JEA-Vol.7-No.2 - Athenian Policy Forum

Transcription

THE JOURNAL OF
ECONOMIC ASYMMETRIES
December, 2010
Volume 7, Number 2
Preface: Volbert Alexander
Articles
Dominick Salvatore
The Global Financial Crisis: Predictions, Causes, Effects, Policies,
Reforms and Prospects ..…………………………………………..... .….……1
Robert W. Kolb
Incentives in the Financial Crisis of Our Time ...…………………..…..…..21
Apostolos Xanthopoulos
Market Value Signal Extraction and the Misapplication of SFAS
133 in the U.S. GSE’s ………………………………………………..….…57
Tao Sun
Identifying Vulnerabilities in Systemically Important Financial
Institutions in a Macro-Financial Linkages Framework………...……..........77
Peter Flaschel, Florian Hartmann, Christopher Malikane, Willi Semmler
Broad Banking, Financial Markets and the Return of the Narrow
Banking Idea..………………………………………………………..……..105
Panagiotis G. Korliras, Yannis A. Monogios
Asymmetric Fiscal Dynamics and the Significance of Fiscal Rules
For EMU Public Finances…………………………………………….….…139
Herbert Grubel
Who is to Blame for the Great Recession?....................................................171
Preface
Volbert Alexander
Goethe University Frankfurt and Goethe Business School
Among many other activities the “Biennial Conference” is a central event in
the academic life of the Athenian Policy Forum (APF). Hosted by the Deutsche
Bundesbank, the 10th Biennial Conference was organized in July 2010 under the
general topic “Regulatory Responses to the Financial Crisis”. During three days,
nearly forty papers were presented and discussed by an international audience. It was
not surprising that, in addition to the main topic, the economic situation in Greece
and its implications for the European Monetary Union and for the Euro as the
common European currency attracted a substantial part of the debates. Many APFmembers have close relationships with Greece and a strong interest in Greek
economic developments.
In the present volume seven papers presented and discussed at the Frankfurt
conference were selected for publication, after a careful process of review. In D.
Salvatore “The Global Financial Crisis: Predictions, Causes, Effects, Policies,
Reforms and Prospects” a broad analysis of important aspects of the financial crisis
is presented from a macroeconomic perspective. The papers by R.W. Kolb
(“Incentives in the Financial Crisis of Our Time”) and A. Xanthopoulos (“Market
Value Signal Extraction and the Misapplication of SFAS 133 in the U.S. GSE’s”)
address the U.S. real estate market and the origin of the crisis. While Kolb shows
how the individual behavior of the market participants driven by wrong incentives
contributed to the crisis, Xanthopoulos emphasized the role of existing regulations
for institutions like Fanny Mae and Freddy Mac, both massively influenced by the
government. The problem of systemically important financial institutions is picked
up by Tao Sun in his paper “Identifying Vulnerabilities in Systemically Important
Financial Institutions in a Macro-Financial Linkages Framework” where
macroeconomic and financial market-related aspects are considered. His conclusions
directly lead to the recent proposals for a new system of banking regulation.
With the help of a macroeconomic model, P. Flaschel, F. Hartmann, C.
Malikane and W. Semmler (“Broad Banking, Financial Markets and the Return of
the Narrow Banking Idea”) analyze stability problems in a system of universal banks
(trading stocks and credit), in contrast to a narrow banking system where banks are
only trading credits. They find out that the last, narrow banking system is far more
stable in absorbing different shocks.
P. Korliras and Y.A. Monogios (“Asymmetric Fiscal Dynamics and the
Significance of Fiscal Rules for EMU Public Finances”) bridge the gap between the
financial crisis and the Greek-Euro issues. They examine the effectiveness of
common fiscal policy rules for Europe with its very different fiscal situations and
developments. Their results are not surprising: Common rules are not optimal for all
European fiscal problems and have to be enriched by more flexible reactions, taking
into account different idiosyncratic national problems.
The volume is closed by a specific macroeconomic look to the causes of the
present financial crisis: H. Grubel (“Who is to Blame for the Great Recession?”)
points out that the non market-oriented exchange rate policy of China significantly
contributed to the financial crisis. In order to achieve high export surpluses, the
policies of oil-exporting countries led to exorbitantly high accumulations of profits
which were directly invested into foreign assets.
The above contributions clearly show the high level of conference
contributions and discussions in terms of theoretical analyses, technical and
statistical skills, empirical research, and policy orientation. For the organizers it is a
great pleasure to look at this positive output and response.
Many people have contributed to the great success of the Frankfurt
conference. I first want to thank the members of the organizing committee who were
responsible for the scientific content. The committee consisted, beside myself, of N.
Baltas (Athens), J. Brox (Waterloo, Canada), A. Malliaris (Chicago), D. Salvatore
(New York) and G. von Furstenberg (Indiana). In particular, I want to thank
“Tassos” Malliaris who really was a great help in all stages of the preparations.
All participants are indebted to the Deutsche Bundesbank which hosted the
conference in a perfect way. President A. Weber supported the initiative from the
beginning and H.H. Kotz, a member of the board in 2010, was our main partner for
all organizational issues. The team around Katrin Gruening helped us with all
administrative problems and the guesthouse crew around Mrs. Grall was open for all
the “small” but important questions from the participants. Together with the luxury
environment of the Bundesbank’s guesthouse, they created a very comfortable
atmosphere so that all conference members could fully concentrate on the sessions
and discussions. We also have to thank the Sal. Oppenheim Bank for financial
support.
The Global Financial Crisis: Predictions, Causes,
Effects, Policies, Reforms and Prospects
Dominick Salvatore1
Fordham University
Abstract. The paper examines the causes, effects, policies, and the prospects for
rapid recovery and growth after the deepest world financial and economic crisis
since the Great Depression. The paper then examines the regulatory and supervisory
systems in the United States and Europe before the crisis, the proposed reforms of
those systems, as well as reforms of the entire world financial system, and the
likelihood that those reforms will succeed in preventing future financial crises.
JEL Classification: E31, E32, F44
Keywords: Financial crisis, Contagion, Stimulus package, Exit strategy, Bank stress
tests, Financial reforms
1. Introduction
Advanced countries faced a serious financial crisis and deep economic recession in
2009 and are now experiencing an anemic recovery. Most emerging markets also
experienced a recession or a growth slowdown. In this paper, I will examine
predictions of the crisis, its causes, effects, policies, reforms and prospects.
2. Predicting the Financial Crisis
The economic profession has failed society by not having predicted the most serious
financial crisis and deepest recession of the post war period. Had the coming crisis
been predicted, policies could have been adopted to prevent the crisis or at least to
soften its impact. Some economists claim to have predicted the crisis. Nouriel Roubini
is one of them. But he had been predicting a crisis for several years before it actually
came and kept changing the cause of the crisis. As Anirvan Banerji, economist with
the New York-based Economic Cycle Research Institute, put it as follows in October
2008: “Roubini started predicting a recession four years ago and saying it was
imminent. He kept changing his justification: first the trade deficit, the current account
deficit, then the oil price spike, then the housing downturn. But the recession actually
1
2
THE JOURNAL OF ECONOMIC ASYMMETRIES
DECEMBER 2010
did not arrive”.2 When the crisis did come it was for the last of the several causes that
Roubini had specified over time.
To be correct and useful, a forecast must specify the time and the cause of the
crisis. If I had said at the end of 2008 that the crisis would end, without specifying
when and how, I would have been correct, but I could not have claimed to have
forecasted the recovery. All crises, for whatever reason, eventually do come to an end.
Alan Greenspan worried aloud in 2000 that the United States faced a
resurgence of rapid inflation, when in fact it was growth that was collapsing under
his very eyes (in fact, the United States fell into recession in 2001).3 Joseph Stigletz,
Jonathan Orszag and Peter Orszag wrote 2002 that on the basis of historical
experience, “the probability of either Fannie Mae or Freddie Mac defaulting would
be close to zero” (2002, p. 5). Paul Krugman stated in 2002 and again in 2003 that
the United States, which had experienced a recession in 2001, would fall into
“double-dip” recession – which also did not happen.4 Jean-Claude Trichet, the
Governor of the European Central Bank, increased the interest rate from 4 per cent to
4.25 percent in July 2008 believing that the European Union would avoid the crisis.5
Making a wrong forecast is dangerous because it could lead to wrong business
decisions and government policies. But forecasting a crisis without clearly specifying
the timing and the reason is not useful. In fact, it is not forecasting all.
3. Causes of the Financial Crisis
The present financial crisis started in the U.S. sub-prime mortgage market in 2007
and then spread to the entire financial and real sectors of the U. S. economy in 2008,
and from there to the rest of the world. The initial causes of the financial crisis are
clear: huge and increasing amounts of home mortgages – often based on weak
underwriting including no down payment or checking credit histories – were given to
individuals and families that clearly could not afford them. These mortgages were
made at variable rates when rates were the lowest in 50 years. It was only to be
expected that a rise in interest rates would cause many homeowners to be unable to
make their mortgage payments and default. The crisis could only have been avoided
if housing prices had continued to rise at the unrealistic high rates of 2000-2005.
These sub-prime home mortgages were then repackaged into mortgagebacked securities (MBS) and sold to credit market investors. Rating agencies, such
as Moody’s and Standard & Poor, gave some of these financial instruments triple A
ratings. Finally, the Securities and Exchange Commission (SEC), which was to
regulate this market, was in fact not highly involved in these transactions.
Although the problem of sub-prime mortgages greatly expanded during the
presidency of George W. Bush, the practice started in 1999 during the Clinton
Administration when Fannie May and Freddie Mac were encouraged to grant home
mortgages to people who clearly could not afford these mortgages in order “promote
the American dream” of owning a home.
VOL.7 NO.2
SALVATORE: THE GLOBAL FINANCIAL CRISIS
3
Two additional and crucial inter-related causes of the sub-prime mortgage
crisis were the easy monetary policy of the Fed and the export–based growth
strategies of some Asian countries, including China, which allowed the Fed to
maintain interest rates at very low levels during the period 2002-2006, enabling
excessive risk-taking and encouraging asset-price bubbles. Undervalued exchange
rates further encouraged financial investment in the United States by Asian
economies, further fueling asset-price bubbles.6
Some economists blame deregulation as the primary cause of the crisis.
Indeed, the repeal of the depression-era Glass-Steagall Act in 1999 (pushed by Alan
Greenspan and Robert Rubin when Larry Summers was Treasury Secretary during
the Clinton Administration) ended the separation of commercial banking from other
financial activities, such as insurance, underwriting and investment banking, and
made possible some of the financial excesses that led to the present crisis. And it was
Greenspan, Rubin and Summers who in 1998 objected to the imposition of any
regulation on credit default swaps (CDS) – which the famed investor Warren Buffett
once called “weapons of financial mass destruction”.
We can thus say that the present financial crisis was caused by deregulation
or inadequate regulation of investment banking, by the inadequate application of
regulations that were already on the books (i.e., by rating agencies and the SEC), by
unfortunate economic policies (granting home mortgages to people who could not
afford them), by too easy monetary policy by the Fed and undervalued exchange
rates by some Asian economies, by economic greed (financial firms caught in a
gigantic profit-seeking scheme with insufficient risk management), and by outright
fraud (such as the incredible $65 billion Bernard Madoff Ponzi scheme).
4. Contagion – The Spread of the Financial Crisis Around the Globe
There was then contagion, by which the crisis in the United States spread first to
other advanced countries through the global financial system and finally to emerging
markets when the former fell into recession and sharply reduced their imports from
and capital investments in the latter.
However, contagion would not have occurred so quickly through the financial
sector in Europe if some even bigger excesses than in the United States had not
occurred in Europe. For example, bank leverage (measuring the risk that a bank
faces) was 31 for Lehman Brothers at the time it failed in March 2008 and 38 at
Citigroup the weakest of the largest U.S. banks, but it was at 42 at UBS, 56 at
Deutsche Bank, and 63 at Barclays. On average, bank leverage was 35 for the largest
12 European banks as compared with 12 for the largest 12 U.S. banks.
The housing bubble was also even greater in some European countries than in
the United States. For example, between 2004 and 2007, the peak housing prices
were 2.58 times higher than their long-run (German) trend in Ireland, 2.10 times in
the United Kingdom, 1.92 times in Spain, as compared with 1.76 times in the United
States (see Figure 1). Yes, the crisis started in the United States but Europe faced
4
DECEMBER 2010
even greater excesses in some sectors, otherwise contagion would not have occurred
as rapidly and as strongly as it did through the financial sector.
Figure 1: Housing Bubble in the United States, United Kingdom, Spain and Ireland
Price-Rent Ratio
(1997=1)
1987 1990
1994
1997
2000
2004
2007
2009
Source: OECD Databank (2005-2010).
The crucial event that triggered the crisis was, of course, the failure of Lehman
Brothers in September 2008. Lehman was allowed to fail presumably because its assets
were less solid than those of Bear Sterns (which was acquired by J.P. Morgan Chase the
previous March to prevent its failure) and there were no buyers after Treasury Secretary
Paulson refused to provide $60 billion of loss guarantees to Barclays and Bank of
America that had shown interest in acquiring Lehman. It is more likely that Secretary
Paulson wanted to use the failure of Lehman Brothers to avoid the accusation of falling
into the moral hazard trap (the situation where profits are private and losses are public)
and to teach a lesson to financial markets. However, he subsequently admitted to having
underestimated the size of Lehman and the problem that its failure would create in the
United States and around the world. At the time of its failure, Lehman had sold nearly
$700 billion in bonds and derivatives, of which about $160 billion was unsecured.
Rescuing Lehman, however, would only have postponed the crisis, not prevented it.
VOL.7 NO.2
5
Some economists blame the operation of the international monetary system for
the crisis. The present crisis, however, is for the most part domestic and not
international in origin. A more efficient and effective international monetary system
(one that exposes financial excesses and excessive risks thus reducing speculative
international capital flows) would not have prevented contagion across the world
because, as we have seen, some even greater financial excesses than in the United
States had occurred in Europe, Japan and elsewhere.
To be sure, China’s exports of huge amounts of capital primarily to the
United States (as the counterpart of its huge trade surplus with the United States)
facilitated and reinforced the financial bubble that was developing in the United
States. Chinese families save a very high percentage of their income because they
have little or no provision of public unemployment insurance and old age pension.
Since China’s financial sector is still rather underdeveloped and cannot absorb a
great deal of its savings, a huge amount of Chinese savings sought foreign (primarily
U.S.) investment outlets (Salvatore, 2010). This supplied excess liquidity to the
United States, which facilitated the financial bubble and increased its size. But a
well-functioning financial sector in the United States could have discouraged such an
inflow and prevented it from further reinforcing the bubble that was developing.
5. Effects of the Crisis
The major effects of the crisis are the following:
1.
Stock markets crashed all over the world during 2008, with declines ranging
from 31 percent in the United Kingdom to 50 percent in Italy (in the U.S. it was
34 percent) among advanced countries, and from 24 percent in Mexico to 65
percent in China and Russia among emerging markets.
2.
The capitalization of banks was cut by more than half from more than $8 trillion
at the end of 2007 to $4 trillion at the end of 2008. As we will see later, between
March and September 2008, the entire U.S. investment banking sector, as we
had known it, disappeared. Going forward, investment banking in the United
States will be conducted mostly by commercial banks under more highly
regulated and less speculative conditions permitted under the Dodd-Frank law
signed by President Obama in July 2010.
3.
All advanced countries fell into the “great recession” (the deepest of the post
war period) with real GDP falling by 2.4 percent in the United States, 4.1
percent in the Euro Area, 4.9 percent in the United Kingdom, and 5.2 percent in
Japan in 2009 (we will come back to this later).
4.
All the most important and largest emerging market economies, with the
exception of China, India and Indonesia fell into recession with real GDP falling
from 0.2 percent in Brazil to 6.6 percent in Mexico and 7.9 percent in Russia in
2009. On the other hand, between 2008 and 2009, the growth rate of real GDP
only slowed down from 9.6 to 9.1 in China, from 7.3 to 5.7 in India, and from
6
DECEMBER 2010
6.0 to 4.5 in Indonesia. While impressive, it must be pointed out that China,
India and Indonesia need very high rates of growth to absorb into the market
economy the still significant segment of their population living at subsistence
level (we will return to this in Section IX).
5.
While the financial crisis spread quickly from the United States to other
advanced countries through the financial sector, the crisis spread to emerging
markets with about a half-year lag primarily through the real sector (i.e., from
the reduction of imports of recessionary advanced countries from emerging
market economies and the sharp fall in cross-border capital flows – see Figures
2 and 3). Figure 2 shows that at the bottom of the recession in advanced
countries in the first quarter of 2009, the growth of real GDP was about minus 2
percent while world trade was down by almost 9 percent. Figure 3 shows that
the net private financial flows to emerging and developing countries declined
from over $700 billion in 2007 to $200 billion in 2008.
6. Monetary Policies, Increased Liquidity, and Bank Rescues
The United Stated and Europe did almost everything possible to avoid the recession,
but their efforts only succeeded in preventing a deeper recession or depression. At
the beginning of 2008, the United States introduced a $168 billion dollar stimulus
package, which contributed to a 2.8 percent growth of real GDP in the second
quarter of last year, but its effect soon faded away.
The United States lowered its interest rate from 5.25 percent in September
2007, to 1 percent in October 2008, and to practically zero in December 2008. The
ECB cut its key policy rate to 1.0 per cent in October 2008 also translated into a fall
in the EONIA -- an overnight (interbank) market rate that is the central focus of
market participants -- to a range of 0.30 to 0.40 per cent. In addition, the ECB
provided liquidity to banks in fixed-rate, full-allotment operations for as much as one
year. (These operations are still available for 3 months’ duration). In other words,
banks could obtain as much liquidity as they wanted (provided they had the
necessary collateral) at a fixed-rate from the Eurosystem, a policy consistent with the
central role of the banking system in the monetary-policy transmission mechanism of
the euro-area (in contrast to the U.S., where capital markets play the central role).
Finally, the ECB embarked, for the first time ever, in purchases of
government securities, a policy still in effect. That is, needing more stimulus, the Fed
also flooded the market with liquidity, as evidenced by the increase in its balance
sheet (and reserves of commercial banks held at the Fed) from $900 billion in the
summer of 2008 to over $2 trillion in 2010. This could potentially generate an
explosive rise in future bank lending and in the money supply, and thus lead to a
huge inflationary spiral.
VOL.7 NO.2
7
Figure 2: Growth of World Real GDP and World Trade, 2007-2009
____________________________________________________________________
____________________________________________________________________
Source: IMF (2010a) and WTO (2010).
Figure 3: Net Private Financial Flows to Emerging and Developing countries,
1985–2011
Source: IMF (2010b).
8
DECEMBER 2010
In March 2008, the Fed helped J. P. Morgan Chase acquire Bear Sterns at a
deeply discounted price (to avoid the accusation of moral hazard -- a situation where
profits are private and costs or losses are public) with a $29 billion debt guarantee; in
May the Treasury acquired a $100 billion of (nonvoting) stock of Fannie Mae, $100
billion of Freddie Mac, and from May to December a total of $185 billion from
American Investment Group (AIG); in September, it encouraged and facilitated the
acquisition of Merrill Lynch by Bank of America and it approved the conversion of
Morgan Stanley and Goldman Sachs into commercial banks. In October it increased
insurance on bank deposit to $250,000 (up from $100,000) and it adopted a $700
billion bank rescue plan, with half of the money spent by the end of the year to
recapitalize the banking sector and purchase money and commercial paper from
firms to make up for the drying up of this crucial lending activity by commercial
banks.
Then in November 2008, the U.S. Treasury injected another $20 billion of
new capital (on top of the $25 billion provided in September) to Citigroup and together
with the Fed provided guarantees against excessive losses on $301 billion of toxic
assets (mostly sub-prime personal and commercial loans owned by Citigroup) to
prevent its collapse. In January 2009, the Treasury injected another $20 billion of new
capital (on top of the $25 billion injected in September) to Bank of America and
Merrill Lynch, and together with the Fed provided guarantees against excessive losses
on $100 billion of toxic assets to prevent Bank of America from withdrawing from the
purchase of Merrill Lynch after it discovered that the latter had even more toxic assets
than it realized at the time Bank of America initially agreed to purchase it. Then, at the
end of February, the U.S. government agreed to become the biggest single shareholder
of Citigroup by taking a 36% stake of the troubled lender to prevent its collapse.
At the same time, many European countries adopted similar but less
ambitious policies to stimulate their economies. In January 2009, the European Central
bank cut the interest rate from 4.25 in July 2008 to 1.0 percent in April 2009, but
indicated that it would not follow the U.S. and Japanese counterparts down the path of
practically zero interest rate. The Bank of England cut the interest more drastically
from 5 percent in October 2008 down to 0.5 percent in April 2009 (the lowest since its
creation in 1694). All of these measures, however, did not prevent an even deeper
recession in Europe than in the United States.
VOL.7 NO.2
9
7. Exit Strategy and the Danger of Inflation
With the resumption of growth, the potential for inflationary explosion becomes a
serious danger. Indeed, the fear in the market that the Fed would not be able to
reverse course in unwinding its huge unconventional monetary stimulus, prompted
Ben Bernanke to outline the Fed’s exit strategy in July 2009. This calmed markets
and led to a sharp decline in U.S. Treasuries. Bernanke testified that he expected the
U.S. economy to start growing again at the end of the year but, with unemployment
likely to reach nearly 10 percent, growth would very likely be slow through 2011, so
that the economy would face little inflationary pressure.
When necessary, Bernanke indicated that an exit strategy could be
established very quickly to mop up the excess liquidity by letting emergency lending
programs wind down or expire, raising the short-term interest rates paid on reserve
balances (to help set a floor under interest rates), letting short-term credits expire,
and selling longer-term assets to the public. He acknowledged that, as always, the
difficulty will be deciding the precise timing to begin to tighten and set the
appropriate pace of the tightening effort. At the same time, Bernanke warned
Congress and the White House to get budget deficits under control or risk damaging
the recovery.
Some economists, including Alan Metzler (2009) of Carnegie Mellon, have
deeper concerns. They believe that it takes about two years for an anti-inflationary
policy to work, which would mean that the Fed needed to implement a policy in
2009 and stick with it. This concern seemed overdone, however because the U.S.
economy is likely to grow well below its potential at least through 2011 so that
demand-pull inflation does not seem a serious threat. Only with another flare up in
the price of petroleum and other primary commodities is inflation likely to become a
serious problem. Be that as it may, it is most unlikely that the Fed and the other
major Central Banks will not start tightening before spring 2011– they stated as
much at their meeting at Jackson Hole, Wyoming in August 2010. Figure 4 shows
that inflation is not now and is not expected to be a serious problem in advanced
countries at least through 2011.
8. The U.S. Stimulus Package, Health Care and Governments’ Indebtedness
In mid-February 2009, the U.S. Congress passed a $789 billion stimulus package of
increased expenditures on infrastructure, education, health, and the environment, as
well as a tax reduction (demanded by Republicans). Together with the hugely
expansionary monetary and other policies, it probably prevented the U.S. economy
from falling into a depression, reminiscent of 1929.
The U.S. Administration pushed through a very ambitious health care
reform plan to provide universal coverage and contain future health care costs by
eliminating waste and introducing more competition (potentially including a
government health plan to keep private health insurance costs down). The cost is
10
DECEMBER 2010
Figure 4: Consumer Price Index in Advanced Countries, 2005-2011
Source: OECD (2010).
expected to be paid mostly by tax increases on high-income earners (defined as those
earning more than $250,000). But as pointed out by the Congressional Budget Office
(CBO) in July 2009, the proposed health care bill could cost from $200 to $300
billion more than the $1 trillion Congressional estimate, and is likely to lead to
higher taxes on all but the lowest income people to pay for it. A strong grassroots
backlash regarding the government option and its cost has also developed, and so the
plan was scaled down to ensure passage.
But even without factoring in the inevitable higher costs and taxes from
potential health care reform, the stimulus package and all other expenditures to bail
out the banking sector will lead to much higher U.S. government debt and taxes in
the years to come. Balancing the CBO-projected out-year budget would require a 44
percent increase in everyone’s taxes. Without an increase in taxes, the U.S.
government debt as a percentage of GDP is expected to increase from 40 in 2008, to
65 in 2010, 70 in 2012, and 103 in 2017. Faced with a drastic decline in their wealth
as a result of the deep recession, and anticipating much higher taxes in the future to
pay for the stimulus package and the other huge government programs to overcome
the crisis, business are investing less and individuals and families are saving more
and spending less, leading to a very low multiplier (barely above one) for every
stimulus dollar spent.
Europe and Japan generally have smaller stimulus packages in the relation
to their GDP than the United States because of their stronger social welfare net and
in order to curtail the growth of their already very high government debts. Despite
VOL.7 NO.2
11
Figure 5: Budget Deficits as a Percentage of GDP, Advanced Countries,
2005-2011
this, the national debt of most advanced countries continues to rise due to still large
(even if declining) budget deficits (see Figure 5). Indeed, it was excessive budget
deficits that got Greece, Ireland, Spain and Portugal in trouble in 2010, with Greece
having to be rescued from bankruptcy by a huge EU financial package.
9. Deep Recession and Slow Recovery in Advanced Countries
Despite the extraordinarily expansionary monetary policy and large fiscal stimuli, the
recovery in most advanced nations is rather slow (see Table 1 and Figure 6). This is
unusual. After previous deep recessions, there was a rapid resurgence of growth in
the year or two after the recession. Not this time. Only Germany grew relatively fast
in the first half of 2010 based primarily on the rapid expansion of its exports due a
low euro and contained labor costs. Growth, however, is slowing down in the second
half of the year. In fact, growth is now (January 2011) expected to be even slower
than indicated in Table 1 and Figure 6 in 2010 (especially in the United States) and
unemployment is expected to remain high (see Figure 7).
12
DECEMBER 2010
Table 1 Real GDP Change in 2009 and Forecasts for 2010-2011,
Percentages, Large Advanced Nations
Forecasts__
2010 2011
Nation/Area
2009
United States
-2.4
2.7
2.5
EURO AREA
-4.1
1.7
1.4
Germany
-4.9
2.8
1.6
France
-2.5
1.4
1.5
Italy
-5.0
1.0
1.1
Spain
-3.6
-0.5
0.6
United Kingdom
-4.9
1.5
2.0
Japan
-5.2
1.9
1.8
Canada
-2.5
3.3
2.4
OECD
-3.3
2.7
2.8
Source: OECD, IMF and European Commission Databank (2010).
Figure 6: Growth and Growth Prospects in OECD Countries, 2005-2011
VOL.7 NO.2
13
Figure 7: Past and Expected Unemployment Rates in Advanced Countries, 20052011
In most large emerging markets (those in the G-20 group) the situation is
different. As Table 2 shows, Russia, Mexico and Turkey faced a deep recession in
2009 with real GDP falling, respectively, by 7.9 percent, 6.6 percent and 4.9 percent.
On the other hand, there was no recession in China, India and Indonesia -- only a
slowdown of rapid growth. The forecast for 2010 and 2011 is for rapid growth to
resume in most countries listed in the table, especially for China and India, but also
for Brazil, Turkey and Indonesia.
10. Bank Stress Tests in the United States and Europe
In order to reassure financial markets on the stability of the U.S. banking system, the
Federal Reserve System conducted a stress test in 2009 in order to determine how
well capitalized and stable the largest 19 American banks were or how much
additional capital each needed to be able to withstand the financial crisis without
possibly collapsing.
The result of the banking stress test made available in May 2009 is shown in Table
3.The second column of the table shows that Bank America needed $33.9 billion to
be adequately capitalized, Wells Fargo needed $13.7 billion, GMAC $11.5 billion,
Citigroup $5.5 billion, and smaller amounts for 6 other large banks, for an overall
total of capital needed of $74.6 billion. The remaining 9 of the 19 large banks large
14
DECEMBER 2010
Table 2: Real GDP Change in Emerging Markets in G-20 in 2009
and Forecasts for 2010-2011 (Percentages)
NATION/AREA
2009
FORECASTS
2010 2011
China
9.1
10.5
9.6
India
5.7
9.4
8.4
Russia
-7.9
4.3
4.1
Brazil
-0.2
7.1
4.2
Korea
0.2
5.7
5.0
Indonesia
4.5
6.0
6.2
Mexico
-6.6
4.5
4.4
Argentina*
0.9
3.5
3.0
Turkey*
-4.9
6.8
4.5
South Africa*
-1.8
3.3
5.0
Saudi Arabia*
0.1
3.7
4.0
Source: IMF, July 2010; * May 2010.
banks examined 9 were regarded to be already well capitalized to withstand the
normal evolution of the financial crisis.
The third column of Table 3 shows instead how much each of the 19 large
U.S. banks would need if the financial crisis became much deeper in 2009 and
continued in 2010. In that case, all 19 large banks would need additional capital for
an overall total of $599.3 billion ($411.9 billion for the 10 banks indicated in the
table plus $187.4 billion for the 9 banks that were regarded as well capitalized if the
financial crisis did not become deeper than it was). The two banks that would need
the largest infusion of capital to be able to withstand the worse-case scenario, were
Bank America (which would need $136.6 billion) and Citigroup needing $104.7
billion. The last column of Table 3 shows Tier 1 common capital ratio.
VOL.7 NO.2
15
Table 3: Stress Test of U.S. Banks (Billion Dollars, May 2009)
Bank/Financial
Institution
Capital
Needed
Loss: More Adverse
Scenario
Tier 1: Common
Capital Ratio (%)
Bank of America
33.9
136.6
4.6
Wells Fargo
13.7
86.1
3.1
GMAC
11.5
9.2
6.4
Citigroup
5.5
104.7
2.3
Regions Financial
2.5
9.2
6.6
SunTrust
2.2
11.8
5.8
Morgan Stanley
1.8
19.7
5.7
KeyCorp
1.8
6.7
5.6
Fifth Third
1.1
9.1
4.4
PNC Financial
0.6
18.8
4.7
74.6
411.9
Av.: 4.9
Total
All Other 9 Banks
0.0
Total: 187.4 (599.3)
Average: 8.7
Source: http://www.federalreserve.gov/newsevents/press/bcreg/bcreg20090507a1.pdf.
With the financial crisis not deepening in 2009 and not expected to become
deeper in 2010, and with the 10 U.S. banks needing additional capital actually
raising that amount of capital by the end of 2009, financial markets became
reassured of the stability of the U.S. banking sector and so that we can say that the
stress test accomplished its aim.
The situation was different in Europe. There, banking regulators originally
intended to have each country conduct its own bank stress test with the results not
made public (for fear of causing a run on the large banks if they were shown to be
inadequately capitalized and weak). Only when financial markets became very
concerned that bank regulators were trying to hide serious banking weaknesses, it
was decided to conduct a European-wide stress test on the 91 largest banks and that
the results would be made public.
Table 4 shows the result of the bank stress test in Europe. The table shows
that only seven of the 91 largest European banks examined failed the stress tests
designed to show whether they could withstand a moderate recession and a fall in the
value of the government bonds they held. Banks whose Tier 1 capital ratio was
16
DECEMBER 2010
Table 4: Stress Test of European Banks (Million Euros, July 2010)
Bank/Financial
Institution
Country
Capital
Needed
Tier 1: Common
Capital Ratio (%)
Diada
Spain
1,032
3.9
Cajasur
Spain
208
4.3
ATEBank
Greece
242.6
4.36
Unnim
Spain
270
4.5
Germany
1,245
4.7
Banca Civica
Spain
406
4.7
Espiga
Spain
127
5.6
3,530.6
Av.: 4.58
Hypo Real Estate
Total
Source: http://stress-test.c-ebs.org/documents/Summaryreport.pdf.
below 6 percent under the test criteria were deemed as having failed the test and
needing to raise more capital. But only 3.5 billion euros ($4,51 billion at the average
exchange rate of 1 euro equal 1.278 dollars in July 2010) were needed to make the
seven banks that failed the stress test be adequately capitalized. This was much less
than the $75 billion that the 10 undercapitalized American needed in May 2009.
Markets, however, remained skeptical about the rigor of the bank stress
tests in Europe as evidenced by the high premia that Greek, Irish, Portuguese and
Spanish banks had continue to pay to borrow fund to increase their capitalization -even after European leaders and the International Monetary Fund agreed in May
2010 to a three-year loan package of €110 billion (about $140 billion) to Greece to
avoid default, and the subsequent establishment of a huge special EU-IMF rescue
fund of €750 billion (about $960 billion) raised by selling bonds guaranteed by eurozone governments to be used to lend money to (i.e., ECB buying bonds of) any
crisis-hit EU government. All this defused the panic but did not snuff the crisis:
unsustainable borrowing continues to pose huge challenges even after the joint EUIMF joint rescue operations of Greece in May and in Ireland in November 2010.
11. Financial Reforms to Prevent Future Crises
Many important financial reforms have been introduced or proposed in the United
States, Europe, and at international organizations in order to strengthen the banking
and financial system and prevent future financial crises.
VOL.7 NO.2
17
In the United States, the Wall Street Reform and Consumer Protection Act
(Dodd Frank Act) was passed in July 20107. This is the most sweeping overhaul
of Wall Street regulations since the 1930s. It has three major components, as
follows:
1. Systemic Risk (Macro-Prudential Regulation). This involves the creation of
a
Resolution Authority to deal with the problems created by financial
institutions “too big to fail”. These are financial institutions that in the
pursuit of higher profits undertake excessive risks in the belief that if
something goes wrong and they instead face huge losses that could cause
the institution to fail, national monetary authorities will come to their rescue
to avoid through contagion the risk of collapse of the entire financial sector
of the nation. To this end, the Fed is to monitor all large financial firms for
systemic risk and given the authority to shut down failing institutions and
recoup the cost from creditors, not taxpayers, by having banks and other
financial institutions hold a new form of capital, known as contingent
capital, to cover losses if the firm is shut down. In addition, more stringent
capital requirements for financial institutions are to be negotiated at Basle
III, but with national flexibility in their application, in order to establish a
level-playing field.
2. Market Regulation (Micro-Firm Supervision). The Fed is also to oversee the
establishment of central clearing of over-the-counter (OTC) derivatives
(such as credit default swaps or CDS) to provide transparency; hedge funds
and investment advisors must register with the Security and Exchange
Commission (SEC); and credit rating agencies must improve their
operation. Furthermore, the so-called Volcker rule, which prohibits U.S.
banks (and U.S. branches of foreign bank)from engaging in proprietary
trading and investing or sponsoring private investment funds, will be
enforced.
3. Consumer Protection. A Consumer Financial Protection Agency is to be
established to regulate and protect consumers from abuses by financial
institutions in the provision of mortgages, credit cards, and other consumer
financial products.
Some of these reforms may take years to implement and may be watered down in
their applications. The banking sector, having lost the battle to weaken the Reform
Act, is now planning to take action to slow down the application of the Act and
weaken its provisions.
In Europe, EU finance ministers approved in September 2010 the proposed
overhaul of the bloc’s patchy system of financial supervision rules by creating three
new EU-wide supervisory authorities for banking, insurance and securities market,
as well as a European Systemic Risk Board housed in the European Central Bank to
warn about threats (such the rise of asset bubbles) to financial stability. The new
rules are to be formally approved by EU member states and the European parliament
18
DECEMBER 2010
before the end of 2010 and to take effect by mid to late 2012. The banking agency is
to be based in London, the insurance regulator in Frankfurt, and the market trading
watchdog in Paris.
The EU must also agree on rules governing hedge funds and short selling
(both of which contributed significantly to the financial crisis), and on whether to
establish European-based credit rating agencies to counter the dominance of
American companies. Standard OTC derivatives are to be processed through clearing
houses and disclosure of short selling (where traders bet on the fall in the price of a
security or share) are to be increased. These proposals will closely align the EU with
the new financial regime which is coming in force in the United States.
Global financial system reforms. There is a growing recognition that the
financial system is global and so it requires global regulations for its smooth
operation. To this end, at its September 12, 2010 meeting, the Basel Committee on
Banking Supervision announced new capital requirements for banks to be presented
to the Seoul G-20 Meeting in November 2010. The Basel III Accord proposes to
increase the amount of reserve capital that banks must keep from 2 percent to 4.5
percent by January 2015. In addition, banks will be required to hold a “capital
conservation buffer” of 2.5 percent to withstand future periods of stress bringing the
total common equity requirements to 7 percent of their total assets by 2019. Basel III
also seeks to transfer OTC derivatives to organized exchanges and to limit short
selling. Although is difficult to establish global consistent financial regulations,
especially now that new financial centers are arising and financial and economic
power is shifting to some of the most dynamic emerging markets, they are essential
to establish a level-playing field and avoid regulatory arbitrage and fragmentation.
Unsustainable structural imbalances – primarily between the United States and
China – must also be reduced before they lead to a new global financial crisis. The
United States has to stop living beyond its means by reducing consumption and
increasing savings, while China must revalue its currency an increase domestic
consumption. Otherwise, the dollar may collapse in the face of continued
unsustainable U.S. trade deficits financed by financial capital inflows from China
(which continues to accumulate huge amounts of dollar reserves) and trigger a new
global financial crisis.
12. Can Financial Crises Be Prevented?
During the past 25 years, there have been many financial crises. World Stock markets
collapsed in October 1987; the U.S. faced the dot.com crisis in 2001 and the current
crisis triggered by sub-prime home mortgages in 2007-2008, which then spread to most
of the rest of the world. The U.S. also faced the saving & loans crisis in the late 1980s.
There have also been several financial crises in emerging markets during the past two
decades, with serious repercussions in advanced countries as well. There was a
financial crisis in Mexico in 1994-1995, South East Asia 1997-199, Russia in summer
VOL.7 NO.2
19
1998, Brazil in 1999, and Turkey and Argentina in 2001-2002 (see, Reinhart and
Rogoff, 2010).
Are financial crises inevitable? Are we set for another crisis before the end of this
decade? Of course, this is what the new Basel III regulations are trying to prevent. But
it will take years before the new regulations are put in place and banks will have
satisfied the higher capital requirements – and a new financial crisis may occur before
then.
There is also the danger that reforms of the international financial system may
take steps that would have prevented or minimized past crises, but will not be able to
anticipate and prevent future crises, which will be different and arising from different
quarters. The Maginot Line would have been useful in World War I but was useless in
World War II. Reforms seem more successful in punishing, not preventing, financial
excesses and fraud after they occur.
Thus, financial reforms must be broad, but general. Broad, because they must
encompass the entire financial sector. The current crisis arose in the less regulated
investment banking sector, not in the better regulated commercial banking sectors.
Financial reforms also need to be general and subject to interpretation by financial
regulators, rather very specific and pointed. The reason is that money is fungible –
closing one avenue leads brilliant financial operators to find other ways around the
specific regulations in trying to earn large profits.
Are financial crises then unavoidable? Are they the price that we must pay to
obtain all the benefits of financial liberalization and innovation? In the end, the ability
of the banking system to avoid a new crisis will depend, in part, on whether regulators
are able, not only to keep up, but to be one step ahead of new financial innovations –
which are often created primarily to avoid capital restraints. Perhaps, the best that we
can hope is for better financial regulations to prevent some crisis and reduce the
severity and cost of others.
20
DECEMBER 2010
Notes
1. Dominick Salvatore is Professor of Economics at Fordham University. Comments
from the participants at the 10th Biennial Conference of the Athenian Policy Forum at
the German Bundesbank and of an anonymous referee are gratefully acknowledged.
