sales contract. Australian Journal of
Agricultural Research 52(1):17-36.
182
Nagy, B., Szentkirályi, F., and Vörös, G. (1999). Changes in the pest status within
maize insect assemblages in the Carpathian Bassin. In E. S. Pego (ed.),
Proceedings of the XIXth Conference of the International Working Group on
Ostrinia. IWGO, 223-235.
Naseem, A. and Pray, C. (2004). Economic impact analysis of genetically modified
crops. In P. Christou and H. Klee (eds.), Handboook of Plant Biotechnolgy.
Chichester, UK: John Wiley and Son, 959-991.
NASS. Acreage. http://www.usda.gov/nass/PUBS/TODAYRPT/acrg0609.pdf. 30-62009.
National Institute for Statistics Romania. Agricultural statistics. www.insee.ro. 2009.
Ndeffo Mbah,M., Forster, G., Wesseler, J., and Gilligan, C. (2010). Economically
optimal timing for pest-control under uncertainty: an options approach. Journal
of the Royal Society Interface, in press.
Nerlove, M. L. (1996). Reflections on the Economic Organization of Agriculture:
Traditional, Modern, and Transitional. In D. Martimort (ed.), Agricultural
Markets: Mechanisms, Failures, and Regulations. Amsterdam: Elsevier, 9-30.
Nielsen, C. P. (2001). Global market effects of alternative European responses to
genetically modified organisms. World Economics Review 137(2):320-348.
Nolte, S. (2008). The future of the world sugar market- A spatial price equilibrium
analysis. European Association of Agricultural Economists 107th Seminar 2008.
OANDA. OANDA: The currency site. www.oanda.com. 2008.
Odening, M., Musshoff, O., and Balman, A. (2005). Investment decisions in hog
finishing: an application of the real options approach. Agricultural Economics
32(1):47-60.
Oehmke, J. and Wolf, C. (2004). Is monsanto leaving money on the table? Monopoly
pricing and BT cotton value with heterogeneous adopters. Journal of
Agricultural & Applied Economics 36(3):705-718.
Oleson, J. D., Park, Y. L., Nowatzki, T. M., and Tollefson, J. J. (2005). Node-Injury
Scale to Evaluate Root Injury by Corn Rootworms (Coleoptera:
Chrysomelidae). Journal of Economic Entomology 98(1):1-8.
Olson, K. D. (2004). Farm Management: Principles and Strategies. Ames, Iowa:
Iowa State Press.
Perry, J. N. (2002). Sensitive dependencies and separation distances for genetically
modified herbicide-tolerant crops. Proceedings of the Royal Society of
London.Series B: Biological Sciences 269:1173-1176.
Phillips, P. W. B. (2003). The economic impact of herbicide tolerant canola in
Canada. In N. G. Kalaitzandonakes (ed.), The Economic and Environmental
Impacts of Agbiotech: A Global Perspective. New York: Kluwer
Academic/Plenum Publishers, 119-139.
Phipps, R. H. and Park, J. R. (2002). Environmental benefits of genetically modified
crops: Global and European perspectives on their ability to reduce pesticide use.
Journal of Animal and Feed Sciences 11:1-18.
Pietola, K. S. and Wang, H. H. (2000). The value of price- and quantity- fixing
contracts for piglets in Finland. European Review of Agricultural Economics
27(4):431-447.
Piggott, R. and Marra, M. (2008). Biotechnology Adoption Over Time In the
Presence of Non-Pecuniary Characteristics that Directly Affect Utility: A
Derived Demand Approach. AgBioForum 11(1):58-70.
183
Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O'Connell, C.,
Wong, E., Russel, L., Zern, J., Aquino, T., and Tsomondo, T. (2001). Economic
and environmental threats of alien plant, animal, and microbe invasions.
Agriculture, Ecosystems & Environment 84(1):1-20.
Polish Statistics. Agricultural Yearbook. 2008. Warsaw, Poland.
