Simplificando o Brasil Propostas de Reforma na Relação

Lecture 15: Cash Transfers
Prof. Eduardo A. Haddad
Outline
 The realm of regional policy
Lula’s regional policy
Pro-vote?
Pro-growth?
Pro-poor?
2
General indicators of regional concentration in Brazil
Table 1 - General indicators of regional concentration in Brazil
Share of
national
territory (%)
North
Northeast
Piaui State
Southeast
São Paulo State
South
Center-West
45.3
18.3
2.9
10.9
2.9
6.8
18.9
Share of national
population (%)
Share of
national GDP
(%)
Per capita income
in relation to
national average
1940
2000
1939
2004
1939
2004
3.5
35.0
2.0
44.5
17.5
13.9
3.1
7.6
28.1
1.7
42.6
21.8
14.8
6.8*
2.7
16.9
0.9
63.0
31.3
15.3
2.1
5.3
14.1
0.5
54.9
30.9
18.2
7.5
0.75
0.48
0.43
1.41
1.80
1.11
0.70
0.67
0.51
0.30
1.29
1.41
1.24
1.07
3
The analysis of different indicators of production
(growth) and welfare (development) in Brazil...
Exploratory analysis of spatial distribution of growth and human
development indicators in Brazil:
Gross Regional Product – GRP
 “Índice FIRJAN de Desenvolvimento Municipal” – IFDM

GRP measures the total amount of final goods and services
produced within the limits of a geographical area (country, state,
municipality), by residents and non-residents.
The IFDM, created by Sistema Firjan to monitor the evolution of
Brazilian municipalities, considers (with the same weights) the three
main dimensions of human development: employment and income;
education, and health.
4
... reveals the existence of a persistent regional
inequality in terms of per capita GRP...
Relative position of per capita GRP with relation to national per capita GDP -- 2008
5
... and human development in its various dimensions
IFDM – 2007
IFDM – Component “education” – 2007
IFDM – Component “employment and income” – 2007
IFDM – Component “health” – 2007
6
In extreme situations, it is clearer the regional dualism
in in the country and the “northeast issue”
Municiaplities with per capita GRP less than
30% below the national average - 2008
Municiaplities with per capita GRP more than
100% above the national average - 2008
7
Regional policy encompasses two distinct dimensions
The realm of regional policy



Physical capital
Human capital
Geographic targeted social policies (transfers)
Fundamental concepts



Sustainability
Endogeneity
Stakeholders participation
8
Where did it stand in the first term? (second as well!)
Human capital
Endogeneity
Physical capital
Sustainability
Transfers
The realm of regional policy
Participation
Key concepts
9
Outline
The realm of regional policy
 Lula’s regional policy
Pro-vote?
Pro-growth?
Pro-poor?
10
Nothing much has been done in the first four years of
Lula’s administration
Nowadays, the regional policy carried out by the central
government consists of isolated subsidies and industrial
incentives to growth centers, in addition to constitutional
transfers to less developed regions and rural areas.
In terms of proper regional policy, central government relied
only on constitutional intergovernmental transfers through
regional funds – FNE, FNO, FCO – and rural pensions.
Key ingredient: lack of coordination!
However, the central government has been engaged in an
effort to design and implement social compensatory policies
with a strong spatial dimension.
11
The Bolsa Família program has a strong spatial
dimension
The pro-poor “Bolsa Família Program” is a program that
provides direct income transfers to poor (with per capita
income between BRL 60,01 and BRL 120,00) and extreme poor
households (with per capita income below BRL 60,00).
Given the geographical distribution of poor households in the
country, targeting benefits to the poor reflects an implicit
concern with regional disparities in the country.
Even though it cannot be considered an explicit strategy of
geographic targeting to reduce poverty, it may achieve the
goal of classical regional policies – namely, the reduction of
regional disparities – through direct income transfers to poor
households, which happen to be concentrated in poorer
regions. However, this remains to be tested.
12
Spatial distribution of poor households
Households eligible for benefits from Bolsa Família, 2000
Region
Number of Poor Households
%
North
1,574,094
0.0917
Northeast
7,140,519
0.4158
South
2,006,596
0.1169
Southeast
5,342,975
0.3111
Mid-West
1,107,909
0.0645
17,172,093
1.0000
Total
Source: Demographic Census, 2000
13
Outline
The realm of regional policy
Lula’s regional policy
 Pro-vote?
Pro-growth?
Pro-poor?
14
What did the ballots tell us?
We evaluate the main determinants of Lula’s voting in the first
round of the 2006 presidential elections (data by municipality)
Lula’s voting (% of Lula’s voting in relation to total voting)
15
Formal tests suggest that Lula’s voting was not
randomly distributed in space
Moran Scatterplot and Global Moran’s I
Statistics for Lula’s Performance in the
First Round of the 2006 Presidential Elections
Moran Significance Map for Lula’s Voting
16
Paradoxical result (?)
Does Lula’s performance reflect his efforts in the first mandate
to fight regional inequality or is there something else behind
this seemingly paradoxical result, which shows that a
government without any concrete regional policy action
achieved its best evaluation in the regions that were left
behind?
Need to investigate the main determinants of Lula’s
performance in the 2006 elections!
Tool: spatial econometric models
17
The model
The dependent variable is the percentage of Lula’s voting in the first
round of the presidential elections.
We considered four groups of covariates in our models:

1. Spatial Structure Variables - Human Development Index; Gini
Coefficient; Per Capita GRP (all for year 2000)

2. Structural Regional Policy Variables - Per Capita Constitutional
Transfers; Per Capita Rural Pension Payments (both in 2006)

3. Social Policy Variables - Per Capita Income from Bolsa Família,
in 2006; Number of Households with Per Capita Income below R$
120,00, in 2000

4. Economic and Political Variables - Dummy for Mayor affiliated
to PSDB; Share of Agriculture in GDP, in 2002
18
The results
Qualitative estimation results (dependent variable: % Lula’s voting)
Coefficients
W_Lula
Constant
HDI
Gini
GDP (per capita)
Constitutional transfers (per capita)
Bolsa Familia income (per capita)
Rural pensions (per capita)
Number of poor households
Dummy_PSDB
Share_Agriculture
Lambda
OLS
(+)
(-)
(+)
(-)
(-)
(+)
(-)
(+)
(-)
(-)
-
Model
Spatial Lag
(+)
(+)
n.s.
n.s.
(-)
(-)
(+)
(-)
(+)
(-)
(-)
-
Spatial Error
(+)
(-)
(+)
(-)
(-)
(+)
(-)
(+)
n.s.
n.s.
(+)
Noteworthy is the robustness of the social policy variables.
Municipalities with higher per capita transfers through the “Bolsa
Família Program” and with the potential for its consolidation in the
future, presented a positive evaluation of the Lula’s first mandate
translated into greater proportion of votes.
19
Outline
The realm of regional policy
Lula’s regional policy
Pro-vote? YES!
 Pro-growth?
Pro-poor?
20
We have assessed the economic impacts of the Bolsa
Família program using an input-output framework
Multi-regional input-output framework, for 2002.
Five regions.
21 sectors in each region.
Full specification of interregional flows.
Ten income brackets were considered (R$/month): from zero income to
400 (5.3% of total national household income); 400 – 600 (5.4%); 600
– 1,000 (11.5%); 1,000 – 1,200 (5.3%); 1,200 – 1,600 (8.9%); 1,600
– 2,000 (8.3%); 2,000 – 3,000 (13.7%); 3,000 – 4,000 (9.7%); 4,000
– 6,000 (11.9%); and 6,000 and over (19.8%).
The household expenditure patterns for each income bracket in each
region are taken into account.
Model captures direct, indirect and induced effects.
21
Patterns of consumption
0,4
0,3
Manufactured goods
0,2
Rent
0,1
Manufactured Food
Transportation
0,0
1
2
3
4
5
6
7
8
9
10
Income Classes
0,3
Services to Households
0,2
Services
0,1
Trade
Communication
0,0
1
2
3
4
5
6
7
8
9
10
Income Classes
22
What were the amounts involved in each region?
Government transfers to families by region - Bolsa Família (*)
2002
2003
(1)
(2)
Mid-West
North
Northeast
South
Southeast
130,797
119,168
1,169,922
231,267
756,846
219,383
242,595
1,676,519
414,757
891,746
226,801
482,206
3,111,165
525,039
1,247,910
300,167
618,950
3,623,624
692,920
1,720,518
376,739
805,087
4,281,900
750,044
1,964,509
1,123,090
2,148,839
12,693,208
2,382,759
5,824,682
Total
2,408,000
3,445,000
5,593,121
6,956,179
8,178,279
24,172,579
5.4%
4.9%
48.6%
9.6%
31.4%
6.4%
7.0%
48.7%
12.0%
25.9%
4.1%
8.6%
55.6%
9.4%
22.3%
4.3%
8.9%
52.1%
10.0%
24.7%
4.6%
9.8%
52.4%
9.2%
24.0%
4.6%
8.9%
52.5%
9.9%