Dominick Salvatore. Fordham University, New York 10458, [email protected]
2. See: http://business.timesonline.co.uk/tol/business/economics/article5014463.ece.
3. See: http//www.fff.org/comment/ed0500d.asp.
4. See: http://www.nytimes.com/2002/08/02/opinion/dubya-s-double-dip.html.
5. See: http://www.ecb.int/press/pressconf/2008/html/is080703.en.html.
6. George Kaufman makes the same points in his excellent paper “The Financial
Turmoil of 2007-2009: Sinners and Their Sins,” also presented at 10th Biennial
Conference of APF Bundesbank, Frankfurt, July 28-31, 2010.
7. http://www.govtrack.us/congress/bill.xpd?bill=h111-4173.
References
Congressional Budget Office (2009) “Lawmakers Warned About Health Costs,”
The Washington Post, July 17, 2009, 1.
IMF (2010a), International Financial Statistics (IMF, Washington, D.C.).
IMF (2010b), World Economic Outlook (IMF: Washington, D.C.), April.
Kaufman, George (2010), “The Financial Turmoil of 2007-2009: Sinners and Their
Sins,” Paper presented at the 10th Biennial Conference of APF Bundesbank,
Frankfurt, July 28-31, 2010.
Metzler, Allan (2009),
http://www.econtalk.org/archives/2009/02/meltzer_on_infl.html
Reinhart, Carmen M. and Kenneth S. Rogoff (2010), This Time Is Different: Eight
Centuries of Financial Folly (Princeton, N.J.: Princeton University Press, 2010).
Salvatore, Dominick (2010), “China’s Financial Markets in the Global Context,”
The Chinese Economy, November-December, 8-21.
Stiglitz, Joseph, Jonathan M. Orszag and Peter R. Orszag (2002), “Implications of
the New Fannie Mae and Freddie Mac Risk-based Capital Standard," Fannie
Mae Papers, Volume 1, Issue 2, March.
WTO (2010), International Trade Statistics (WTO, Geneva).
OECD (2005-2010), OECD Economic Outlook (OECD, Paris).
Incentives in the Financial Crisis of Our Time
Robert W. Kolb 1
Loyola University Chicago
Abstract. This article traces the incentives that infected every part of the chain of
relationships that constitute the Originate-To-Distribute (OTD) Model of mortgage
production. At every step, the OTD Model introduced incentive conflicts that were
absent or largely ameliorated in the old Originate-To-Hold (OTH) Model. The article
shows how these incentive conflicts helped to create the bubble in housing prices in
the United States that started to deflate in 2007 and that led to the financial crisis of
2007-2009.
Further, the article emphasizes the moral dimension of the setting of incentives,
which has been largely neglected in economic thought, and which was surely ignored
in the establishment and use of the OTD Model. The article argues that setting
incentives and acting upon the incentives set by others both have a moral dimension
that is generally neglected, but one that should have a central role in economic policy
generally and in the financial crisis particularly.
JEL Classification: D02, D14, D18, G18, G21, G24, H11, H12, H23, H81, K23,
L22, L51, L85, L88
Keywords: Credit agencies, Ethics, Executive compensation, Financial crisis,
Incentive alignment, Incentives, Mortgages
“We hope to do to this industry what Wal-Mart did to theirs,
Starbucks did to theirs, Costco did to theirs and Lowe’s-Home
Depot did to their industry. And I think if we’ve done our job, five
years from now you’re not going to call us a bank.”— Kerry K.
Killinger, chief executive of Washington Mutual, 20032
1. Introduction
There is no shortage of suggested causes for the financial crisis that originated in
2007. Perhaps the most prominent single candidate for the honor of prime cause is
the collapse in housing prices that followed the price peak of 2006. Close contenders
for pride of place among the causes of the crisis are personal avarice, corporate
greed, “global imbalances,” and the loosening of lending standards. Other potential
causes also receive their share of the blame, including: predatory lending, predatory
borrowing, excessive liquidity in the financial system, and too much, too little, and
21
22
DECEMBER 2010
the wrong kind of governmental regulation. In addition, the industrial organization of
mortgage lending as it has emerged in recent years, with mortgage brokers and
arrangers, the institutional design of securitizing mortgages, and mortgage
derivatives, also are assigned a causal role in many accounts.
In one way or another, the idea of incentives plays a critical role in almost all
accounts of the housing crisis. There appears to be a virtually universal recognition
that “bad incentives” on the part of many actors played a central role in the financial
crisis that has grown out of a housing finance system run amok. Because managing
and correcting incentives will play an important role in restoring order and efficiency
to credit and housing markets, conceptual clarity that distinguishes among
incentives, awards, rewards, opportunities, reasons, interests, temptations, and so on,
will be quite important.
In the housing finance system of the United States, recent financial
innovations generated a movement from a quite simple model of mortgage
production (the originate-to-hold model) to a model that was much more complex.
This new originate-to-distribute system of mortgage finance resulted in a new and
complex structure of mortgage finance that created an almost bewildering system of
incentives. This paper analyzes the systems of incentives that pervade the originateto-distribute model of mortgage origination, which involves eleven principal actors:
mortgagors (homeowners), mortgage brokers, appraisers, initial lenders, mortgage
servicers, warehouse lenders, due diligence firms, Fannie Mae and Freddie Mac,
investment banking firms, rating agencies, and ultimate private investors. This paper
provides a fairly detailed analysis of the incentives at play at the head of this chain
by focusing on the mortgagor, mortgage broker, appraiser, and initial lender.
This paper argues that participants in the originate-to-distribute model
operated within a system of incentives that were perverse in many ways. In some
cases, incentives for unethical behavior appear to have been created intentionally. In
other instances, careless contracting provided incentives that induced some of these
actors to behave in ways that were unethical. In still other cases, the incentives
encouraged actors to operate in a manner that was explicitly contrary to their
fiduciary duties. In other situations, these incentives functioned in an environment
where duties were either denied or unclear, but the environment encouraged an
unethical behavior that transgressed decent business behavior.
In sum, the paper investigates the chain of perverse incentives that
characterized the originate-to-distribute model of housing finance in the early 21st
century. As such, it attempts to provide a systematic analysis of how the originate-todistribute model functions, but it does so from a perspective that focuses on ethical
and unethical behavior in an environment of powerful incentives.
Section II reviews an earlier model of mortgage finance, the so-called
originate-to-hold model to use as a contrast to the “new, improved” model of
mortgage production, the originate-to-distribute model. Against the background of
the originate-to-hold model of Section II, Section III of the article discusses
incentives among mortgagors, mortgage brokers, appraisers, and initial lender.
VOL.7 NO.2
KOLB: INCENTIVES IN THE FINANCIAL CRISIS
23
Section IV of the paper discusses the theory of incentives, especially as they
play a central role and briefly discusses the application of incentive alignment in
typical executive compensation contracts. Section V analyzes the ethics of
incentives, especially as incentives arise in business. As we will see, incentives and
incentive-like arrangements in this originate-to-distribute process often provided a
framework in which individuals and institutions engaged in unethical and socially
destructive behavior. Section VI concludes the article.
2. The Originate-to-Hold Model of Mortgage Production: A Perspective on
Incentives
Within the living memory of many in the United States, there was no subprime
mortgage market and no securitization of mortgage financing in the contemporary
sense. Instead the creation of mortgages was once a fairly simple matter with few
participants and a structure that was easy to understand, and the market was
dominated or even entirely constituted by an originate-to-hold model of industrial
organization. The principal actors were the prospective mortgagor, or home buyer,
and a lending institution that funded the mortgage loan, which was usually a local
savings and loan association. The entire process was circumscribed by a somewhat
stifling regulatory regime, with savings and loan associations (S&Ls) being tightly
regulated at both the state and federal level. The key federal regulators were the
Federal Home Loan Bank Board (FHLBB) and the Federal Savings and Loan
Insurance Corporation (FSLIC). 3
Most mortgages involved the application of a prospective home buyer for a
mortgage that would be issued by a financial institution, usually a savings and loan
association (S&L). In the typical scenario, the S&L would test the creditworthiness
of the mortgage applicant and investigate the quality of the property by hiring its
own appraiser. If it decided to grant the loan, the property would be in the
geographically confined service area of the S&L. Further, the S&L would typically
hold the mortgage in its own portfolio of assets for the life of the loan and would
service the loan itself. The S&L would fulfill this role within a structure of elaborate
regulation carried out by both state and local regulatory authorities. Figure 1 depicts
the structure of this relatively simple institutional arrangement—the originate-tohold model of mortgage production.
In a typical transaction the prospective homebuyer would approach the S&L
to apply for a loan, with the S&L typically being a local institution of long standing
in the community. The S&L would carefully review the application and verify the
representations made by the home buyer, especially with regard to the applicant’s
creditworthiness, sufficiency of assets, adequacy of down payment, and verifiability
of employment. In addition, the S&L would investigate the property that was to
serve as collateral for the loan, typically hiring an appraiser to report on the condition
and value of the property. Once satisfied with the financial probity of the borrower
and the value of the collateral, the S&L would issue the mortgage. Typically a
mortgage was a level-payment, self-amortizing loan, with monthly payments, and a
24
DECEMBER 2010
maturity of 30 years. The S&L would collect payments on the loan (including
escrow payments for taxes and insurance) and would hold the mortgage as an asset
on its balance sheet for the entire life of the mortgage.
Because S&Ls were the local, less sophisticated cousins of commercial
banks, and because their primary business was mortgage lending, they were allowed
to pay slightly higher deposit rates 25 basis points (one-fourth of one percent) higher
than commercial banks. One of the perceived advantages of this differential was to
help S&Ls secure deposits that could be used to stimulate home ownership, long a
public policy goal in the United States.
This system of mortgage finance grew up in the aftermath of the Great
Depression, came into full flower with the growing wealth of the United States in the
1950s-1970s, and reached its full potential before 1980. Nothing if not conservative,
this institutional arrangement for housing finance served the public fairly well, but
by the mid-1970s some important limitations had become quite obvious. So long as
the S&L used the deposits it received to fund housing purchases and held those
mortgages to their maturity, the growth in housing finance was constrained to match
the grown in deposits at the S&L, and this deposit growth roughly paralleled the
growth of the economy as a whole. In a certain sense, S&L deposits tied up in
funding a particular mortgage represented dead money during the life of the loan, in
the sense that those deposits could not be used to expand home financing.
The analysis of incentives in the originate-to-hold model is rather
straightforward as there are only three key participants, the mortgagor, the lender,
and the appraiser, and the relationships among these parties are also quite clear and
relatively free of conflicts. The mortgagor had three principal reasons to seek a home
loan. In most cases, the mortgagor sought a loan simply to fund the purchase of his
or her own home. In addition, a homeowner might apply for a new loan to obtain a
better financing rate, or to secure additional funds on a second mortgage to enhance
the value of the property through renovation. The mortgagor’s interest in the
character of the loan were equally straightforward—simply to secure the best
financial terms available.
The lender had an incentive to lend funds at the highest feasible rate and
charge as much as possible. Of course, this desire was not unconstrained. With the
typical lender operating in the same community as the borrower, the lender, like any
merchant, faced some competition and feared being perceived as too rapacious, so
the ordinary constraints that many merchants feel held for the lender as well. Further,
the lender had a strong incentive to ensure that the mortgagor could actually meet the
obligations of the mortgage agreement. Eviction and repossession is costly and
unpleasant—even for heartless financial institutions. After all, these financial
institutions want to operate in the financial industry, not in the property management
or real estate workout industry. To ensure that the mortgagor would honor the terms
of the mortgage agreement, the lender would want to assess the financial capacity of
the mortgagor carefully and confirm that the mortgagor has sufficient cash flow to
cover the loan payments. Beyond that, the lender would usually demand a substantial
VOL.7 NO.2
25
down payment so that the mortgagor has sufficient “skin in the game” to militate
against abandonment of the property. To protect against the worst outcome of
foreclosure, eviction and repossession, the lender would want to establish terms to
ensure that the property will be worth the loan value. A significant down payment
and a property fully worth the sale price both help to ensure that the lender cannot
lose. If the mortgagor pays as promised, the best situation is achieved. If the
mortgagor defaults and the lender takes over the property, the down payment and
high value of the property relative to the loan amount assure that the lender can
recoup the principal balance of the loan, along with payment for the expenses of
repossession and re-marketing the home.
One part of the lender’s due diligence in this process was to secure an
appraisal of the property. While the mortgagor undoubtedly ultimately paid for the
appraisal if the loan goes through, the lender hires the appraiser and has good reason
to instruct the appraiser to value the property accurately or perhaps even
conservatively, because the lender ultimately looked to the true value of the property
to recoup its principal in the event of default by the mortgagor. The appraiser must
serve the financial institution, which can choose among many appraisers, so the
appraiser has an incentive to provide the kind of appraisal that the financial
institution desires. Happily for the appraiser, the financial institution had every
reason to want an honest appraisal, freeing the appraiser from the conflicts that arise
in other contexts.
3. The Originate-to-Distribute Model of Mortgage Production: A Perspective on
Incentives
While the originate-to-hold model involved simple relationships with few
incentive conflicts or incompatibilities, the overall institutional arrangement
effectively set an upper bound on the expansion of homeownership. However, this
upper bound was increasingly seen to be too stultifying and incommensurate with a
national housing policy that sought to stimulate homeownership and especially to
bring the perceived benefits of homeownership to segments of the population with
low rates of homeownership, most notably minorities. This recognized limitation of
the originate-to-hold model of mortgage financing played a significant role in the
liberalization of financial institutions in the United States that began in the early
1980s. The securitization of mortgages and the originate-to-distribute model of
mortgage production can reasonably be seen, at least in significant part, as products
of the desire to stimulate homeownership and the social goal of creating a more
inclusive social structure that would bring minorities and other disadvantaged
segments of the population closer to parity with those groups already enjoying high
levels of homeownership. Significant progress toward this goal of expanding home
ownership was largely achieved. Figure 3 shows the growth of homeownership in the
United States with its significant and rapid expansion in recent years to its peak in
2006.
26
DECEMBER 2010
In the earlier period, say before 1980, the process of qualifying for and
consummating a home loan was lengthy, expensive, and cumbersome. Application
fees, title insurance, origination fees, and so on were quite common and were
collectively quite high compared to the principal being financed. Further, these fees
typically had to be paid in cash. So these high fees combined with large down
payments served as a substantial barrier to homeownership, particularly to minorities
and persons of lower financial capacity. Considered just by themselves, the high fees
relative to the principal of the loan also discouraged rapid acquisition and disposal of
property simply because transaction costs were too high to make such rapid
purchases and re-sales financially productive.
Over the years, the originate-to-distribute model slowly supplanted the
originate-to-hold model and had surely reached dominance before 2000. During this
same period competition among many more financial institutions reduced the total
costs of securing a loan for many borrowers. But some potential mortgagors
continued to face quite high costs, as we shall see. However, lower transaction costs
for some potential mortgagors meant that some participants in the market were able
to contemplate the purchase of a house with a rapid subsequent resale if market
conditions permitted. Rather than chronicle all the phases in the transition from a
high transaction fee/originate-to hold model to a low transaction fee (for
some)/originate-to-distribute model, this discussion of the originate-to-distribute
model focuses on the fully developed model as it played such an overwhelming role
in expanding mortgage financing and as it helped to stimulate the current financial
crisis.
Figure 2 shows the fully developed originate-to-distribute model of mortgage
production, and the contrasts between it and the originate-to-hold model of Figure 1
could not be more striking. Figure 2 clearly shows many more participants covering
the same total distance between initial borrower and ultimate lender. The substantial
increase in the number of participants and linkages also indicates a much greater
complexity for the originate-to-distribute model. But perhaps most importantly, the
linkages in Figure 2 are generally between independent parties with their own
interests. This means that contractual arrangements between these various parties
must seek to overcome the disparity in their interests so that both parties to any
agreement participate in a process of mutual benefit. That is, each of these
contractual arrangements must create a high degree of incentive alignment if the
overall structure is to succeed.
The full elucidation of all of the linkages shown in Figure 2 is beyond the
scope of this article. Instead, this study focuses on the incentives near the beginning
of the chain from the mortgage borrower to the initial lender or originator. In a
certain sense, this is the same ground covered in Figure 1, with several exceptions.
First, the initial lender in the originate-to-hold model of Figure 1 is also the final
lender and the ultimate provider of funds. By contrast, the initial lender in the
originate-to-distribute model of Figure 2 is just one of the potentially many providers
of funds, and this initial lender makes funds available to the borrower with the
VOL.7 NO.2
27
intention of very quickly recapturing those extended funds by re-selling the
originated mortgage to the next participant in the chain. Thus, the initial lender faces
radically different incentives in the two models. In particular, the new lender will not
suffer directly if the loan he makes but then sells should happen to go into default.
Second, the new world of mortgage lending (reduced transaction fees and the
originate-to-distribute model) widened the array of possible incentives for
mortgagors. Third, the relevant portion of Figure 2 introduces a new actor, the
mortgage broker, who functions as an information intermediary between the
mortgagor and the initial lender. As we will see, the role of the mortgage broker
brings considerable complexity to the range of available behaviors for the various
participants, including the mortgage borrower and the initial lender. This new world
of mortgage lending transfigured the incentives of the key participants—mortgagor,
appraiser, and initial lender. In addition, the entry of the mortgage broker with her
own conflicted incentives further complicated the mix of potentially toxic incentives.
We consider the mortgagor, lender, and mortgage broker in turn.
The mortgagor had two sets of incentives. The old incentives still pertain: to
secure a home, to refinance to better terms, to secure a home improvement loan. But
the homeowner now had other reasons to consider obtaining a loan. If the mortgagor
already owned a home with substantial equity, the homeowner could refinance and
gain access to that equity from a “cash-out refinancing.” The equity in the home that
was thus converted to cash could be used for any purpose the homeowner desired.
The lower transaction costs for obtaining a mortgage and the greater liquidity the
originate-to-distribute model brought to the entire market created these new
possibilities. In the old world of mortgage finance, lenders virtually limited their
financing only to owner-occupied housing. Any other financing would be offered on
much less favorable terms, if at all. This earlier policy was a reflection of the much
more conservative nature of the industry and was induced in part by the relative
scarcity of funds available for mortgage lending.
In a world of consistently and rapidly rising home values (per Figure 4), the
easier financing terms available in the originate-to-distribute model made the
prospect of buying residential real estate as a pure investment property attractive in a
way it never had been before. The prospective home buyer might contemplate the
purchase of a house as a pure investment with the intention of quickly reselling it for
profit—the process of “home flipping.” In some cases, mortgagors might also be
induced to secure a loan with no special incentive of their own and no independent
urge to enter the housing market, but were in some cases merely the victim of a sales
job in which other parties (for example, a mortgage broker) might approach them
and convince them to refinance an existing property or to acquire a new one.
If the lender plans to retain the mortgage in its permanent portfolio, it has the
same essential incentives that prevailed under the originate-to-hold model. But, if the
lender plans to originate and then sell the mortgage straightaway, the originate-todistribute model completely alters the lender’s incentives. For a lender planning to
sell a mortgage soon after origination, the lender mainly desires to create a highly
28
DECEMBER 2010
salable mortgage with the characteristics that potential purchasers demand. As in the
originate-to-hold model, the lender wants to create a mortgage with an attractively
high interest rate and features that hold out the possibility of additional fee income to
the purchaser, such as prepayment penalties that will accrue to the ultimate investor.
For the lender planning to sell a mortgage, some factors that were crucially
important in the originate-to-hold model now hold little interest. First and most
crucially, the originate-to-distribute model frees the lender from concern whether the
mortgagor can make the promised payments. For the initial lender, the typical terms
of sale are made on a no-recourse basis after an initial period of 45-60 days. So if the
mortgagor avoids immediate default, the originator’s sale of the mortgage is final.
Further, the lender now has virtually no concern about the value of the property
relative to the loan amount. After the sale is final, any problem of default,
repossession, and extraction of value from the house will be the financial problem of
the subsequent purchaser of the mortgagor. (As we will see, the initial lender may
well be involved in the process as the servicer of the mortgage, but it will not bear
the principal risk if the mortgagor defaults.)
While the initial lender may be indifferent to the mortgagor’s ability to pay or
to the ultimate investor’s ability to recover the principal lent in the event of a
repossession, some factors that were important to the lender planning to hold the
mortgage now are not merely a matter of indifference, but the lenders incentives
have changed completely. Consider a house sold for $200,000 that is worth the sales
price exactly. The lender in the originate-to-hold model might demand a 20 percent
down payment, making the mortgage balance $160,000, and this would be the
maximum the originator might finance if it planned to retain the mortgage. However,
if the intention is to sell the mortgage, these scruples are not merely a matter of
indifference, but they actually become adverse to the initial lender. The initial lender
can make more money by securing an inflated appraisal and requiring a lower down
payment. Let us assume that the lender secures an inflated appraisal stating that the
property is worth $220,000 and reduces the down payment to 5 percent, both of these
being fairly conservative assumptions compared to many actual practices. With these
assumptions, the mortgage balance will be $209,000, or 31 percent more than the
amount the lender would be willing to finance if planning to hold the mortgage.
Assuming the same interest rate on the mortgage, the lender would receive 31
percent more for selling this risky mortgage than it would for selling a mortgage that
was more conservative—such as a mortgage it would be willing to hold in its own
portfolio.
In addition to inflating the appraisal amount and reducing the down payment
to increase the principal balance for resale, the initial lender also has a strong
incentive to originate a mortgage with a higher interest rate. These high interest rates
above the prevailing market rate are said to bear a yield-spread-premium. In the
originate-to-hold model this same incentive exists, but it is constrained by the fact
that the lender does not want the mortgagor to fail. So there is less benefit from
inflating the interest rate to a point that the mortgagor cannot meet the payments. In
VOL.7 NO.2
29
the originate-to-distribute model, subsequent defaults will not be the problem of the
lender, but will fall on investors further down the chain, and the higher the interest
rate on a given mortgage, the more valuable it is for the initial lender to sell.
There are other ways for the initial lender to profit by originating a mortgage.
It is quite common for the initial lender to service the mortgage—that is to collect the
monthly payments and to manage the escrow account for taxes and insurance.
Typically, the annual fee for this service is 25 basis points or one-fourth of one
percent of the mortgage initial principal amount. Initiating a mortgage assures this
future business for the originator. Further, many subprime mortgages were structured
in a way to encourage periodic refinancing, which brings repeat business to the
originator with more fees and more opportunities for profit. 4 All of these factors just
discussed benefit the originator under the originate-to-distribute model by offering
more income and less risk than the originator would bear under the originate-to-hold
model.
Given these much more complex and varied incentives for both mortgagor
and lender, the introduction of a mortgage broker into the relationship creates many
interesting and potentially toxic opportunities. As the name implies, a mortgage
broker arranges or facilitates mortgages, rather than being a lender or borrower. In
essence, a mortgage broker acts as an information intermediary who uses her
knowledge to connect prospective mortgage borrowers with prospective lenders.
Possessed of valuable knowledge about prospective borrowers and lenders, the
mortgage broker is in a position to provide a valuable service to both the borrower
and lender. For the prospective borrower, the mortgage broker could help the
borrower complete the mortgage application and could direct the borrower to the
financial institution most likely to fund the loan at a reasonable rate. In the happiest
situation, the mortgage broker might even use her knowledge of lending institutions
to help the borrower secure the most favorable terms available in the market. The
mortgage broker could also benefit the lender by helping the lender make more loans
than would otherwise be possible. From the lender’s perspective the mortgage broker
could also provide a valuable service by pre-screening prospective borrowers and
helping them to get all necessary information in order, thereby reducing the lender’s
cost of vetting the borrower and completing the loan. By providing such services, the
mortgage broker would earn a reasonable compensation.
The mortgage brokerage industry is a fairly recent development. It was only
in 1960 that the first trade association for mortgage brokers was formed, 5 with the
National Association of Mortgage Brokers not being established until 1973. In 1991,
there were about 14,000 brokerage firms, but by 2004-2006, approximately 53,000
mortgage brokerage firms employed somewhere in the range of 200,000 to 420,000
people. By the early 2000s, brokers originated about two-thirds of all residential
loans. 6
Virtually without exception, mortgage brokers are paid by the lending
institution at the time the loan is funded and the transaction closes, but the source of
all revenue for any mortgage stems ultimately from the mortgagor. The mortgage
30
DECEMBER 2010
broker typically receives a fee expressed in “points” or one-hundredths of the
principal balance on the mortgage, so that a fee of two points on a $100,000
mortgage would bring the mortgage broker a payment of $2,000. The broker receives
payment if and only if the mortgage loan is consummated, an incentive arrangement
that encourages mortgage brokers to ensure that the deal goes through. In the
subprime market, the financial institution that funded the loan would pay more to the
mortgage broker for a mortgage with more desirable financial terms, with the
broker’s compensation ranging from one to as many as four or, in extreme cases,
seven points.
As the name implies, a mortgage broker, is not a principal in the lending
transaction, but a facilitator, who owes no fiduciary duty to either the prospective
borrower or the initial lender. At least that was the conceit of the mortgage broker
industry. The recommended disclosure practice of the National Association of
Mortgage Brokers of 1997 includes the following language:
In connection with this mortgage loan we are acting as an
independent contractor and not as your agent… we do not
distribute the products of all lenders or investors in the market and
cannot guarantee the lowest price or best terms available in the
market….The lenders whose loan products we distribute generally
provide their loan products to us at a wholesale rate. The retail
price we offer you—your interest rate, total points and fees—will
include our compensation. In some cases, we may be paid all of
our compensation by either you or the lender. Alternatively, we
may be paid a portion of our compensation by both you and the
lender. For example, in some cases, if you would rather pay a
lower interest rate, you may pay higher up-front points and fees.
Also, in some cases, if you would rather pay less up-front, you
may be able to pay some or all of our compensation indirectly
through a higher interest rate in which case we will be paid directly
by the lender. We also may be paid by the lender based on (i) the
value of the Mortgage Loan or related servicing rights in the
market place or (ii) other services, goods or facilities performed or
provided by us to the lender. 7
Of course, whether such disclosures were made is a different question. But the
disclosure, if made and understood, certainly makes the potentially adversarial
relationship between the prospective mortgagor and mortgage broker quite clear. In
recent months, some states have passed laws to insist that mortgage brokers owe a
fiduciary duty to borrowers.
Even in the heyday of the originate-to-distribute model, many mortgages
were originated without any mortgage broker. For such mortgages, creating a
mortgage began in the old-fashioned way—a prospective home buyer applied
VOL.7 NO.2
31
directly to a financial institution for a loan. Such mortgagors were, for the most part,
wealthier and more financially sophisticated. One of the most distinctive features of
the mortgage finance crisis turns on the role played by the mortgage broker, so the
subsequent discussion considers only those mortgages that were originated with the
assistance of a mortgage broker.
As we have seen, from the point of view of the originating financial
institution at least in the originate-to-hold era, a given mortgage proposition is more
attractive if the mortgagor is wealthier, has a greater income, and has a higher credit
rating; if the property is more valuable; if the down payment is larger; if the fees that
the mortgagor will pay are higher; and if the interest rate on the mortgage is higher.
In the subprime market, as it developed in the early twenty-first century, the
originating lender did not, or even could not, verify the information about the
mortgagor or the property, due to the interposition of the mortgage broker. Thus, the
financial institution often funded mortgages based on representations made by the
mortgagor or the mortgage broker.
The financial institution’s lack of direct knowledge of the mortgagor means
that the mortgage broker occupies a crucial position in the chain running from the
mortgagor to the ultimate investor. In many instances, only the mortgage broker
really knows the mortgagor and his capacity to fulfill his obligations. Because the
mortgage broker stands between the mortgagor and the originator, the originator may
feel less concern about how it treats the mortgagor. The mortgage broker’s privileged
information position and the greater remove of the originator from the mortgagor
gives the mortgage broker opportunities and incentives to play two dishonest games.
First, the mortgage broker might cooperate with the originator to create a mortgage
that harms the mortgagors, a practice usually known as predatory lending. Second,
the mortgage broker might cooperate with the mortgagor to abuse the lender, a
practice known as mortgage fraud or predatory borrowing.
“Predatory lending” is a controversial term subject to competing definitions,
some of which are absurdly expansive and include as predatory many practices and
loan features that can be quite desirable for particular borrowers and quite injurious
to others. Without striving for precision, the following can serve as a working
definition of predatory lending: “knowingly creating a mortgage that the mortgage
broker and originator know, or should know, is financially injurious to the
mortgagor.” Without trying to give an exhaustive account, two common practices
seem clear examples of predatory lending. First, creating a mortgage with a yield
spread premium—an interest rate above the going market rate of interest for which
the borrower could qualify—clearly injures the borrower. Second, creating a
mortgage with a principal balance having implied payments that the borrower cannot
financially sustain also clearly injures the borrower, as such a practice can be
expected to lead to default. These practices harm the borrower but benefit the
mortgage broker and the originator by improving their income prospects from the
mortgage. As the mortgage broker receives a number of points for helping to
originate the loan, a higher principal balance benefits the mortgage broker. Further,
32
DECEMBER 2010
the number of points received will be larger, and sometimes significantly larger, for
mortgages with a yield spread premium.
There is a second set of perverse incentives that may influence the mortgage
broker, and that is to cooperate with the borrower to defraud the lender. Two of the
most common abuses are to commit owner-occupancy fraud or to falsify financial
resources. Based on decades of experience, defaults tend to be lower for owneroccupied dwellings, so conscientious lenders prefer to lend on homes that are to be
occupied by the mortgagor, and they give better terms for such mortgages. In the
now seemingly distant and happy world of steadily rising home prices, many
participants sought to buy homes as pure investment properties with no intention of
ever occupying them, so lying about occupancy intentions was financially
advantageous for the borrower as it helped to secure better terms for the mortgage,
and so brokers would sometimes assist in this fraud.
Compared to typical borrowers, particularly those in the subprime market,
mortgage brokers have a superior understanding of what financial resources the
borrower must have to secure a particular mortgage. The mortgage broker can use
this knowledge to help the borrower secure a mortgage for which they might not
actually qualify. Two examples illustrate the point. First, the mortgage broker might
encourage the borrower to secure down payment monies by borrowing from a
relative and then cooperate with the borrower to hide this additional indebtedness
from the lender. Second, it became a frequent practice for lenders to grant mortgage
loans based on the mere statement of income by the borrower. (These loans were
known as “stated-income loans” or more commonly as “liars’ loans.”) Knowing
what the lender needs to hear, the mortgage broker could guide the borrower in
making a false statement sufficient to qualify for the desired loan.
In the case of mortgage fraud, there is a question as to which party is
defrauded. If the initial lender merely intends to sell the loan, the initial lender might
not object to these practices of mortgage fraud or predatory borrowing. After all, the
initial lender will profit just by getting the deal done, so the initial lender might
connive with the mortgage broker to help the mortgage fraud along, confident that a
subsequent investor in the mortgage will be the party that actually bears the cost of
the fraud.
Perhaps in many cases the mortgagor, mortgage broker and initial lender all
cooperated to secure what each wanted: the mortgagor lied about his occupancy
intentions and financial resources; the mortgage broker helped to fashion the lies and
then transmitted them to the lender; and the lender pretended to believe the mortgage
broker’s representations. In this case, the mortgagor got the loan she wanted, the
mortgage broker got a fat fee, and the lender originated a mortgage that it
immediately resold for a significant profit.
This conflicted incentive relationship between initial lender and mortgage
broker relates to the interpretation of the payment scheme that initial lenders offered
to mortgage brokers. Did these lenders offer incentives to mortgage brokers with the
understanding that they were inducing brokers to engage in predatory lending and
VOL.7 NO.2
33
predatory borrowing? Or was the pay design merely the lazy and witless creation that
inadvertently pointed mortgage brokers toward predatory lending and borrowing?
There seems little doubt that both behaviors were in play, with different lenders
falling into each category. For some lenders it now seems clear that they consciously
engaged in a game to generate new mortgages of poor quality that they could sell
into the securitization pipeline. Yet when one views the wreckage of so many
financial institutions destroyed by their own actions, it also seems equally clear that
some institutions had little understanding of the ultimate effects of their actions.8
As this discussion of the part of the originate-to-distribute model shows, the
incentives facing participants in the mortgage market were radically transformed as
the originate-to-hold model was supplanted. A similar story of radically altered
incentives and the introduction of powerful incentive conflicts could be replicated for
nearly every one of the actors shown in Figure 2. However, with the conflicts
involving mortgagor-mortgage broker-appraiser-originator before us, it is best to
now consider some of the social and ethical dimensions of these conflicts.
4. Incentives in Financial Theory and Practice
In recent decades, and only in recent decades, the idea of an incentive has become
extremely important in economic thought, with a great stress on “incentive
alignment” and “incentive compatibility” between contracting parties. More than in
any other area, this idea has been most fully elaborated and tested in the realm of
incentive compensation for corporate executives. Two events stimulated an
increasing reliance on incentive compensation. First, Michael Jensen and Kevin
Murphy, in a 1990 article on CEO incentives, maintained that executives were paid
like bureaucrats and called for a movement to an incentive compensation scheme. 9
Second, in an effort to limit the total scale of CEO compensation, Congress passed a
law in 1993 allowing no more than $1 million of annual compensation to be treated
as a tax-deductible business expense, unless the additional compensation were
performance-based. 10 This new law and the increasing acceptance of the line of
thought advanced by Jensen and Murphy set the stage for incentive- or performancebased pay to become a huge portion of executive compensation. Consequently, stock
options soon became a large portion of CEO pay, in many instances constituting
fully half of total compensation (which would include salary, bonus, option grants,
retirement benefits, perquisites, and other monetary benefits limited only by human
ingenuity).
All of these developments were consonant with the dominant theory of the
firm, which regards a corporation as a nexus of contracts. That is, the firm is
essentially a contracting party that acquires through contract the goods and services
it needs to effect the firm’s mission. Said another way, the board of directors and
senior management act as the agents of the principals of the firm, who are the firm’s
shareholders.
Whenever a principal contracts with an agent to provide some service or
perform some task, the interests of principal and agent are virtually certain to diverge
34
DECEMBER 2010
to some degree. The agent, of course, has her own interests, so the agent will likely
pursue those interests rather than devote herself fully to the interests of the principal.
For example, a CEO commands the resources of the firm which belong to the
shareholders. In the case of a CEO who receives only salary as compensation, the
CEO might choose to consume a high level of perquisites (corporate jet, fine meals,
sumptuous offices with fine French furniture, and corporate apartments, etc.) which
benefit the executive exclusively, but for which the shareholders pay.
Aware of this problem, the savvy principal must structure a compensation
agreement that attempts to overcome this “incentive incompatibility” and that strives
to achieve “incentive alignment.” For example, the theory goes, if the CEO holds a
significant equity position in the firm, then the CEO will come to regard her interests
as being congruent with those of other shareholders. Within this frame of reference,
the CEO regards the dissipation of the firm’s resources on perks as a squandering of
her own funds. Thus, incentive alignment in contracting strives to reduce the
dissonance between the interests of the principal and the agent. An ideal contract
would induce the CEO to manage the firm just as the shareholders would if they
possessed the CEO’s expertise and were to operate the firm themselves. In all of
these contracts, however, there is the realization that no perfect incentive alignment
is likely ever to be possible.
CEO incentive compensation has spawned hundreds of academic articles,
thousands of news reports, and perhaps millions of hours of discussion, yet the CEO
contract is fairly simple. CEO incentives are simple in that there are essentially two
parties involved, a firm and an individual. The firm explicitly contracts with an
individual CEO and consciously tries to provide incentives that will encourage
management policies that benefit the firm. Yet the executive compensation literature
is replete with hundreds of examples of misfiring incentives and outrageously
unethical and anti-social outcomes, all against this background of intensively
analyzed and managed effort to provide just the right incentives to stimulate just the
right managerial behavior. 11
Consider, in contrast, how much more complicated the housing finance
industry had become at its zenith about 2006, with borrowers, mortgage brokers,
appraisers, lenders, securitizers, investors, rating agencies, regulators, and ultimate
investors all working on their piece of the market, all facing their own incentives,
and all pursuing their own interests. The sprawl of incentives at play in this market
has been well-recognized as the following selection of quotations indicates:
“… it is likely that flaws in the design and workings of the systems
of incentives within the financial sector have inadvertently
produced patterns of behavior and allocations of resources that are
not always consistent with the basic goal of financial stability.” 12
VOL.7 NO.2
35
“…we need to determine how incentive structures in the homemortgage market have fueled predatory lending and how these
incentives can best be countered.” 13
“In the event of delinquency, the servicer has a natural incentive to
inflate expenses. . .” 14
“Asset managers had an incentive to reach for yield . . .” 15
“an originator can have the incentive to collaborate with a
borrower in order to make significant misrepresentations on the
loan application . . .” 16
“High profit margins from rating RMBS and CDOs may have
provided an incentive for a rating agency to encourage the
arrangers to route future business its way.” 17
“Originating brokers had little incentive to perform their due
diligence and monitor borrowers’ credit worthiness, … This
phenomenon was aggravated by the incentive compensation
system for brokers…” 18
“… there is a prepayment penalty, creating an incentive not to
refinance early.” 19
“Incentive conflicts explain how securitization went wrong, why
credit ratings proved so inaccurate, and why it is superficial to
blame the crisis on mark-to-market accounting, an unexpected loss
of liquidity or trends in globalization and deregulation in financial
markets.”20
So almost all commentators agree that incentive problems played a significant
role in generating the financial crisis. Of course, the idea of incentive conflicts as a
cause of the financial crisis cuts across and remains compatible with the crisis having
other causes as well, such as technical flaws in the process of securitization or
defective governmental regulation. However, reflection on these quotations also
shows that these authors collectively use the word ‘incentive’ and its variants in a
wide variety of meanings.
In some instances, conscious design of compensation contracts provided
particular individuals with certain incentives—this would be analogous to incentives
in CEO incentive compensation contracts. In other instances, quotations make it
appears that the state-of-the-world provided the incentives—as in the case that
falling rain gives me an incentive to go indoors or hunger gives an incentive to find a
36
DECEMBER 2010
job. Some commentators stress that this or that institutional feature of the mortgage
market provided the incentive. In such instances, extremely complex institutions
have emerged without the conscious design of an individual or even a team of
individuals, and these institutions somehow provide incentives, or, at least, actors
find themselves with incentives that can only be traced to broad institutional
arrangements. In some usages, ‘incentive’ might be replaced with the word
‘opportunity’ without a loss, or perhaps even a gain, in precision so that we might
say that “loose lending standards provided an opportunity to submit false income
information.”
Keeping in mind this common lack of precision in talking about incentives,
including a similar lack of precision in the discussion of mortgage origination in
sections II and III above, we now turn to examine the ethical dimensions of
incentives with reference to the financial crisis.