Poonyth, D. (1998). A structural econometric model of European Union sugar
industry and the potential implications of the GATT WTO. Faculty of the
Graduate School, University of Missouri-Columbia.
Poonyth, D., Westhoff, P., Womack, A., and Adams, G. (2000). Impacts of WTO
restrictions on subsidized EU sugar exports. Agricultural Economics 22(3):233245.
Pray, C. E., Huang, J., Hu, R., and Rozelle, S. (2002). Five years of Bt cotton in China
- the benefits continue. The Plant Journal 31(4):423-430.
Pretty, J., Brett, C., Gee, D., Hine, R., Mason, C., Morison, J., Rayment, M., van der
Bijl, G., and Dobbs, T. (2001). Policy challenges and priorities for internalizing
the externalities of modern agriculture. Journal of Environmental Planning and
Management 44(2):263-283.
Protection, S. (2009). References to other topics. Environmental Policy and Law
39(2):128.
Purvis, A., Boggess, W. G., Moss, C. B., and Holt, J. (1995). Adoption of emerging
technologies under output uncertainty: an ex ante approach. American Journal
of Agricultural Economics 77(3):541-551.
Qaim, M. (2009). The economics of gentically modified crops. Annual Review of
Resource Economics 1:665-693.
Qaim, M. and de Janvry, A. (2003). Genetically modified crops, corporate pricing
strategies, and farmers' adoption: The case of Bt cotton in Argentina. American
Journal of Agricultural Economics 85(4):814-828.
Qaim, M., Subramanian, A., Naik, G., and Zilberman, D. (2006). Adoption of Bt
cotton and impact variability: Insights from India. Review of Agricultural
Economics 28(1):45-58.
Qaim, M. and Traxler, G. (2005). Roundup Ready soybeans in Argentina: Farm level
and aggregate welfare effects. Agricultural Economics 32:73-86.
Qaim, M. and Zilberman, D. (2003). Yield Effects of Genetically Modified Crops in
Developing Countries. Science 299:900-902.
Rahim, A. H., van Ierland, E. C., and Wesseler, J. (2007). Economic incentives for
abandoning or expanding gum arabic production in Sudan. Forest Policy and
Economics 10(1-2):36-47.
Rasmusson, A. (1998). Vad finns i betodlarens ryggsack? [What is in the beetgrowers rucksack?]. Betodlaren 4:21-23.
Rice, C. A., Mesbah, A., and Miller, S. D. (2001). Economic evaluation of weed
management systems in sugarbeets. Proceedings of the American Society of
Sugar Beet Technologists 31:64.
Rice, M. E. and Oleson, J. D. (2005). Three year summary of corn rootworm control
products. Integrated Crop Managment 494(26):215-217.
Ripka,G. 2008. State Phytosanitary institute, Hungary. Personal Communication.
Rizov, M. (2003). An essay on the agricultural production organization in former
communist countries. The Social Science Journal 40(4):665-672.
Romaneckas, K., Romaneckiene, R., and Pilipavicius, V. (2009). Non-chemical weed
control in sugar beet under intensive and conservation soil tillage: I. Crop
weediness. Agronomy Research 7:457-464.
184
Rosca, I. (2006). Corn and Diabrotica virgifera virgifera Le Conte in Romania. In G.
Governatorti (ed.), Integrated Pest Management for Corn: a Necessity or an
Opportunity? The Case of Western Corn Rootworm. Rome: FAO, 56-58.
Roumasset, J. (1995). The nature of the agricultural firm. Journal of Economic
Behavior & Organization 26(2):161-177.
Sanvido, O., Romeis, J., and Bigler, F. (2007). Ecological Impacts of Genetically
Modified Crops: Ten Years of Field Research and Commercial Cultivation.
Advances in Biochemical Engineering/Biotechnology 107(235):278.
Sanvido, O., Widmer, F., Winzeler, M., Streit, B., Szerencsits, E., and Bigler, F.