24.1%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
2004
2005
2006
Lula's
Administration
R$ 1,000 of 2002
Regional shares
Mid-West
North
Northeast
South
Southeast
Total
(1) Based on "interests, profits, dividends and others" received by individuals with total earnings below R$ 120 in the Norttheast, R$ 130 in the North, and
R$ 140 in the other three regions
(2) PNAD 2003, special segment on social programs
(*) Untill 2003 payments to auxílio-gás, bolsa alimentação, bolsa escola and cartão alimentação were summed. These separate programs were unified in
2004 under the Bolsa-Família program. Official data from 2004 on.
23
Simulation strategy
1.
Extra money was introduced in the region as an absolute
increase in government expenditure.
2.
Extra money had to come from reduced government current
expenditures.
The size of the shock simulated is of R$ 24.172 billions,
which represents 0.45 % of national GDP, 0.82% of national
disposable income, and 13.4% for the poorest income
bracket.
24
Not pro-growth, but pro-regional equity!
Impacts on production, by sector and region
Increased government expenditure
North
Agriculture
Mining
Metal
Machinery
Vehicles
Wood & Furniture
Chemicals
Textiles
Manfactured Food
Other Manufacturing
Public Utilities
Construction
Commerce
Transportation
Communication
Financial Services
Services to Families
Services to Business
Rent
Public Administration
Non-mercantile Services
All Sectors
Northeast Mid-West Southeast
South
Constant government expenditure
Brazil
North
Northeast Mid-West Southeast
South
Brazil
5.31%
1.53%
1.20%
1.54%
1.06%
3.97%
4.10%
4.89%
8.22%
1.68%
5.47%
0.20%
4.31%
7.46%
5.17%
2.69%
3.34%
3.14%
11.85%
0.10%
5.61%
17.1%
4.7%
2.3%
6.0%
4.6%
9.0%
5.8%
7.4%
20.9%
8.2%
9.4%
0.4%
7.6%
14.4%
6.9%
3.5%
6.8%
6.1%
14.0%
0.2%
7.7%
4.21%
1.46%
1.45%
2.46%
1.32%
2.90%
3.68%
4.34%
4.25%
2.36%
2.95%
0.17%
1.89%
3.00%
2.01%
0.51%
1.34%
2.07%
5.59%
0.03%
1.02%
4.40%
1.91%
1.35%
1.52%
1.49%
2.27%
3.20%
4.19%
4.63%
2.20%
3.06%
0.20%
1.95%
2.60%
2.17%
0.72%
1.69%
1.00%
3.36%
0.15%
1.49%
4.65%
1.10%
1.64%
1.20%
1.53%
2.48%
3.06%
2.97%
5.34%
2.27%
3.70%
0.24%
2.65%
2.97%
2.81%
0.93%
2.38%
2.31%
4.23%
0.16%
2.61%
6.26%
2.13%
1.45%
1.56%
1.53%
2.75%
3.62%
4.32%
6.62%
2.43%
4.27%
0.23%
2.92%
4.66%
3.06%
1.02%
2.65%
1.65%
4.95%
0.15%
2.55%
3.21%
0.32%
0.08%
0.40%
0.01%
1.36%
1.21%
2.37%
5.20%
-0.46%
2.77%
0.06%
1.29%
4.72%
2.26%
0.92%
0.97%
-0.40%
7.57%
-2.94%
1.21%
14.1%
2.2%
1.3%
4.1%
2.3%
5.4%
3.5%
4.9%
17.3%
4.4%
6.9%
0.2%
4.9%
11.5%
4.3%
2.1%
4.1%
2.1%
10.6%
-3.4%
4.3%
-0.27%
-1.63%
-1.68%
-8.39%
-4.59%
-8.19%
-5.78%
-7.35%
-1.11%
-8.52%
-11.04%
-1.03%
-8.61%
-8.79%
-12.31%
-3.68%
-17.37%
-22.74%
-20.32%
-21.77%
-18.29%
1.34%
0.21%
0.00%
0.05%
-0.78%
-0.95%
0.05%
0.45%
1.69%
-1.30%
-0.52%
-0.08%
-0.92%
-0.43%
-1.11%
-0.42%
-1.90%
-1.25%
-0.02%
-7.24%
-1.46%
2.14%
0.11%
0.22%
0.26%
-0.40%
0.22%
0.45%
0.87%
2.93%
-0.30%
1.16%
0.04%
0.20%
0.91%
0.33%
0.07%
-0.30%
-0.70%
1.88%
-4.24%
0.19%
3.19%
0.40%
0.12%
0.19%
-0.62%
-0.32%
0.60%
1.10%
3.41%
-0.80%
0.16%
-0.09%
-0.60%