5. Ethical Dimensions of Incentives in the Financial Crisis
The preceding discussion of incentives in the originate-to-hold and originate-todistribute models of mortgage production spoke of incentives very casually, but to
begin exploring the ethical dimensions of incentives in the financial crisis, this article
makes a basic distinction between incentives and incentive structures. Here incentive
will mean “a non-coercive inducement intentionally offered by one party that is
designed to elicit certain behaviors from the recipient.” If the inducement is coercive,
it is not an incentive, but is something else, such as a threat. To be non-coercive,
declining an inducement must leave the person offered the inducement no worse off
than she was before. By contrast, an incentive structure is “a state of affairs that is
not intentionally created to induce a particular behavior, yet which provides payoffs
or inducements for actors to behave in a certain manner.” Heavy roadway traffic
provides an incentive structure that should induce people to cross roads with care,
but drivers (unless playful or diabolical) are not providing incentives to pedestrians
to walk carefully.
In ordinary discourse “incentives” and “incentive structures,” in the sense
defined above, are often conflated, and we would find nothing amiss with someone
commenting that “crazy drivers provide a strong incentive for walkers to tread
carefully.” The same conflation of incentives and incentive structures occurs with
great frequency in discussions of the current financial crisis that began in the housing
sector, particularly with subprime mortgages, and that has now spread throughout the
international financial system and into the world’s real economy.
Incentive structures as defined above may be created by natural conditions,
by persons, by institutional design, or by some combination of nature and human
effort. Thunderstorms create an incentive structure that encourages one to seek
shelter, but nature does not seek to offer incentives to humans, although in an earlier
time people generally may have ascribed such agency to nature. Careless cash
management creates an incentive structure that encourages theft, but the sloppy
manager did not provide an incentive for employees to steal.
VOL.7 NO.2
37
Human agency often sets out systems that are designed to provide certain
incentives, but that actually provide unforeseen incentive structures that operate in
unintended ways. Thus, these systems turn out to be a mixture of both incentives and
incentive structures. For example, the Corporate Average Fuel Economy (CAFE)
standards enacted by Congress in 1975 were intended to improve fuel economy for
manufacturers’ car fleets, but as they did not apply to “light trucks,” a category that
turned out to include SUVs, this law also created an incentive structure that gave
manufacturers inducements to shift production from smaller cars to larger and less
fuel-efficient SUVs. Thus, in attempting to provide an incentive for car
manufacturers to craft more fuel-efficient vehicles, the law created an incentive
structure that induced them to produce gas-guzzling SUVs.
Viewed externally, there is an epistemological problem of distinguishing
incentives from incentive structures, because the distinction turns on the intention of
the person or institution that created the incentive or incentive structure. This is a
significant problem in assigning blame for some of the consequences of mortgage
lending in the financial crisis. For example, did initial lenders establish payment
policies that ignorantly create incentive structures that led to deceit and dishonest
actions by mortgage brokers? Alternatively, did they intentionally establish a system
that they could reasonably foresee would induce mortgage brokers to misrepresent
the borrowers to lenders? That is, did they provide an incentive for mortgage brokers
to engage in predatory lending and to assist in mortgage fraud? In the second
scenario, the initial lenders would knowingly make loans based on inflated
appraisals, falsely stated income, and dishonest declarations of occupancy intentions.
Given the importance of incentives and incentive structures in economic
thought and public policy, the literature focusing on incentives from a conceptual
standpoint is surprisingly scarce. In the Western tradition of philosophical thought,
the great masters of antiquity scarcely addressed the issue, and it certainly was not
conceptualized as the pervasive system within which we weigh our options and
choose our actions today. Part of the reason for this ancient neglect requires a brief
detour into the history of the idea of an incentive, a detour that shows our current
conception of ‘incentive’ to have rather recent origins.
In antiquity, philosophical psychology focused on a bipartite view of the
mind, with a division between the soul and body, between the noble and ignoble, and
between the rational and the irrational. This outlook is perhaps most vividly
encapsulated in Plato’s Phaedrus with the metaphor of two steeds that pull a chariot.
One horse is noble, the other ignoble, and while the noble steed pulls the chariot
straight ahead and under the direction of the charioteer, the unruly steed ignores the
charioteer and periodically plunges off the path and toward a ditch. Its context
encourages readers to understand this metaphor as contrasting reason with the
passions, with the charioteer representing reason, and the two steeds the noble and
ignoble passions.
This essential bifurcation of human motivation into reason and passions
persisted, but as Albert O. Hirschman argues, philosophers came to see passion as
38
DECEMBER 2010
destructive and reason as ineffectual, leaving only a bleak view of human nature.21
As Hirschman goes on to explain, in the seventeenth and eighteenth century,
“interest” quickly came to be seen as a third motivating force occupying a middle
ground between reason and the passions: “Interest was seen to partake in effect of
the better nature of each, as the passion of self-love upgraded and contained by
reason, and as reason given direction and force by that passion.”22 Interest combined
the rational with the appetitive, and Hirschman sees interest as the motivating spirit
of capitalism—the calculative and sober pursuit of advantage, contrasted with the
stasis of reason or the unbridled destructiveness of the wilder passions.
Thus, the pursuit of interest was seen as leading to an improvement in actual
behavior because it moderated the tendency to act from pure passion, and the pursuit
of financial gain was seen not only as innocent, but as morally improving. The
gentling effect of commerce on morals and manners was encapsulated in the thesis of
doux commerce or “sweet commerce,” and found one of its earliest and perhaps most
famous expressions in Montesquieu’s Spirit of the Laws, as this celebrated passage
forcefully states: “Commerce is a cure for the most destructive prejudices; for it is
almost a general rule that wherever we find agreeable manners, there commerce
flourishes; and that wherever there is commerce, there we meet with agreeable
manners.” 23 Or as Jerry Muller summarizes the same point: “A moral advantage of
commercial society, therefore, was that it channeled self-interest into less morally
corrupt forms than the society that preceded it.”24 This idea was adopted and
elaborated by other prominent writers in the following decades, such as Adam Smith,
Francis Hutcheson, and David Hume, and it finds a modern restatement in Deirdre
McCloskey’s The Bourgeois Virtues.
Given its introduction as a motivating force, interest quickly came to be seen
as the wellspring of almost all human action, even though people often fell short of
acting in accordance with their interests when they were overcome by the passions.
In effect it was a short intellectual ride from the introduction of the idea of interest
and Mandeville’s Fable of the Bees to the famous tag from Adam Smith’s Wealth of
Nations: “It is not from the benevolence of the butcher, the brewer, or the baker, that
we expect our dinner, but from their regard to their own interest.” (I.ii.2)
However, in the literature of the Enlightenment, amid a concentrated focus on
interest, one finds hardly a reference to incentives. Ruth W. Grant notes that the
word ‘incentive’ in any usage scarcely appears in the works of John Locke, Bernard
Mandeville, Adam Smith, David Hume, Jeremy Bentham, James Mill or John Stuart
Mill. 25 Yet surely the two ideas are intimately connected, especially since an
incentive is intentionally given in order to shape the perception of interests of that
party to whom the incentive is given.
Instead of being used in its modern manner, Grant explains that, during the
Enlightenment, through the nineteenth century, and into the twentieth century,
‘incentive’ was used in the context of “inciting or arousing to feeling or action,
provocative, exciting.” Thus Lord Shaftesbury made an incentive speech—a speech
inciting action—in the House of Lords in 1734. By contrast, the first recorded use of
VOL.7 NO.2
39
‘incentive’ in its modern context dates only from World War II. As Grant points out,
the Oxford English Dictionary records this new usage in the phrase, “Mr. Charles E.
Wilson . . . is urging war industries to adopt “incentive pay.” 26 As Grant goes on to
further document, the use of incentive discourse in this modern sense grew out of
three discourses: scientific management, the “practical efficacy of socialist
economics,” and behavioral psychology, particular in the area of motivation. 27
Thus, for Grant the initial approach to understanding the ethics of incentives
comes out of the historical origins of our modern discourse about incentives, and
incentives are to be understood in the first instance as a technique of control:
“Incentives are one of the various ways in which people can get other people to do
what they want them to do. They involve relations of power.” 28 And this perspective
is reflected in her formal definition of an incentive: “An incentive is an offer of
something of value, sometimes with a cash equivalent and sometimes not, meant to
influence the payoff structure of a utility calculation so as to alter a person’s course
of action.” 29 So described, it seems that incentives and the responses elicited by them
are to be understood within a context of exchange.
But should incentives be viewed exclusively from within a market
perspective? If so, then bribery and blackmail can be cast as just another morally
unproblematic incentive. Grant criticizes the viewing of incentives solely from
within a market context, because such a perspective makes the use of incentives
appear morally unproblematic. Instead, she insists, incentives as a means of getting
people to do what we want can also be understood in contrast to the alternatives of
coercion and persuasion. In many respects, incentives are positioned as a morally
preferable alternative to coercion. But if we compare the giving of incentives with
persuasion, the supposed moral superiority of incentives tends to evaporate. On
Grant’s analysis, forms of persuasion include rational argument, the inducement of
personal conviction, and the fostering of intrinsic motivation. For example,
attempting to get someone to do what we want them to do by offering rational
arguments for why they ought to pursue a course of action is very different from
providing them with incentives so that they will choose to act as we wish.
While Grant essentially sees incentives and coercion as two alternative means
of getting someone to act as desired, she is aware of significant points of contact
between the two methods, as her mention of blackmail as a potential kind of
incentive makes clear. There is the possibility that one could offer incentives or
incentive-like inducements to induce people to act unethically, and in some cases it
seems that incentive-like inducements can be coercive. A gangster’s threat to break
an arm might be seen as an incentive to prompt payment, yet such a threat is surely
coercive. On a more innocuous level, a parent’s threat to withhold ice cream unless a
child behaves nicely might be cast as a kind of incentive as well. 30 While both of
these examples involve a threat of diminishing a right or an entitlement–the right to
be free of violence from others or an “entitlement” to the expected ice cream—some
commentators have seen coercion in other kinds of offers.
40
DECEMBER 2010
For example, should an offer to an indigent person of a high level of pay to
participate in a clinical trial be seen as a generous incentive or a coercive offer? A
significant literature has grown up around the idea of a coercive wage offer and this
has been explored fairly widely in the context of paying human research subjects. 31
However, it may well be doubted that an offering positive benefits without
threatening to reduce a subject’s current standing, could be regarded as coercive.
While an exploration of this topic is beyond the purview of this article, it seems clear
that the evolution of the current financial crisis did not turn in any significant way on
offers with strong coercive elements. However, it is equally clear that the crisis did
stem in significant part from people accepting offers they should not have accepted,
or by responding to incentives or incentive structures that they should have ignored.
If incentives are given by one party to channel the actions of the recipient into
certain pathways, what is the ethical impact of acting on an incentive that has been
offered? Insofar as one is acting on an incentive, one acts in a manner that differs
from the way they would have acted had no incentive been offered. Using the
terminology of the early economists, the person who responds to an incentive
responds to an altered evaluation of her interests.
The moral appraisal of acting in response to an incentive cannot be complete
without a fuller account of the action in question and some idea of what actions are
moral—that is one needs a relevant description of the action against the background
of some moral theory. For example, an action that arises in response to an incentive
would be judged as moral or immoral by a utilitarian based on the outcome, the
likely outcome, or the intended outcome of the action. For the utilitarian, the
incentives do not really matter to the appraisal of the action.
One who acts on an incentive responds to a new perception of her interests as
altered by the inclusion of the incentive in the frame of reference for making a
decision. Such an action, therefore, accords with the actor’s interest, or perhaps is
even performed out of interest. We might say that the incentives altered someone’s
inclination to act in a certain way, such that taking the incentive into account, one is
inclined to act as one then does act.
One philosophical tradition is extremely clear, however, on the moral
appraisal of an action performed in response to incentives. For Immanuel Kant, an
action can have moral worth only if it is done from duty, not because it accords with
one’s interests or inclinations: “Thus the first proposition of morality is that to have
moral worth an action must be done from duty.” 32 Actions performed in accordance
with duty but contrary to inclination have clear moral worth. However, the analysis
of actions that are done in accordance with both duty and inclination presents a
greater difficulty of appraisal. Kant is extremely clear that an action that conforms to
one’s duty, but that is taken because the action accords with one’s inclinations, also
lacks moral worth. Kant’s example of such an action in accordance with duty, but
also matching one’s inclinations, is the action to preserve one’s own life, and he
explicitly states that such an action “…has no intrinsic worth...” However, he goes
on to say: “On the other hand, if adversity and hopeless sorrow have completely
VOL.7 NO.2
41
taken away the relish for life; if the unfortunate one, strong in mind, indignant at his
fate rather than desponding or dejected, wishes for death, and yet preserves his life
without loving it—not from inclination or fear, but from duty—then his maxim has a
moral worth.” 33 Just as with an action performed from inclination, so with an action
performed in response to an incentive—such actions can never have moral worth
according to Kant, because the action is undertaken because the attendant incentive
changes the payoffs from the action.
Thus, the major ethical traditions may disagree on the appraisal of acting in
response to an incentive as a general description of the action. However, areas of
agreement between such traditions can perhaps be found in appraising the morality
of acting in response to an incentive or an incentive structure if the incentive or an
incentive structure creates an interest to act in a manner that is immoral. For
example, if an incentive induces one to act contrary to one’s duty, then such an
action is clearly immoral for Kant, not because it is responding to an incentive, but
because the action is contrary to duty. If an incentive stimulates one to act in a
manner that increases utility in a relevant way, utilitarians would presumably find the
resulting action to be a moral one, but if an incentive leads one to behave in a utilityreducing manner, then the action would be immoral on utilitarian grounds.
Perhaps a more interesting ethical question arises if we consider the provider
of the incentive. If one grants an incentive to induce unethical behavior, then the
moral appraisal of providing such an incentive seems straightforward. On Kantian
grounds, if I provide an incentive for someone to act contrary to his duty, then such
an action is immoral—it is against my duty to induce someone to act contrary to their
duty. Similarly, if I provide an incentive to someone to behave in a utility reducing
manner, my act itself destroys utility, or at least it can reasonably be expected to
destroy utility, and would therefore be unethical on utilitarian grounds.
One person may provide an incentive to induce someone to behave in a
particular way, but the granting of the incentive may actually stimulate some
behavior that was not intended or even contemplated by the grantor of the incentive.
This is an extremely frequent occurrence of incentive creation, so much so that it is
captured in the cliché of “unintended consequences.” This appears to be exactly the
situation in the case of the fleet minimum fuel efficiency standards discussed
above—the incentive for manufacturers to produce more fuel efficient vehicles
actually created an incentive structure that encouraged them to manufacture SUVs
instead. The ethical import of granting incentives is difficult to appraise in many
cases because of the frequently vast slippage between the behavior the grantor of the
incentive sought to encourage and the effect of the incentive on the actually resulting
behavior. In granting incentives, the grantor’s behavior can be subject to ethical
appraisal under the usual requirement for any actor to exercise due care in
considering the effect of an action.
Similar ethical problems arise in the creation of incentive structures. As the
term is used in this article, an incentive structure can either arise naturally or through
human agency. Incentive structures with purely natural origins are mere “facts in the
42
DECEMBER 2010
world” and are presumably ethically neutral. However, incentive structures that arise
through human action have a moral dimension. If I leave an old refrigerator on my
vacant lot near a school playground, I may create an incentive structure that will
induce a child to become trapped inside. Of course, I did not offer an incentive for a
child to play in the refrigerator—such a result was no part of my intention. In this
example I was merely negligent. Nonetheless, my negligence creates an “attractive
nuisance” for which I may be legally culpable and morally blameworthy.
The creation of incentives may well be rarer than the creation of incentive
structures, and many persons in authority must routinely act in ways that create
incentive structures. For example, an owner of a small firm must devise some pay
plan, and even if it is done with no intention of creating any particular incentives, it
will certain implement some incentive structure. For example, a simple hourly pay
plan might be expected to foster a certain set of work habits while a piece-rate wage
will create a different incentive structure. For persons creating incentive structures,
there is a duty to attend to the kinds of incentive structures that are being created, just
as is the case in the example of the old refrigerator.
The legislative process is a particularly powerful creator of both incentives
and incentive structures. Many laws specifically aim at changing or channeling
behavior, thus making the passage of such laws an instance of incentive creation. But
in addition to creating incentives, new laws also create incentive structures that may
encourage entirely unwanted behavior. This is especially likely to be an outcome of
complexity, when one new law interacts with an existing institutional setting or other
extant laws. Thus, there may be grounds for thinking that the legislative process,
because of its far reaching impact, may be particularly vulnerable to this kind of
ethical appraisal. That is to say, government may have a special obligation to attend
to the incentives it implements and the incentive structures it creates.
The creation of incentives is not a simple matter and is actually much more
difficult than one might imagine. This is due in part to the variety of human
responses to a given set of rules or conditions. Parents routinely set rules for their
several children to follow, yet they must constantly be astounded at the variety of
behavioral responses a particular set of rules engenders. Similarly, a given
compensation plan may encourage one person to work hard and well in just the
intended manner, while another might respond with entirely non-productive
behavior. For example, the granting of health benefits might lead one person to
exploit sick leave for amusement, while another might not change her behavior in
any way. As the setting for the creation of incentives and incentive structures
becomes more complex, and as they apply to more people, the effective
establishment of such incentives and incentive structures becomes ever more
difficult.
Incentives also often work in settings with other morally charged principles in
play, and one of these is the interaction of duty and incentives. For example, the
CEO of a corporation receives a salary and undertakes a fiduciary duty to operate the
firm on behalf of its shareholders. Yet CEO compensation packages typically include
VOL.7 NO.2
43
incentive compensation as a component. How are such incentives to be understood?
On one interpretation, the firm might be providing the incentive compensation in
order to achieve incentive alignment, that is, to make it in the CEOs interest to do his
duty. By contrast, one might reason that the CEO can perform his duty in a fully
adequate manner, but that the incentive compensation might induce the CEO to
exceed the mere fulfillment of his duty and to lead the firm in a way that exceeds his
duty. In such a case, the incentive compensation might induce a superogatory
performance of leadership.
The first case, in which the incentive is designed to encourage merely the
performance of an already accepted duty, is highly problematic morally. If one has a
duty but must have incentives to be induced to perform that duty, the idea of duty
and the moral obligation to perform a duty is eviscerated and the idea of a duty does
no conceptual work. Having a duty implies the moral obligation to act in a certain
manner. To offer incentives to perform a duty that one has already accepted tacitly
endorses the immorality of the person already under the obligation of duty. It is to
say: “I know that you have accepted this position with its attendant duties; I know
that you will not perform those duties for the compensation already granted as we
have already agreed and as you have already promised in accepting the
compensation and its attendant duties; therefore, I am going to give you additional
incentives to make it in your interest to act in a way that is consistent with your
duty.”
Introducing incentive compensation into a context in which the person who
receives the incentive already has a duty can make sense from an ethical perspective
only to induce a superogatory performance that goes beyond the duty. To draw on
some awkward terminology from the management literature, the executive might
adequately discharge a duty by “satisficing,” but the incentive compensation is
designed to move the performance beyond the merely satisfactory performance of
the duty to a more nearly optimal level.
Many compensation arrangements prevail in which a clear principal-agent
relationship does not exist, so that the person being hired or compensated does not
have a clear duty to perform in a certain manner. For example, an employment-atwill agreement with a non-professional employee generally has such a feature. If one
hires a cashier, there are certainly expectations of a certain kind of performance, but
these expectations are not codified as professional duties in the same way that might
prevail for the employment of a medical doctor, an attorney, or even a CEO. (For
that matter, the duties of a CEO are not codified as strongly as are those for a
medical doctor or attorney, whose duties are regulated by law and by codes adopted
by their professional associations.)
In virtually any employment contract, the employee or agent operates in an
environment in which incentive structures militate against the desires of the
employer or principal. This is virtually certain to be the case, because only a perfect
employment contract can secure complete incentive alignment between employer
and employee, or between principal and agent. While it may be impossible to avoid
44
DECEMBER 2010
all incentive incompatibility, the reduction of incentive incompatibility is obviously
critical. Sloppy contracts have greater incentive incompatibility, and in the subprime
mortgage market, much of the problems originated from poor employment contracts
with great incentive compatibility. Further, poor management of employees and
agents, allowed situations in which perverse incentive structures encouraged bad
behavior. In many cases, the inadvertent creation of perverse incentive structures
played a large role in creating financial disaster.
Aside from the self interest of the employer or principal in mitigating
incentive conflicts, there is a normative issue in the creation of defective contracts.
That is, does an employer or principal have an obligation to create contracts that
mitigate the creation of perverse incentive structures? For example, if I hire a
salesman and compensate her based totally and exclusively on the dollar volume of
sales, I create an incentive structure that encourages her to neglect honest sales
practices and to pursue any avenue toward increasing sales. 34 The offering of such an
employment contract may well be morally culpable because it encourages unethical
behavior, and this can be the case even if such encouragement is an incidental or
unanticipated feature of the contract. I am suggesting that the employer or principal
has an ethical obligation to design contracts that reduce perverse incentive structures
and that do not encourage an employee or agent to act unethically.
Beyond being a merely practical and effective contract that achieves intended
outcomes, an ethical contract must also recognize the power of incentives and
incentive structures to override the clearest of duties and the firmest of ethical
principles. There can be no doubting the power of incentives in influencing human
conduct. Thus, offering a contract that provides powerful incentives or that creates
significant incentive structures for unethical behavior is itself unethical.
The ethical appraisal of behavior of mortgagors, mortgage brokers, initial
lenders, and appraisers in the financial crisis turns on understanding how they
responded to incentives and incentive structures that were available in the mortgage
market under the originate-to-distribute model. And the same is true for appraising
the setting of incentives and the creation of incentive structures.
Under the originate-to-distribute model, prospective mortgagors found
themselves confronting a situation with powerful inducements for unethical
behavior. Mortgagors obviously did not set the terms that were on offer from lenders,
so they essentially confronted a world with large temptations to unethical behavior.
We have seen that a considerable portion of poorly performing mortgages exhibited
excessive loan-to-value ratios and various forms of mortgage fraud. There can be
considerable debate as to whether such mortgagors confronted incentives or
incentive structures in a particular case, but overall it does seem that both incentives
and incentive structures were in play. Given the numerous anecdotal accounts that
have appeared in the press, there can be little doubt that some lenders and some
mortgage brokers consciously offered incentives for prospective mortgagors to
participate in predatory borrowing or mortgage fraud schemes. Similarly, in many
instances thoughtless contract design and heedless participation in prevailing market
VOL.7 NO.2
45
arrangements led some mortgage brokers and lenders to create incentive structures
that unscrupulous mortgagors could exploit.
In pure cases of mortgage fraud, the mortgagor fools and exploits both the
mortgage broker and the initial lender. As a result, the mortgage broker suffers
potential reputational damage for bringing bad mortgages to the lender, but at least
the mortgage broker receives payment when the mortgage closes, which mitigates
the possible reputational damage, or perhaps even fully compensates for that
damage. If a mortgage broker is fooled too many times, the reputational impact could
be severe and could result in financial harm to the mortgage broker in the long run.
The brunt of the harm in mortgage fraud falls on the provider of funds. If the
initial lender retains the mortgage in its own portfolio (per the originate-to-hold
model), the lender will be left holding the bag when the mortgage fails. However, if
the lender sells the mortgage into the securitization process, the initial lender’s
financial harm is mitigated, eliminated, or even converted to a benefit. In granting
the mortgage and selling it to a securitizer, the initial lender secures a profit on the
sale and probably acquires rights to service the mortgage, so there is likely to be a
profit on the entire process, even if the mortgage should never have been made. In
fact, some of the worst mortgages carried the highest profits for the initial lender.
Therefore, as a general rule, in the case of the sale of a fraudulent or predatory
mortgage, the lender likely secures an immediate financial benefit, but may develop
a reputation as a source of bad mortgages.
Like mortgagors, appraisers typically did not set incentives or create incentive
structures, but they did respond to opportunities presented to them (whether those
opportunities were based on incentives offered or incentive structures inadvertently
created). Appraisers found themselves in a particularly vulnerable situation with their
entire livelihood depending on satisfying the desires of lenders. In many markets
appraisers might have relatively few potential lenders as clients which heightened
that vulnerability. Perceiving that lenders were intent on closing a deal and making a
mortgage go through, many appraisers faced a choice of submitting a series of
dishonest appraisals or losing their livelihood, and there is considerable evidence that
they responded to their situation by inflating appraisals. While mortgagors and
lenders might look at dishonest practices as a form of income or wealth
enhancement, appraisers held a more vulnerable position. If the environment was
generally corrupt, appraisers faced a situation of submitting dishonest appraisals or
giving up their careers entirely and seeking alternative employment.
Compared to mortgagors, appraisers, and initial lenders, mortgage brokers
faced a more complex set of incentives and incentive structures. In the most sordid
instances of predatory lending, some mortgage brokers actively sold loans to
mortgagors that they did not need and could not afford. Other actions were more
ambiguous. Given a compensation plan that rewarded the mortgage broker if and
only if a mortgage loan was made, the purely financial interests of the mortgage
broker were fairly clear. However, they operated in an environment in which they
received various pressures and signals both from lenders and prospective
46
DECEMBER 2010
mortgagors. Imagine a mortgage broker who gets repeated and unmistakable signals
from a lender that the lender’s highest priority is getting deals done, no matter
whether some corners need to be cut. Washington Mutual, for example, emphasized
this point with its “Power of Yes” mortgage promotion program and its extreme
willingness to make any kind of loan. 35
At the same time, the mortgage broker may have been dealing with a variety
of prospective home buyers who wanted loans. Consider a family of limited means
that very much wanted a starter home that was somewhat beyond its means, but for
which they could qualify if they stretched the truth about their income and got a few
thousand dollars from a non-visible loan by a relative for the down payment. This
kind of situation creates a powerful alignment of incentives and incentive structures
for a mortgage broker: Helping the loan go through satisfies the anxious home buyer,
meets the clearly signaled demands of a lender, and results in a fat payday for the
broker. Such an approach satisfies the brokers own financial interest and meets the
perceived needs of the parties with whom the mortgage broker is in immediate
contact, while the parties injured, if any, are the distant subsequent participants in the
mortgage origination chain.
In sum, the mortgage broker held a particularly powerful and conflicted
position in the chain of participants that constituted the originate-to-distribute model.
Armed with the view that they were merely independent contractors with no
fiduciary obligations to either borrower or lender, mortgage brokers operated in an
orchard of low hanging fruit of predatory lending and predatory borrowing with the
temptation of both types appearing virtually every day.
6. Conclusion
This article has discussed the process of mortgage origination in the United States
with a focus on the incentives available to the various actors. The originate-todistribute process involved many parties—mortgagors, mortgage brokers, appraisers,
initial lenders, mortgage servicers, warehouse lenders, due diligence firms, Fannie
Mae and Freddie Mac, investment banking firms, rating agencies, and ultimate
private investors—the relationships among whom were all mediated by contracts,
whether formal or informal. For this complex system to function well, the chain of
contracts needed to succeed in aligning the incentives of the contracting parties.
It has been the argument of this article that the incentive relationships at
virtually every juncture in the process were not adequately aligned. This view has
been illustrated by a fairly detailed analysis of those parties at the head of the
process—the mortgagor, mortgage broker, initial lender, and appraiser. But this is
only a representative slice of the full story of incentives in the originate-to-distribute
model, because grotesque misalignments occurred elsewhere in the design of
securities.
The use of incentive to mean a form of inducement or the offer of an item of
value designed to elicit certain behavior is a linguistic innovation of quite recent
introduction, yet incentives have become extremely important in economic analysis
VOL.7 NO.2
47
in recent decades. The recent economic literature, particularly in finance, sees
contracting as a critical relationship constitutive of the firm. As such, these contracts
must secure adequate incentive compatibility between the contracting parties. Yet
there is a widespread realization that no contract can be expected to secure perfect
incentive alignment.
Given that incentive alignment will be imperfect, a contract generally
succeeds in giving some parties particular incentives that are desired, but it also
inadvertently creates other unforeseen or undesired incentives. To distinguish these,
this article introduced a distinction between an incentive and an incentive structure,
the essential difference being that an incentive is a non-coercive, consciously
designed, and offered inducement, but that an incentive structure is a state of affairs
that is not intentionally created, but induces some behavioral response. As this article
has argued, the originate-to-hold model can be understood fully only by
distinguishing incentives and incentive structures. For an external party,
distinguishing an incentive from an incentive structure can be difficult, because the
difference turns on the intentions of the contracting parties.
Assessing the originate-to-distribute model from an ethical point of view
turns on understanding the incentives that contracts among the participants offered
and received, but it also requires considering the incentive structures that were
inadvertently created in the contracting process. While it has long been clear that the
offering of incentives has a strong ethical dimension, this article argues that the
ethics of contracting also require close attention to avoid the creation of perverse
incentive structures. Creating perverse incentive structures is a fault of negligence
that can be extremely destructive. Further, in some cases contracts can be designed
that appear to induce bad behavior inadvertently but that actually may be designed to
induce that behavior without acknowledging the intention to do so. That is, an
apparent incentive structure viewed externally actually may be the offering of a
corrupt incentive.
48
Figure 1: The Originate-to-Hold Model of Mortgage Production
DECEMBER 2010
VOL.7 NO.2
Figure 2: The Originate-to-Distribute Model of Mortgage Production
49
50
DECEMBER 2010
Figure 3: Percentage of Americans Living in Their Own Homes
% Living in Own Homes
U. S. Home Ownership Rates
70.0
65.0
60.0
55.0
50.0
45.0
40.0
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Decade
2000
2004
2008
67.5
69.2
67.9
VOL.7 NO.2
51
Figure 4: U.S. National Home Price Index, 1987—2008
S&P/Case-Shiller U.S. National
Home Price Index
200.00
150.00
100.00
0.00
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
50.00
Notes
1
Robert W. Kolb is a Professor of Finance and the Frank W. Considine Chair of
Applied Ethics at Loyola University Chicago, 1 East Pearson Street, Chicago, IL
60611, [email protected].
All of the figures in this article appear in his book, The Financial Crisis of Our Time
published by Oxford University Press, 2011, and they appear here through the
courtesy of the Press.
2
Quoted in Goodman, Peter S. and Morgenson, Gretchen, (2008), “By Saying Yes,
WaMu Built Empire on Shaky Loans,” New York Times, December 28, accessed on
December 28, 2008 at:
http://www.nytimes.com/2008/12/28/business/28wamu.html
3
The FHLBB and FSLIC were abolished in 1989 by the Financial Institutions
Reform, Recovery and Enforcement Act. The Federal Deposit Insurance Corporation
assumed the duties previously performed by the FSLIC, and the functions of the
FHLBB were transferred to the Office of Thrift Supervision.
4
See Gorton, Gary, (2009), “The Subprime Panic,” European Financial
Management, January, 15:1, 10-46. Gorton argues that a key structural element of
the subprime market was structured to require periodic refinancing, which effectively
gave the lender the opportunity to call the loan by refusing to refinance.
52
5
DECEMBER 2010
Muolo, Paul and Padilla, Matthew, (2008), Chain of Blame: How Wall Street Cause
the Mortgage and Credit Crisis, Hoboken, NJ: John Wiley & Sons, Inc., 59.
6
Figures on the size of the mortgage broker industry vary considerably, and it is
certain that the number of mortgage brokers has been declining recently. Muolo,
Paul and Padilla, Matthew, (2008), Chain of Blame: How Wall Street Cause the
Mortgage and Credit Crisis, Hoboken, NJ: John Wiley & Sons, Inc., 66-67. Bitner,
Richard, Confessions of a Subprime Lender, (2008), Hoboken, NJ: John Wiley &
Sons, Inc., 48, reports 250,000 mortgage brokers in 2000. The National Association
of Mortgage Brokers reported total employees of 418,700 in 2004. See:
http://www.namb.org/namb/Mission.asp?SnID=1149970333
accessed on February 1, 2009.
7
National Association of Mortgage Brokers, (1997), “Model Disclosure Form,” June
22.
8
Of course in some instances CEOs of financial institutions may have led their
institutions down an ultimately destructive path because excessive mortgage lending
fattened their own pay checks. The incentives faced by such managers could be quite
compelling. For example one observer noted that at Washington Mutual “Kerry
[Killinger] has made over $100 million over his tenure based on the aggressiveness
that sunk the company.” See Goodman, Peter S. and Morgenson, Gretchen, (2008),
“By Saying Yes, WaMu Built Empire on Shaky Loans,” New York Times, December
28, accessed on December 28, 2008 at:
9
See Jensen, Michael C. and Murphy, Kevin J., (1990), “CEO Incentives: It’s Not
How Much You Pay, But How,” Harvard Business Review, May-June, 138-149.
10
This is section 162(m) of the Internal Revenue code.
11
Of course there are many other actors involved besides a single individual one side
of the negotiation and a monolithic institution on the other. For example, individual
members of the board often have their own values and interests that affect the
contractual outcome.
12
Counterparty Risk Management Policy Group III, (2008), “Containing Systemic
Risk: The Road to Reform,” August 6, 5.
13
Engel, Kathleen C. and McCoy, Patricia A. (2002), “A Tale of Three Markets: The
Law and Economics of Predatory Lending,” Texas Law Review, 80(6), 1255-1367,
1258.
14
Ashcraft, Adam B., and Schuermann, Til, (2008), “Understanding the
Securitization of Subprime Mortgage Credit,” Federal Reserve Bank of New York
Staff Reports, no. 318, March, ii.
15
Ashcraft, Adam B., and Schuermann, Til, (2008), “Understanding the
Staff Reports, no. 318, March, ii.
VOL.7 NO.2
16
53
Ashcraft, Adam B., and Schuermann, Til,
(2008), “Understanding the
Staff Reports, no. 318, March, 5.
17
Securities Exchange Commission, (2008), “Summary Report of Issues Identified
in the Commission Staff’s Examinations of Select Credit Rating Agencies,” 32.
18
Crouhy, Michel G., Jarrow, Robert A. and Turnbull, Stuart M., (2008), “The
Subprime Credit Crisis of 2007,” The Journal of Derivatives, Fall, 1-30, 12.
19
See Gorton, Gary, (2009), “The Subprime Panic,” European Financial
Management, January, 15(1), 10-46, especially 5.
20
Caprio, Jr., Gerard, Demirgüç, Asli and Kane, Edward J., (2008), “The 2007
Meltdown in Structured Securitization: Searching for Lessons not Scapegoats,
November, Working Paper. See abstract.
21
Hirschman, Albert O., (1977), The Passions and the Interests, Princeton: Princeton
University Press, 43.
22
University Press, 43.
23
Montesquieu, Baron de, Charles de Secondat, (1914), The Spirit of Laws,
translated by Thomas Nugent, revised by J. V. Prichard, London: G. Bell & Sons,
Ltd. Book XX., 1.
24
Jerry Z. Muller, (2002), The Mind and the Market, New York: Alfred A. Knopf,
73. The even more corrupt society that was replaced was feudal society with its
domination by an aristocracy oriented toward violence.
25
Grant, Ruth W., (2002), “The Ethics of Incentives: Historical Origins and
Contemporary Understandings,” Economics and Philosophy, 18, 115. See also
Grant, Ruth W., (2006), “Ethics and Incentives: A Political Approach,” American
Political Science Review, February, 100:1, 29-38.
26
Grant, Ruth W., (2002) “The Ethics of Incentives: Historical Origins and
Contemporary Understandings,” Economics and Philosophy, 18, 114.
27
Contemporary Understandings,” Economics and Philosophy, 18, 115-116.
28
Grant, Ruth W. and Sugarman, Jeremy, (2004), “Ethics in Human Subjects
Research: Do Incentives Matter?” Journal of Medicine and Philosophy, 29:6, p. 721.
29
Contemporary Understandings,” Economics and Philosophy, 18, 111.
30
Note that the gangster’s threat would not be an incentive as defined in this paper
because of its coercive element. Also, the threat to withhold ice cream would not be
an incentive as it presumes it leaves the child worse off than would otherwise have
been the case.
31
For examples, see Zimmerman, David (1981), “Coercive Wage Offers,”
Philosophy and Public Affairs, Spring, 10:2, 121-145; Macklin, Ruth, (1981), “’Due’
and ‘Undue’ Inducements: On Passing Money to Research Subjects,” IRB: Ethics
54
DECEMBER 2010
and Human Research, May, 3:5, 1-6, and Newton, Lisa, (1982), “Inducement, Due
and Otherwise,” IRB: Ethics and Human Research, March, 4:3, 4-6.
32
Kant, Immanuel, (1959), Foundations of the Metaphysics of Morals, translated by
Lewis White Beck, Indianapolis: Bobbs-Merrill Educational Publishing, 16.
33
Lewis White Beck, Indianapolis: Bobbs-Merrill Educational Publishing, 14.
34
The perverse incentive structure created by such a contract could be mitigated by
monitoring the actual conduct of the employee. But such monitoring is costly and
difficult to perform. The worst situation is created with such a compensation plan
accompanied by no monitoring of sales practices actually employed.
35
Goodman, Peter S. and Morgenson, Gretchen, “By Saying Yes, WaMu Built
Empire on Shaky Loans,” New York Times, December 28, 2008, accessed on
December 28, 2008 at: http://www.nytimes.com/2008/12/28/business/28wamu.html
References
Ashcraft, Adam B., and Til Schuermann, (2008), “Understanding the Securitization
of Subprime Mortgage Credit,” Federal Reserve Bank of New York Staff
Reports, March (no. 318).
Bitner, Richard, (2008), Confessions of a Subprime Lender, Hoboken, NJ: John
Wiley & Sons, Inc.
Caprio, Jr., Gerard, Asli Demirgüç, and Edward J. Kane, (2008), “The 2007
Meltdown in Structured Securitization: Searching for Lessons not Scapegoats,”
Working Paper, November.
Counterparty Risk Management Policy Group III, (2008), “Containing Systemic
Risk: The Road to Reform,” August 6, 5.
Crouhy, Michel G., Robert A. Jarrow, and Stuart M. Turnbull, (2008), “The
Subprime Credit Crisis of 2007,” The Journal of Derivatives, Fall, 1-30.
Engel, Kathleen C. and Patricia A. McCoy, (2002)“A Tale of Three Markets: The
Law and Economics of Predatory Lending,” Texas Law Review, May, 80 (6),
1255-1367, 1258.
Goodman, Peter S. and Gretchen Morgenson, (2008) “By Saying Yes, WaMu Built
Empire on Shaky Loans,” New York Times, December 28, 2008, accessed on
December 28, 2008 at:
Gorton, Gary, (2009), “The Subprime Panic,” European Financial Management,
January.15(1), 10-46.
Grant, Ruth W., (2006), “Ethics and Incentives: A Political Approach,” American
Political Science Review, February, 100(1), 29-38.
Grant, Ruth W., (2002) “The Ethics of Incentives: Historical Origins and
Contemporary Understandings,” Economics and Philosophy, 18, 111-139.
VOL.7 NO.2
55
Grant, Ruth W. and Jeremy Sugarman, (2004), “Ethics in Human Subjects Research:
Do Incentives Matter?” Journal of Medicine and Philosophy, 29(6), 717-738.
Hayek, Friedrich A., (1960), The Constitution of Liberty, Chicago: University of
Chicago Press.
University Press.
Jensen, Michael C., and Kevin J. Murphy, (1990), “CEO Incentives: It’s Not How
Much You Pay, But How,” Harvard Business Review, May-June, 138-149.
Lewis White Beck, Indianapolis: Bobbs-Merrill Educational Publishing.
Macklin, Ruth, (1981), “’Due’ and ‘Undue’ Inducements: On Passing Money to
Research Subjects,” IRB: Ethics and Human Research, May, 3(5), 1-6.