(2008). Definition and feasibility of isolation distances for transgenic maize
cultivation. Transgenic Research 17(3):317-335.
Scatasta,S., Wesseler,J., and Nillesen,E. Real options and the adoption of transgenic
crops: An intertemporal perspective. 2006. 10th ICABR International
Conference, Ravello (Italy), June 29-July 2.
Schaafsma, A. W., Baufeld, P., and Ellis, C. R. (1999). Influence of cropping
practices on corn rootworm in Canada as a basis for assessment of the potential
impact of Diabrotica virgifera in Germany. EPPO Bulletin 29(1-2):145-154.
Schäufele, W. R. (2000). Chemische Unkrautbekämpfung in Zuckerrüben im Wandel
- Ergebnisse einer Befragung in der IIRB-Arbeitsgruppe "Unkrautregulierung".
IIRB 63rd Congress Proceedings. Brussels: IIRB, 93-109.
Schmalensee, R. (1981). Output and Welfare Implications of Monopolistic ThirdDegree Price Discrimination. The American Economic Review 71(1):242-247.
Schmitt, G. (1991). Why is the agriculture of advanced Western economies still
organized by family farms? Will this continue to be so in the future? European
Review of Agricultural Economics 18(3/4):443-458.
Schmitz, A. (2002). The European Union's high-priced sugar support-regime. In A.
Schmitz, T. H. Spreen, W. A. Messina, and C. B. Moss (eds.), Sugar and
Related Sweetener Markets: International Perspectives. CABI publishing, 193213.
Scholderer, J. (2005). The GM foods debate in Europe: history, regulatory solultions,
and consumer respons research. Journal of Public Affairs 5:263-274.
Scot, J. (2000). Rational choice theory. In G. Browning, A. Halcli, N. Hewlett, and F.
Webster (eds.), Understanding Contempory Society: Theories of the Present.
London: Sage.
Sheingate, A. D. (2009). Federalism and the Regulation of Agricultural
Biotechnology in the United States and European Union. Journal of
Comparative Policy Analysis: Research and Practice 11(4):477-497.
Shelton, A., Naranjo, S., Romeis, J., Hellmich, R., Wolt, J., Federici, B., Albajes, R.,
Bigler, F., Burgess, E., Dively, G., Gatehouse, A., Malone, L., Roush, R., Sears,
M., and Sehnal, F. (2009). Setting the record straight: a rebuttal to an erroneous
analysis on transgenic insecticidal crops and natural enemies. Transgenic
Research 18(3):317-322.
Short, C. and Gitu, K. W. (1991). Timeliness costs for machinery selection. Canadian
Journal of Agricultural Economics 39:457-462.
Simic, D., Ivezic, M., Brkic, I., Raspudic, E., Brmez, M., Majic, I., Brkic, A.,
Ledencan, T., Tollefson, J. J., and Hibbard, B. E. (2008). Environmental and
Genotypic Effects for Western Corn Rootworm Tolerance Traits in American
and European Maize Trials. Maydica 52(425):430.
Sivcev,I. 2008. WCR problem in Serbia. Personal Communication.
185
Sivcev, I., Stankovic, S., Kostic, M., Ladic, N., and Popovic, Z. (2009). Population
density of Diabrotica virgifera virgifera Leconte beetles in Serbian first year and
continous maize fields. Journal of Applied Entomology 133(6):430-437.
Skevas, T., Wesseler, J., and Fevereiro, P. (2009). Coping with Ex-ante Regulations
for Planting Bt Maize: The Portuguese Experience. AgBioForum 12(1):60-69.
Smith, G. L. (2007). Sugar Policy Reform in the European Union and in World Sugar
Markets. Paris, France: OECD.
Sobolevsky, A., Moschini, G., and Lapan, H. E. (2005). Genetically modified crop
innovations and product differentiation: Trade and welfare effects in the
soybean complex. American Journal of Agricultural Economics 87(3):621-644.