1.23%
-0.92%
-0.41%
-1.80%
-2.02%
0.74%
-8.16%
-1.75%
3.35%
7.2%
2.15%
2.13%
2.77%
2.96%
1.15%
4.6%
-9.51%
-0.82%
0.51%
-0.48%
25
Outline
The realm of regional policy
Lula’s regional policy
Pro-vote? YES!
Pro-growth? NO!
 Pro-poor?
26
Pro-poor? Yes,...
Impacts on household income by region
27
... and pro-income equity
Impacts on income inequality
After Shock
Observed 2002
Gini
Increased
government
expenditure
Gini
Change
Constant
government
expenditure
Gini
Change
North
Northeast
Mid-West
Southeast
South
0.4659
0.4988
0.5353
0.4666
0.4580
0.4661
0.4962
0.5351
0.4662
0.4579
0.04%
-0.51%
-0.05%
-0.08%
-0.02%
0.4655
0.4961
0.5317
0.4661
0.4576
-0.07%
-0.54%
-0.67%
-0.10%
-0.10%
Brazil
0.5280
0.5266
-0.26%
0.5259
-0.39%
28
How regions are affected
Brazil
Northeast Southeast Mid-West
North
South
Production
-0,48%
4,6%
-0,82%
-9,51%
1,15%
0,51%
Income
-1,78%
3,3%
-2,01%
-16,55%
0,15%
-0,39%
1,71%
0,33%
-0,77%
5,96%
4,96%
3,60%
-0.97%
-1,13%
-1,48%
-12.77%
-13,13%
-14.61%
1,02%
0.69%
0.45%
0.37%
0.21%
-0.07%
1st Poorest
2nd Poorest
3rd Poorest
All others
lose more
All others All others All others All others All others
gain less lose more lose more gain less lose more
29
In summary
Pro-vote? YES!
Pro-growth? NO!
Pro-regional equity? YES!
Pro-poor? YES!
Pro-income equity? YES!
30
But how about long-run considerations?
Ranking
Total
Poverty
Infrastructure
Education
Transfers
Urban
poverty
Education
Infrastructure
Transfers
Rural poverty
Infrastructure
Education
Transfers
31
Challenge: to promote development in depressed areas
What is an economically depressed area?
 Low
rates of economic growth
 Lack
of labor absorption (high rates of unemployment,
underemployment and disguised unemployment)
 High
poverty rates and social problems
 strong
socioeconomic and intra-regional (between rural
and urban zones) unbalances
 Economic
and social infrastructure in bad conditions
 High
degree of dependence on government
transfers, both for residents and local governments
32
Where are they?
IFDM – Component “employment and income” –
Northeast and Southeast, 2007
33
What happens to the possibilities of endogenous
growth of a society in which more than one third of its
members depend upon compensatory social policies?
“No Brasil, desde a implantação dos direitos sociais previstos
na Constituição Federal de 1988, tem-se mais de 11 milhões
de famílias no programa Bolsa-Família, mais de 7 milhões de
benefícios da Previdência Social e mais de 5 milhões de idosos
e deficientes na Lei Orgânica da Assistência Social (LOAS). Aos
programas sociais do Governo Federal, somam-se aqueles de
menor escala coordenados e implementados por governos
Estaduais e Municipais do País. Não é difícil, pois, chegar a
um número superior a 70 milhões de brasileiros que
sobrevivem, direta ou indiretamente, sob a cobertura
das políticas sociais compensatórias, recorrentes do ponto
de vista fiscal e inflexíveis do ponto de vista político.”
34
Typology of products that will lead growth in Brazilian
micro-regions, 2007-2030
35