Macleod, Alistair, (1985), “Economic Inequality: Justice and Incentives,” in Kipnis,
Kenneth and Meyers, Diana T. (eds.), Economic Justice: Private Rights and
Public Responsibilities, Totowa, New Jersey: Rowman & Allanheld, 176189.
McCloskey, Deidre N., (2007), The Bourgeois Virtues: Ethics for an Age of
Commerce, Chicago: University of Chicago Press.
Montesquieu, Baron de, Charles de Secondat, (1914), The Spirit of Laws, translated
by Thomas Nugent, revised by J. V. Prichard, London: G. Bell & Sons, Ltd.
Muolo, Paul and Padilla, Matthew, (2008), Chain of Blame: How Wall Street Cause
the Mortgage and Credit Crisis, Hoboken, NJ: John Wiley & Sons, Inc.
Muller, Jerry Z. (2002), The Mind and the Market, New York: Alfred A. Knopf.
National Association of Mortgage Brokers, “Model Disclosure Form,” June 22,
1997.
Newton, Lisa, (1982), “Inducement, Due and Otherwise,” IRB: Ethics and Human
Research, March, 4(3), 4-6.
Securities Exchange Commission, (2008) “Summary Report of Issues Identified in
the Commission Staff’s Examinations of Select Credit Rating Agencies”.
Smith, Paul, (1998), “Incentives and Justice: G. A. Cohen’s Egalitarian Critique of
Rawls,” Social Theory and Practice, Summer, 24:(2), 205-235.
Zimmerman, David, (1981), “Coercive Wage Offers,” Philosophy and Public
Affairs, Spring, 10(2) 121-145.
Market Value Signal Extraction and the
Misapplication of SFAS 133 in the U.S. GSE’s
Apostolos Xanthopoulos1
Illinois Institute of Technology
Abstract. The misapplication of the Statement of Financial Accounting Standards 133
in the U.S. Government-Sponsored Enterprises had obscured the first signals of trouble
in mortgage securities. Still, regulators should have detected signals using their market
value of equity measure, after statistical modification of related balance sheet accounts.
The nonlinearity in this measure would imply that any implementation by management
could significantly understate the exposure of portfolio equity to interest rate changes.
JEL Classification: C53, G18, H83
Keywords: Logistic Regression, Prepayments, Principal Components, SFAS 133.
1. Introduction
In efficient markets, investors are able to see through an accounting veil and factor-in
relevant information about the financial condition of an organization. Before the 2008
financial crisis, however, some institutional practices of hedging for financial risk had
obscured the signals of the upcoming trouble in mortgage related assets. Pivotal events
of the crisis remained the default of Fannie Mae (FNMA) and Freddie Mac (FHLMC),
and the decline in market value of equity in the Federal Home Loan Banks (FHLBS).
The three housing U.S. Government-Sponsored Enterprises (GSE’s) exhibited
lack of corporate governance in applying Statement of Financial Accounting Standards
133 (SFAS 133). Their contention that hedging was ideal had helped stabilize balance
sheets, while reducing earnings fluctuation. Their response to early signals of the crisis
was to raise interest rate exposure, exerting downward pressure on equity market value.
Their misapplication of SFAS 133 should have prompted regulatory agencies at the
time to scrutinize the significant daily changes in the GSE balance sheets, four years in
advance of the 2008 financial crisis. In the process, regulators should have uncovered
that (i) prepayments in 2004 substituted for default in pools of underlying mortgages,
(ii) GSE balance sheets could have not been hedged against large prepayment events,
(iii) the managerial implementation of any regulatory measure would have understated
the exposure to a decline in GSE portfolio equity. Arguably, had regulators recognized
these issues, they would have mandated the reduction in GSE mortgage asset holdings.
57
58
DECEMBER 2010
2. The Balance Sheet of Government-Sponsored Enterprises
The rationale for the close supervision of the Federal National Mortgage Association
(FNMA), Federal Home Loan Mortgage Corporation (FHLMC) and Federal Home
Loan Bank System (FHLMS) was generally accepted. Investor perception of implicit
guarantees by the federal government had allowed the enterprises to seek exposures
that ultimately imposed a burden on U.S. taxpayers. Borrowing at favorable rates and
investing in mortgage related assets had provided higher spreads than other activity.
Unfortunately, the sub-prime and Alt-A loans that found a route into GSE portfolios
were tailored to the ritual of refinancing, and housing markets that were expanding.
The process of revaluing a mortgage pool had never involved the clear distinction
between bona-fide prepayments and mere deferrals of default.2 In the last quarter of
2004, prepayments accelerated irrespective of interest rates, as increased refinancing
activity had substituted for mortgage default.3 GSE reporting of hedges had certainly
veiled the impact of prepayments on the market value of equity. In the process of
covering such accounting irregularities, management had generally diverted attention
away from this en-masse refinancing activity. Based on a 2006 lawsuit by the SEC,
bonus maximizing behavior by management at FNMA assured that little fluctuation
in financial information was ever revealed.4 However, the regulators should have
attempted to extract signals from the balance sheets of GSE’s, after realizing that
loan refinancing and mortgage default could have behaved as close substitutes.
The large prepayments that preceded the 2008 financial crisis by three to four
years had altered the sensitivity of portfolio equity values to interest rates, and made
the interpretation of commonly accepted measures of risk difficult. In particular, the
Market Value of Equity (MVE) measure of risk had declined across GSE institutions
during this period. The impact of prepayments was hard to figure out. Mysteriously,
derivatives had appeared to hedge against unexpected events. The stability in balance
sheets and reduction in earnings variability was artificially conjured through use of
the short-cut method of SFAS 133. This misapplication had spawned the subsequent
investigations and lawsuits by the Securities and Exchange Commission (SEC).
2.1 Statement of Financial Accounting Standards 133 (SFAS 133)
Rule SFAS 133 entails the recording of derivatives at fair value, and recognition of
their unrealized gains and losses in the current period. When transactions are eligible
for hedge accounting, the rule permits the gains and losses on the underlying assets
or liabilities to be recognized as well, and that reduces income variability. SFAS 133
treats prepayments as a call, embedded in the mortgage related asset, and permits the
market price of the whole option to qualify as hedged risk. The effectiveness of such
hedges is generally assessed by the changes in the fair value of the bifurcated option
against changes in the fair value of derivatives. Before 2005, the housing GSE’s had
elected the short cut method of SFAS 133, which basically implied an ideal matching
between derivatives and hedged items, and mostly eliminated variability in earnings.
GSE’s had issued consolidated and agency debt to acquire mortgage related assets,
and used derivatives to cover the mismatch in duration between asset and liability
Vol.7 No. 2
XANTHOPOULOS: MARKET VALUE SIGNAL EXTRACTION
59
values caused by fluctuations in interest rates. For the short-cut method, the hedging
derivatives were apparently offsetting the rest of the balance sheet, indicating ideal
hedge effectiveness. However, the prepayments in 2004 could have not been hedged
against, since they were unrelated to interest rates. This irregularity became apparent
during a statistical decomposition of accounts that comprised a regulatory measure of
risk. The misapplication of SFAS 133 had made recognition of the unanticipated
prepayments easier, as two components of the balance sheet had moved against each
other in lockstep throughout this shock. Instead of examining prepayments in detail,
the regulators had paid attention to correcting the errors related to SFAS 133.
In one branch of the Federal Home Loan Bank System, the unimpeded use of
such methods and subsequent corrections of financial statements had obscured such
prepayment shocks in late 2004. During 2008, the Office of Finance of these FHL
Banks reported in its Combined Financial Information for 2004 that most branches
had had issues with implementing rule SFAS 133. The regulators had required these
Banks to register a class of equity securities with the SEC, whose rigorous review
exposed these shortcomings. Namely, derivative trades had not been documented at
inception, and had not qualified for hedge accounting. The Chicago branch entered
into an agreement with the Federal Housing Finance Board to review its hedging
activities. Later, the branch restated its financial statements for 2001, 2002 and 2003,
since it had misapplied fair value and cash flow hedging. Specifically for cash flow
hedges, the effective and ineffective portions should have been realized in Other
Comprehensive (OCI) and current income, respectively. OCI should have been
reclassified into income, as the hedged cash flows affected earnings. In March 15 of
2005, the Federal Home Loan Bank of Chicago announced the corrected results for
2004. By that time, market value of equity had declined to 85 cents of book value.
All three housing GSE’s matched the duration of assets to that of liabilities.
The large change in assets caused by prepayment shocks did not correspond to any
equivalent liquidation of liabilities. However, the mismatched portion of liabilities
could have not been hedged by derivatives. According to this 2006 case “Securities
and Exchange Commission v. Federal National Mortgage Association,” FNMA had
failed to comply with SFAS 133 between 1998 and 2004, as debt had proven hard to
hedge with the offsetting derivatives. This inappropriate use of the short-cut method
obligated FNMA to finally exclude derivative transactions from hedge accounting.
Moreover, the Federal Home Loan Mortgage Corporation (FHLMC) had followed
similar practices in the same time period, before the crisis. Thus, the SEC charged
“Steady Freddie” in 2007 with improper management of reported earnings, and for
deceiving investors about its true profitability through use of the short-cut method.5
Most likely, all three housing GSE’s were exposed to a decline in equity that
resulted from the large prepayments in late 2004. The misapplication of SFAS 133
both preceded and followed this event, and helped stabilize its effect as derivatives
appeared to exactly offset resulting mismatches between GSE assets and liabilities.
Using the measure of market value of equity, Federal Housing and SEC regulators
should have discerned that these were early signals of the impending financial crisis.
60
DECEMBER 2010
2.2 Market Value of Equity (MVE)
The Financial Institutions Reform, Recovery and Enforcement Act established the
Office of Thrift Supervision (OTS) as the regulator of all thrift institutions. Recently,
GSE supervision fell under the Office of the Comptroller of the Currency (OCC).
Thrift Bulletins TB13 and TB13a of the OTS required examination of the sensitivity
of the balance sheet to floating interest rates, such as the London Inter-Bank Offer
Rate (LIBOR). This sensitivity referred to the Market Value of Equity. Market Value
of Portfolio Equity (MVE) was the net economic value of assets, liabilities, and offbalance sheet obligations. Net Portfolio Value (NPV) was the net present value of all
assets, liabilities and off-balance sheet items. According to TB13 and TB13a, the
sensitivity to interest rates was the largest negative change that resulted from shocks
of +/-100, 200, 300 and 400 basis points. Thrift Bulletins 13 and 13a had required
institutions to compare the effect of changes in interest rates on future income and
equity against established limits. The Office of Thrift Supervision had prescribed the
Market Value of Equity (MVE) and Net Portfolio Value (NPV) measures leaving
significant latitude over the technical details of their institutional implementation to
management. One implementation of this measure involved the ratio of the market
value of equity divided by the difference between asset and liability book values, or
MVE Ratio. Based on this implementation, the balance sheet data of GSE’s, which
reported ideal hedging, should have helped discern the reasons behind the steady
decline in the market value of equity. The ripple on the balance sheets at arrival of
the event would have been small. Nonetheless, orthogonal analysis of the correlation
between accounts that comprised the MVE Ratio would have shown that the balance
sheets only appeared to be hedged, which was likely due to the misuse of SFAS 133.
2.3 Principal Component Analysis of the Balance Sheet (PCA)
The data used were high level account classifications from a Government Sponsored
Enterprise.6 These account classifications were the Market Value of Assets (AMV),
Market Value of Liabilities (LMV), Market Value of Hedges (HMV), Book Value of
Assets (ABV), and the Book Value of Liabilities (LBV). Six months of data were
available for 127 business days between 9/1/2004 and 3/7/2005, before corrections
related to SFAS 133. Thus, account balances still reflected ideal hedging. As a result,
mortgage prepayment shocks that occurred in mid-December 2004 were rationalized
as hedged phenomena. The short-cut method had insulated the measure of balance
sheet sensitivity to interest rates from prepayment shocks. This measure was the
Market Value of Equity Ratio, in (1). The correlation between account classifications
produced the eigenvectors of Table 1 below and generated these linear combinations:
(i) An Increase in All Accounts, (ii) Hedging with Derivatives, (iii) Market Value
Decline, (iv) Market Value Mismatch, and (v) Book Value Mismatch.
MVER =
AMV − LMV + HMV
ABV − LBV
(1)
Vol.7 No. 2
61
Component values were rescaled between $0 and $1 for simplicity. In the first
eigenvector of Table 1, the weights for the market value and book value of assets and
liabilities were approximately 0.49 and that of the hedge market value was 0.22. This
component captured the duration-matched increases in assets and liabilities, and a
portion of derivative trades that accompanied mismatches. The component explained
83.04 percent of the variability in all account balances. In the second eigenvector,
however, the market value of the hedge account was a large and negative -0.97. This
component portrayed the reduction in market value for such interest rate caps/floors,
swaps and swaptions. This negative weight for the market value of hedging seemed
to counter the first component in near perfect stabilization, illustrating the effect of
misapplying the rule SFAS 133. The third eigenvector had negative weights for
market value of assets and liabilities, positive weights for book values of same, and a
very small weight for the hedge account. This component portrayed the reduction in
market value and the increase in book value of assets and liabilities, and was not
affected by hedging. Mortgage asset and callable liability market values declined in
tandem, and corresponding book values increased, as management had continued to
acquire assets at lower prices. The regulators had difficulty in identifying this aspect
of the balance sheet, since it explained a small portion of account variability, only
0.14 percent. Strictly accounting inquiries by regulators pointed to components 1 and
2, which collectively explained 96.85 percent of variability in account classifications.
These two had entirely overshadowed the interest rate effects of the third component.
The fourth and fifth components revealed the separation of assets from liabilities in
market and book values and explained a small portion of account variability. These
effects were not addressed in detail. Their impact was included in model estimation.
Table 1: Eigenvectors of the Correlation Matrix of Balance Sheet Accounts
Balance Sheet
Increase
Hedging
Market
Market
Book
Account
in All
with
Value
Value
Value
Classifications Accounts Derivatives
Decline
Mismatch
Mismatch
AMV
0.4868
0.1282
-0.5718
0.6476
-0.0160
LMV
0.4872
0.1235
-0.4160
-0.7578
0.0072
HMV
0.2200
-0.9750
-0.0314
-0.0002
-0.0030
ABV
0.4886
0.0922
0.4915
0.0661
0.7119
LBV
0.4884
0.0961
0.5073
0.0445
-0.7020
Variability
83.0396%
16.8103%
0.1393%
0.0091%
0.0018%
5
 MVERt 
= α 0 + ∑ α i PCi ,t + ut = 1.82 + 4.28 PC1,t

(344.42 ) (78.67 )
i =1
1 − MVERt 
Logit (MVERt ) ≡ ln 
− 4.36 PC 2 ,t − 0.30 PC3 ,t + 0.86 PC 4 ,t − 0.54 PC5 ,t + ε t
( −78.79 )
( −52.81)
(129.05 )
( −49.09 )
(2)
62
DECEMBER 2010
Large prepayments represented the arrival of a discrete event which imposed
a binary or dichotomous structure on the MVE Ratio in relation to the balance sheet.
Prepayments introduced a regime switch in the relation between the market value of
equity ratio and the components of the balance sheet. Equation (2) shows the logistic
regression of the ratio Logit(MVERt) against components of the balance sheet, PCi,t,
for i = 1,…,5. Equation (2) has adjusted R-squared of 0.997 and significance 0.0000
for F and for estimated coefficients (t – statistics in parentheses). Thus, increases by
$1 in components 1 and 2 subtracted $4.28 and added $4.36 to Logit(MVERt) leaving
it largely unaffected. A $1 increase in component 3 reduced Logit(MVERt) by $0.30.
These results had demonstrated that the GSE balance sheets remained paradoxically
hedged against large prepayments, which were not anticipated interest rate scenarios.
Meanwhile, the continued mortgage acquisitions were draining the portfolio equity.
3. The Effect of Interest Rates on the Market Value of Equity
Thrift Bulletins TB13 and TB13a focused primarily on the ‘exposure’ to a decline in
the market value of equity, caused by the parallel shift in interest rates. In view of the
large prepayments that preceded the crisis and owing to the misuse of SFAS 133, it
should be expected that MVE implementations understated the bank’s exposure to an
equity decline. By linking the level of interest rates to balance sheet components, this
PCA method had isolated managerial dimensions along which decisions resulted in
such decline. The tracking of these decisions through time could have provided
regulators with the clues as to the actual reasons that led to the steady waning of the
MVE Ratio, even as the latter appeared insulated from prepayment shocks. Equation
(2) applied to all 127 days of the sample, while an internal risk monitoring process
revealed that the apparent exposure of PCi,t components to interest rates had changed
dramatically over time. Thus, the MVE Ratio decline could inadvertently be tied to
recent managerial decisions that altered the portfolio sensitivity to interest rates.
3.1 Balance Sheet Components and the Level of Interest Rates
Principal components also described movements in the term structure of swap rates.
Data were the daily changes in the rates of fixed interest rate swaps against floating
London Inter-bank Offer Rates (LIBOR) for 3 and 6 months and 1, 2, 3, 5, 10 and 30
years. Table 2 lists the eigenvectors for these rates. The multiplication of changes in
the swap rates by these eigenvectors produced changes in the ‘Level’, ‘Slope’ and
‘Curvature’ of the swap curve. An increase in Level (Lt) described higher swap rates
across all maturities. Slope (St) illustrated higher long, and lower short maturities.
Curvature (Ct) raised long and short swap rates and lowered intermediate maturities.
While TB13a of 1998 had alluded to changes in the slope of rates, the original TB13
of 1989 referred to rate levels only. The level corresponded to the first component
which explained 75.08 percent of the variability in the term structure. A basis point
move in any part of the curve was explained in the framework of these components.
Changes in the 10-year swap rate closely related to changes in 30-year asset yields.
A +/- 300 basis point change in this rate could be replicated through equation (3).
Vol.7 No. 2
63
Table 2: Eigenvectors of the Correlation Matrix of U.S. LIBOR Swap Rates
Swap Rates
Level (Lt)
Slope (St)
Curvature (Ct)
3 Month
0.1070
-0.8658
0.3033
6 Month
0.3429
-0.4012
-0.2881
1 Year
0.3805
-0.0332
-0.4017
2 Year
0.3882
0.0661
-0.3189
3 Year
0.3974
0.1130
-0.1486
5 Year
0.3974
0.1540
0.1006
10 Year
0.3754
0.1559
0.4177
30 Year
0.3444
0.1526
0.5960
Variability
75.0846%
14.2256%
8.5407%
dr (10Yr )t = − 16.06 + 236.66 Lt + 16.82 St + 84.36 Ct
( −140.00 )
[
]
(58.14 )
( 4.90 )
( 40.62 )
E PCi , s = β 0, i , s + β1, i , s Ls , i = 1,...,5, s = 1,...,98
(3)
(4)
Equation (3) is the linear regression of the basis point changes in the 10-year
swap rate against the Level (Lt), Slope (St) and Curvature (Ct), based on eigenvectors
of Table 2 (t-statistics in parentheses). Assuming that Slope (St) and Curvature (Ct)
stay at their 127-day averages of 0.0544 and 0.0467 respectively, the +300 basis
point move in the 10-year swap rate amounts to a Level (Lt) equal to 1.3150 in this
case (-11.21 + 236.66 x 1.3150 = +300). This Lt of 1.3150 can be divided by the
square root of 252 trading days, since daily data are used. Thus, the +300 basis
points are roughly equivalent to a +2.23 volatility move for the component variable
Level (Lt), after the standardization by the mean µ(Lt), and standard deviation σ(Lt).
The internal risk monitoring process (4) links a principal component i of the
balance sheet to the level of interest rates Ls in rolling sample, s. These equations are
estimated with maximum likelihood across 98 samples of 30 days each, and rolling
forward one day at a time. Thus, there are 196 coefficients for each component, PCi.
Coefficient β0,i,s in the i-th component captures the idiosyncratic risk introduced
through any decision along dimension i, and is not dependent on the swap rate level.
Term β1,i,sLs describes the market effect of the decision along dimension i, affected
by the swap rate level. The terms represent the ‘risk premium’ and ‘market effects’
of a managerial dimension, defined as a principal component of the balance sheet.
The large prepayments of this period severely impacted the risk premia β0,i,s,
for i = 1, 2, as shown in Figures 1 and 2. Risk dropped from 80.5 to 3.9 ‘cents’ in the
first component, and from 79.8 to 7.5 ‘cents’ in the second one. Corresponding rate
effects appeared small. The daily changes in the first two components took place in
lockstep, reflecting this presupposition of ideal hedging. The reduction in these risk
premia apparently accommodated the ripple-less passing of prepayments through the
balance sheet of GSE’s, even as interest rates had stayed unchanged in this period.7
64
DECEMBER 2010
The boundaries between hedging and speculation had become very muddy.
Prepayments may have reduced assets, liabilities and hedging along dimension PC1,s.
In response, management had curtailed speculation through a reduction in the risk
premium of derivative positions along ‘hedging’ dimension PC2,s. In the past, higher
premia along this second dimension could have served as a method to raise profits.
The surmise of ideal hedging had accommodated such practice, but should have also
made the recognition of prepayment signals easier. The regulators clamped down on
proper implementation, glossing over the implication of the reduction in risk premia.
Figure 3 reveals that GSE spread-seeking was now funneled into buying mortgage
assets at reduced prices, and that raised the risk premium in the third component.
3.2 Understatement of Exposure in Implementing the MVE Measure
The Office of Thrift Supervision had allowed the technical details of the exposure of
GSE equity to be determined during the managerial implementation of the measure.
Most likely, management would have reported this exposure in some models that
were linear in coefficients to swap rate levels. Otherwise, GSE management would
have either openly anticipated the arrival of such events, or altruistically revealed the
magnitude of exposure through correct implementation of SFAS 133. The regulators
should have known that reported limits of equity decline understated such exposure.
Linear relationships do not capture the full effect of default masked as prepayments.
Duration of equity, for example, is based on a linear approximation.8 The resulting
magnitude of exposure understatement can be gauged by substitution of equation (4)
into (2). Results are compared to linear approximation (7), of logistic equation (5).
Equation (7) represents a linear relation likely to be implemented by management.
Logit (MVERs ) = α 0 + γ 0, s + γ 1, s Ls ⇒ MVERs =
[
= [α
][
α ] ⋅ [β
where γ 0, s = α1 α 2 ... α 5 ⋅ β 0,1, s
and γ 1, s
dMVERs
dLs
=
1
γ 1,s ⋅ e
(1 + e
α 2 ...
5
1,1, s
α 0 +γ 0 , s +γ 1, s Ls
)
α 0 +γ 0 , s +γ 1, s Ls 2
P[MVERs ] = MVERs +
=
dMVERs
dLs
(
α 0 + γ 0 , s + γ 1, s L s
e
α 0 + γ 0 , s + γ 1, s L s
1+ e
]
]
β 0, 2, s ... β 0,5, s
T
β1, 2, s ... β1,5, s
T
(1 2 )γ 1,s
1 + cosh α 0 + γ 0,s + γ 1,s Ls
Ls (± 300bp )
(5)
)
(6)
(7)
Vol.7 No. 2
65
Coefficients γ0,s and γ1,s in (5) are obtained by multiplying those of balance
sheet components in (2) by those of Level (Lt) in (4), for rolling samples s = 1,…,98.
Figure 4 confirms that the estimated MVE Ratio in (5) had declined, even while the
change in the ten-year swap rate had remained close to zero. There are two clusters
of observations in which large changes in rates do not affect the ratio. In between,
there is a sharp decline that is attributed only to prepayments. This pattern confirms
that equity was declining at the time, for reasons other than changes in interest rates.
Equation (6) is the derivative of Logit(MVERt) with respect to the rate level.
This derivative changes over time in each rolling sample s = 1,…,98 depending on
γ1,s, the coefficient of rate level Ls. For the nonlinear response of this analysis, this
derivative comes closest to linear ‘sensitivity’ to interest rates. Figure 5 reveals the
dynamic adjustment of the MVE Ratio. The ratio declined as management raised this
sensitivity of the portfolio to the level of rates through additional acquisitions of
mortgage assets. During the nonlinear response of equity to rates, this higher rate
sensitivity only reduced portfolio value at a constant pace. The sensitivity to rates
was causing steady declines in the equity of portfolios, which falsely appeared as
hedged against a nonlinear event. The steady decline was the result of mortgage asset
acquisitions during a period of strong refinancing of Alt-A, and sub-prime loans.
The linear approximation P[MVERs] in (7) captures the decline from a +/-300
basis point move, which management would have offered. This approximation states
the predicted change in the MVE Ratio for rolling sample s, P[MVERs]-MVERs, as a
function of derivative (6) at s, and level Ls. The predicted ratio equals the actual one
plus the derivative with respect to level of rates, times the level that corresponds to a
+/-300 basis point move in 10-year rates. Management implementation would not
account for the arrival of large prepayment events. Nonetheless, GSE management
and regulators would be puzzled as ratios declined by amounts larger than predicted.
The gap between the predicted P[MVERs] from (7) and simulated MVER from (5)
provides an estimate of the understatement of a potential decline in the market value
of equity ratio, for such interest rate scenario. When sensitivity to interest rates was
close to zero, the understatement of the exposure was nearly negligible. However,
during (after) prepayments, the ratio became exposed to a decline in falling (rising)
rates, when the sensitivity to interest rates was changed to positive (negative).
In the sample, MVERs sensitivity to swap rate levels for October to November
2004 averaged around -0.025 and exposure was low, as shown in Figure 6. A +/-300
basis point move in the 10-year swap rate caused no gaps between the P[MVERs]
predicted by management and the realistic MVERs, which was simulated through (5).
Management raised this sensitivity to a high 0.2781 in the period of December 2004
to January of 2005, when the prepayments were processed. Figure 7 reveals the large
understatement of 29 cents on the dollar for the exposure of the MVE Ratio to falling
rates during positive sensitivity, when prepayments passed through balance sheets.
But while the decline in falling rates was large, the increase in rising rates would be
capped at one. Thus, because of an asymmetry in MVE Ratio changes caused by
prepayments, the increased sensitivity to rates could have only lowered this ratio.
66
DECEMBER 2010
After processing the large prepayments, GSE’s had an MVE Ratio sensitivity
to swap rates as low as -0.3218, between February and March 2005. Figure 8 shows
a large understatement of 26 cents in rising rates during the negative sensitivity to
rates, after the prepayment shock. Again, the potential decline of the MVE Ratio in
rising rates was large, but the increase was capped at the value of one. Because of the
asymmetry introduced through prepayments, a decrease in sensitivity to rates would
have lowered the ratio, as well. Overall, the linear approximation of the decline in
the MVE Ratio by management would have understated this potential exposure by a
third to a fourth of a dollar, for a +/-300 basis point move in the 10-year swap rate. It
would have proven thorny for the regulators to enforce measures that assured GSE’s
stayed within acceptable corridors of portfolio sensitivity to interest rates, had this
understatement of exposure been addressed with management.
The changes in the sensitivity of the balance sheet to interest rates imposed a
downward pressure on the MVE Ratio during this time of large prepayments. Figures
7 and 8 reveal that the arrival of this event extended the exposure to a decline in the
ratio, and limited the potential for its increase, in either interest rate scenario. At the
0.85 value of that time, the MVE Ratio was still fairly close to its upper bound, and
was exposed to a sharp decline in either rising, or falling rates. The decision to alter
the sensitivity to either rate scenario exerted downward pressure on the MVE Ratio.
Instead of curtailing, regulators allowed management to augment mortgage
holdings, thus helping erode the portfolio equity of Government-Sponsored
Enterprises.
Detecting signals in hindsight is easy, one may argue. However, the complete
lack of rigor in applying any regulatory measure of risk would have only understated
the risks before the 2008 financial crisis. Any interpretation of exposure of mortgage
portfolio equity to a decline due to interest rates should have explicitly incorporated
the arrival of some nonlinear or regime switching event, such as the processing of
large prepayments or default, as surely had been originally intended by TB13 and
TB13a. An examination of what regulators had to gain by not expending resources to
scratch underneath the surface of their own measure of risk was beyond the scope of
this analysis. On the managerial side, bonus maximizing behavior would have led to
understating, or not even reporting the GSE exposure of equity obtained through the
use of nonlinear techniques. Implementations of the original MVE measure would
either be marred by the wrong application of the accounting standard, or be restricted
from pointing out that large prepayments had impacted the sensitivity of the balance
sheet to interest rates. It would be up to the regulators to measure the degree to which
the arrival of these events had caused an understatement of the potential exposure to
declines in equity. In the case of the decline in market value of equity experienced by
GSE’s before the crisis, the sheer magnitude of understatement should have alerted
regulators to the possibility of systemic collapse. Apparently, that had not happened.
Vol.7 No. 2
67
4. Conclusion
The luxury of hindsight allows for some after-the-fact recognition of the early
signals of the crisis in mortgage related securities. Nonetheless, an opportune claim
of complete lack of any indication of trouble regarding the decline in markets cannot
be extended to GSE’s, where the crisis had started. The misapplication of SFAS 133
by Government Sponsored Enterprises should have alerted regulators three to four
years in advance, steering their attention toward signals of the upcoming trouble. The
regulators should have recognized that large prepayments, which preceded the crisis,
were the forerunners of a collapse. Refinancing and default had acted as substitutes,
while mortgage valuations depended on an upward trend in the housing market.
The supposition that GSE balance sheets stayed ideally hedged through an
event unrelated to rates was not the best practice of GSE management. The increase
in mortgage holdings during a market decline was reckless. These practices should
have made the recognition of crisis signals easier. Although the misapplication SFAS
133 obscured the decline in equity, regulators should have foretold the impending
collapse in mortgage assets. The measure of Market Value of Equity was designed to
capture the exposure of equity to a decline from changes in interest rates. A simple
statistical decomposition of accounts that comprised this measure as a ratio to book
values would have indicated that large prepayments from the en-masse refinancing
of Alt-A, and sub-prime loans could have not passed through the balance sheet of
GSE’s and left it unaffected. The market value of hedges, which were put in place to
protect against movements in interest rates, had artificially offset the equity decline
due to prepayments, even as swap rates had not shifted. The largest components of a
balance sheet had moved in lockstep, while the third component had revealed the
increased exposure to interest rates, which ultimately reduced the MVE Ratio. It is
unknown whether, at that time, the supposition that GSE balance sheets were ideally
hedged had made the effect of large prepayments apparent to regulators, including
the Securities and Exchange Commission. In any case, these regulators should have
viewed changes in GSE balance sheets as the signals of upcoming trouble in the
financial markets, instead of only focusing on correcting errors for SFAS 133.
The Office of Thrift Supervision (Office of the Comptroller of the Currency)
should explicitly specify the technical aspects of implementing its MVE measures,
and not leave seemingly unrelated details up to management of regulated institutions.
In the absence of their control over such details, the regulators should expect a severe
understatement of exposure by implementations spearheaded through management.
Such implementations would not take into consideration the regime switching effect
of the arrival of large prepayments, in the eve of a crisis. An internal risk monitoring
process would isolate this effect, and confirm the acquisition of mortgage securities.
Combined with a regime switching model, such methodology would have provided a
more accurate, if not dramatic statement of the potential equity decline.
Generally, the discernment of early signals of the crisis by regulators was at
the very least possible, three to four years in advance of 2008. Recognition of trouble
could have come from a simple modification in the Market Value of Equity measure.
68
DECEMBER 2010
Notes
1. Apostolos Xanthopoulos has Ph.D.-Finance from Illinois Institute of Technology.
He has industry experience in risk management of portfolios & financial derivatives
and taught at Argosy, Aurora, Dominican, Loyola and Northern Illinois University.
Apostolos Xanthopoulos, 3140 Autumn Lake Drive, Aurora, IL 60504, USA.
630-236-7830 Fax, 630-606-7830 Voice, [email protected], [email protected].
2. The processed prepayments included the amounts prepaid voluntarily, liquidation
amounts from foreclosure and sale of properties and disqualified loan amounts.
3. Geetesh Bhardwaj and Rajdeep Sengupta of the Fed of Saint Louis connected the
high default rates in the post-2004 loan originations with the prepayments in the pre2004 period in “Did Prepayments Sustain the Subprime Market” (2008). Borrowers
were able to avoid default by prepaying loans during times of rising house values.
4. This legal case against FMNA was Securities and Exchange Commission versus
Federal National Mortgage Association. In his April 2010 testimony to the Financial
Crisis Inquiry Commission, Armando Falcon, former director of OPHEO, referred to
the many instances of incomprehensible bonus maximizing behavior at the GSE’s.
5. Green and Wachter (2005) discussed the fact that both FHMA and FHLMC had
failed to report earnings according to Generally Accepted Accounting Principles
(GAAP), regarding derivatives they used to manage interest rate and duration risk.
The authors pointed out that risk taking at member-bank-owned Federal Home Loan
Banks was equal to, or higher than that of the privately owned FNMA and FHLMC.
6. In 2005, this author presented a study to management during his employment as
senior quantitative analyst at the Federal Home Loan Bank of Chicago. The results
reached Federal Housing Finance Board examiners. The author was then terminated.
7. Note that the weight for the hedge account in the second component is negative.
A reduction in the component would mean that the hedge account balance went up.
Thus, the first two components together created the artificial stability in the market
value of equity of this, and most likely the rest of enterprises which held mortgage
related assets, and had misapplied Statement of Financial Accounting Standards 133.
8. The measure ‘duration of equity’ has been reported by GSE’s as the sensitivity of
bank equity to interest rate levels. The measure was an application of bond cash flow
duration, a mathematical extension of the first derivative of the price function with
respect to interest rates. This measure was inadequate in capturing nonlinear effects.
Vol.7 No. 2
Appendix
Figure 1: Effect of Interest Rates on All Balance Sheet Accounts, PC1
40%
30%
Market Effect
20%
10%
October 14, 2004
March 7, 2005
0%
-10%
-20%
-30%
-40%
-20%
0%
20%
40%
60%
80%
100%
120%
Risk Premium
Figure 2: Effect of Interest Rates on Hedging with Derivatives, PC2
40%
30%
Market Effect
20%
October 14, 2004
March 7, 2005
10%
0%
-10%
-20%
-30%
-40%
-20%
0%
20%
40%
60%
Risk Premium
80%
100%
120%
69
70
DECEMBER 2010
Figure 3: Effect of Interest Rates on Market Value Decline, PC3
50%
45%
Market Effect
40%
35%
30%
March 7, 2005
25%
20%
15%
10%
5%
October 14, 2004
0%
0%
10%
20%
30%
40%
50%
Risk Premium
Figure 4: Change in 10-Year Swap Rate and Decline in MVE Ratio
MVE Ratio
0.895
October 14, 2004
0.890
0.885
MVE Ratio
0.880
0.875
0.870
0.865
0.860
0.855
March 7, 2005
0.850
0.845
-15
-10
-5
0
5
Basis Point Change in 10-Year Swap Rate
10
15
Vol.7 No. 2
71
Figure 5: Sensitivity to Interest Rates and Decline in the MVE Ratio
MVE Ratio
0.895
0.890
October 14, 2004
0.885
MVE Ratio
0.880
0.875
0.870
0.865
0.860
0.855 March 7, 2005
0.850
0.845
-0.40
-0.30
-0.20
-0.10
0.00
0.10
0.20
0.30
0.40
First Derivative of MVE Ratio with Respect to Interest Rate Levels
Figure 6: Exposure of MVE Ratio from October to November 2004
1.00
0.90
0.80
MVER
0.70
0.60
0.50
0.40
0.30
0.20
Simulated MVER
0.10
Predicted MVER
0.00
-400
-300
-200
-100
0
100
200
300
Simulated Basis Point Change in the 10-year Swap Rate
400
72
DECEMBER 2010
Figure 7: Exposure of MVE Ratio from December 2004 to January 2005
1.00
0.90
0.80
MVER
0.70
0.60
0.50
0.40
0.29
0.30
0.20
Simulated MVER
0.10
Predicted MVER
0.00
-400
-300
-200
-100
0
100
200
300
400
Figure 8: Exposure of MVE Ratio from February to March 2005
1.00
0.90
0.80
MVER
0.70
0.60
0.50
0.40
0.26
0.30
0.20
Simulated MVER
0.10
Predicted MVER
0.00
-400
-300
-200
-100
0
100
200
300
400
Vol.7 No. 2
73
References
Archer, Wayne R., David Ling, and Gary McGill (2001), “Prepayment Risk and
Lower Income Borrowers”, Joint Center for Housing Studies, 1-41.
Bhardwajy, Geetesh, and Rajdeep Sengupta (2008), “Did Prepayments Sustain the
Subprime Market?”, Research Division, Federal Reserve Bank of St. Louis
Working Paper 2008-039A, 1-44.
Campbell, John Y., Andrew W. Lo, and A. Craig MacKinlay (1997), The
Econometrics of Financial Markets, Princeton: Princeton University Press.
Committee on Capital Markets Regulation (2009), The Global Financial Crisis: A
Plan for Regulatory Reform.
DeLiban, Nancy and Brian P. Lancaster (1995), “Understanding and Valuing
Mortgage Security Credit”, The Handbook of Mortgage-Backed Securities, ed.
Frank J. Fabozzi, Chicago: Probus Publishing, 449-487.
Dillon, William R. and Matthew Goldstein (1984), Multivariate Analysis: Methods
and Applications, New York: John Wiley & Sons.
Dynan, Karen E. (2009), “Changing Household Financial Opportunities and
Economic Security”, Journal of Economic Perspectives, 23 (4), 49-68.
Fabozzi, Frank J., and Franco Modigliani (1992), Mortgage and Mortgage-Backed
Securities Markets, Boston: Harvard Business School Press.
Falcon, Armando (2010), “Testimony Submitted to the Financial Crisis Inquiry
Commission, April 9, 2010”, 2-11.
Federal Home Loan Bank of Chicago (2004), Annual Report, 2004.
Federal Home Loan Bank of Chicago (2005), Financial Statements and Notes for the
Quarter Ended March 31, 2005.
Federal Home Loan Banks (2004), “Delay in Publication of the FHLBanks Third
Quarter Combined Financial Report: Eleven FHLBanks Expected to Post Third
Quarter Results Today”, Office of Finance.
Federal Home Loan Banks (2008), Unaudited Combined Financial Information2004, Office of Finance.
Federal Housing Finance Agency (2008), Report to Congress.
Feldhaus, David (2005), “News Release, March 2005”, Chicago Federal Home Loan
Bank.
Frame, Scott W. and Lawrence J. White (2004), “Thoughts on Institutional Structure
and Authorities”, Federal Reserve Bank of Atlanta Economics Review, 87 – 102.
Gangwani, Sunil (1998), “MBS Structuring: Concepts and Techniques”, The
Securitization Conduit, (1), 26-37.
Gerardi, Kristopher, Andreas Lehnert, Shane Sherlund, and Paul Willen (2008),
“Making Sense of the Subprime Crisis”, Brookings Papers on Economic
Activity, 1-61.
Golub, Bennett W. and Leo M. Tilman (1997), “Measuring Yield Curve Risk Using
Principal Component Analysis, Value at Risk, and Key Rate Durations”,
Journal of Portfolio Management, (24), 72-84.
74
DECEMBER 2010
Green, Richard K. and Susan M. Wachter (2005), “The American Mortgage in
Historical and International Context”, The Journal of Economic Perspectives, 19
(4), 93-114.