SPA. State Phytosanitary Administration Czech Republic. 2009.
Statistical Office of the Republic of Serbia (2007). Statistical Yearbook. Belgrade,
Serbia.
Stein, A. J. and Rodriguez-Cerezo, E. (2010). International trade and the global
pipeline of new GM crops. Nature Biotechnology 28(1):23-25.
Stiglitz, J. E. (1974). Incentives, Risk , and Information: Notes Towards a Theory of
Hierarchy. The Bell Journal of Economics 6(2):552-579.
SUBEL. 2007. Personal Communication.
Swanton, C. J., Weaver, S., Cowan, P., Van Acker, R., Deen, W., and Shreshta, A.
(1999). Weed Thresholds -- Theory and Applicability. Journal of Crop
Production 2(1):9-29.
Swinnen, J. F. M. and De Gorter, H. (1998). Endogenous commodity policies and the
social benefits from public research expenditures. American Journal of
Agricultural Economics 80(2):107-115.
Szõke, C., Zsubori, Z., Pók, I., Rácz, F., Illés, O., and Szegedi, I. (2002). Significance
of the European corn borer (Ostrinia nubilalis Hübn.) in maize production. Acta
Agronomica Hungarica 50(4):447-461.
Takacs,J., Balogh,P., and Nadasy,M. Rapid forecasting of the Western Corn
Rootworm (Diabrotica virgifera virgifera LeConte), a pest of sweetcorn in
Hungary. Bulletin OILB/SROP.International Organization for Biological and
Integrated Control of Noxious Animals and Plants (OIBC/OILB), West
Palaearctic Regional Section (WPRS/SROP) 2007. Dijon, France.
Thelen, M. (2004). What are the main forces operating on the world market? Courier
1.
Toth, V. (2005). Protection against Western Corn Rootworm Adults (Diabrotica
vigifera vrigifera) in Banranya county (Hungary). Journal of Central European
Agriculture 6(3):309-316.
Traxler, G., Godoy-Avila, S., Falck-Zepeda, J. B., and Espinoza-Arellano, J. d. J.
(2003). Transgenic cotton in Mexico: A case study of the Comarca Lagunera. In
N. G. Kalaitzandonakes (ed.), The Economic and Environmental Impacts of
Agbiotech: A Global Perspective. New York: Kluwer Academic/Plenum
Publishers, 183-202.
Trnka, M., Muška, F., Semeradova, A., Dubrovsky, M., Kocmankova, E., and Zalud,
Z. (2007). European Corn Borer life stage model: Regional estimates of pest
development and spatial distribution under present and future climate.
Ecological Modelling 207:61-84.
Urban, J., Pulkrabek, J., Valenta, J., Beckova, L., and Kviz, Z. (2008). Influence of
lower and more frequent doses of herbicides on yield and technological quality
of sugar beet. 43rd Croatian and 3rd International Symposium on Agriculture
Proceedings. Opatija, Croatia: 646-649.
186
Urias-Lopez, M. and Meinke, L. J. (2001). Influence of Western Corn Rootworm
(Coleoptera: Chrysomelidae) Larval Injury on Yield of Different Types of
Maize. Journal of Economic Entomology 94(1):106-111.
USDA. Adoption of Genetically Engineered Crops in the U.S. United States
Department of Agriculture. 2006a.
USDA (2006b). EU25 sugar semi anual. Gain Report.
Useche, P., Barham, B., and Foltz, J. (2009). Integrating technology traits and
producer heterogeneity: A mixed-multinomial model of genetically modified
corn adoption. American Journal of Agricultural Economics 91(2):444-461.
Van Mellor,T., Alexander,C., Bledsoe,L., and Krupke,C. An economic analysis of the
western corn rootworm. American Agricultural Economics Association Anual
meeting 2006. Long Beach, California, USA.
von Braun,J. The world food situation: new driving forces and required action. Food
Policy Report 2007. Washington DC, International Food Policy Research
Institute.