Hentschel, Ludger, and S. P. Kothari (2001), “Are Corporations Reducing or Taking
Risks with Derivatives?”, The Journal of Financial and Quantitative Analysis,
36 (1), 93-118.
Hosmer, David W. and Stanley Lemeshow (1989), Applied Logistic Regression,
New York: John Wiley & Sons.
Hull, John C. (2000), Options, Futures, and Other Derivatives, Upper Saddle River:
Prentice-Hall.
Levine, Michael E., and Jennifer L. Forrence (1990), “Regulatory Capture, Public
Interest, and the Public Agenda: Toward a Synthesis”, Journal of Law,
Economics, and Organization, (6), 167-198.
Lockhart, James B. (2010), “Testimony of James B. Lockhart, Financial Crisis
Inquiry Commission, April 9, 2010”, 2-11.
McCool, Thomas J. (2000), “Government-Sponsored Enterprises, Creating a Single
Housing GSE Regulator”, United States General Accounting Office, 1-8.
Office of Regulatory Activities (1989), “Thrift Bulletin 13, Responsibilities of the
Board of Directors and Management with Regard to Interest Rate Risk”, Federal
Home Loan Bank System, 1-6.
Office of Thrift Supervision (1998), “Thrift Bulletin 13a, Management of Interest
Rate Risk, Investment Securities, and Derivative Activities”, Department of the
Treasury, 1-24.
Pennington-Cross Anthony (2002), “Patterns of Default and Prepayment for Prime
and Nonprime Mortgages”, OFHEO Working Papers, 1-25.
Pollock, Alex J. (2006), “FASB Fesses Up to Derivatives Disaster”, American
Banker.
Report of Survey Results (2002), “The Impact of FAS 133 on the Risk Management
Practices of End Users of Derivatives”, Association for Financial Professionals.
Schwartz, Eduardo S. and Walter N. Torous (1992), “Prepayment, Default, and the
Valuation of Mortgage Pass-Through Securities”, The Journal of Business, 65
(2), 221-239.
Sharpe, William (1963), “A Simplified Model for Portfolio Analysis”, Management
Science, 9, 277-293.
Sharpe, William (1964), “Capital Asset Prices: A Theory of Market Equilibrium
Under Conditions of Risk”, The Journal of Finance, 19, 425-442.
Strauss, Mel (1997), “The Danger in Focusing on Market Value”, US Banker.
Taff, Lawrence G. (2003), Investing in Mortgage Securities, St. Lucie PressAMACOM.
Taleb, Nassim N. (2007), The Black Swan: The Impact of the Highly Improbable,
New York: Random House, 2007.
Tuckman, Bruce (2002), Fixed Income Securities: Tools for Today’s Markets, New
York: John Wiley & Sons.
Vol.7 No. 2
75
United States District Court, District of Columbia (2006), Securities and Exchange
Commission versus Federal National Mortgage Association.
Identifying Vulnerabilities in Systemically
Important Financial Institutions in a MacroFinancial Linkages Framework
Tao Sun 1
International Monetary Fund, Washington, D.C.
Abstract. This paper attempts to identify the indicators that can demonstrate the
vulnerabilities in systemically important financial institutions by: (i) investigating the
differences between the intervened and nonintervened financial institutions during
the subprime crisis with balance sheet data; and (ii) detecting the domestic/global
macroeconomic and financial driving factors of financial institutions’ expected
default frequencies with panel specifications and panel cointegration techniques. The
paper finds that: (i) basic leverage, return on assets, provision for loan losses, equity
prices, and business scope can help identify the differences between the intervened
and nonintervened financial institutions;(ii) the expected default frequencies reacts
positively to shocks to basic leverage, inflation, global financial stress, and global
excess liquidity, while negatively to return on assets and equity prices; and (iii) basic
leverage has been the most robust factor with a long-run causal effect on the
expected default frequencies. Therefore, these results suggest that the global
regulators and policy makers could monitor the specific components of capital, basic
leverage, return on assets, equity prices, and at the same time, create a stable
macroeconomic and financial conditions in the context of maintaining price stability,
reducing global excess liquidity and global financial stress. In particular, measures to
set up basic leverage constraints could pay significant dividends in strengthening
systemically important financial institutions.
"If there is one common theme to the vast range of crises we consider in this book, it
is that excessive debt accumulation, whether it be by government, banks,
corporations or consumers, often poses greater systemic risks than it seems [to do]
during a boom".
－Carmen Reinhart and Kenneth Rogoff, 2009
1. Introduction
During the subprime crisis, central banks and governments worldwide have taken
unprecedented policy actions to stabilize banks’ financial condition. One
distinguishing policy action is government rescue of some troubled large financial
institutions (FIs). Two questions naturally arise: Why are some institutions
77
78
DECEMBER 2010
intervened while others are not? What are the macro-financial driving forces of the
vulnerabilities in the systemically important FIs? A more detailed consideration of
those questions involves a response to the following questions:


What are the common factors among the FIs that have required public
intervention? Did balance sheet data, especially traditional financial soundness
indicators (FSIs), provide meaningful warnings?
Can bank-specific indicators explain the development over time of the expected
default frequencies (EDF) for the systemically important FIs? What role does
the macroeconomic and global situation play in this process? Can we find
robust indicators indicating rising EDF?
This paper responds to these questions by: (i) investigating balance sheet
data well beyond the widely-used FSIs, and trying to find more “good” indicators
that capture the key features of FIs; (ii) constructing a group of panel models
(pertaining to different scenarios), which link the measures of the expected default
frequencies to a set of domestic and global macroeconomic and financial variables.
In particular, we use panel cointegration to test the long-run causal effect of some
important indicators, such as basic leverage, on the EDF2.
The results, which are based on data from selected global FIs, demonstrate
that traditional balance sheet data are only partially able to detect, ex ante,
institutions at risk of failing. 3 In addition, panel specifications show that
macroeconomic variables (CPI inflation), bank-specific fixed effects, bank-specific
variables (basic leverage, equity prices and ROA), and global variables (global
excess liquidity and global financial stress index) can help explain the EDF. There
are some intuitive variations to these results when intervened and nonintervened FIs
are investigated separately.
Identifying the reasons behind the relative immunity of some FIs to
government intervention during the subprime crisis could be a part of any financial
stability monitoring exercise. Characterizing the nature of the FIs is of considerable
interest for analytical reasons as well as for understanding the implications of these
differences between intervened and nonintervened FIs. In addition, these indicators
could be also helpful in identifying macro-financial linkages, promote ongoing
financial reforms and design crisis prevention initiatives.
This paper proceeds as follows. Section II gives an overview of the
literature on the FSIs and macro-financial models using EDF as a proxy for
vulnerabilities in FIs. Section III presents a detailed picture of the evolution of the
balance sheet data before and during the subprime crisis. Section IV discusses the
methodologies and results of the panel specifications and panel cointegration.
Section V concludes.
2. Literature review
A substantial amount of theoretical and empirical work has documented how FSIs
VOL.7 NO.2
SUN: IDENTIFYING VULNERABILITIES
79
are used to capture vulnerabilities in firms and economies. The financial crises of
the late 1990s prompted the search for indicators of financial system soundness.
Various studies have proposed early warning indicators of impending turmoil in
banking systems (e.g., Demirgüç-Kunt and Detragiache, 1998, 1999, 2005; Hardy
and Pazarbaşioğlu, 1999; Gonzalez-Hermasillo,1998; Hutchinson and McDill, 1999;
Hutchinson, 2002; European Central Bank, 2005). The need for appropriate tools to
assess strengths and weaknesses of financial systems led to efforts to define sets of
so-called “core” and “encouraged” financial soundness indicators (FSIs), designed to
monitor the health and soundness of FIs and markets, and of their corporate and
household counterparts (Sundararajan and others, 2002). The precise definitions of
the core and encouraged FSIs were laid down in the Compilation Guide on Financial
Soundness Indicators (IMF, 2004). In 2004, the IMF spearheaded a Coordinated
Compilation Exercise (CCE), which was designed to coordinate the efforts of
national authorities to compile and disseminate internationally comparable FSI data
(and the related metadata).
Despite these strengths, there is increasing evidence that some FSIs might
not fully capture the sources of risk. For instance, by incorporating FSIs in an early
warning model of banking crises, Cihak and Schaek (2007) illustrate that crosscountry variation in regulatory capital cannot send a strong signal in the run-up to a
banking crisis. In addition, Poghosyan and Cihak (2009) further illustrate that
relating regulatory thresholds only to capital adequacy is insufficient, and one needs
to include combinations of several relevant variables (notably asset quality and
profitability) to capture the level of risk of individual institutions. Similarly, country
experiences have been gradually indicating that a set of FSIs only for the banking
sector is too narrow. Problems may eventually show up clearly in the simple FSIs,
but it is useful to know when potential problems are mounting before they are
evident in the banks' accounts (Bergo, 2002). Moreover, since each FSI is designed
to capture the sensitivity of the financial system to a specific risk factor (credit or
market risk), none of these “piecewise approach” indicators can provide in and of
itself a comprehensive assessment of the various sources of risk to which the
financial sector is exposed (Sorge, 2004).
Rojas-Suarez (2001) provides evidence that the traditional CAMELS
system has limitations in predicting bank failure, and needs to be complemented by
other indicators. Several studies based on U.S. bank data complement the FSI
analysis by suggesting that market-price based indicators contain useful predictive
information about bank distress that is not contained in the CAMELS indicators (e.g.,
Flannery, 1998; Curry, Elmer, and Fissel, 2001). 4
Besides the research on traditional balance sheet data, there is a growing
body of literature that analyzes the macroeconomic determinants of banks’ credit
risks. A more data-intensive approach is to examine the impact of macro factors on
the corporate and/or household sector default risk and map these developments into
banks’ loan losses using various techniques. Chan-Lau (2006) reviewed a number of
different fundamentals-based models—including macroeconomic-based models,
80
DECEMBER 2010
credit scoring models, ratings-based models and hybrid models—for estimating EDF
for firms and/or industries, and illustrates them with real applications by practitioners
and policy making institutions.
There are generally three approaches that can be used to link EDF with
macro-financial indicators: (i) the VAR framework, (ii) probit and logit models, and
(iii)panel models.
2.1 VAR Framework
Among the more recent contributions that use the VAR model to analyze the links
between the macro economy and the corporate sector credit quality are Alves (2005)
and Shahnazarian and Åsberg-Sommer (2007), who incorporate the Moody’s KMV
EDF data in cointegrated closed-economy VAR models. They find cointegration
relationships between the macro and EDF variables and identify significant
relationships between EDF on the one hand and short-term interest rates, GDP and
inflation on the other. Sommar and Shahnazarian (2008) use a vector error correction
model to study the long-term relationship between aggregate expected default
frequency and macroeconomic development, i.e. CPI, industry production and the
short-term interest rate. Aspachs and others (2006) use a VAR model which includes
the banking sector EDF and macroeconomic data on seven industrialized countries.
They show that shocks to banks’ EDF and equity values can have an impact on GDP
variables. Jacobson, Lindé, and Roszbach (2005) use the VAR approach to study the
interactions between Swedish firms’ balance sheets and the evolution of the Swedish
economy. They find that macroeconomic variables are relevant for explaining the
time varying default frequency in Sweden. Drehmann, Patton, and Sorensen (2005)
analyze corporate sector defaults in a non-linear VAR framework for the UK
economy and find that non-linearities matter for the shape of the impulse response
functions. Finally, Pesaran, Schuermann, and Weiner (2006) adopt the Global Vector
Autoregressive (GVAR) model to generate the conditional loss distributions of a
credit portfolio of a large number of firms in various regions of the world. Castren,
Dees and Zaher (2008) use the GVAR model to construct a linking satellite equation
for the firm-level EDF. Their results show that the median EDF react most to shocks
to GDP, the exchange rate, oil prices and equity prices.
2.2 Probit and logit models
The second approach is the use of probit and logit models to assess EDF. Virolainen
(2004) provides a good summary of this approach. Bunn and Redwood (2003)
examine the determinants of failure among individual UK companies with a probit
model to assess risks arising from the UK corporate sector. In addition to firmspecific factors like profitability and financial ratios, their explanatory variables also
include macroeconomic conditions (proxied by the GDP growth rate). GDP growth
proves to have a negative effect on the failure rate even after controlling for the firmlevel characteristics. They find that the measure which uses firm-level information
performs better in predicting actual debt at risk (ex post sum of all debt of failed
VOL.7 NO.2
81
firms) than a simple estimate that involves multiplying the average probability of
failure by the total debt stock. Tudela and Young (2003) analyze the performance of
a “hybrid model” by adding Merton-based default probability measures into a
company account data based probit model for individual firm failures. They find that
the implementation of the Merton approach clearly outperforms a model based solely
on company account data. Interestingly, they also find that even after controlling for
a Merton type default probability measure and company account variables, GDP has
a significant effect on firm default. Virolainen (2004) uses data on industry-specific
corporate sector bankruptcies and estimates a macroeconomic credit risk model for
the Finnish corporate sector. The results suggest a significant relationship between
corporate sector default rates and key macroeconomic factors including GDP,
interest rates and corporate indebtedness.
2.3 Panel models
The third approach is the use of panel models. Pain and Vesala (2004) employ a
dynamic factor model to analyze the determinants of firm default risk, as measured
by the Merton-based Moody’s KMV EDF, using a large panel of quoted EU area
companies. Although the factor analytic approach does not allow them to identify the
explanatory factors, Pain and Vesala conclude that EU-wide country and industrial
sector effects seem to play only a minor role in explaining EDF.
2.4 Assessing systemic risks
The fourth approach is the discussion of the framework to assess the systemic risk
and address the systemic importance of financial institutions (FIs). For instance,
Goodhart (2006) suggests that financial stability analysis should relate to the system
as a whole, not just to individual institutions. It needs to assess the probability,
virulence and speed of occurrence of potential shocks. Alexander and Sheedy (2008)
propose a methodology for stress testing in the context of market risk models that
can incorporate both volatility clustering and heavy tails. When applied to major
currency pairs using daily data spanning more than 20 years they find that stress test
results should have little impact on current levels of foreign exchange regulatory
capital. Huang, Zhou and Zhu (2009) propose a framework for measuring and stress
testing the systemic risk of a group of major financial institutions. Using realized
correlations estimated from high-frequency equity return data can significantly
improve the accuracy of forecasted correlations. Tarashev, Borio and Tsatsaronis
(2009) proposes a general and flexible allocation methodology and uses it to identify
and quantify the drivers of systemic importance. It illustrates how the methodology
could be employed in practice, based on a sample of large internationally active
institutions.
3. Differences between Intervened and Nonintervened Financial Institutions
Regulators and supervisors typically use a set of FSIs to assess the stability of their
financial system. Indeed, the International Monetary Fund has promoted their
82
DECEMBER 2010
construction and collection over the last several years. As a starting point for the
analysis, a small sample of major institutions is used to examine whether traditional
FSIs and other balance sheet data were able to discriminate between institutions that
would eventually require government intervention and those that were not
intervened. 5 This section seeks to identify the key indicators that are useful in
differentiating between the intervened and nonintervened FIs.
The advantage of this approach is that some indicators are readily available and are
widely used by financial regulators. In addition, we also investigate more indicators
related to the characteristics of the subprime crisis, such as subprime products and
business scope. However, these indicators are reported at low frequencies, are
generally static and backward-looking, and focus on an individual FI without much
regard for the spillovers from other institutions.
The sample comprises 36 key commercial and investment banks across the world.6
This sample of FIs is divided into nonintervened commercial banks (NICBs),
intervened commercial banks (ICBs) and intervened investment banks (IIBs) (Annex
A). The periods covered are: (i) 1998Q1–2008Q1 (before the wave of government
interventions), (ii) 2005Q1–2007Q2 (before the start of the current cycle and the
beginning of the subprime crisis), and (iii) 2007Q3–2009Q1 (during the subprime
crisis). A comparison of these indicators during 2007Q3–2009Q1 will enable us to
capture the possible differences among the three groups of FIs and to see if the crisis
has significantly changed the business and behavior of the FIs. Table 1 shows the
following features of the intervened and nonintervened FIs.
 Capital adequacy ratios were unable to clearly identify institutions
requiring intervention. In fact, contrary to the common belief that a low
capital adequacy ratio signals the weakness of an FI, all four capital adequacy
ratios examined for ICBs were significantly higher than (or similar to) the
NICBs as a whole (Figure 1). 7 In addition, the retained earnings to equity ratio
can also indicate differences among these three groups of institutions. During
all subsample periods, the retained earnings to equity ratios for intervened FIs
are much higher than for nonintervened FIs. This shows that the higher retained
earnings to equity ratio could demonstrate higher risks in FIs.8
 Several basic leverage indicators appear to be informative in identifying
the differences in the institutions 9 . The higher ratios of debt to common
equity, debt to assets, long-term debt to capital, short-term and current portfolio
long-term debt to total debt, and cost of debt in the ICBs and IIBs all indicate
that these measures of basic leverage are especially informative about the
differences (Figure 2).10 This echoes the fact that many FIs borrow far more
than the capital they had on hand to make additional investments in mortgagebacked securities, pocketing the 2%-3% difference between mortgage rates and
their cost of short-term capital. However, the formal leverage ratios, which take
general capital as denominator, such as assets (debt) to capital ratio could not
indicate any major difference.
VOL.7 NO.2
83
Table 1. Selected Indicators on Fundamental Characteristics of Financial Institutions
Capital/assets (%)
common equity/assets (%)
Tier 1 Capital/risk-weighted assets (%)
Tier1 and 2 capital/risk-weighted assets (%)
Retained Earnings/Equity (%)
cost of equity
NPL ratio (%)
reserve loan losses to capital ratio (%)
provision for loan losses/loans (%)
debt/assets
debt to common equity
long-term debt/capital
short-term debt /total debt (%)
cost of debt (%)
debt/capital
Loans/deposits
deposits/assets (%)
loans/assets (%)
ROA(%)
ROE(%)
Total Interest expenses/total deposits (%)
PE
EPS
book value per share
Mortgage loans/total loans (%)
net interest margin (%)
commision fee/operating income (%)
Interbank loans/loans
Noninterest Expense/income before taxes (%)
Operating Expense/operating income (%)
Nonintervened banks
Intervened commercial banks
Intervened investment banks
98Q1-08Q105Q1-07Q207Q3-09Q198Q1-08Q105Q1-07Q207Q3-09Q198Q1-08Q105Q1-07Q207Q3-09Q1
Capital adequacy
16.55
19.39
17.91
18.24*** 21.42* 24.74*** 17.27**
19.44
25.17***
3.99
4.36
4.44
5.98*** 5.66***
5.22
3.70
3.72*** 3.32***
7.19
9.61
8.66
8.91***
8.95
10.06***
10.65
14.12
13.52
12.23*** 12.45
13.83
37.73
45.89
51.82
60.97*** 60.81*** 59.3*** 75.54*** 90.64*** 74.51**
1.92
4.69
-5.02
0.15
3.57
-15.82 13.39*** 16.33*** -6.16
asset quality
2.38
2.24
1.64
1.43*** 0.86**
1.85*
11.16
6.53
6.19
6.11*** 4.17*** 4.57***
0.09
0.06
0.17
0.2*** 0.15*** 0.55***
leverage
0.29
0.28
0.26
0.34*** 0.35*** 0.36*** 0.47*** 0.48*** 0.47***
7.48
7.56
7.27
8.38*** 9.02*** 10.47*** 13.36*** 13.67*** 14.94***
58.64
62.15
63.88
61.8*** 66.38*** 72.11*** 74.95*** 79.74*** 80.12***
45.63
51.06
57.35
66.62*** 66.28*** 59.90
68.36*** 70.13*** 59.19
16.55
19.39
17.78
18.24*** 21.42* 23.92*** 17.27**
19.44
25.17***
2.17
1.99
1.82
2.12
2.01
1.86
2.9*** 2.51***
1.96
liquidity
1.25
1.33
1.22
1.19
1.31
1.23
49.01
45.22
42.65
41.88*** 38.7*** 37.37***
0.55
0.49
0.48
0.51*** 0.51***
0.5**
earning and profit
1.17
1.24
1.05
1.78*** 1.6***
1.44** 3.89*** 4.26*** 3.46***
3.86
4.77
2.38
4.05
5.34
-2.88*
4.09
5.32
-14.74**
0.02
0.02
0.02
0.02** 0.02*** 0.03***
0.00
0.00
0.00
stock market performance
15.45
12.60
11.77
15.98
11.66
9.23
15.58
13.08
14.37
0.55
0.96
0.34
0.55
0.94
-0.73*** 1.25*** 2.43*** -1.85**
14.61
21.40
26.65
13.56** 17.93*** 19.49*** 33.85*** 50.45*** 53.37***
Business scope
0.22
0.28
0.24
0.33*** 0.38*** 0.36***
1.84
1.79
1.76
2.92*** 3.2*** 2.63***
0.20
0.17
0.24
0.25*** 0.2***
0.29
contagion
0.12
0.09
0.09
0.16*
0.15*** 0.12**
management level
1.88
23.37
2.05
19.47
2.15
25.79
2.96
19.84
2.13
20.63
0.96
-13.36
-
-
-
Sources: Thomson Reuters; and IMF staff estimates.
Note: The ratios of nonintervened banks, intervened banks and intervened U.S.
investment banks are the average of all institutions in each category.

Traditional liquidity ratios are partially indicative of the differences
between intervened and nonintervened institutions. This may be partly due
to the fact that the loan to deposit ratio may not be able to measure fully the
wholesale funding risks. However, the ratio of deposits to assets for the
intervened institutions are much lower than those in the NICBs, suggesting that
elevated risks are associated with less dependence on retail deposits (or more
dependent on wholesale funding), thus undermining the banks’ capability to
fend off liquidity shocks. In addition, the ratio of loans to assets for the
intervened institutions is higher than that for the NICBs, suggesting that
84






DECEMBER 2010
elevated risks are associated with a higher loans-to-deposits ratio.
Asset quality indicators show a mixed picture. Similar to the capital
adequacy ratios, the ratio of nonperforming loans (NPL) to total loans for the
ICBs has been lower than for the NICBs, indicating that NPL ratio is not very
reliable indicator of the deterioration in asset quality(Figures 3). However, the
lower provisions for loan losses to loans ratio and the higher reserve loan losses
to capital ratio for the NICBs suggest that NICBs are more prudential in dealing
with possible loan losses, and these two indicators are better than the NPL ratio.
The standard measures of earnings and profits show a mixed picture. The
return on assets (ROA) for the intervened institutions is much higher than those
for the NICBs, suggesting that elevated risks are associated with higher returns
(Figure 4). However, return on equity (ROE) has not captured any major
differences between the FIs that were intervened and those that were not. This
contrast between the effectiveness of the ROA and ROE likely reflects the high
basic leverage ratio of intervened FIs, which typically rely on higher levels of
debt to produce profits.11
Some stock market indicators are able to capture some differences. The
book value per share of the IIBs has been generally higher than those of the
NICBs, which suggests that higher book value does not necessarily reflect
healthier institutions, but perhaps concomitant higher risks.
The indicators on possible contagion show a mixed picture. The ratios of
interbank loans to total loans for the ICBs are much higher than those for the
NICBs, suggesting that elevated risks are also associated with higher interbank
borrowing for the intervened banks, which might be more dependent on
wholesale funding from other banks.
The management quality indicators do not reveal differences. The
noninterest to income ratio and operating expense to operating income ratio for
the ICBs is the same as those for the NICBs, reflecting that the level of
management quality don’t make much difference.
The indicators on business scope are also able to capture the differences.
Net interest margin and the ratios of commission fees to operating income for
ICBs are much higher than those for the NICBs, suggesting that elevated risks
are associated with higher revenues from both off- and on-balance sheet
businesses. This reflects the fact that intervened banks are more aggressive in
doing off- and on- balance sheet business, which is naturally associated with
higher risks. In addition, The ratio of mortgage loans to total loans for the ICBs
are much higher than those for the NICBs, suggesting that elevated risks are
associated with a higher mortgage loans ratio in the banks’ portfolios, echoing
one of the features of the current crisis.
Our analysis finds that: (i) risk-weighted capital adequacy ratios have generally not
been informative in identifying financial firms that eventually required intervention
(in fact, the intervened institutions sometimes had higher capital adequacy ratios than
VOL.7 NO.2
85
the nonintervened institutions); (ii) several indicators, such as debt related leverage,
ROA, equity prices, management quality, and business scope have been better at
discriminating between intervened and the nonintervened institutions.
Moreover, further comparison among the three subgroups during the period of
2007Q3-2009Q1 shows that most of these indicators did not experience a significant
change in trend after the breakout of the crisis, reflecting the FIs’ difficulties in
dealing with their long-existing problems in their business models. However, some
indicators did experience great changes after the crisis. For instance, the cost of
equity for the intervened institutions is much negatively lower than that for the
NICBs, suggesting that the intervened institutions have lowered their dividends since
the crisis.
In sum, based on the sample of institutions examined, it would be useful to
include on the regulatory radar screen indicators on basic leverage, profit , reserve
loan loss to capital ratio, and business scope, since they could provide a starting
point for a deeper analysis of vulnerable institutions. Also, the center-stage focus on
regulatory capital adequacy ratios may need to be redefined, especially if it can be
shown that FIs were able to shift risks to off-balance sheet vehicles, which receive
lower risk weights, and thus the risks on the balance sheet are under-representing
those of the FI. Though the analysis here has been partial and cursory, others have
found similar issues with the application of FSIs, calling for further improvement in
the collection and usage of FSIs. On the other hand, for less sophisticated institutions
and general financial sector analysis, the current FSIs are useful, since the ratios are
the most available ones to meaningfully represent the FIs’ level of risks. Finally,
even those variables that can identify vulnerabilities do not necessarily mean they
can be separately used. We need to check their usefulness in a macro-financial
framework by putting them together with other bank-specific variables,
macroeconomic and global conditions. This will be the task in section IV.12
4. Methodologies and results of the panel specification and panel cointegration
As a robustness check on the usefulness of the indicators identified in section III, this
section attempts to take these indicators as the driving factors of EDF in a macrofinancial framework by using the panel specification and panel cointegration
techniques. Specifically, we estimate an econometric model that relates the EDF—
our main object of interest—to the macro-financial variables, and then test the longrun causal effect of key factors (i.e., basic leverage) on EDF.
We make two contributions to the empirical literature on the driving forces
of EDF. First, we employ quarterly data on three sets of factors as determinants of
EDF: (i) domestic macroeconomic factors, including inflation, GDP growth and real
effective exchange rate; (ii) bank-specific indicators, including basic leverage (i.e.,
debt to common equity ratio), capital ratio, return on assets, and equity prices; and
(iii) global factors, including global excess liquidity and IMF’s Financial Stress
Index.
86
DECEMBER 2010
Second, we use a conditional EDF, which is derived from nonstationary
techniques of panel cointegration. In particular, the cointegrated panel specification
framework provides us with a broader and more flexible approach, by which the
statistical proxies, such as the fixed effects and heterogeneous trend components, can
serve to capture a broad class of unobserved mechanisms.
The data set for the panel models consists of 45 FIs from different regions in the
world— the Euro area, NonEuro area, Asia, and the United States—covering
banking, securities, and insurance (Annex B). The data we use as a measure of
corporate sector credit quality are the EDF (both the one-year and the five-year
EDF)13, which are provided at the firm level by Moody’s KMV. The EDF, which are
publicly available measures of a firm’s probability of default, are a measure of the
probability that a firm will default over a specified period of time. EDF are dynamic
and forward-looking measures and are actual probabilities. When incorporated in the
panel models, the shifts in EDF provide a measure of the conditional expectation of
the FIs’ default intensities. We also use quarterly observations during 1998Q12009Q1. In order to get quarterly frequencies for all data, the daily data was
collapsed by taking the average of the observations of the quarter.
4.1. Panel specification
We define a fixed-effects panel data specification to examine the factors driving
EDF. Specifically, the three groups of factors are as follows:
Domestic macroeconomic factors include inflation, real effective exchange
rates, and real GDP growth.
Bank-specific factors include basic leverage, capital ratio, return on assets,
equity prices
Global factors include proxies for global excess liquidity (the difference
between broad money growth and estimates for money demand in the euro area,
Japan, and the United States) and the financial stress index.
The model is specified in terms of (log) differences of all macroeconomic and all
global variables.
The two alternative specifications for the panel data are as follows:
D EDFit = C + b 1INFLATION it + b 2D REERit + b 3D GDPit + b 4D CAPRATIOit + b 5 D LEVERAGERATIO it
+ b 6ROAit + b 7D MSCI it + b 8EXCLIQit + b 9D FSI it + b 10CONDEDFit + b 11INTERVENTION it + eit
Where “D” denotes log differences
D INFLATION = Inflation rate
D EXCHRATE = Real Effective Exchange rate
D GDP = GDP growth
D TCTART = Total capital to total assets ratio
(1)
VOL.7 NO.2
87
D LEVERAGERATIO = Debt to common equity ratio
ROA =Return on assets
D MSCI = Morgan Stanley Capital International world index
EXCLIQ = Global excess liquidity
D FSI = the change of Financial Stress Index
CONEDF = Conditional Expected Default Frequencies, which are derived from
panel cointegration among EDF, basic leverage (debt to common equity ratio) and
inflation
Dummy = Government intervention
ε = Residual
4.2. Panel cointegration
The study employs nonstationary panel techniques to deal explicitly with the
nonstationarities that are present in some individual time series that constitute the
members of the panel. Then the regressions of the EDF and nonstationary
explanatory variables are run to obtain conditional EDF, which are taken as inputs to
the specification of the panel estimations. This combination of conventional and
nonstationary panel techniques therefore allows us to focus explicitly on the
stochastic and nonstochastic long-run trend features of the data and filter out the
effects of short-run transitional dynamics.
The panel cointegration specification is as follows:
EDFit =
α1i ,t + β1i ,t CPI1i ,t + β 2i ,t LEVERAGERATIO2i ,t + eit
(2)
Where
EDFit = log Expected Default Frequencies
CPI1i,t = log CPI
LEVERAGERATIO2i,t = log Debt to Common Equity Ratio
If EDFit has a unit root (t=1,….,T, i represents the member of financial institutions),
so that EDFit~ I(1). And if CPI1i,t and DTCERT2i,t have a unit root (t=1,….,T), so that
CPI1i,t ~ I(1), DTCERT2i,t ~ I(1). The EDF, CPI, and Debt to common equity ratio are
cointegrated if the residual, eit= EDFit -αit-β1i,t CPI1i,t - β2i,t DTCERT2i,t, is stationary,
so that eit~ I(0).
In this cointegrated panel specification framework, the combination of the
extra dimension (the cross-sectional added to the time-series dimension) and the long
run properties of the cointegrating relationship provides us with a broader and more
flexible approach, by which the statistical proxies such as the fixed effects and
heterogeneous trend components can serve to capture a broad class of unobserved
mechanisms.
88
DECEMBER 2010
Moreover, the nonstationary panel framework allows us to relax many of
the strong assumptions that have typically been required in cross-sectional-based
approaches. This framework relaxes the exogeneity assumptions and picks up the
long-run relationships between the variables in a manner that is robust to the
presence of short run dynamics, and the steady state relationships even in the
presence of endogeneity among the right-hand side variables. Overall, this
cointegration framework allows for a broad set of mechanisms that may explain EDF
across institutions.
4.3. Unit root tests and panel cointegration test
Unit root tests show that the indicators used in panel cointegration tests—log EDF,
log CPI, and log Debt to Common Equity Ratio—are nonstationary (Table 2).
According to the Pedroni panel cointegration tests performed on the log EDF, log
CPI, and log Debt to Common Equity Ratio, the statistics point to the conclusion that
the variables are cointegrated (Table 3) (Pedroni 1995, 1999). Based on this
cointegration relationship, we obtain conditional EDF from the panel cointegrations
among log EDF, log CPI, and log Debt to Common Equity Ratio. After we obtain
the conditional EDF, we incorporate it into the panel estimation.
Table 2. Unit root tests
LOGEDF5 LOGDTCERT
LOGCPI
Levin-Lin rho-stat
-6.11
-2.06
2.55
Levin-Lin t-rho-stat
-0.73
0.50
1.83
Levin-Lin ADF-stat
-0.86
0.81
0.95
IPS ADF-stat
-4.74
-0.67
0.67
Source:Thomson Reuters; Moody's KMV; and IMF staff estimates.
Note: The critical values are -1.28 (10 percent) and -1.64 (5 percent).
VOL.7 NO.2
89
Table 3. Pedroni Heterogeneous Panel Cointegration
LOG Expected Default Frequencies, LOG CPI and LOG Debt
to common equity ratio
panel v-stat
6.25
panel rho-stat
-2.40
panel pp-stat
-2.05
panel adf-stat
-1.73
group rho-stat
-2.20
group pp-stat
-1.95
group adf-stat
-1.86
Source:Thomson Reuters; Moody's KMV; and IMF staff estimates.
Note: The first four tests are pooled within-dimension tests and the last three tests are
group mean between-dimension tests. Specifically, the first three statistics correct for
serial correlation, the fourth parametric test similar to the ADF-type test allows the
number of lags in the model to be estimated directly. The last three statistics treat the
parameter of interest as varying across the members of the panel. The critical values for
the variance statistic (v-stat) are 1.28 (significant at 10 percent level, denoted by *) and
1.64 (significant at 5 percent level, denoted by **), and those for all others are –1.28
(significant at 10 percent level, denoted by *) and –1.64 (significant at 5 percent level,
denoted by **).
4.4. Panel regressions
The estimation results for the full sample of 45 global FIs over the 45-quarter period
suggest that, for a given institution, EDF are positively associated with the inflation14,
basic leverage 15 , global excess liquidity, and global financial stress index, while
having a negative relation to equity prices and ROA.16 Moreover, the conditional
EDF is also significant across samples. A comparison of the two main groups of
intervened and nonintervened FIs indicates that these factors can explain around 50
percent of the change in EDF of intervened and nonintervened FIs. Moreover, there
appear to be stronger spillover effects for intervened FIs, as the two global market
factors remain significant and with higher positive coefficients than in the full
institution sample and the nonintervened FIs sample. However, the capital to assets
ratio, REER, and GDP are generally insignificant (Table 4). 17
In addition, we also test the significance of a dummy of government
intervention. This is significant for the full sample panel specification and intervened
institutions as well (Table 5).18
Given the increasing spillover (i.e., liquidity shock in the subprime crisis)
among global FIs, a global macroeconomic model is well placed to capture the
various shocks and interlink ages that might affect FIs’ EDF. By taking into account
a large set of linkages across macroeconomic and financial variables, the panel
model is particularly suitable for analysis of the transmission of real and financial
shocks across regions and institutions.
90
DECEMBER 2010
4.5 . Long-run causality tests
This section exploits a cointegrated panel framework to check the direction of longrun causality and the sign of the long-run effect between basic leverage and EDF. As
evidenced by equation (2), table 3, basic leverage is positively cointegrated with
EDF. This section endeavors to further detect these relations.
To do this exercise, we follow three steps: first, we estimate the cointegrating
relationship between log basic leverage and log EDF given in equation; second, we
then estimate the error correction model; finally, we calculate the long-run causal
effect of basic leverage on EDF following Pedroni (2008).
The results for each of these panel tests for the direction of long-run causality and the
sign of the long-run causal effect as described are presented in Table 5, which reports
the results for the direction of long-run causality between basic leverage and EDF.
The results are reported for the panel as a whole, as well as for intervened and
nonintervened subgroups.
In Table 6, the group mean tests indicate that the average long-run effects
are zero for the 45 FIs, the intervened and the nonintervened FIs. However, the
lambda-pearson tests clearly indicate that the long-run effects are pervasively nonzero individually for the 45 FIs and the two subgroups. Furthermore, the group
median sign ratio tests in column 8 indicate that the level of basic leverage is
associated with a positive causal effect pervasively among all FIs and the subgroups.
The implication of these results is that basic leverage level is positively associated
with permanent long-run causal effects on EDF through the FIs. These results can be
taken as further evidence of the damaging impact of higher basic leverage on EDF.
In addition, Table 6 shows that the sign of the "estimate" for three groups are all
negative, indicating that, EDF have a long-run negative causal effect on basic
leverage. That is, higher (lower) EDF tend to reduce (increase) basic leverage. The
implication is that there will be a tendency for the basic leverage to rise as long as
the default risks decline. Therefore, designing a mechanism to control basic leverage,
among others, would be vital to reduce EDF.
VOL.7 NO.2
Table 4. Fixed-Effects Panel Least-Square Estimation of the Determinants of EDF-Quarterly observations
(1998Q1-2009Q1), 45 financial institutions.
45 financial institutions
Intervened
non-intervened
Constant
-15.67
-18.11
-15.54
(0.00)***
(0.00)***
(0.00)***
Macroeconomic factors
Inflation
3.56
6.29
6.18
(0.06)*
(0.02)**
(0.03)**
REER
0.19
0.79
-0.28
(0.65)
(0.19)
(0.64)
GDP
-0.91
-7.06
1.38
(0.56)
(0.01)**
(0.50)
Bank-specific factors
Capital ratio
-0.02
0.06
-0.1
(0.52)
(0.27)
(0.05)*
leverage
0.1
0.1
0.08
(0.00)***
(0.01)**
(0.01)**
ROA
-1.5
-0.54
-1.75
(0.00)***
(0.68)
(0.00)***
equity prices
-0.74
-0.39
-0.86
(0.00)***
(0.08)*
(0.00)***
Global market conditions
global excess liquidity
12.35
12.94
11.81
(0.00)***
(0.00)***
(0.00)***
financial stress index
3.9
7.06
1.35
(0.00)***
(0.00)***
(0.14)
Dummy
Conditional EDF
9.89
10.67
9.03
(0.00)***
(0.00)***
(0.00)***
Adjusted R2
0.49
0.54
0.47
Time-series sample
(Quarterly)
1998Q1-2009Q1
1998Q1-2009Q1
1998Q1-2009Q1
No. of cross-section
institutions
45
18
27
No. of observations
816
434
382
Sources: Bloomberg L.P.; Thomson Reuters; IMF WEO Database and Moody's KMV.
Note: Probability values are in brackets (***significant at 1 percent level; **significant at 5 percent level; *significant
at 10 percent level).
91
92
DECEMBER 2010
Table 5. Fixed-Effects Panel Least-Square Estimation of the Determinants of EDF-Quarterly observations
(1998Q1-2009Q1), 45 financial institutions.
45 financial institutions
Intervened
non-intervened
Constant
-18.75
-28.53
-15.95
(0.00)***
(0.00)***
(0.00)***
Macroeconomic factors
Inflation
6.11
9.94
5.99
(0.00)***
(0.00)***
(0.03)**
REER
0.07
0.51
-0.15
(0.86)
(0.37)
(0.80)
GDP
0.7
-1.03
1.48
(0.65)
(0.68)
(0.47)
Bank-specific factors
Capital ratio
-0.05
0.02
-0.1
(0.14)
(0.65)
(0.04)**
leverage
0.1
0.16
0.1
(0.01)**
(0.15)
(0.02)**
ROA
-1.85
0.16
-1.78
(0.00)***
(0.90)
(0.00)***
equity prices
-0.73
-0.25
-0.88
(0.00)***
(0.22)
(0.00)***
Global market conditions
global excess liquidity
12.6
13.25
12.14
(0.00)***
(0.00)***
(0.00)***
financial stress index
2.7
5.37
1.21
(0.00)***
(0.00)***
(0.18)
Dummy
Governemnt Intervention
55.54
61.48
(0.00)***
(0.00)***
Conditional EDF
6.71
5.57
8.83
(0.00)***
(0.00)***
(0.00)***
0.53
0.6
0.47
Adjusted R2
Time-series sample
(Quarterly)
1998Q1-2009Q1
1998Q1-2009Q1
1998Q1-2009Q1
No. of cross-section
institutions
45
18
27
No. of observations
816
434
382
Note: Probability values are in brackets (***significant at 1 percent level; **significant at 5 percent level; *significant
at 10 percent level).
VOL.7 NO.2
Table 6. Long-run causality of leverage to EDF
λ2: Leverageit→EDFit
λ1:EDFit →Leverageit
estimiate
test
p value
estimiate
test
p value
All 45
Group mean
0.26
0.57
(0.72)
-0.17
-1.48
(0.07)
Lambda-Pearson
169.26
(0)
241.75
(0)
Intervened
Group mean
0.3
1.15
(0.88)
-0.19
-1.23
(0.11)
Lambda-Pearson
89.41
(0)
90.89
(0)
Nonintervened
Group mean
0.23
0.19
(0.57)
-0.15
-1.64
(0.05)
Lambda-Pearson
79.85
(0.01)
150.87
(0)
93
−λ2/λ1
median
0.22
(0.45)
0.14
(1)
0.26
(0.46)
Note: For each of these subgroups there are two rows, one for the group mean based tests, and one
for the lambda-Pearson based tests. Columns 2–4 report these for tests based on the parameter λ2i,
which reflects the presence or absence of long-run causality running from leverage to EDF. The
second column reports the panel point estimate, which exists only for the group mean, not for the
lambda-Pearson. The third column reports the corresponding panel test statistics and the fourth
column reports the p value for outcome of the panel test statistic. The next three columns repeat
this same pattern for analogous tests based on the parameter λ1i, which reflects the presence or
absence of long-run causality running from EDF to leverage. Finally, the last column reports the
group median point estimate of the sign ratio in the first row, with the simulated standard error
reported in parentheses in the second row.
5. Conclusions
This paper, by using both balance sheet data and panel approach in a macro-financial
framework, has provided the following key conclusions:
 Mixed results were found regarding the balance sheet data to highlight those
firms that proved to be vulnerable in the current financial crisis. Basic leverage
ratios were most reliable, and ROA, and business scope can also provide
predictive power. However, capital-to-asset ratios (including risk-adjusted
ratios), formal leverage ratio, and nonperforming loan data proved to be of little
predictive power. In the current crisis, key vulnerabilities were unanticipated
due to off-balance-sheet exposures and lenders’ dependence on wholesale
funding. Indeed, many “failed” institutions still met regulatory minimum capital
requirements.19 The monitoring of the specific components of capital, such as
retained earning to equity ratio, could be more helpful in detecting
vulnerabilities. In particular, caution should be taken to encourage banks to
increase retained earnings when boosting capital.
 Further econometric work using panel specifications and panel cointegration
further strengthen the importance of some bank-specific indicators, including
basic leverage, ROA, stock market performance indicators (equity prices and
book value per share) in driving the changes in the EDF.20 In addition, basic
94


DECEMBER 2010
leverage21 has a long-run causal effect on EDF. These evidences also suggest
that measures to set up basic leverage constraints could pay significant
dividends in restraining the rise in EDF when designing a new regulatory
framework.22 Once again, some indicators that are widely taken as important to
strengthen FIs and push forward future financial reforms, such as capital ratio
and formal leverage ratio, do not provide a useful indication of the rising
EDF.23
Price stability matters. As the panel specifications show, inflation can exert an
influence on EDF 24 . This further underscores the importance of maintaining
price stability, which is vital not only for monetary stability but financial
stability as well.
Global macroeconomic conditions also matters. There is evidence that global
excess liquidity 25 and the financial stress index 26 are significantly associated
with EDF. This appears to suggest that global FIs are highly vulnerable to
changes in the global conditions. This calls for better macroeconomic and
global policies to achieve lower expected default frequencies.
Figure 1 Capital to Assets Ratio
40
Capital to Assets Ratio
35
30
25
20
15
10
5
Non-intervened banks
0
Q1Y1998
Q2Y1999
Q3Y2000
Q4Y2001
Intervened banks
Q1Y2003
Q2Y2004
Intervened U.S. investment banks
Q3Y2005
Q4Y2006
Q1Y2008
Overall, the panel specifications and cointegration approach appears to be a useful
tool for analyzing plausible global macro-financial shock scenarios designed for
financial sector stress-testing purposes. The empirical analysis highlights several
VOL.7 NO.2
95
factors that would account for the vulnerabilities in the systemically important FIs.
The results discussed above and the policy challenges associated with them point to
the need to enhance the bank-specific institutional framework and reduce the
vulnerabilities emanating from the macroeconomic and global environment.
Figure 2. Debt to Common Equity Ratio
18
Debt to Common Equity Ratio
16
14
12
10
8
6
4
Intervened banks
2
0
Q1Y1998
Q2Y1999
Q3Y2000
Q4Y2001
Q1Y2003
Q2Y2004
Q3Y2005
Q4Y2006
Q1Y2008
Figure 3. Nonperforming Loan Ratio
0.045
Non-performing Loan ratio
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
Q1Y1998
Q2Y1999
Q3Y2000
Q4Y2001
Q1Y2003
Q2Y2004
Intervened banks
Q3Y2005
Q4Y2006
Q1Y2008
96
DECEMBER 2010
Figure 4. Return on Assets (%)
7
Return on Assets
6
5
4
3
2
1
0
Q1Y1998
Q2Y1999
Q3Y2000
Q4Y2001
Intervened banks
Q1Y2003
Q2Y2004
Q3Y2005
Q4Y2006
Q1Y2008
Notes
1
Tao Sun: Monetary and Capital Markets Department, International Monetary Fund.
Thanks are due to Laura Kodres and Brenda Gonzalez-Hermosillo for their advice.
Peter Pedroni provided help in improving the econometric work. Thanks are also due
to the participants (particularly Abol Jalilvand and Volbert Alexander) of
“Regulatory Responses to the Financial Crisis” in July 2010 for their constructive
comments. Yoon Sook Kim and Ryan Scuzzarella provided data support. All
remaining errors are my own.
2
Expected Default Frequency (EDF) is the probability that a firm will default within
agiven time horizon. Default is defined as failure to make a scheduled payment or
the initiation of bankruptcy proceedings. The main drivers of EDF credit measures
are the market value of the firm (asset value), the level of its debt obligations
(default point), and the volatility of firm value (asset volatility).
3
The 45 FIs are selected with an intention of their being systemically important in
the context of size, business scope and possible regional/global impact, though
proving this is beyond the reach of this paper.
4
CAMELS refers to capital adequacy, asset quality, management quality, earnings,
liquidity and sensitivity to market risk.
VOL.7 NO.2
97
5 In this paper, intervened institutions are assumed to be those that have gone
bankrupt, or that have received government capital injections or loans, or that have
had assets purchased by government, or that have received official loans to facilitate
a merger or acquisition. Central bank temporary liquidity injections are not
considered to be a type of intervention.
6
The insurance companies were excluded from the analysis given their different
business lines. The rationale for choosing these FIs is based on their systemic
importance while keeping a balanced sample that is representative of the various
regions around the world. Data constraints also played a role, as the sample chosen
was limited to FIs for which balance sheet and market-based data were available.
7
The reasons that capital adequacy ratios are not always useful indicators of distress
may reflect: (i) difficulties in determining the actual riskiness of assets; (ii)
deficiencies in mark-to-market accounting practices; and (iii) locating assets and
contingent claims (e.g., derivatives) in off-balance sheet vehicles where they can
receive lower risk-weights.
8
The higher risks associated with the higher retained earning to equity ratio indicate
that it would not be a safe way to rebuild their capital via retained earnings, as being
encouraged by European regulators (The Wall Street Journal, Sep 28, 2009 ).
9
Here we check more indicators on basic leverage rather than formal basic leverage
ratio－total assets to capital ratio and debt to capital ratio. The reason is this i)
capital includes too many items and can’t tell the difference in its specific
composition, although the capital in general reach the regulatory standards; ii) formal
basic leverage ratio may prove overly-optimistic since basic leverage migrates to
balance sheets requiring less capital but with higher risk.
10
Short-term debt and current portfolio long-term debt refers to that portion of debt
payable within one year.
11
The ratio of ROE has to be interpreted with caution, since a high ratio may
indicate both high profitability as well as low capitalization, and a low ratio can
mean low profitability as well as high capitalization (IMF, 2000). This caveat further
encourages the use of ROA as a better measure of earnings.
12
It should be noted that in a fast changing dynamic environment, this balance sheet
approach may not work well when there are nonlinearity and feedback effects. In
addition, this approach can not be applied to high-frequency data and multipleinstitutions with forward-looking feature.
13
The difference between the two EDF is that there is a higher relative variance of
the one-year EDF compared to the five-year EDF.
14
Theoretically speaking, the link between inflation and EDF is mainly twofold,
through factor prices and the prices that companies charge for their goods and
services. On the one hand, higher factor prices lead to increased production costs and
tend to impair credit quality, thus leading to higher EDF. On the other hand, higher
product prices can boost earnings and thereby improve creditworthiness, thus
98
DECEMBER 2010
resulting in lower EDF. In this case, the empirical evidence shows that the effect of
higher product prices outweighs that of higher factor prices, at least in the short run.
15
Here again, the formal basic leverage ratio－－total assets to capital ratio is
insignificant, further indicating it is relatively less useless than some other basic
leverage ratio.
16
The negative association between ROA and EDF in the panel regressions is not in
conflict with the fact that the ICBs have a higher ROA. This is because: the ROA for
the intervened institutions have declined quickly since late 2007, reflecting the rising
EDF, consistent with the panel analysis. The higher ROA value across 1998-2009 in
table 1 disguise the decline of ROA since the breakout of the subprime crisis. In
addition, this negative association shows the advantage of panel regressions, which
incorporate the combined effects of various indicators during the long time span, and
provide a general guidance of its impact on EDF.
17
The general insignificance of GDP growth, though significant for intervened FIs,
is not in line with Bunn and Redwood (2003)’s research using UK companies. This
could be due to the fact that we use more country samples, and the variation in GDP
growth could be large enough to offset each other.
18
As a robustness check, we also put those useful indicators identified in section III
into the panel regressions. The results show that book value per share (stock
performance) is significant, while deposits-to-assets ratio (liquidity), and mortgage
loans-to-total loans ratio (business scope) are generally insignificant.
19
However, FSIs are still helpful in assessing individual and systemic vulnerabilities
when reliable market data may not be available—particularly in less-developed
financial markets—as they can provide both an indication of rising vulnerabilities
and as a check when other information reveals weaknesses. For countries with more
sophisticated sources of information, FSIs could be usefully reevaluated, perhaps
refocusing them on basic basic leverage ratios and ROA as a proxy for risk-taking.
Of course, FSIs should be complemented by other measures and systemic stress tests,
and be broadened to better capture off-balance-sheet exposures and liquidity
mismatches.
20
The long-run causality relations between the basic leverage ratios and EDF further
confirm the importance of leverage in identifying risks.
21
In theory, debt is a disciplining device because default allows creditors the option
to force the firm into liquidation and thus exert pressure on the management to avoid
borrowing too much. However, the tremendous gain from leverage could impose
strong incentives for the management’s borrowing to achieve excessive returns. The
moral hazard associated with Too-Big-To-Fail would strengthen the incentives.
22
Given the fact that deleveraging process could trigger downward spirals in asset
prices, regulators must consider leverage constraints when designing policies for
capital regulation.
VOL.7 NO.2
99
23
Higher capital ratios, on their own, do not necessarily stop banks from financing
frothy asset purchases, and becoming vulnerable when a crisis occurs.
24
Higher inflation could create a room for more leverage and risk-taking behavior.
25
This is in line with the theory that Inflation is always and everywhere a monetary
phenomenon (Friedman, 1970). Global excess liquidity creates more searching-for-yield
behavior, thus more likely leading to higher risks and defaults.
26
Financial stress is often associated with the volatilities in banking, equity, bond,
exchange markets, which could trigger more losses and defaults.
References
Alexander and Sheedy, 2008, “Developing a stress testing framework based on
market risk models,” Journal of Banking and Finance, 32, 2220-2236
Alves, I. (2005), “Sectoral Fragility: Factors and Dynamics,” BIS Papers 22.
Aspachs O., C.A.E. Goodhart, D.P. Tsomocos and L. Zicchino (2006), “Towards a
Measure of Financial Fragility,” Working Papers, Oxford University,
http://www.finance.ox.ac.uk/file_links/finecon_papers/2006fe04.pdf
Bongini, P., L. Laeven, and G. Majnoni, 2002. “How Good is the Market at
Assessing Bank Fragility? A Horse Race Between Different Indicators,” Journal
of Banks and Finance, Vol. 26, pp. 1011–31.
Bunn, P. and Redwood, V. (2003) “Company Accounts Based Modelling of
Business Failures and the Implications for Financial Stability.” Bank of England
Working Paper No. 210.
Carmen Reinhart and Kenneth Rogoff, “This Time is different–-Eight Centuries of
Financial Folly ”, (Princeton), 2009
Curry, Timothy, Peter Elmer, and Gary Fissel, 2003, “Using Market Information to
Help Identify Distressed Institutions: A Regulatory Perspective.” FDIC Banking
Review, Vol. 15, no.3, pp. 1–16.
Canning, David, and Peter Pedroni, 2008, “Infrastructure, Long Run Economic
Growth and Casuality Test for Cointegrated Panels,” The Manchester School,
Vol. 76, No. 5, pp. 504–27. http://www3.interscience. Wiley.com/
cgiin/fulltext/121381553/PDFSTART
Drehmann, M., J. Patton and S. Sorensen (2005), “Corporate Defaults and Large
Macroeconomic Shocks,” Mimeo, Bank of England.
Demirgüç-Kunt, A., and E. Detragiache, 1998, “Financial Liberalization and
Financial Fragility,” IMF Working Paper 98/83 (Washington: International
Monetary Fund).
———, 1999, “Monitoring Banking Sector Fragility: A Multivariate Logit
Approach.” IMF Working Paper 99/147 (Washington: International Monetary
Fund).
100
DECEMBER 2010
———, 2005, “Cross-Country Empirical Studies of Systemic Bank Distress: A
Survey.” IMF Working Paper 05/96 (Washington: International Monetary Fund).
European Central Bank, 2005, “Financial Stability Review June 2005” (Frankfurt:
European Central Bank).
Flannery, Mark J., 1998, “Using Market Information in Prudential Bank Supervision:
A Review of the US Empirical Evidence,” Journal of Money, Credit and
Banking, August, pp. 273–302.
Goodhart, 2006, “A framework for assessing financial stability?”, Journal of
Banking and Finance, 30, 3415-3422
Hardy, D. and C. Pazarbaşioğlu, 1998, “Leading Indicators of Banking Crises: Was
Asia Different?” IMF Working Paper 98/91 (Washington: International
Monetary Fund).
Hermosillo-Gonzalez, Brenda, 1999, “Determinants of Ex-Ante Banking System
Distress: A Macro-Micro Empirical Exploration of Some Recent Episodes”,
IMF Working Paper 99/33, (Washington: International Monetary Fund).
Huang, Zhou and Zhu, 2009, “A framework for assessing the systemic risk of major
financial institutions,” Journal of Banking and Finance, 33, 2036-2049
Hutchinson, M. M., and K. McDill, 1999, “Are All Banking Crises Alike? The
Japanese Experience in International Comparison,” Journal of the Japanese and
International Economies, Vol. 13, pp. 155–180.
Hutchinson, M. M., 2002, “European Banking Distress and EMU: Institutional and
Macroeconomic Risks,” Scandinavian Journal of Economics, Vol. 104 (3), pp.
365–389.
Jarle Bergo, “Using Financial Soundness Indicators to Assess Financial Stability”,
Deputy Central Bank Governor of Norges Bank, at an IMF Conference,
September 17, 2002.
Jacobson, T., J. Lindé and K. Roszbach (2005), “Exploring Interactions between
Real Activity and the Financial Stance,” Journal of Financial Stability 1, 308341.
Jarrow, R. and S. Turnbull (1995), “Pricing Derivatives on Financial Securities
Subject to Credit Risk,” Journal of Finance, 50 (1).
Jorge A.Chan-Lau, “Fundamentals-Based Estimation of EDF: A Survey”
IMF,WP/06/149.
Virolainen Kimmo, “Macro Stress Testing with a Macroeconomic Credit Risk Model
for Finland”. Bank of Finland, Discussion Papers, Research Department
12.10.2004.
Marco Sorge, “Stress-testing Financial Systems: an Overview of Current
Methodologies”, BIS Working Papers, No 165, December 2004.
Martin Čihák and Klaus Schaeck, “How Well Do Aggregate Bank Ratios Identify
Banking Problems?”, WP/07/275, December 2007.
Olli Castrén, Stéphane Dées, and Fadi Zaher, “Global Macro-financial Shocks and
Expected Default Frequencies in the Euro Area”, ECB Working Paper Series No
875, February 2008.
VOL.7 NO.2
101
Owen Evans, Alfredo M. Leone, Mahinder Gill, and Paul Hilbers, 2000,
Macroprudential Indicators of Financial System Soundness, IMF Occasional
Paper, No 192 (Washington: International Monetary Fund).
Pain, D. – Vesala, J., 2004, “Driving Factors of Credit Risk in Europe”. Mimeo,
European Central Bank.
Per Åsberg Sommar and Hovick Shahnazarian,”Macroeconomic Impact on Expected
Default Freqency”, January 2008, Sveriges riksbank working paper series.
Pedroni, Peter, 2001, “Purchasing Power Parity Tests in Cointegrated Panels,” The
Review of Economics and Statistics, Vol. 83, No. 4, pp. 727–31.
———, 2007, “Social Capital, Barriers to Production and Capital Shares:
Implications for the Importance of Parameter Heterogeneity from a
Nonstationary Panel Approach,” Journal of Applied Econometrics, Vol. 22, No.
2, pp. 429–51.
Rojas-Suarez, L., 2001, “Rating Banks in Emerging Markets: What Credit Agencies
Should Learn from Financial Indicators,” Institute for International Economics
Working Paper, 01-6, May.
Shahnazarian, H., and P. Asberg-Sommer, 2007, “Macroeconomic Impact on
Expected Default Frequency,” Mimeo, Sveriges Riksbank.
Tarashev, Borio and Tsatsaronis, Sep 2009, “The systemic importance of financial
institutions,” Bank of International Settlements Quarterly Review
Tigran Poghosyan and Martin Čihák, 2009, “Distress in European Banks: An
Analysis Based on a New Data Set”, WP/09/09, January.
Tudela, M. – Young, G. (2003), “A Merton-model approach to assessing the default
risk of UK public companies”. Bank of England Working Paper No.194.
Simon Nixon, “Raising the capital stakes in Europe”, The Wall Street Journal, Sep
28, 2009.
102
DECEMBER 2010
Annex A. List of Intervened Financial Institutions
Date
(s)
of
Intervention
Intervened
institutions-banks
9/29/2008
9/29/2008
Country
Institution
Wachovia
Fortis
10/3/2008
10/13/2008
United States
Belgium/ Netherlands/
Luxemburg
Belgium/Netherlands
United Kingdom
10/16/2008
10/20/2008
10/28/2008
10/28/2008
11/24/2008
1/19/2009
Switzerland
Korea
United States
United States
United States
United Kingdom
Fortis
Royal Bank of Scotland,
HBOS, LloydsTSB
UBS
Industrial Bank of Korea
JPMorgan Chase & Co.
Bank of America
Citigroup
Royal Bank of Scotland
1/9/2009
Intervened
investment banks
3/14/2008
9/15/2008
9/15/2008
10/28/2008
10/28/2008
Intervened
insurance
9/16/2008
Germany
Commerzbank
United States
United States
United States
United States
United States
Bear Stearns
Lehman Brothers
Merrill Lynch
Goldman Sachs
Morgan Stanley
United States
AIG
VOL.7 NO.2
103
Annex B. List of Selected Financial Institutions
Regions
Insurance
Asia/United States
Companies
AIG (AIG)
Asia
Australia & New Zealand Banking Allianz (ALV)
Intesa Sanpaolo (ISP)
Group (ANZ)
Ambac Financial
BNP Paribas (BNP)
Bank of China (BOC)
(ABK)
Commerzbank (CBK)
DBS Group (DBS)
AXA (AXA)
Deutsche Bank (DBK)
ICICI Bank (IBN)
MBIA (MBI)
Munich
Re
Fortis (FORB)
Industrial Bank of Korea (IBK)
(MUV)
Mitsubishi UFJ Financial
ING Group (INGA)
(MUF)
PMI (PMI)
Santander
Hispano
Prudential
Plc
Group (SAN)
Nomura (NOM)
(PRU)
Société Generale (GLE) State Bank of India (SBIN)
Swiss Re (RUKN)
UniCredito (UCG)
Sumitomo Mitsui Financial (SUM)
Europe
Euro area
Non-Euro area
Barclays (BARC)
Credit Suisse (CSGN)
Danske (DANSK)
HBOS (HBOS)
HSBC (HSBA)
LloydsTSB (LLOY)
Nordea (NDA)
Royal Bank of Scotland
(RBS)
UBS (UBS)
United States
Bank of America (BAC)
Bear Stearns (BSC)
Citigroup (C)
Goldman Sachs (GS)
JPMorgan Chase & Co. (JPM)
Lehman Brothers (LEH)
Merrill Lynch (MER)
Morgan Stanley (MS)
Wachovia (WB)
Broad Banking, Financial Markets and the
Return of the Narrow Banking Idea
Peter Flaschel
Bielefeld University
Florian Hartmann
University of Osnabrück
Christopher Malikane
University of the Witwatersrand
Willi Semmler 1
The New School for Social Research
Abstract. We use a dynamic Keynesian multiplier and rate of return driven
adjustment for stock prices to study the role of commercial banks when embedded
into such an environment. We first consider a broad banking system where
commercial banks are trading in stocks and credit. We show that such a scenario is
likely to be unstable. We then consider narrow banking defined by Fisherian 100
percent reserves for checkable deposits and the exclusion of trade in stocks. It is
shown that in such a scenario stability is guaranteed by some weak assumptions. We
also study the efficiency properties of such a system.
JEL Classifications: E12, E24, E31, E52.
Keywords: Broad banking, Financial markets, Credit, Portfolio choice, Narrow
banking, Stability, Efficiency
1. Introduction
The role and extent of commercial banking and the issue whether it adds to
macroeconomic instability is currently in the focus of a large body of literature. 2
105
106
DECEMBER 2010
There are also a lot of historical studies that demonstrate that many of the historical
financial crises may have originated in adverse shocks to firms, households, foreign
exchange, stock market or sovereign debt. Yet, as has been shown 3 the banking
sector could seldom escape the crises. In fact most of crises ended up as a meltdown
of the banking sector and the banking sector has usually exacerbated and amplified
the crisis whatever origin it had and this even more since traditional banks have been
turned into investment banks. As Gorton (2010) shows in earlier times loan losses
and bank runs where usually the way the crises were triggered, but in recent times
banking crises seem to be strongly related to adverse shocks in asset prices. This is
occurring when banks have significantly invested in capital assets. One might want
to show of how such asset accumulation of banks can lead to a channel through
which some exacerbating or even destabilizing effects on the macroeconomy can be
generated.
The issue is whether we do have proper models to explain this. Do we have
models that help to understand this central aspect of the instability of the banking
system? There are the earlier non-conventional studies by Kindleberger and Aliber
(2005) and Minsky (1986, 1982) that view the role of credit as significantly
amplifying forces. In Kindleberger it is the instability of credit and in Minsky it is
the way financing becomes de-linked from collaterals that contributes to a downward
spiral once large real or financial shocks occur. This is surely an important tradition
that captured many of the aspects of the boom-bust scenarios that we have seen
historically.
On the other hand, recent vintages of the DSGE model, for example of the
Bernanke et al. (1999) type, have considered financial markets as accelerating force.
In principle such models can explain amplifications of the macroeconomy through
the financial side, the financial accelerator, but those models are locally stable. The
amplification through shocks is there, but the financial and real sides are mean
reverting: After an amplified shock the variables revert back to their mean level. This
is shown through local linearizations where the locally approximated linear system
shows the mean reverting tendencies in spite of some amplifications of shocks. 4
Moreover, in those DSGE models with a built-in financial accelerator the banking
system is often not specifically modeled.
Here we pursue a rather traditional root and model the banking system as
commercial banks that can accumulate capital assets in particular equity. In this
paper we use a minimal structure of assets to reconsider the issue of broad versus
narrow banking. Broad banking means that the bank can accumulate capital assets,
but in our set up there is only one risky asset (equities E ) and no further tradable
financial asset, but only two types of deposits (checkable deposits D1 and time
(saving) deposits D2 ) besides high powered money H supplied by the central
bank. The central bank can therefore only perform open market policies by trading in
equities and it can enforce reserve requirements in our model. We assume in this
respect that these requirements are only made for checkable deposits (commercial
VOL.7 NO.2
FLASCHEL-HARTMANN-MALIKANE-SEMMLER: BANKING
107
banks‘ money creation).
Policy actions are therefore narrowly defined, but open market policies do reach the
financial markets here in a direct way (and not indirectly via the federal funds rate).
Financial markets are modeled as portfolio choice of households between E and
M 2 (money and deposit holdings of households), who thereafter adjust the
structure of their M2 money holdings as described by the textbook money multiplier
(concerning high powered money and checkable deposits). Time deposits are treated
in a fairly standard way and are used to balance certain operations of the commercial
banks in this paper. The goods market is modeled via a textbook multiplier approach
and the labor market is assumed to be simply appended to what happens on the
market for goods.
We show that there are two sources of instability in the financial markets
(first, a Tobinian investment accelerator and secondly, accelerating capital gains or
losses and expectations about them). Moreover, there is a destabilizing credit channel
effect which comes into operation if commercial banks are strongly stock market
oriented in their decision on new loan supplies. 5 These feedback channels make the
considered situation of broad banking a fairly unstable one if the parameter that
characterizes their stock market orientation becomes large enough. Central banks can
influence this situation through wealth effects in the financial markets and through
the goods market dynamics if changes in high powered money have an influence on
economy activity.
Taken together it is however questionable if broad commercial banking can
be influenced to such a degree that instability, the occurrence of banking crises and
bank runs can be safely excluded from the working of the financial part of the
economy. We therefore propose – based on Fisher‘s (1935) 100 percent money
proposal – that money creation (in the form of checkable deposits) should be
excluded, at least to a greater extent, from the operations performed by commercial
banks. This would mean that the banks had to reduce proprietary trading
significantly. Noting that this has already become a major corner stone of the
Obama financial market reform.
We explore what it means if the banking sector of the economy is simply a
narrowly defined depository institution with respect to pure money holdings and is
primarily concerned with channeling the flow of savings (time deposits) into
investment flows where they act as credit creators, generating endogenous credit, but
not endogenous money. As we will show, such an economy is characterized by
strong stability features. In our view this case is to be preferred to the situation of
broad or excessive banking. Commercial bank money and credit creation may
sometimes be more flexible with respect to large upturns in investment booms, but
may be dangerous in opposite situations, where risk management has failed to work
and in cases where large bankruptcy scenarios (banks, firms and also governments)
can have dramatic chain effects on the working of the national and the world
economy.
We consider the model first from the perspective of stock-flow consistency
108
DECEMBER 2010
and thereafter study, through the introduction of laws of motions for the real and the
financial markets, the stock-flow interactions generated by the model under the
assumption of a ‗broad banking‘ scenario. Due to the analytical difficulties that pile
up when the model is too rapidly extended we are limiting our analysis to a set of
special cases here, before we contrast the obtained results with a ‗narrow banking‘
scenario. The paper is closed by comparing the obtained results with actual financial
market reforms of the past and the presence. Longer proofs are collected in an
appendix to the paper.
2. The basic accounting framework for investment, credit, and consumption
behavior
In this section we introduce the model by way of balance sheets and flow accounts
for the four sectors: firms, commercial banks, households and the central bank. We
first model the economy with a completely passive central bank and commercial
banks that can create deposits (ink stroke money) by purchasing equities on the stock
market from the household sector. We denote in the following by x the time
derivative of a variable x and by x̂ the growth rate of x and by f the derivative
f
of a function
2.1. Production and Investment
Table 1: Firms (f, loan and equity financing):
Balance Sheet:
Assets
Capital Stock
pK [ p 1 in the following ]
Inventories V
Liabilities
Loans L
Equities pe E
Net Worth
Flow Account:
Uses
Wages
Resources
wN (Y ) N (Y ) 0
Interest Payments
il (Y )(1
) L il (Y ) 0
Dividends
r (Y )(1
)E
r (Y ) 0
VOL.7 NO.2
Retained Profits or Losses
f
Unintended Inventory Changes
V
Output and Demand Y
Yd
X
X
Investment Function
I
109
i1Y i2 pe E al ( Lo
Investment Funds
L) I L
L
f
pe E sf E sf
E
E
The balance sheet of firms is a simple one. Firms have issued equities E and have
used credit L as external sources to finance their past investment into the capital
stock K We do not consider goods price inflation and normalize the corresponding
price level by 1. The only variable price of the model is the share price pe We
ignore the accumulation of assets E and K I in this paper. We will use a
dynamic multiplier process later on for the description of output dynamics which
means (since the Metzlerian inventory adjustment process is still absent) that
inventories V are adjusted passively by just the difference between aggregate
d
demand and aggregate supply Y
Y
V
We next consider the flow account of firms concerning production and their
investment behavior. We assume that the level of economic activity determines the
loan rate il and also the dividend rate r in a positive way. The only thing in the
production account that needs further explanation is given by the relationship
K
L KL
E KE and il (Y )(1 ) L , which we by and large assume to work
in the background of the model. We assume that capital depreciation occurs due to
bankruptcy which makes this part of the capital stock just disappear and which also
reduces the interest payments of firms and their loans by a corresponding amount.
As for the investment function, we assume that it depends positively on
capacity utilization and thus the activity level and also positively on the state of
confidence in the economy which we measure by the deviation of the share prices
from their steady state value. There is in addition a negative leverage effect in the
investment function. The investment function will be suitably extended later on.
Investment is financed through retained earnings
f (to be determined residually),
through new credit L and residually through the issue of new equities (depending
on the amount of retained earnings that firms can realize). Concerning the income of
firms we get the expression Y f
Sf
X which assumes that unintended
f
inventory changes and output (not sales) are used in the income calculations made in
this paper. There is moreover the following transfer of income from firms to the
) Eh based on the portion of dividends
household sector Yh wN (Y ) r (Y )(1
that go into this sector.
110
DECEMBER 2010
We stress again that the amount of investment that is financed by loans
depends on what is supplied by commercial banks (so that there credit rationing
occurring) and that the new equity issue is determined on this basis in a residual way
in order to get investment demand realized.
2.2. Banking and Credit
The balance sheet of commercial banks is also a simple one: Banks can provide
loans L out of checkable and time deposits D1 D2 but they can also invest these
deposits or the contract based returned principal on loans into stock holdings pe Eb
The interest rate on time deposits is id and considered as a given magnitude in this
paper, while the loan rate il was already assumed to depend positively on economic
activity Y There is no interest on checkable deposits which represent money
endogenously generated by the commercial banking system.
Table 2: Broad Commercial Banking (b, private ownership):
Balance Sheet:
Assets
Reserves R (
Hb
Loans L
Liabilities
Households‘ C-Deposits D1
b D1 )
Households‘ T-Deposits D2
Net Worth
Equities (from firms) pe Eb
Flow Account:
Uses
Interest Payments id D2
Reserve Adjustment
Resources
Interest Payments
il (Y )(1
)L
Change in C-Deposits
R 0
L Ebs 0
Dividends r (Y )(1
D1
Defaults
L (retained profits)
Distributed Profit
bh
il (Y )(1
)L
r (Y )(1
) Eb id D2
Change in Equity Holdings
b
0
) Eb
Net Loans
L
L
[bl (il (Y ) il (Yo )) be (ree reoe )]L
Change in T-Deposits
VOL.7 NO.2
L
pe Ebs Eb
Ebs
Eb
D2
111
0
We assume the simple textbook multiplier relationship between M
Hh
D1
where H denotes the high powered money issued by the central bank. This
multiplier formula is given as follows:
M
D1 H h
1
h
b
based on the relationships H h
h
H
m
H
h
D1 and R
b
D1 which represent the cash
demand of households and the reserve requirements of commercial banks,
respectively. This money multiplier is however assumed as inactive in the flow
account of banks. We also ignore changes in time deposits in this account for the
time being.
The first part of the flow account is then largely self-explanatory. We stress
however that it contains credit default (at rate
) and the corresponding loss of
interest on these loans. Moreover the amount of central bank money is considered a
given magnitude here. We assume finally that there is a positive reserve requirement
ratio b 0 on C-Deposits, but none with respect to T-Deposits.
If commercial banks intend to provide additional loans L (or intend to
reduce the number of outstanding debt) the following sequence of events is assumed
to happen. They sell (purchase) equities of amount
pe Eb
L . The means for the
intended supply of loans therefore lead to a reduction in the asset holdings of banks.
The opposite of course occurs when they find equities more interesting than loans
from the perspective of profit maximization (under uncertainty). 6 We assume also as
given a loan supply function L which depends on a comparison between the loan
rate (in its deviation from the steady state) and the rate of return of equities (again in
its deviation from the steady state). This later rate will be introduced when the
dynamics of stock markets are considered in the next section. We assume finally that
the profits (assumed to stay positive in this paper) made by the banking systems are
transferred to their owners, the sector of households.
In the flow account of banks we could allow in addition to the profit-oriented
reallocation between their new loan supply (which can also be negative if they do not
turn returned principals back into the credit market) and their equity holdings for the
loan generation sequence where commercial banks create new loans L which give
rise to new deposits D1 or D 2 through the circuit of money.
We summarize the above structure by pointing to its crucial elements again.
The amount of credit is assumed to be determined by commercial banks by their
comparing of the return on loans with the expected rate of return on their equities.
Additional loans are here generated solely through the sale of some of the equities of
112
DECEMBER 2010
firms owned by banks, i.e., there is not yet a supply of credit through the creation of
commercial bank deposits, since the income circuit and money per se does not
increase the stock of money. We here only discuss the possibility that there may be
credit rationing by commercial banks, in the extreme simply because they find it
more profitable from to invest in additional equities the principal they receive from
those firms that have to repay their contracted debt.
2.3. Households and Consumption
The flow account mirrors what was already discussed in the previous flow account.
It however adds now a consumption function to the investment function already
provided which uses as main determinants the income of households and thus the
activity level of the economy and the measure of the state of confidence we are
using. The account moreover shows again how loans are financed through the
creation of time deposits via the purchase of equities by commercial banks. Due to
these operations we assume that the savings of households goes into new equity
demand at first, subject to reallocations when financial markets are considered in the
next section. The income of households consists of wage income, dividend income
and loan rate income (which comprise time-deposit income, but is of reduced by the
defaulting loans).
The balance sheet of households is on the basis of what has already been said
and is self-explanatory.
Table 3: Households (h, bank and firm owners):
Balance Sheet:
Assets
Cash H h
Liabilities
C-Deposits D1
T-Deposits D2
Equities pe Eh
Flow Account:
Uses
Consumption Function
C
Resources
Wages wN (Y )
c1Y c2 pe E C
Change in Cash Holdings
Hh
0
Reallocation of Equity Holdings
Interest on T-Deposits id D2
Dividends r (Y )(1
) Eh
VOL.7 NO.2
pe E dh
pe E bs
Change in C-Deposits
D1
113
L pe Ehd
Extra Dividend Payments r (Y )(1
) Ec
0
D2
Households‘ Savings S h
Distributed Profit
0
Change in T-Deposits
bh
il (Y )(1
)L
r (Y )(1 ) Eb id D2
L
wN (Y ) r (Y )(1 ) E
il (Y )(1 ) L L
Income Yh
Note that we simplify dividend distribution by assuming that all dividends are
channeled back (one way or the other) into the household sector. Note also that the
savings of households is directed towards the demand of new equities solely and that
his portfolio is also modified by the loan – equity exchange of commercial banks.
Note finally that dividends are paid per equity unit and not per value unit of the
stocks and are thus independent of the occurrence of stock marked rallies.
2.4. The Monetary Authority
It is currently assumed that the monetary authority is completely inactive, but has
accumulated equities in the past, through its open market operations, which in this
model can only concern the equity market. All dividends that could accrue to the
central bank are assumed to be paid to or transferred into the household sector (for
reasons of simplicity), see their flow account. 7
Table 4: The Central Bank (c):
Balance Sheet:
Assets
Liabilities
High Powered Money
H
Equities of Firms pe Ec
Hh
R
CB: Net Worth
Monetary Policy (Flows):
Uses
Resources
Open Market Policies
0
114
0
) Ec
DECEMBER 2010
Equity Demand
CB Surplus: r (Y )(1
The accumulation effects
HH
Dividends r (Y )(1
) Ec
s
E f E h on the stocks of equities held and the reallocation
of the existing stock will be ignored as accumulation equations in this first version of
the model, just as the capacity effects on the capital stock through investment I and
the capacity effect on inventories through unintended inventory investment X .
The assumed major determinants of consumption and investment imply as
aggregate demand function the expression:
Yd
with a y
c1 i1 ae
a yY
ae pe E al ( L Lo ) A
ay
1
c2 i2 A C I The aggregate demand function is thus
based on income and activity level effects (on households‘ consumption and firms‘
investment), state of confidence effects on firms and households, and self-discipline
or enforced discipline of firms with respect to debt levels.
The laws of motions that flow from this section are:
Y
y
(Y d Y )
y
((a y 1)Y
ae pe E al ( L Lo ) A)
L [bl (il (Y ) il (Yo )) be (ree reoe )]L
with
ree
r (Y ) pe
e
e
r (Y ) 0 the expected rate of return on equities – to be
considered in the next section – and Yo
[ A ae peo Eo ] (1 a y ) Lo as the
steady state levels of economic activity and debt. The matrix of partial derivatives of
the Jacobian of this system at the steady state is given by:
(a y 1)
r
(bil l be ) Lo
pe
y
Jo
a
y l
0
0
We consider this subsystem of the full model as describing the credit channel
of it. The matrix of partial derivatives in this respect shows that the credit channel
(the interaction of firm‘s debt with economic activity) can be stable ( bl il
be
r
pe
)
VOL.7 NO.2
or of unstable saddlepoint type ( bl il
be
r
pe
115
). Increasing sensitivity of loan supply
to rates of return on the financial markets thus destabilizes the credit channel on the
real side of the economy. This may for example occur in the form of a Minsky
(1982, 1986) moment during periods of tranquil progress which may induce bank
management to bear more speculative risk than is acceptable from a pure banker‘s
point of view.