Ward, D., DeGooyer, T., Vaughn, T., Head, G., McKee, M., Astwood, J., and
Pershing, J. (2005). Genetically enhanced maize as a potential mangament
option for corn rootworm: YieldGard rootworm maize case study. In S. Vidal,
U. Kuhlmann, and C. R. Edwards (eds.), Western Corn Rootworm: Ecology and
Management. Wallingford: CABI publishing, 239-262.
Weaver, R. D. (2004). R&D incentives for GM seeds: restricted monopoly, nonmarket effects, and regulation. In R. E. Evenson and V. Santaniello (eds.), The
Regulation of Agricultural Biotechnology. 143-151.
Weaver, R. D. and Wesseler, J. (2004). Monopolistic pricing power for transgenic
crops when technology adopters face irreversible benefits and costs. Applied
Economics Letters 11:969-973.
Wesseler, J. (2005). Environmental costs and benefits of transgenic crops. In J.
Wesseler (ed.), Environmental Costs and Benefits of Transgenic Crops.
Dordrecht, NL: Springer.
Wesseler, J. and Fall, E. H. (2010). Potential damage costs of Diabrotica virgifera
virgifera infestation in Europe - the "no control" scenario. Journal of Applied
Entomology 134(5):385-395.
Wesseler, J., Scatasta, S., and Nillesen, E. (2007). The Maximum Incremental Social
Tolerable Irreversible Costs (MISTICs) and other benefits and costs of
introducing transgenic maize in the EU-15. Pedobiologia 51(3):261-269.
West, T. O. and Post, W. M. (2002). Soil organic carbon sequestration rates by tillage
and crop rotation: a global data analysis. Soil Science Society of America
Journal 66:1930-1946.
Wevers, J. D. A. (1998). Agronomic and environmental aspects of herbicide-resistant
sugar beet in the Netherlands. Aspects of Applied Biology 52:393-399.
Winter-Nelson, A. and Amegbeto, K. (1998). Option Values to Conservation and
Agricultural Price Policy: Application to Terrace Construction in Kenya.
American Journal of Agricultural Economics 80(2):409-418.
World Bank. World development indicators. www.worldbank.org. 2008.
Wu, F. (2006). Bt corn's reduction of mycotoxins: regulatory decisions and public
opinion. In R. Just and D. Zilberman (eds.), Regulating Agricultural
Biotechnology: Economics and Policy. New York: Springer, 179-200.
Zilberman, D., Ameden, H., and Qaim, M. (2007). The impact of agricultural
biotechnology on yields, risks, and biodiversity in low-income countries.
Journal of Development Studies 43:63-78.
187
Zorya,S. and von Cramon-Taubadel,S. (When) will Ukraine be a global player on
world agricultural markets. Xth EAAE congress. 2006. Zaragoza Spain.
188
List of Publications
International Scientific Journals
Dillen, K., P.D. Mitchell, T. Van Looy & E. Tollens. “The Western Corn Rootworm, a
new threat to European agriculture: Opportunities for biotechnology?”, Pest
Management Science, in press.
Demont, M., K. Dillen, W. Daems, C. Sausse, E. Tollens & E. Mathijs. 2010. “On the
Proportionality of EU Spatial ex ante Coexistence Regulations: Reply.” Food Policy,
35(2), 183-184.
Dillen, K., P.D. Mitchell & E. Tollens. “On the Competitiveness of Diabrotica virgifera
virgifera Damage Abatement Strategies in Hungary: a Bio-economic Approach”,
Journal of Applied Entomology, 134(5):395-408.
Dillen, K., M. Demont & E. Tollens. 2009. “Corporate Pricing Strategies with
Heterogeneous Adopters: The Case of Herbicide Resistant Sugar Beet”, AgBioForum,
12(3&4): 334-345.