As we approach the last decade of the twentieth century, our economic
world is in apparent disarray. After two secure decades of tranquil
progress following World War II, in the late 1960s the order of the day
became turbulence - both domestic and international. Bursts of
accelerating inflation, higher chronic and higher cyclical
unemployment, bankruptcies, crunching interest rates, and crises in
energy, transportation, food supply, welfare, the cities, and banking
were mixed with periods of troubled expansions. The economic and
social policy synthesis that served us so well after World War II broke
down in the mid-1960s. What is needed now is a new approach, a
policy synthesis fundamentally different from the mix that results when
today’s accepted theory is applied to today’s economic system.
Minsky (1982, p.3)
3. Portfolio choice and the dynamics of financial markets
We consider next the financial markets of the economy which in this paper is simply
described by the portfolio choice (desired portfolio readjustment) of households
between money plus T-Deposits M D2 and equities Eh . We use a dynamic
approach here in place of a Tobinian equilibrium determination of share prices,8 by
assuming that stock imbalances in households‘ liquid portfolio: 9
pe E d
pe E
pe Ehd
ree
r (Y )
pe
pe Eh
f e (ree )( M
pe Eh )
pˆ e
in
e
the
e
(E
d
e
pe E d pe E
pe E
stock
E)
e
market
e
d
h
(E
Md
M
e
e
lead to a fractional flow demand for assets of amount
which in turn leads to
pe Eh
share price inflation
e
(E d
E)
e
(0 1)
or deflation of amount
the adjustment speed of share prices whereby equilibrium
is
reestablished
d
( Eh
Eh ).10
Excess
demand
e
e
Eh ) depends on the rate of return on equities r which is
composed of the dividend rate of return
r (Y )
pe
and expected capital gains
e
e
We
116
DECEMBER 2010
assume that there holds
fe (ree ) (0 1) fe
0 fe (reoe )
pe E
around the steady state value of the rate of return on equities.
Expected capital gains are based here on chartist behavior solely which is
modeled on the theoretical level by a simple adaptive expectations formation
mechanism. One could use nested adaptive expectations (humped shaped
explorations of the past) or other backward looking mechanism as well, but this
would increase the dimension of the considered dynamics, without leading really to
an increase in insight. Adding fundamentalists‘ behavior on the other hand could be
used to add stabilizing elements to the considered expectations formation, but again
not to a real change in what we shall show below.
The laws of motion shown below thus represent our modeling of the
dynamics of financial markets, primarily driven by the interaction between actual
capital gains and expected ones.
pe E d
pe
e
E
r (Y )
e [ fe (
pe
e
e
e
pe E
e
e
e
e
e
( pˆ e
(
e
e
e
e
e
e
)( M
pe E )
Whn
pe E ] E
M
pe E
)
[ fe (
r (Y )
pe
e
e
)( M
pe E )
e
e
pe E ] E
The Jacobian of these dynamics is given (at the steady state
)
peo
e
eo
0)
by:
e [ fe ( )
e
Jo
e
e
e
r ( )Whn
pe2
e[ f e ( )
( f e ( ) 1) E ] E
r ( )Whn
( f e ( ) 1) E ] E
pe2
e
e
[
f ( )Whn
e
e e
e
e e
f ( )Whn 1]
Stability analysis is simple in this case since the determinant of the matrix
is always positive and the trace of J gives rise to the critical stability condition
J
VOL.7 NO.2
n
e
H
e
e
e
[ f e ( ) r ( p)W2 h
(1 f e ( )) E ]
e
[
f ( )Whn 1]E
e
117
0
e e
if the entry J 22 is positive and thus representing a danger for asymptotic stability.
This asymptotic stability gets lost at the Hopf-bifurcation point
H
where the
e
e
system looses its stability in a cyclical fashion, in general through the disappearance
of a stable corridor around the steady or the birth of an attracting limit cycle
(persistent fluctuations in share prices) if the system is a non-linear one (where
degenerate Hopf-bifurcations are of measure zero in the considered parameter
space).
The considered Hopf-bifurcation represents in general however only a local
phenomenon, around the considered bifurcation parameter. We expect therefore that
the systems tends to become globally unstable when the adjustment speed of capital
H
gain expectations
e
e
e
becomes larger and larger.
This instability can be suppressed by introducing a Tobin type capital gain tax
(not as he has proposed it: a transaction tax) with respect to the stock market.
This modifies the second law of motion, for capital gain expectations, as follows:
e
e
e
e
((1
e
H
e
)
e
e
[ fe (
r (Y )
pe
e
e
)( M
pe E )
e
e
pe E ] E
)
and leads to
n
e
e
e
[ f e ( ) r ( p)W2 h
(1 f e ( )) E ]
e
[(1
e
)
e
f ( )Whn 1]E
0
e e
or
n
H
e
1
e
e
[ f e ( ) r ( p)W2 h
e
e
n
1
(1 f e ( )) E ] E
e
[ f e ( ) r ( p)W2 h
e
e
e
e
n
h
f ( )W
e e
(1 f e ( )) E ] ( E
f e ( )Whn
e
e
) 1 (
e
e
)
1
The destabilizing financial market accelerator can therefore always be tamed through
the introduction of an appropriate level of a Tobin capital gain tax. We assume now
in fact that this tax is operated as a stock tax, meaning that existing equities (on the
secondary markets) are taxed in this way (but not the issue of new equities by firms
on the primary markets). The change in taxation at time t is therefore given by
118
T
e
DECEMBER 2010
p eE which in the case of capital losses represents a subsidy to equity
holders. On this basis one can assume that the parameter c2 is affected (lowered) by
such a tax, but not the qualitative form of the consumption function.
Note here however that such a tax introduces a new type of income into the
economy, administered by an independent fiscal authority, which is assumed to raise
or deliver funds T according to the rule
T
e
pe E
We assume that this fiscal authority has an initial endowment that is large enough
such that this endowment remains positive during the business fluctuations that are
implied by the model.11
4. The core real-financial market feedback interactions
We consider first the interaction of share prices with the credit channel of the
economy by keeping capital gains expectation at their steady state value. The
resulting feedback chains are mathematically determined through the products of the
partial derivatives of the laws of motion that appear in the calculation of the principal
minors of the 3D Jacobian of the dynamics at the steady state of the model. The 3
principal minors of order 2 represent in this way the credit channel (if the third law
of motion is excluded), the financial accelerator (if the second is excluded) and a
Tobin-type real-financial market interaction in the last case.
Y
y
((a y 1)Y
L [bl (il (Y ) ilo ) be (
pe
e
e
([ f e (
r (Y ) e
reo )]L
pe
r (Y )
)( M
pe
pe E )] E
pe )
Note that this steady state is uniquely determined and given by
Yo
A peo E
1 ay
Lo
peo f e (r (Yo ) peo )( M
peo E )
peo E
Note also that we have to assume for the functions il r that there holds
il (Yo )
in the steady state.
The determinant of the Jacobian
holds:
r (Yo ) peo
a3 of this dynamical system is zero iff there
VOL.7 NO.2
f e Whn pr2
r
e
[1
]be
il pe
[ f e Whn pr2 1 f e ]
a3
l
b
119
r
il pe
e
For values of bl below this value we have a positive determinant and thus the
instability of the steady state of the dynamics.
The Routh-Hurwitz coefficient a1a2 a3 on the other hand is zero iff:
b
l
b
if
e
r
be
il pe
( J11
J 33 ) J 2
il
a Lo
y l
y
e
(1 a y )
r r
e e p 2 pe
e
b
f e Whn
a
y l
r
il pe
sufficiently large. The opposite holds true if this parameter is chosen
sufficiently small. For values of bl below this critical value we have a negative
a1a2
a3 expression and thus the instability of the steady state of the dynamics.
It is obvious that the conditions
a3
0 a1a2 a3
0
imply
a2
since a1
trace
0
0 holds true, so that stability will be given for all
bl
max{bla3 blb }
while there is instability below this maximum, which there represents the critical
stability condition for these dynamics. The details of the proofs are provided in
appendix I.12
There may be a Minsky (1982,1986) moment present in this type of an
economy whereby the parameter be is increasing relative to bl over time, since
equity markets become more and more the focus of interest of banks in relatively
prosperous and tranquil phases of economic evolution. The economy may therefore
become more and more fragile and volatile over time. Minsky type moments can be
introduced into the dynamics of this section by the systematic change in some
parameters of the model towards more volatile parameter constellations.
5. Open market policy
We now consider the possibilities for the central bank to steer the economy in the
context of broad banking. Since the rate of interest on T-Deposits does not influence
economic activity as well as financial markets there remains in the context of the
model only the possibility to conduct open market operations through the purchase
120
DECEMBER 2010
or selling of equities on the market for stocks (through trade with the household
sector). This policy is assumed to react to the state of confidence in a negative way
and is therefore characterized as being countercyclical in nature, and shown in the
flow account of the central bank below.
Uses
Resources
H
d
c
pe E
) Ec
CB Surplus: r (Y )(1
Equity Demand
cm ( peo
pe ) E
H
Dividends r (Y )(1
H-sector
) Ec
Additional credit supply is now generated through the shown open market operations
of the central bank, leading to the following sequence of events with respect to
money and credit:
M
Hh
D1
m
D1
H (1
m
H
h
)
R
H
Hh
Hh
b
D1
Hb
since the changes in the reserves of commercial banks and the high powered money
holdings of households are automatically adjusted by means of the money multiplier,
creating deposits of amount D1 which can be totally transformed into loans by the
commercial banks (but not into T-Deposits), since the reserves of the banks have
already been adjusted.
Table 6: Broad Commercial Banking (b, private ownership):
Flow Account:
Uses
Resources
il (Y )(1
id D2
R
Hb
b
D1
D1
L
H (1 h )
r (Y )(1 ) Eb
m
D2
bh
e
e
L [bl (il (Y ) il (Yo )) be (r
e
eo
)L
r )]L pe E
s
b
(1
0
b
) D1
VOL.7 NO.2
121
In the flow account of commercial banks we have now the presence of a money
multiplier process. We ignore here further circuit effects of the money supply for
simplicity. Note that this is not yet a situation with endogenously generated credit,
since the impulse for money creation comes from the central bank. The result is
therefore of a conventional textbook multiplier type. The changes implied in the
household sector are shown in their flow account as follows:
Table 7: Households (h, bank and firm owners):
Flow Account:
Uses
C
wN (Y )
c1Y c2 pe E C
pe E hs
D1
Resources
pe ( E bd E cd )
m H (1
h)
Hh
h D1
D2 0
Sh
r (Y )(1
) Eh
r (Y )(1
) Eb
r (Y )(1
) Eb
id D2
bh
wN (Y ) r (Y )(1
Yh
) E il (Y )(1
)L
L
Taken together the structure of the model is only modified in the equity demand
function of commercial banks (which is based on their intended loan supply
function). This does not change the laws of motion of the model and thus implies
that monetary policy is completely ineffective in this case.
This however is not completely true since we have neglected here the effect
of changes in H on the definition of private wealth
canceling balancing terms by:
Whn
(
m
1) H
Whn which is given by
pe E This implies that the
monetary policy is feeding back into this term and thus into stock price dynamics
such that the original 3D Jacobian is augmented as follows:
Jo
J11
J 21
J 31
J12
J 22
J32
0
0
The determinant of this matrix is given by
J13
J 23
J33
0
0
0
122
Jo
J11
J 21
0
J12
J 22
0
0 0
0 0
0
0
0
0
DECEMBER 2010
which is positive if and only if the shown 2D subsystem has a positive determinant.
This shows that monetary policy and the implied endogenous money creation is
adding stability to the considered 3D dynamics, at least for small values of cm
since negative real parts of the three eigenvalues of the 3D system must then be
augmented by a fourth eigenvalue which is negative.
6. Credit demand and extended goods market dynamics
We are here reconsidering the supply schedule of bank loans by explicitly adding a
d
d
l f (il ) L l f (il ) 0 . This gives as equilibrium
demand side expression to it now:
condition for the credit market the relationship:
0 [bl (il ilo ) be (ree reoe )]
if we specify loan demand by assuming
d
l f (il ) L
fl (il ilo )
fl (il ilo ) L Note that we no
longer postulate a relationship between economic activity and the loan rate, since this
relationship is to be derived now. The equilibrium condition for the credit market
implies:
il
be (ree reoe )
bl fl
ilo
The new law of motion for loans therefore now is
L
fl
bl
fl
be (ree reoe ) L
The Jacobian is in this case characterized by
VOL.7 NO.2
y
Jo
(a y 1)
fl
bl
e
fl
e fe
a
aE
y l
r
pe
0
Whn r
pe
0
be
123
y e
fl
e
be
r
pe2
bl
fl
e [ fe
Whn r
1 fe ]
pe2
0
0
It is again possible to derive the type of maximum condition we have considered
beforehand. Since a1 0 holds true again, stability will be given for all
max{bla3 blb } while there is instability below this maximum. Yet, in the
present situation, we observe that the system becomes unstable if the parameter be is
chosen sufficiently large, since the parameter bl is no longer available to
bl
compensate for this and since the credit channel is now always an instable one. This
holds, since the coefficient a2 from the Routh Hurwitz condition can now easily be
made positive by increasing the parameter be since there is no more much stability
resistance present in the terms that make up the coefficient a2 (while the
determinant is still composed by opposing effects of the parameter be ).
Assuming a Minsky bankers‘ carelessness increasing moment at work in the
sizing of the parameter be may therefore lead to instability when the always
destabilizing credit channel becomes sufficiently dominant. Instead of going into the
details of a stability analysis of the present case we extend it further by recognizing
d
that the aggregate demand function Y did not allow for an explicit role of credit.
However since part of the investment is credit financed it should be explicitly
augmented by the credit volume currently provided, implying that the dynamic
multiplier should – on this basis – in fact be formulated as follows:
124
Y
L
y
[Y d Y )
fl
bl
fl
y
((a y 1)Y
DECEMBER 2010
ae pe E al ( L Lo )
L L A]
be (ree reoe )
It is obvious that the determinant of the Jacobian of these subdynamics is unchanged
through this extension. This implies that the arguments of the preceding case remain
intact, in particular the one concerning the Minsky moments in the credit channel.
7. Narrow banking and efficient credit supply
The return to the narrow banking idea, related to what Fisher (1935) proposed after
the Great Depression in his book 100% Money, has recently been discussed again by
de Grauwe (2008). In the mainstream textbook literature, however, see for example
Freixas and Rochet (2008), this idea lives at best a shadowy existence, though of
course the topic of bank runs is definitely of importance in this mainstream literature,
see for example Rochet (2008) and Sinn (2009).
For simplicity we now assume that
0 holds true and that an inflow of
checkable deposits is reallocated in equal proportions into such deposits and timedeposit increases. The circuit of credit and money then implies that the loans
commercial banks intend to provide are exactly backed up by (reserve-free) timedeposits.13
Resources
Uses
pe E
CB Surplus: r (Y ) Ec
Equity Demand
d
c
H
HH
H
cm ( peo
pe ) E
Dividends r (Y ) Ec
Narrow Commercial Banking (b, private ownership):
Flow Account:
Uses
Resources
id D2
il (Y ) L
Reserve Adjustment R
D1
New C-Deposits
D1
(1
h
)H 2
VOL.7 NO.2
R( L)
125
D1 ( L)
il (Y ) L id D2
bh
Loan Supply
L cb (Y Yo ) (1
New T-Deposits
h
)H
D2 cb (Y Yo ) (1
h
)H
We reconsider in this section Fisher‘s (1935) 100%-money proposal as modification
of our modeling framework of a commercial banking system that acts on the credit
market and the financial markets without any institutional barrier. We therefore now
assume – to limit such a behavior from an ideal perspective of Fisher (1935) – that
checkable deposits are secured by a reserve requirement of 100 % ( b 1 ), so that
commercial banks are reduced to purely depository institutions in this respect, while
there are no reserve requirements on T-Deposits D2 , which are safeguarded by other
means (including contract lengths, withdrawal penalties) against bank runs. Time
deposits earn an interest rate that is interrelated with the loan rate received by firms
and manipulated in order to initiate that granted loans are backed up by time deposits
through the circuit of money when these loans reappear at first as checkable deposits
in the money holdings of the household sector. We are thus now allowing (which can
also be added to what we discussed beforehand) for the endogenous creation of
commercial bank money, in addition to what we discussed when the textbook money
multiplies was considered. By contrast there are now no equity holdings of
commercial banks anymore.
This type of money creation concerns the difference M 2 M 1 of the
conventional measures of money supply only and thus does not allow banks to get
interest income out of the money deposits for which they pay no interest. These
money holdings are thus always checkable central bank money and can therefore not
be subject to bank runs, since they are purely passive in the balance sheet of the
banks and not at their disposal should they become insolvent.
Table 9: Households (h, bank owners and firm stock owners):
Flow Account:
Uses
C
Resources
wN (Y )
c1Y c2 pe E C
id D2
Change in Equity Holdings
pe E hs
pe E cd
Change in Cash Holdings
r (Y ) Eh
H
Hh
h
H
r (Y ) Ec
126
DECEMBER 2010
C-Deposits Change
D1
(1
h
)H 2 L
T-Deposit Change
D2
L
bh
il (Y ) L id D2
wN (Y ) r (Y ) E il (Y ) L
Yh
The view thus is that commercial banks should not be allowed to endogenously
create money out of the central bank money in their balance sheet and also not by
purchasing equities through ink stroke money. The full control of the M1 money
supply process – in our view – should remain in the hands of the central bank which
not only eliminates bank runs on checkable deposits. The primary role of the
commercial banking system then becomes to channel not only the interest bearing
savings of households into the investment projects of firms – besides the creation of
T-Deposits through their autonomous lending decisions through the circuit of money,
supported in addition by the money supply or withdrawal rule of the central bank.
The changes implied in the household sector are shown in their flow account.
Taken together the dynamics of the model we considered so far is then modified in
the loan supply function L (plus the correction of the aggregate demand function
discussed in the previous section).14 This gives rise to:
Y
y
((a y 1)Y
(1
L
pe
h
)cm ( peo
cb (Y Yo ) (1
e
e
([ f e (
ae pe E al ( L Lo ) cb (Y Yo )
pe ) E
h
)cm ( peo
r (Y ) n
)Wh
pe
This gives as Jacobian (if we assume that
Jo
A)
pe ) E
pe E ] E
a y cb 1 holds true):
0
0
Assuming again that the Tobinian real-financial market interaction, the interaction
between pe and Y is a stable one, i.e., J 2 0 which holds here a fortiori, since
monetary policy is influencing the output dynamics in a stabilizing way, then implies
stability also for the remaining feedback interactions, if the parameter cm is chosen
VOL.7 NO.2
not too large, since a1 a2 a3 are then positive. Moreover, a1a2
127
a3 must be
positive then too, since the problematic term in the determinant is part of the positive
items in the product a1a2 while the other one adds to the positivity of the remaining
terms in the product a1a2
The assumed type of narrow banking therefore not only eliminates the
discontinuities created by the occurrence of bank runs, but also makes the economy a
stable one if the real financial market interaction (the product of the coefficients
J13 J 31 ) is not allowed to work in a too pronounced way by a proper choice of
monetary policy). This shows that Narrow Banking is dynamically seen more
reliable and robust than the model of broad banking we have used beforehand.
But is it also as efficient in the supply of credit as the broad banking system
(which as we know can be plagued by credit rationing if banks are too much focused
on financial markets instead). This will indeed be the case – ignoring the financial
market focus of broad banking already – if the interest rate on T-Deposits can be
managed by commercial banks effectively such that their loans (supplied in view of
the credit demand of creditworthy firms) are channeled into time deposits in place of
checkable deposits to a sufficient amount (we have assumed a fifty-fifty rile above).
It is then a matter of the variables il id to achieve such a result with however
rationing occurring if there are limits to the interaction of these two interest rates.
In our view the considered institutional change outperforms possible
efficiency gains of non-credit based risk taking broad banking. Loan supply does not
depend negatively on the rate of return on the stock markets, but now positively on
the level of economic activity. Moreover, a countercyclical monetary policy with
respect to the state of confidence (and the level of economic activity) of the economy
further improves the stability of the dynamics. And the channeling of sufficient
checkable deposits into time deposits allows to serve not only the autonomous
supply of loans by the banking system, but would also allow to support a loan
demand from firms through the creation of sufficient time-deposits.
8. Implications for Financial Market Reform
Our model of narrow banking could be considered an extreme case, where there is no
involvement of the commercial banks in security underwriting and security trading.
This is just banking in the traditional sense with accepting time deposits and
providing loans, and with a depository role solely as far as checkable deposits are
concerned. These checkable deposits are safe due to the fact that they are – in the
ideal case postulated by Irving fisher (1935)– fully covered by reserves. Bank runs
are thereby prevented since the time deposits cannot be withdrawn without accepting
a (significant) penalty. On the other hand our concept of broad banking, represented
by investment banks, implies that equity purchase can substitute the supply of loans,
and thus, under broad banking loans may be reduced by investment banks and capital
128
DECEMBER 2010
assets purchased. Though in practice investment banks might borrow from capital
markets to extend loans, for example from the money market or through carry trade,
the emphasis being here on security trading. In other words, what we have basically
stylized here in an extreme form through the two types of banking — narrow and
broad banking — is commercial banks and investment banks.
In some way, this was the vision of the 1930 banking reform, inacted by the
Glass-Steagall act of 1933, where commercial banks were not allowed to underwrite
securities and to trade securities. The Glass-Steagall act was repealed under the
Clinton administration in the US in 1999. This opened the door for commercial
banks to engage in the security sector either in-house or through affiliations. From
early on the conflict of interest was pointed out when regular banking and loan
business are mixed with security issuing and trading, see Puri (1994) and later the
evaluation by Gande (2008). This way, it was argued, commercial banks would
become investment banks and they would have superior knowledge about firms to
whom they lend. This would give them an unfair advantage that could result in
monopolizing the market. On the other hand the strict separation of banking and
security sectors never existed in its pure form under the Glass-Steagall act and many
excessive practices with respect to risk taking, leveraging and bonus payments took
place. It is thus not quite clear whether investment banking is the main cause for the
financial meltdown of the years 2007-09, see Shin (2009).
What rather appeared to be major problems were the lack of leverage
regulation, proprietary trading, excessive risk taking and bonus payments,
unregulated derivative trading, and rise of unequal size of banking and investment
firms, see Kaufman (2009).15 In particular for the investment banks the leveraging,
as measured as capital assets over equity, rose from 22.7 percent in 2001 to 30.4
percent in 2007, see Shin (2009).
Moreover, the issue of too big to fail came up with respect to the investment
firms. According to Kaufman (2009), in the last 15 years the 8 biggest investment
firms could increase their market share from 10 to 50 percent. So one might argue
that a pure separation of commercial banking and investment banking will not
generate a persistent solution for a stable banking sector.
The US administration under President Obama was aware of this and passed
the Dodd-Frank Act in 2010 on a more comprehensive financial reform. A core part
is the investment bank sector with the banning of proprietary trading. Narrow
banking will however not be introduced, the reforms are far away from this. Solely
to do this may not be so effective, but rather comprehensive reforms seem to be
needed. The reforms are aiming at preventing again a meltdown of the sort that has
occurred 2007-9. The banking system, mainly investment banking, but also
commercial banking, had created excessive risk taking through proprietary trading,
issuing of complex securities and excessive bonus payments.
Major points of the legislation were to avoid future bailouts and the cost that it has
imposed on the tax payers through the establishment of a fund. A system risk council
would be empowered to require that financial institutions that may pose risks to the
VOL.7 NO.2
129
financial system be regulated by the Federal Reserve, and it would also make
recommendations to the Federal Reserve regarding capital, leverage, liquidity, and
risk management requirements. Furthermore, what is planned is a regulation of overthe-counter derivatives and a tougher regulation of credit rating agencies, a
regulation regarding corporate governance and executive compensation by
shareholders, and a consumer protection agency which is a new agency that is
supposed to monitor credit-card fees, credit agreements and mortgage offerings to
make sure consumers are protected from predatory lending.
So there is a more comprehensive regulatory reform intention, but crucial in
our context is the following: One important new provision is that the Act
significantly restricts proprietary operations undertaken by commercial banks
(provision known as the Volcker rule). Banks can place up to 3 percent of their Tier
1 capital in hedge fund and proprietary trading investments. The other aspect of the
rule is that banks are prohibited from holding more than 3 percent of the total
ownership interest of any private equity investment or hedge fund. This falls short of
a complete disallowance of proprietary desks, which had been originally suggested
and would have been equivalent to restoring the Glass-Steagall act. Further, there are
some notable exceptions to the ban. There is a list of permitted activities, including
investments in U.S. government securities, transactions made in connection with
underwriting or market making related activities, transactions on behalf of
customers, and ―risk-mitigating hedging activities‖ in connection with individual or
aggregated holdings of the banking entity. So overall, there is some move to avoid an
extensive broad banking with excessive and uncontrolled security and derivative
trading, but the financial reform seems conceptionally broader than this. Yet how
much will finally be implemented remains to be seen.
9. Conclusions
We discussed in this paper, monetary and fiscal policy measures aimed at preventing
the financial market meltdown that started in the US subprime sector. This meltdown
has spread worldwide and developed into a great recession. Although some slow
recovery appears to be on the horizon, it is worthwhile exploring the fragility and
potentially destabilizing feedbacks of the banking sector and the macroeconomy in
the context of Keynesian macro models.
We have used a simple dynamic multiplier approach on the market for goods
and also a simple rate of return driven adjustment rule for stock prices to study the
role of commercial banks and credit when embedded into such an environment. We
first considered the implications of a broad banking system where commercial banks
are allowed to trade in capital assets (here equities) as a substitute for lending. We
showed that such a scenario is likely to be an unstable one, even if an appropriate
monetary policy of the central bank is added to the considered dynamics. Though in
our simplified model of broad banking asset purchases and credit expansion are
substitutes, as we have indicated, the model can be extended to accommodate more
the empirical fact of comovements of asset purchases and credit expansion.
130
DECEMBER 2010
We then considered a situation of narrow banking which is defined by a
Fisherian 100 % reserve ratio for checkable deposits and the exclusion of trade in
stocks for commercial banks. This would imply a significant reduction of proprietary
trading of the banking sector. It was shown that: a) in such a scenario stability is
guaranteed by some weak assumptions on the behavior of economic agents, b) a
sufficient loan supply to the entrepreneurs is guaranteed in such a framework, and c)
disastrous bank runs are no longer possible, in contrast to what is possible under
broad and also traditional banking.
Narrow banking thus not only provides systemic stability in place of systemic
crises, but also dynamic stability as well as sufficient efficiency of the credit creation
process. Though narrow banking appears a too extreme case to be implemented
realistically, it shows the improved stability properties when broad banking is
constrained.
In this paper we have concentrated the consideration of broad commercial
banking on the case where the supply of credit versus investment in financial assets
are in the focus of interest of commercial banks. This is however only a partial view
on their activities which moreover can include in particular the channeling of
households savings in form of time or checkable deposits into credit for firms.
Moreover, there also exists a channel – working in the opposite direction – that leads
from firms‘ credit demand to the generation of household deposits that back up this
demand. This mechanism of endogenous money creation has been introduced,
contrasted with broad banking and investigated in the section on narrow banking, but
can of course also be active under broad banking. In the interaction of savings and
investment we therefore have from the viewpoint of the circuit of money causalities
that run from saving to investment, but also a circuit that is working the other way
round. In addition there is the interaction of credit supply with investment in
financial markets under broad banking. Such an extension of the models of this paper
is needed if one wants to discuss the stylized fact of the comovement of credit and
stock markets, an observation that must however be left here for future research.
VOL.7 NO.2
131
Appendix: Proofs of propositions (section 4)
We consider first the interaction of share prices with the credit channel of the
economy by keeping capital gains expectation at their steady state value.
Y
y
((a y 1)Y
L [bl (il (Y ) ilo ) be (
pe
e
e
([ f e (
r (Y ) e
reo )]L
pe
r (Y )
)( M
pe
D2
pe E )
pe Eb
pe Ec ] E
pe )
The Jacobian is in this case characterized by
y
Jo
(bl il
e
(a y 1)
be
r
) Lo
pe
Whn r
e fe
pe
0
0
This gives for the determinant of J o
a
aE
y l
y e
0
0
be
e
r
Lo
pe2
Whn r
1 fe ]
e [ fe
pe2
132
(a y 1)
r
(bl il be ) Lo
pe
a
y
Jo
e
y
e
e
y
e
0
r
f e Whn
pe
0
0
1
r
pe
0
r
f e Whn
pe
0
Lo al [[
which gives for J o
a3
l
b
be
Whn r
1 fe ]
e [ fe
pe2
al
0
be
L a bl il
e
e
r
pe
Lo bl il
e o l
e
y e
0
ay 1
y
aE
r
be 2 Lo
pe
y l
Whn r
f
e e
pe
DECEMBER 2010
ae E
be
r
pe2
r
f Whn 1 f e
2 e
pe
0
be
r
pe2
r
f e Whn (1 f e )
pe2
r
f e Whn 1 f e ][bl il
pe2
be
r
]
pe
r
r
f e Whnbe 2 ]
pe
pe
0 the parameter relationship:
f e Whn pr2
r
e
[1
]be
n r
il pe
[ f e Wh p2 1 f e ]
e
r
il pe
For values of bl below this value we have a positive determinant and thus the
instability of the steady state of the dynamics. Note that this steady state is uniquely
determined and given by
VOL.7 NO.2
A peo E
1 ay
Yo
Lo
f e (r (Yo ) peo )( M
D2
peo E )
133
peo E
Note also that we have assumed for the functions il r that there holds
il (Yo )
r (Yo ) peo
in the steady state.
The determinant of J o is given by
Jo
y
be
e
r
]
pe
e
r
f Whn 1 f e ][bl il
2 e
pe
Lo al [[
r
r
f e Whnbe 2 ]
pe
pe
a3
and the trace by
trace J o
1)
y (a y
e [ fe
e
Whn r
1 fe ]
pe2
a1
For the sum of the minors of order two we get:
a2
y (1 a y )
a1a2 a3
y
Whn r
1 fe ]
pe2
be
r
]
pe
(1 a y )
y l
y al Lo [bl il
b
e
e [ fe
y ae E
a [bl il
e
e fe
Whn r
pe
r
]
pe
be
r r
f eWhn
2
pe pe
This gives, when solved under the critical stability condition b a1a2 a3
( J11 J 33 ) J 2
a Lo
y l
b
e
e e
0 the
expression:
b
l
b
if
e
r
be
il pe
( J11
J 33 ) J 2
il
a Lo
y l
y
e
(1 a y )
r r
e e p 2 pe
e
b
a
y l
f e Whn
r
il pe
sufficiently large. The opposite holds true if this parameter is chosen
sufficiently small. Compare to
134
a3
l
b
DECEMBER 2010
f e Whn pr2
r
e
[1
]be
n r
il pe
[ f e Wh p2 1 f e ]
e
For values of bl below this critical value we have a negative a1a2
a3 expression
and thus the instability of the steady state of the dynamics. It is obvious that the
conditions
a3
0 a1a2 a3
0
imply
a2
since a1
0
0 holds true, so that stability will be given for all bl
max{bla3 blb }
while there is instability below this maximum.
There may be a Minsky moment present in this type of an economy whereby the
parameter be is increasing relative to bl over time, since equity markets become
more and more the focus of interest of banks in relatively prosperous and tranquil
phases of economic evolution. The economy may therefore become more and more
fragile and volatile over time.
VOL.7 NO.2
135
Notes
1
Peter Flaschel is Professor Emeritus of Economic Theory at Bielefeld University,
Germany (email: [email protected]), Florian Hartmann is Research
Assistant at the Institute of Empirical Economic Research, University of Osnabrück,
Germany (email: [email protected]), Christopher Malikane is Associate Professor of
Economics at the Univerity of the Witwatersrand (University of Johannesburg),
South Africa (email: [email protected]) and Willi Semmler is
Professor of Economics at the New School for Social Research, New York (email:
[email protected]).
We have to thank Karl Betz, Martin Ehret, Jan Priewe and Peter Spahn for helpful
comments during the time the paper took shape. Of course, the usual caveats apply.
2
See Adrian et al. (2010), Brunnermeier and Sannikov (2010), Gorton (2009, 2010),
and Shleifer and Vishny (2010).
3
See Reinhard and Rogoff (2009) and Gorton (2009, 2010)
4
For a details of such an evaluation, see Brunnermeier and Sannikov (2010).
5
Note we will see in our model that as banks go into capital assets they reduce the
loan supply. One might argue that empirically one might observe a comovement of
credit expansion and rising asset or equity prices. We will come back to this issue at
the end of the paper.
6
The role of equities to act as collateral or bank capital is neglected in this paper.
7
Total savings are Sh S f Sb (Yh C ) ( f I )
L This gives after
some restructuring of such expressions the consistency result that total savings equal
total investment if and only if there is flow consistency on the equity market.
8
Significantly more elaborate versions of the dynamics of the financial sector (and
also of the real sector) are provided in Asada et al. (2010a,b,2011), there however on
the basis of Tobin‘s portfolio equilibrium approach in place of the delayed
disequilibrium adjustment processes we consider in the present section.
9
Since households are ultimately receiving – by assumption – all dividend payments,
we use only an aggregate excess demand function as driving the price of stock and
reserve a detailed treatment of the distribution of stocks and its implications for a
later extension of the model.
10
Note here that banks (including the central bank) are assumed in this paper of only
adjusting their equity stock by way of time derivatives, that is not instantaneously.
11
We remark that capital gains are only realized when equities are moving between
the three sectors of the economy.
12
a2 J1 J 2 J 3 the sum of the three principal minors of order 2 of the matrix
J The index in these minors shows the index of the excluded rows and columns.
13
Since the time-deposit multiplier is then given by 0.5 / (1-0.5) = 1.
Changes in stocks are again excluded from consideration.
15
For details of this and the subsequent points, see Semmler (2011)
14
136
DECEMBER 2010
References
Adrian, T., Moench, E. and Shin, S.S., (2010), ―Macro risk premium and
intermediary balance sheet quantities‖, Federal Reserve Bank of New York Staff
Reports, no. 428.
Asada, T., Chiarella, C., Flaschel, P., Mouakil, T., Proaño, C. and Semmler, W.,
(2010a), ―Stabilizing an Unstable Economy: On the Choice of Proper Policy
Measures‖, Economics, The Open-Access, Open-Assessment E-journal, 3, 201021,
July
16,
2010:
http://www.economics-ejournal.org/economics/
journalarticles/2010-21.
Asada, T., Flaschel, P. Mouakil, T. and Proaño, C.R., (2010b), Macroeconomic
Activity, Asset Accumulation and Portfolio Choice: A Keynesian Perspective,
Basingstoke, Hampshire: Palgrave / Macmillan , forthcoming.
Asada, T., Chiarella, C., Flaschel, P., Mouakil, T., Proaño, C. and Semmler, W.,
(2011), ―Stock Flow Interactions and Disequilibrium Macroeconomics: The
Role of Economic Policy‖, Journal of Economic Surveys, doi: 10.1111/j.14676419.2010.00661.x
Bernanke, B., Gertler, M. and Gilchrist, S., (1999), ―The financial accelerator in a
quantitative business cycle framework‖, in Taylor, J.B. and Woodford, M.,
(eds.) Handbook of Macroeconomics, Vol. 15, Amsterdam: North-Holland,
1341–1393.
Brunnermeier, M. and Sannikov, Y., (2010), ―A macroeconomic model with a
financial sector‖, Working Paper , Princeton University.
Chiarella, C., Flaschel, P., Proaño, C. and Semmler, W., (2010), Income
Distribution, Portfolio Choice and Asset Accumulation. Tobin’s Legacy
Continued, Book Manuscript, New School University, New York.
De Grauwe, P., (2008), ―Returning to narrow banking‖, Center for European Policy
Studies: CEPS Commentary (14.November, 2008).
Fisher, I., (1935), 100 % - Money. New York: Adelphi.
Freixas, X. and Rochet, J.-C., (2008), Microeconomics of Banking,Cambridge, MA:
The MIT Press.
Gande, A., (2008), ―Commercial Banks in Investment Banking‖, in Thakor, A. and
Boot, A. (eds.) Handbook of Financial Intermediation and Banking,
Amsterdam: North-Holland, 163-188.
Gorton, G.B., (2009), Slapped in the Face by the Invisible Hand: Banking and the
Panic of 2007, New York: Oxford University Press.
Gorton, G.B., (2010), ―Questions and Answers About the Financial Crisis‖, NBER
Working Paper No. w15787
Kaufman, H., (2009), The Road to Financial Reform, New York: Wiley
Kindleberger, C.P. and Aliber, R.Z., (2005), Manias, Panics, and Crashes: A History
VOL.7 NO.2
137
of Financial Crises Hoboken, New Jersey: Jon Wiley & Sons.
Minsky, H., (1982), Can it Happen Again? Essays on Instability and Finance. New
York: M.E. Sharpe.
Minsky, H., (1986), Stabilizing an Unstable economy, New Haven, Conn.: Yale
University Press.
Puri, M., (1996), ―Commercial Banks in Investment Banking: Conflict of Interest or
Certification Role?‖, Journal of Financial Economics, 40 (March) 373-401.
Rochet, J.-C., (2008), Why are there so many Banking Crises? The Politics and
Policy of Bank Regulation, Princeton: Princeton University Press.
Semmler, W., (2011), Asset Prices, Booms and Recessions, 3rd edition, Berlin:
Springer Publishing House.
Shin, B., (2009), ―The Future of Investment Banking‖, Capital Market Weekly,
Korea Capital Market Institute, 1(1).
Shleifer, A. and Vishny, R.W., (2010), ―Unstable Banking‖, Journal of Financial
Economics, 97(3), 306-318.
Sinn, H.-W., (2009), Risk-Taking, Limited Liability and the Banking Crisis. Munich:
CESifo.
Asymmetric Fiscal Dynamics and the Significance
of Fiscal Rules for EMU Public Finances
Panagiotis G. Korliras
Athens University of Economics and Business and Centre
of Planning and Economic Research, Greece
Yannis A. Monogios1
Centre of Planning and Economic Research, Greece
Abstract. Since the onset of the international financial and economic downturn, the
issue of the sustainability of public finances has strongly repositioned itself at the
center of economic policy debates, as a number of chronic fiscal ailments for many
EMU countries still need to be effectively addressed for they pose new challenges
for future economic policy. In this work we take stock of the fiscal situation in the
EMU and attempt to dissect some of the most salient aspects of fiscal performance
for a number of member states during the last decade, by concentrating on the
evolution of key fiscal variables in a unified framework of analysis. We focus on the
asymmetric evolution of public finances, in order to assess whether this has been the
outcome of discretionary measures (including the massive stimulus packages as a
response to the crisis) and the effects of automatic stabilizers or also due to poor
fiscal management and lack of fiscal discipline within the existing institutional
framework. The analysis is then taken a step further by estimating the requirements
for long-term sustainability of public finances in the EMU member states in our
sample. Within this context, the issue of the effectiveness of existing rules and
institutions is addressed, as institutional arrangements are crucially related to fiscal
performance. We conclude by highlighting the importance of adopting a cohesive
and enabling framework for public finance stabilization and adjustment purposes.