Dillen, K., M. Demont & E. Tollens. 2009. “Potential Economic Impact of GM Sugar
Beet in the Global Sugar Sector.” International Sugar Journal, 111 (1330): 638-643.
Demont, M., K. Dillen, W. Daems, C. Sausse, E. Tollens & E. Mathijs. 2009. “On the
proportionality of EU spatial ex ante coexistence regulations”, Food Policy, 34: 508518.
Dillen, K., M. Demont & E. Tollens. 2009. “Global Welfare Effects of GM Sugar Beet
under Changing EU Sugar Policies”, AgBioForum,12(1):119-129.
Devos, Y., M. Demont, K. Dillen, D. Reheul, M. Kaiser & O. Sanvido. 2009.
“Coexistence of GM and Non-GM Crops in the EU. A Review” Agronomy for
Sustainable Development, 29:11-30.
Dillen, K., M. Demont & E. Tollens. 2008. “European Sugar Policy Reform and
Agricultural Innovation.” Canadian Journal of Agricultural Economics, 56:533-553.
Demont, M. & K. Dillen. 2008. “Herbicide Tolerant Sugar Beet: The Most Promising
First-Generation GM Crop?” International Sugar Journal, 110(1318): 613-617.
Demont, M., M. Cerovska, W. Daems, K. Dillen, J. Fogarasi, E. Mathijs, F., Muška, J.
Soukup and E. Tollens. 2008. "Ex ante impact assessment under imperfect
information: Biotechnology in New Member States of the EU." Journal of
Agricultural Economics 59(3):463-486.
Demont, M., W. Daems, K. Dillen, E. Mathijs, C. Sausse & E. Tollens, 2008,
“Regulating Coexistence in Europe: Beware of the Domino-effect!” Ecological
Economics, 64(4):683-689.
Demont, M., K. Dillen, E. Mathijs & E. Tollens. 2007. "GM crops in Europe: How much
value and for whom?" EuroChoices. 6(3): 46-53 .
190
Book Chapters
Bezlepkina, I., R. Jongeneel, F. Brouwer, K. Dillen, A. Meister, J. Winsten, K. de Roest
& M. Demont. 2010. “Dairy” The Economics of Regulation: Compliance with Public
and Private Standards in Agriculture. Brouwer, F., G. Fox & R. Jongeneel, eds.,
Wallingford, UK: CABI Publishing, in press.
De Roest, K., J. Winsten & K. Dillen. 2010. “Assessing the Impact of Nutrient
Management Policies and Growth Hormone Bans on International Trade in Beef:
An EU/US Perspective” The Economics of Regulation: Compliance with Public
and Private Standards in Agriculture. Brouwer, F., G. Fox & R. Jongeneel, eds.,
Wallingford, UK: CABI Publishing, in press.
De Roest, K., J. Winsten, P. Rajsic & K. Dillen. 2010. “The impact of standards on the
competitiveness of the EU with respect to pigs and poultry” The Economics of
Regulation: Compliance with Public and Private Standards in Agriculture.
Brouwer, F., G. Fox & R. Jongeneel, eds., Wallingford, UK: CABI Publishing, in
press.
Devos, Y., M. Demont, K. Dillen, D. Reheul, M. Kaiser & O. Sanvido. 2009.
“Coexistence of Genetically Modified (GM) and Non-GM Crops in the EU. A
Review.” Sustainable Agriculture, Vol. 1. Lichtfouse, E., Navarrete, M., Debaeke,
P., Souchère, V., Alberola, C., eds., pp. 203-228. Les Ulis, France: EDP Sciences
– Springer.
Demont, M., K. Dillen & E. Tollens. 2008. “Economics of Spatial Coexistence: Isolation
Distances versus Pollen Barriers.” Theory in Ecology. Breckling, B., H. Reuter &
R. Verhoeven, p159-163, ed. Frankfurt: Peter Lang, Europäischer Verlag der
Wissenschaften.
191