Our view is that since the root causes of fiscal problems in the Eurozone are diverse,
responding to the diversity of fiscal challenges in a symmetric way i.e. applying a
„one-size fits all‟ EMU strategy, may prove sub-optimal for individual country fiscal
sustainability objectives. In fact, fiscal policy responses will inevitably be
asymmetric. This asymmetric evolution of public finances puts into question the
sustainability objective and thus convergence of the EMU as a whole.
JEL Classification: H30, H6, H87
Key Words: Fiscal Policy, Debt Dynamics, EMU
139
140
DECEMBER 2010
1. Introduction
The recent financial and economic crisis which led to an abrupt reversal of the
favorable economic and financial conditions that prevailed until 2007 represents a
symmetric shock with asymmetric implications for the EMU economies, whose
public finances were hit disproportionately. In addition to the negative crisis impulse,
internal and external imbalances exacerbated the cyclical swings in fiscal positions.
To make matters worse, perceived budgetary and macro-financial imbalances
surfaced in weaker economies catapulting as a result, sovereign risk premia. This led
to ongoing sovereign debt-tensions; a fact that serves as a constant reminder that
fiscal developments are under continuous markets’ watch. Arguably, also due to
different shock-absorbing capacities, individual country responses to the crisis were
largely asymmetric.
As a consequence, public finances (government debt and budget deficits)
have deteriorated sharply and virtually everywhere in the EMU. Public indebtedness
and budget deficits are expected to rise even further amidst subdued growth
projections (EC 2010, [2], [3]).
The rapid weakening in fiscal positions has to be attributed to reasons other
than the effects of the economic cycle and the workings of ‘automatic stabilizers’ or
the fiscal stimulus interventions alone. Weak initial structural positions prior to the
crisis combined with lack of appropriate fiscal frameworks (fiscal rules, institutions,
and budgetary processes) in some countries, have also contributed to the pronounced
fiscal slippage.
Under the auspices of EERP (European Economic Recovery Plan, launched
in December 2008) most EMU countries introduced discretionary support packages
(consisting mostly of expenditures increases and support funds) to mitigate the
negative effects of the downturn (EC 2010, [3]). In that juncture, the effects of the
automatic stabilizers account for the smaller fraction of the total size of the recorded
deficits. The remaining is attributed to the adoption of discretionary policies.
This upward trend in fiscal dynamics raises serious concerns and casts
considerable doubts for the sustainability of the EMU public finances in the medium
to long run. For some EMU member states, sizeable fiscal consolidation efforts are a
sine qua non in order to reverse the rising trajectory of debts and deficits and put
their public finances back to a sustainable path. The adjustment path to a new steadystate will be (time and pace) asymmetric among EMU economies.
In Section 2 of this article, by means of a comparative analysis, we evaluate
fiscal performance in a sample of twelve EMU countries for the period 2000-2010
based on their fiscal record (public debt and budget deficits). For this purpose we
examine the observed trends in fiscal dynamics and we decompose overall budget
balances into cycle-induced and policy-led components to get a clear idea of the
factors behind fiscal developments. As the observed fiscal deterioration raises
concerns about the sustainability of the public debt/GDP ratio for many EMU
VOL.7 NO.2 KORLIRAS- MONOGIOS: ASYMMETRIC FISCAL DYNAMICS
141
economies, we then assess EMU fiscal sustainability in a partial equilibrium
framework of analysis.
Next, in Sections 3 and 4, we built upon the previous analysis to explore
further the argument that in the presence of appropriate fiscal frameworks (with
emphasis on numerical fiscal rules) budgetary outcomes are improved. Our analysis
provides support to the above argument, suggesting the crucial importance of
institutional arrangements, such as integrated fiscal frameworks, for the cohesion and
sustainability of EMU public finances. Section 5 concludes.
2.
Asymmetric Fiscal Dynamics in Selected EMU Countries
2.1
Evolution of key fiscal variables in the EMU
In this section we present in three Tables the evolution of the basic macroeconomic
variable (real GDP, Table 1) along with the two key fiscal variables i.e. fiscal deficits
and public debt for twelve selected EMU member states 2 (Tables 2 and 3). These
tables are the raw material and background for the ensuing analysis, exhibiting prima
vista the asymmetries in fiscal performance during the last ten years, which is the
main subject of the analysis.
During the 2000-2010 period real GDP in the EU recorded an average growth
of 1.56 percentage points (1.38% in the EMU), with some countries outperforming,
on average almost six-fold vis-à-vis the least good performers (e.g. IE =3.51%
versus IT = 0.59%). The period from 2000 to 2008 is characterized by positive rates
of growth for all member states.
Nonetheless, in 2009 a broad-based and steep recession took a heavy toll
predominantly on growth. All EMU countries experienced a severe downturn in real
growth ranging from -2.3% (EL) to -8.0% (FI) in 2009. Member states were hit
disproportionately (in 2009 for instance growth slumped markedly in IE by -7.6%
and in IT by -5.0%) for a number of reasons but mainly as a result of preexisting
weak fundamentals, while some other countries have shown resilience and are now
back on a fast-track to recovery (BE, DE, LU, AT).
However, in 2010 there are signs of improvement, although real GDP is
expected to slow down again mid-2011 onwards (EC 2010 [2]) as Europe enters a
new era of fiscal consolidation. The aggregate picture however, conceals marked
differences in developments across member states.
While the growth outlook remains uncertain, a differentiated pace of recovery
within the EMU seems most likely, reflecting the policy challenges individual
economies face, the most important of which are fiscal sustainability and
competitiveness, especially in some euro-area Member States (such as EL, IE, PT,
ES, IT) that remain under intense market scrutiny.
142
DECEMBER 2010
Table 1. Real GDP growth rate (y-o-y % change)
EU 27
EMU 16
Belgium
Germany
Ireland
Greece
Spain
France
Italy
Luxembourg
Netherlands
Austria
Portugal
Finland
2000 2001
3.9
2.0
3.9
1.9
3.7
0.8
3.2
1.2
9.7
5.7
4.5
4.2
5.0
3.6
3.9
1.9
3.7
1.8
8.4
2.5
3.9
1.9
3.7
0.5
3.9
2.0
5.3
2.3
2002
1.2
0.9
1.4
0.0
6.5
3.4
2.7
1.0
0.5
4.1
0.1
1.6
0.7
1.8
2003
1.3
0.8
0.8
-0.2
4.4
5.9
3.1
1.1
0.0
1.5
0.3
0.8
-0.9
2.0
Source: AMECO-EUROSTAT - e Estimates for 2010,
f
2004
2.5
2.2
3.2
1.2
4.6
4.4
3.3
2.5
1.5
4.4
2.2
2.5
1.6
4.1
2005
2.0
1.7
1.8
0.8
6.0
2.3
3.6
1.9
0.7
5.4
2.0
2.5
0.8
2.9
Forecasts for 2011
2006
3.2
3.0
2.7
3.4
5.3
4.5
4.0
2.2
2.0
5.0
3.4
3.6
1.4
4.4
2007
3.0
2.8
2.9
2.7
5.6
4.3
3.6
2.4
1.5
6.6
3.9
3.7
2.4
5.3
2008 2009
0.5
-4.2
0.4
-4.1
1.0
-2.8
1.0
-4.7
-3.5
-7.6
1.3
-2.3
0.9
-3.7
0.2
-2.6
-1.3
-5.0
1.4
-3.7
1.9
-3.9
2.2
-3.9
0.0
-2.6
0.9
-8.0
2010 e 2011 f
1.8
1.7
1.7
1.5
2.0
1.8
3.7
2.2
-0.2
0.9
-4.2
-3.0
-0.2
0.7
1.6
1.6
1.1
1.1
3.2
2.8
1.7
1.5
2.0
1.7
1.3
-1.0
2.9
2.9
143
Table 2. General Government deficit/surplus (% of GDP)
EU 27
EMU 16
Belgium
Germany
Ireland
Greece
Spain
France
Italy
Luxembourg
Netherlands
Austria
Portugal
Finland
2000
0.6
0
0
1.3
4.8
-3.7
-1.0
-1.5
-0.8
6.0
2.0
-1.7
-2.9
6.8
2001
-1.4
-1.9
0.4
-2.8
0.9
-4.5
-0.6
-1.5
-3.1
6.1
-0.2
0
-4.3
5.0
2002 2003
-2.5 -3.1
-2.6 -3.1
-0.1 -0.1
-3.7 -4.0
-0.3
0.4
-4.8 -5.6
-0.5 -0.2
-3.1 -4.1
-2.9 -3.5
2.1
0.5
-2.1 -3.1
-0.7 -1.4
-2.8 -2.9
4.0
2.4
2004
-2.9
-2.9
-0.3
-3.8
1.4
-7.5
-0.3
-3.6
-3.5
-1.1
-1.7
-4.4
-3.4
2.3
2005
-2.5
-2.5
-2.7
-3.3
1.6
-5.2
1.0
-2.9
-4.3
0
-0.3
-1.7
-6.1
2.7
2006
-1.5
-1.3
0.2
-1.6
2.9
-5.7
2.0
-2.3
-3.4
1.4
0.5
-1.5
-4.1
4.0
2007
-0.8
-0.6
-0.3
0.3
0
-6.4
1.9
-2.7
-1.5
3.7
0.2
-0.4
-2.8
5.2
2008
-2.3
-2.0
-1.3
0.1
-7.3
-9.4
-4.2
-3.3
-2.7
3.0
0.6
-0.5
-2.9
4.2
Source: AMECO-EUROSTAT
Net lending (+)/Net borrowing (-) under the EDP (Excessive Deficit Procedure), e Estimates for 2010,
2009
-6.8
-6.3
-6.0
-3.0
-14.4
-15.4
-11.1
-7.5
-5.3
-0.7
-5.4
-3.5
-9.3
-2.5
f
2010 e 2011 f
-7.2
-6.5
-6.6
-6.1
-4.8
-4.6
-3.7
-2.7
-32.3 -10.3
-9.6
-7.4
-9.3
-6.4
-7.7
-6.3
-5.0
-4.3
-1.8
-1.3
-5.8
-3.9
-4.3
-3.6
-7.3
-4.9
-3.1
-1.6
Forecasts for 2011
144
DECEMBER 2010
The EU headline budget deficit in 2007 from less than 1% climbed to almost
7% of GDP in 2009 (a similar trend is observed for the EMU countries over the same
time span). Even though fiscal positions have deteriorated virtually everywhere in
the EMU (and in the EU) the distribution of the increases in fiscal deficits, however,
is uneven. Some countries in particular, experienced quite dramatic budgetary
developments. IE, ES and EL posted double digit deficits (in excess of 10 percentage
points of GDP) in 2009.
Remarkably in the same year ten out of the twelve EMU countries in our
sample (BE, DE, IE, EL, ES, FR, IT, NL, AT and PT) were placed by the European
Council (on recommendation of the Commission) in the Excessive Deficit Procedure
(EDP). The negative developments in the EU (and the EMU) budgetary positions are
expected to continue in 2010, as all EMU countries (except LU) in our sample are
expected to breach the Maastricht deficit/GDP reference value, of -3.0%.
The sharp deterioration in the budget positions as a result of the systemic
crisis led to an upswing in the debt/GDP ratio as well, for all the EU/EMU member
states (Table 3).
Between 2007 and 2010 the EMU consolidated gross debt of the general
government increased by almost 18 percentage points (by more than 20% in the EU).
However, there is considerable variation in debt dynamics in individual countries, as
the debt distribution among the EMU is profoundly asymmetric. In many cases in
our sample the debt/GDP ratio increased dramatically in the reference period by
more than 30 percentage points (IE, EL, ES, FR, LU, NL, PT, FI), while in some
other the rise in the debt/GDP ratio was less than 20% (BE, DE, IT, AT).
Against the backdrop of high primary budget deficits and growing interest
expenditures, the debt/GDP ratio is projected to remain on a rising course in 2011
and beyond (EC 2010, [2]). Calculations based on partial equilibrium debt
projections, suggest that by 2015 the average EU debt will exceed 100% of GDP
(assuming a no policy change scenario), and will remain on a rising trajectory
thereafter, skyrocketing to levels well in excess of 130% of GDP by 2020 (EC 2010,
[3], Deutsche Bank 2010).
As a consequence, widespread concerns about the long-run sustainability of
public finances in some countries (e.g. IE, EL, PT, ES) caused turbulence in
sovereign bond markets which led to sharp increases in government bond yields and
relevant risk premia. Apart from growth concerns, the issue of sustainability of
public finances with all its ramifications occupies now the centre stage in the
ongoing discussions regarding the future of the Eurozone.
2.2
Cyclical versus discretionary effects: an assessment
The deterioration in fiscal positions discussed above (i.e. budget deficits and public
debt) has been partly attributed to cyclical factors i.e. the normal operation of the
automatic stabilizers (EC 2009, [9]). However, once the economy returns to recovery
this effect should be reversed, although in the aftermath of the crisis the echoing
145
Table 3. General Government Consolidated Gross Debt (% of GDP)
2000
61.9
EU 27
69.2
EMU 16
107.9
Belgium
59.7
Germany
37.8
Ieland
103.4
Greece
59.3
Spain
57.3
France
109.2
Italy
6.2
Luxembourg
53.8
Netherlands
66.5
Austria
50.5
Portugal
43.8
Finland
2001
61.0
68.2
106.6
58.8
35.6
103.7
55.5
56.9
108.8
6.3
50.7
67.1
52.9
42.5
2002
60.4
68.0
103.5
60.4
32.2
101.7
52.5
58.8
105.7
6.3
50.5
66.5
55.6
41.5
2003
61.9
69.1
98.5
63.9
31.0
97.4
48.7
62.9
104.4
6.1
52.0
65.5
56.9
44.5
2004
62.2
69.5
94.2
65.8
29.7
98.6
46.2
64.9
103.8
6.3
52.4
64.8
58.3
44.4
Source: AMECO-EUROSTAT - e Estimates for 2010, f Forecasts for 2011
2005
62.8
70.1
92.1
68.0
27.4
100
43.0
66.4
105.8
6.1
51.8
63.9
63.6
41.7
2006
61.5
68.6
88.1
67.6
24.8
106.1
39.6
63.7
106.6
6.7
47.4
62.1
63.9
39.7
2007
58.8
66.2
84.2
64.9
25.0
105.5
36.1
63.8
103.6
6.7
45.3
59.3
62.7
35.2
2008
61.8
69.8
89.6
66.3
44.3
110.3
39.8
67.5
106.3
13.6
58.2
62.5
65.3
34.1
2009
74.0
79.2
96.2
73.4
65.5
126.8
53.2
78.1
116.0
14.5
60.8
67.5
76.1
43.8
2010e
79.1
84.1
98.6
75.7
97.4
140.2
64.4
83.0
118.9
18.2
64.8
70.4
82.8
49.0
2011f
81.8
86.5
100.5
75.9
107.0
150.2
69.7
86.8
120.2
19.6
66.6
72.0
88.8
51.1
146
DECEMBER 2010
negative effects on potential growth could add further stress on public finances (EC
2009, [9], [10]).
However, for some countries the deterioration in underlying fiscal positions
dates back to well before the crisis. In many countries, tax buoyancy, low interest
rates, rapid credit growth and asset price booms had all led to improvements in fiscal
circumstances in the years prior the crisis (EC 2010, [3]), a fact that partly masked
the gradual erosion in underlying positions. The burst, however, of the crisis revealed
the true state of public finances in many countries. Once the crisis unfolded, tax
revenues fell precipitously, leading public finances to rapid deterioration. On the
other hand, crisis-evoked stimulus measures also added to the free-fall in fiscal
positions. In addition, most EU member states implemented counter-cyclical fiscal
policies under the common framework provided by the European Economic
Recovery Plan (EERP)3, giving in this way aggregate demand a ‘Keynesian boost’
(on average expansionary stimulus in 2009 reached 1.5% of GDP and 1.4% of GDP
in 2010 in the EU). Given the tight credit conditions prevailing at the time and the
fact that monetary policy was constrained by the zero lower bound on nominal
interest rates, the choice of a discretionary fiscal policy was deemed more
appropriate in order to address the crisis, but also in view of the fact that the
workings of automatic stabilizers were considered insufficient to mitigate the
deterioration in demand conditions.
In light of the aforementioned discussion on fiscal developments, the aim of
the analysis put forth in this section is to evaluate the fiscal performance of the EMU
countries in our sample for the period 2000-2009. In order to do so, we concentrate
on the budget in order to first decompose the observed changes in the overall fiscal
balances into ‘automatic’ and ‘discretionary’ effects 4. We then proceed to examine
the changes in the ‘automatic’ and the ‘discretionary’ components. Table 4
summarizes the evolution of the main components of Total Fiscal Balance ( TFB ) in
the EMU member states, while Table 5 presents the changes in these components for
the periods 2005-2007 and 2007-2009, i.e. the two years preceding and respectively
following the beginning of the crisis in 2007.
Arguably, a more precise measure to assess the underlying fiscal stance in the
EMU countries during the last decade is the Cyclically Adjusted Balance ( CAB )
(also known as the ‘structural budget balance) 5, as it removes transitory elements
from Total Fiscal Balance ( TFB ) figures and is presumably less volatile.
We take the CAB as the more appropriate indicator in assessing the SGP
(Stability and Growth Pact) Medium Term Budgetary Objectives ( MTBO ), and thus
for rules-based fiscal surveillance. Governments aiming at a stable cyclicallyadjusted balance however, they do so in recognition of fully operational automatic
stabilizers.
147
TFB 6 is decomposed to a) ‘automatic’ responses of fiscal variables to output
changes CB (Cyclical Balance) and b) reaction of fiscal variables to changes in
discretionary policy, CAB (Cyclically Adjusted Balance):
so that:
TFB  CB  CAB
(1)
TFB  CB  CAB
(2)
We compare the pre-crisis 2005-2007 with the post-crisis period 2007-2009,
when the average growth real GDP rates were (3.34%) and (-0.03%) respectively, in
order to assess fiscal stance in the twelve EMU countries in the sample.
Decomposing fiscal balances into their constituent components we observe that:
During the pre-crisis 2005-2007 period, all EMU countries (except IE, EL)
improved their total fiscal balance positions ( TFB  0 ). In four countries (BE, IT,
LU, PT) the improvement is attributed more to the discretionary than to the cyclical
effect on TFB (i.e. CAB  0 ), while in four countries (DE, ES, FR, and FI) the
improvement in TFB was mainly due to the workings of ‘automatic stabilizers’
( CB  CAB  0 ). Within the same period in four countries (IE, EL, NL, AT)
discretionary policy had a negative impact ( CAB  0 ) on TFB in the sense that it
reduced the positive effect of the cyclical stabilizers ( CB  0 ), while maintaining
TFB positive in two cases (NL, AT), and in two other cases (IE, EL) it even
reversed TFB to negative (in absolute terms CAB  CB ).
During the post-crisis 2007-2009 period, in all EMU countries we observe a
significant deterioration in their budget positions ( TFB  0 ). In six countries (DE,
FR, IT, NL, AT, FI) the deterioration was mainly due to the contribution of the
negative cyclical effect on the budget (in absolute terms CB  CAB ), although in
LU, CAB had no discretionary policy impact at all ( CAB  0 ). In five countries
(IE, EL, ES, PT and BE) the deterioration of fiscal balances was mainly due to the
negative effect of expansionary discretionary fiscal policy (in absolute terms
CAB  CB ).
The preceding presentation points to a number of interesting asymmetries in
the conduct of fiscal policy among the EMU countries: In six countries (BE, DE, ES,
IT, PT, FI), there was a uniform change in fiscal stance (regime switch) between the
two periods from a contractionary policy regime ( CAB  0 ) before the crisis, to an
expansionary one ( CAB  0 ), after the crisis. FR with CAB  0 in the first
period, engaged in an expansionary policy regime in the second period ( CAB  0 ),
whereas LU from CAB  0 in the first period switched to a zero – i.e. policy
neutral - ( CAB  0 ) in the second period. This is clearly attributed to the economic
downturn. On the other hand, four countries (IE, EL, NL, AT), maintained their
expansionary fiscal stance ( CAB  0 ) in both periods, which implies that during
the crisis those countries engaged in an increasingly expansionary policy, adding in
this way to the deterioration of fiscal outcomes.
148
DECEMBER 2010
Table 4. Decomposition of Total Fiscal Balance in the EMU (♦)
Total Fiscal Balance (TFB )
Cyclical Balance (CB )
Cyclically Adjusted Balance (CAB )
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010* 2011* 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010* 2011* 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010* 2011*
BE
-0.1
0.4 -0.2 -0.1 -0.3 -2.8
0.4
0.9
1.3
0.8 -1.4
-1.1
-0.9 -1.0
DE
1.3 -2.8 -3.7 -4.0 -3.8 -3.3 -1.6
0.3
IE
4.8
0.6
0.0 -0.6 -0.9 -1.0
0.1
0.9
0.8 -2.1
-0.9
1.1
1.2
0.7 -0.3
0.1
0.7
1.7
0.0 -2.6
-2.1
EL
-3.7 -4.5 -4.8 -5.6 -7.5 -5.2 -5.7 -6.4 -9.4 -15.4
ES
-1.0 -0.7 -0.5 -0.2 -0.3
1.9 -4.2 -11.1
-9.6
-7.4 -0.5 -0.3 -0.5
0.4 -15.5
0.0
0.7
1.3
1.1 -0.2
-2.2
-3.3 -3.2 -4.2 -4.3 -6.0
-7.4
-4.1
-9.3
-6.4
0.6
0.5
0.1 -0.1
0.1
0.1
0.4
0.6
0.0 -1.9
-1.9
-1.5 -1.6 -1.1 -0.6 -0.1 -0.4
1.3
-4.2
-9.2
-7.4
-4.9
FR
IT
-1.5 -1.6 -3.1 -4.1 -3.6 -2.9 -2.3 -2.7 -3.3 -7.5
-7.7
-6.3
1.0
0.9
0.5
0.2
0.9
0.8
0.9
1.0
0.2 -1.7
-1.7
-1.7 -2.4 -2.4 -3.6 -4.3 -4.5 -3.7 -3.2 -3.7
-3.5
-5.8
-6.0
-4.6
-0.9 -3.1 -3.0 -3.5 -3.5 -4.3 -3.4 -1.5 -2.7 -5.3
-5.0
-4.3
0.5
0.8
0.5
0.0
0.2
0.3
1.0
1.5
0.6 -1.8
-1.3
-0.8 -1.3 -3.9 -3.4 -3.5 -3.7 -4.6 -4.4 -3.0
-3.3
-3.5
-3.7
-3.5
LU
6.0
0.0
1.4
3.7
3.0 -0.7
-1.8
-1.3
1.9
0.8
0.6 -0.7
0.2
0.6
1.0
2.0
0.7 -2.4
-2.2
-2.1
4.1
1.7
2.3
1.7
0.4
0.8
NL
2.0 -0.3 -2.1 -3.1 -1.7 -0.3
0.5
0.2
0.6 -5.4
-5.8
-3.9
1.3 -0.8
1.4 -1.2 -0.9 -0.8 -0.1
1.0
0.9 -2.0
-1.7
-1.6
0.7 -1.0 -1.7 -1.9 -0.8
0.6 -0.8
-0.3
-3.4
-4.1
-2.3
AT
-1.7
0.0 -0.7 -1.4 -4.4 -1.7 -1.5 -0.4 -0.5 -3.5
-4.3
-3.6
1.1
0.1 -0.1 -1.8 -0.6 -0.5
0.3
1.2
1.3 -1.2
-0.9
-0.7 -2.7 -0.1 -0.5 -0.6 -3.8 -1.2 -1.8 -1.6
-1.8
-2.3
-3.4
-2.9
PT
-3.0 -4.3 -2.9 -2.9 -3.4 -5.9 -4.1 -2.8 -2.9 -9.3
-7.3
-4.9
1.0
0.9
0.4 -0.6 -0.5 -0.5 -0.4
0.3
0.1 -1.1
-0.6
-1.1 -4.0 -5.2 -3.2 -2.3 -2.9 -5.4 -3.7 -3.1
-3.0
-8.2
-6.7
-3.8
FI
6.8
-3.1
-1.6
1.3
0.5 -0.3 -0.7
2.8
1.8 -3.1
-2.5
-2.0
2.4
0.6
-0.6
0.4
0.9 -0.3
6.1
5.0
2.1
4.0
0.4
1.4
0.5 -1.1
2.4
2.3
1.6
1.0
2.7
0.2 -0.3 -1.3 -6.0
2.9
2.0
4.0
-4.8
-4.6
1.0
0.2 -1.0 -0.6
-3.7
-2.7
0.7
0.0 -7.3 -14.4 -32.3 -10.3
1.7
5.2
0.1 -3.0
4.2 -2.5
0.4
0.6
0.6
Source: EC-DG ECFIN Cyclical Adjustment of Fiscal Balances, Autumn 2010
1.5
(♦)
0.5 -0.7 -3.2 -0.7 -1.6
-2.1
-4.6
-3.7
-3.7
-0.5
0.6 -3.5 -3.7 -3.4 -2.9 -2.3 -1.7 -0.6
-0.7
-0.9
-2.8
-2.2
-1.2
3.0 -0.2 -1.5 -0.3
-7.3 -11.8 -30.2
-9.1
5.5
0.2
5.3
4.5
0.1
1.5
4.2
1.7
1.5
8.0 -5.2 -6.4 -7.7 -10.5 -15.2
0.9
1.2 -1.3 -0.6
3.2
2.2 -1.7
1.7
0.5
2.1
1.6
0.4
2.5
2.4
in % of GDP, * Commission forecasts 2010-2011
149
Table 5. Decomposition of changes in Total Fiscal Balance in the EMU
Change in Total Fiscal Balance (ΔTFB )*
Belgium
Germany
Ireland
Greece
Spain
France
Italy
Luxembourg
Netherlands
Austria
Portugal
Finland
Change in Cyclical Balance (ΔCB )*
Change in Cyclically Adjusted Balance (ΔCAB )*
ΔTFB 07-05
ΔTFB 09-07
ΔCB 07-05
ΔCB 09-07
ΔCAB 07-05
ΔCAB 09-07
2.5
3.6
-1.6
-1.2
0.9
0.2
2.8
3.7
0.5
1.3
3.1
2.5
-5.7
-3.3
-14.4
-9.0
-13
-4.8
-3.8
-4.4
-5.6
-3.1
-6.5
-7.7
0.9
1.9
1.6
1.3
0.5
0.2
1.2
1.4
1.8
1.7
0.8
2.2
-2.7
-3
-4.3
-1.5
-2.5
-2.7
-3.3
-4.4
-3.0
-2.4
-1.4
-5.9
1.6
1.7
-3.2
-2.5
0.4
0
1.6
2.3
-1.3
-0.4
2.3
0.3
-3.0
-0.3
-10.1
-7.5
-10.5
-2.1
-0.5
0
-2.6
-0.7
-5.1
-1.8
Authors' calculations, *Changes over referen e year figures.
150
DECEMBER 2010
The analysis of fiscal responses also indicates asymmetries in the contribution
of discretionary versus the cyclical component in the composition of the overall
fiscal outcomes: in five countries (DE, NL, FR, AT, FI) in both periods, the cyclical
component is consistently greater than the discretionary component (in absolute
terms CB  CAB ), indicating strong cycle effects. In contrast, in four countries
(BE, IE, EL, PT,) in both periods, the discretionary component was greater than the
cyclical component (in absolute terms CB  CAB ), indicating thus the
dominance of discretionary policy. However, in one country (ES) the combination of
these components changed between the first and the second period and turned from a
dominant cyclical component in the first period (in absolute terms CB  CAB ) to
a dominant discretionary component in the second period (in absolute terms
CB  CAB ), while in two countries (IT, LU) that change run in the opposite
direction; from a dominant discretionary to a dominant cyclical effect.
The significant deterioration in fiscal outcomes in virtually all EMU countries
raises serious questions about the sustainability of public finances. This apparent
deterioration in budget terms, in part as a consequence of the crisis, but also as a
consequence of the lack of fiscal discipline and thus poor fiscal performance - in
some EMU countries- in the pre-crisis period, has led to current conditions which
necessarily differentiate the path of adjustment among these countries. As mentioned
before, according to the Commission forecasts, in 2010 all EMU countries will be
below the -3% Maastricht threshold in their total fiscal balance (except LU), as well
as in their Cyclically Adjusted Balance (except LU, FI and DE). For those countries
whose projected cyclically adjusted balance in 2010, is above -4% (BE, DE, IT, LU,
NL, AT, FI) the fiscal adjustment appears to be ‘relatively easier’, as opposed to
those countries whose projected cyclically adjusted balance is below -4%, especially
so, for countries with higher structural imbalances (IE, EL, ES, FR and PT). This
differentiation among EMU countries concerning their overall (or structural) deficits,
which implies a varying degree of adjustment, can be examined in conjunction with
the debt/GDP ratios which are also differentiated across the twelve EMU countries
under scrutiny.
The crux of the above analysis was to demonstrate that depending on the
economy’s initial position, available fiscal space and policies pursued, country
responses to the crisis varied widely among EMU member states. And although
policy reactions to the challenges brought forward by the crisis were grosso modo
synchronized, they yielded an interesting mix of budgetary outcomes. All things
considered, the observed deterioration in current EMU fiscal positions can partly be
attributed to cyclical factors and to the effects of the crisis. Past patterns in fiscal
policy and budgetary positions have also played a key role in explaining the rapid
decline in public finances.
2.3. Sustainability analysis of public finances
The size of the fiscal deterioration across the EMU has, inter alia, given rise to
concerns over the sustainability of public finances7. Complementing the previous
151
analysis, the aim of this part of the study is to examine whether (based on current
policies and projected trends in public finances) the debt/GDP ratio for the twelve
EMU countries in our sample, is in a self-sustained path. In order to do so, we resort
to the analysis of the debt dynamics.
The analysis of debt dynamics8 refers to the size of primary balances required
to achieve debt/GDP target levels within specified time horizons. Debt dynamics are
affected by a number of factors (the initial level of debt, the primary balance, the
snow-ball effect - i.e. the interest growth differential - the stock-flow adjustment).
Nonetheless, some parameters (such as the government’s contingent liabilities
stemming from age-related pension entitlement schemes and demographic dynamics)
are expected to add significant pressure to the budget for many EMU countries in the
years to come9 (EC 2009, [7]). Failure to consolidate budgetary positions in the
medium-term can lead the debt/GDP ratio to an explosive path.
The dynamics of debt accumulation for assessing fiscal sustainability can be
summarized in the following formula, known as „the sustainability identity‟ (see for
instance Monogios 1998, Balassone et al. 2002, Sturzenegger 2002, Ley 2009):
 1 r 
d t 1  pbt
d t  
1 g 
where, in time
(3)
t
d = debt level
g = growth rate of real GDP
r = real interest rate
pb = primary balance
By transforming eq. (3) it can be shown that the debt-stabilizing primary
balance pb * is given by:
rg
d 0
pb *  
 1 g 
(4)
where in this case: d 0 = current debt level.
It follows that the debt-reducing primary balance pb ** (to desired debt-target levels)
in the next T periods is given by:
T
 1 r 
  d*
d 0 
1  g 
**

pb 
j
T 1  1  r 


j 0 1  g



where: d * debt-target level.
(5)
152
DECEMBER 2010
Based on the above formulas, the results from the sustainability exercise for
the EMU countries in our sample are presented in Table 6.
This table documents the required fiscal effort, in terms of creation of primary
fiscal surpluses pb, in order to achieve certain debt/GDP target levels d * in specified
time spans (over five or ten years, T  5, 10 starting from 2011 onwards), based on
exogenous assumptions about r , g , in the EMU countries in our sample. Given
initial conditions the aim is, first, to estimate the primary surplus required to stabilize
the debt/GDP ratio in each country to its current (2010) level. We then proceed to
estimate the size of the primary surplus necessary to reduce the debt/GDP ratio to set
target levels. In this case, one option is to consider the pre-crisis 2007 debt/GDP
level as the target, while another is to consider as a debt/GDP target level the
Maastricht benchmark of 60%.
Turning to Table 6, columns (1) and (2) document the pre-crisis 2007 and
post-crisis 2010 debt/GDP ratios for the EMU countries in our sample 10, while
columns (3) and (4) depict the combinations of the –exogenously determinedgrowth rate of GDP and interest rate for the countries under scrutiny 11. On the basis
of adopted assumptions and based on text formula (4), column (5) provides estimates
of the size of the debt-stabilizing primary surplus to levels prevailing in 2010. The
results indicate a moderate variation across the EMU countries’ primary surplus
generation. One should bear in mind however, that the size of the public surplus
required for debt-stabilization purposes, is time-sensitive and relevant to the debttarget pursued. According to our scenarios, sovereign fiscal efforts range from less
than one percent of GDP (BE, ES, FR, NL, AT, FI), to a few percentage points of
GDP (IE, EL, IT). For instance, (AT) has to permanently deliver a primary surplus of
0.20% in order to preserve a debt/GDP ratio of 70%, which is half as much as that of
(EL) of 140% in 2010, and who needs a primary surplus of 2.62% to sustain it. For
debt-ridden governments, engaging in such a fiscal effort would prevent further
increases in the debt/GDP ratio, but most importantly would transmit a positive
signal about fiscal prudency, which is a precondition to restoring fiscal credibility.
Next we discuss the permanent primary fiscal balances required to bring
current (2010) public debt/GDP ratios back to pre-crisis levels (2007) within five or
ten years’ time.
As columns (6) and (7) show, all EMU countries need to achieve primary surpluses
in the order of 3.0% to over 19% of GDP, in order to lower their debt/GDP ratios
down to the pre-crisis 2007 levels in a period of five years. Alternatively, if spread
over ten years, the required adjustment in terms of primary surpluses, is reduced to
half, running the gamut from 1.4% to 9.1% of GDP.
153
Table 6. Debt-target analysis
pb* = Primary Surplus
required to
stabilize the Debt/GDP
d o = Debt as
% of GDP
Country
Growth/Interest rate Scenarios
Growth Rate
Interest Rate
ratio to 2010
level (% of GDP)
pb** = Primary Surplus required to reduce
the Debt/GDP ratio to target levels
(% of GDP)
d* = pre-crisis 2007
debt level
Maastricht level
d* = 60%
2007
2010
g
r
T=5
T=10
T=5
T=10
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Belgium
84.2
98.6
2.1
2.7
0.579
4.27
2.20
10.2
4.81
Germany
64.9
75.7
1.6
3.1
1.118
3.99
2.35
5.17
2.85
Ireland
25.0
97.4
2.0
3.6
1.528
19.3
9.12
10.9
5.53
Greece
105.5
140.2
1.5
3.4
2.624
11.53
6.40
22.4
11.0
Spain
36.1
64.4
1.6
3.1
0.951
8.00
3.97
2.24
1.50
France
63.8
83.0
2.1
2.6
0.406
5.25
2.53
6.19
2.94
Italy
103.6
118.9
1.6
3.1
1.755
5.87
3.52
16.4
8.01
Netherlands
45.3
64.8
1.6
3.0
0.893
5.82
3.01
2.27
1.48
Austria
59.3
70.4
3.0
3.1
0.205
3.01
1.44
2.84
1.37
Portugal
62.7
82.8
1.2
3.3
1.555
6.72
3.74
7.37
4.02
Finland
35.2
49.0
2.0
2.6
0.288
3.76
1.81
do < d*
do < d*
Sources: AMEC0, EUROSTAT - Authors’ calculations
Data in italics: calculations on g are based on extrapolation of projections of WEO (May 2010) to 2020. Calculations for r for the
Netherlands are based on equivalent rates for Italy-Spain-Germany and for Finland on France-Belgium.
154
DECEMBER 2010
However, assessing debt sustainability by targeting the pre-crisis debt levels
may not be a meaningful exercise either because for some countries the difference
between the 2007 and 2010 debt/GDP levels is relatively small (e.g. DE, AT), or
because for some other the pre-crisis stock of debt was already high in 2007 (e.g. EL,
IT). For this purpose we utilize the Maastricht reference value of 60% debt/GDP
ratio as a meaningful benchmark to repeat the calculations. The results out of this
debt sustainability exercise which are based on text equation (5), are presented in
columns (8) and (9) and virtually speak the same language. With the exception of
(FI) which is already on target, as the debt/GDP was less than the Maastricht
threshold in 2010 (FI = 49.0%), all the other EMU countries in the sample would
have to achieve constant annual primary surpluses of 1.3% - 11.0% of GDP over the
next ten years, to reach a public debt to GDP ratio of 60%. If adjustment is envisaged
within a period of five years, then the debt-reducing primary surpluses become
extremely demanding, if not prohibitive for some countries.
The above results convey a pristine message: EMU economies with high
debt/GDP ratios will have to alter their fiscal policy stance substantially by
introducing austere and bold fiscal policy adjustments, if current debt/GDP levels
are to become sustainable in the long run. Moreover, the debt target analysis offers
some tentative insight on potential pressures on sovereign creditworthiness in the
absence of credible consolidation plans. In that sense, these results provide a glimpse
at the future of the Eurozone. The results obtained from the sustainability exercise
performed in this section are broadly consistent with those of others studies
investigating the issue of sustainability (see IMF 2009, [2], Deutsche Bank 2010).
Most EMU countries will need to consistently generate positive and sizeable
primary surpluses for many years, in order to achieve a meaningful decumulation in
their debt/GDP ratios. That of course, presupposes a credible and lasting
commitment by those governments embarking on an aggressive and sustained fiscal
consolidation odyssey.
From the above analysis it is obvious that those countries which are not
currently (2010) far from the target level (e.g. NL, AT, FI), could start fiscal
consolidation with more favorable initial conditions, which allow them to achieve
the target both sooner and with an attainable fiscal effort. On the contrary, countries
with unfavorable initial conditions (BE, EL, IT), face the prospect of an enormous
fiscal adjustment for an extended period of time. However, this higher consolidation
effort entails a serious risk of „fiscal fatigue‟, which might eventually undermine the
prospects of fiscal adjustment. The effort of a country engaging in a debt reduction
scheme which is overly ambitious, (too lengthy, too intensive and likely
unattainable), may prove a short-lived consolidation episode, and on this account
looks like an exercise doomed to failure. Countries must then, be committed to
targets that are achievable both in terms of intensity of effort and time horizon,
otherwise they put at stake their credibility, which in the course of adjustment may
adversely affect their efforts towards fiscal consolidation.
155
3. Fiscal Frameworks in the EMU
This part of the study reviews the experience on fiscal frameworks and their
constituent components. The aim here is to assess the relationship between budget
performance and fiscal rules, the most prominent element in an all-encompassing
fiscal framework. To sharpen our understanding on the issue, we devised a simple
analysis that correlates budgetary outcomes to the strength of numerical fiscal rules
for the group of the EMU countries in our set, over a number of years. Our results
corroborate previous findings that associate improved fiscal performance with the
existence of fiscal rules.
3.1. An overview
According to the EC 2009, [8]: “A domestic fiscal framework can be defined as the
set of elements that form the basis of national fiscal governance, i.e. the
country‑specific institutional policy setting shaping fiscal policy making at national
level”. The main elements of domestic fiscal frameworks are numerical fiscal rules,
independent public institutions acting in the field of budgetary policy, medium‑term
budgetary frameworks for multiannual planning, and budgetary procedures
governing the preparation, approval and implementation of budget plans. All these
elements interplay with each other, as complements rather than substitutes,
influencing the working of the whole system of fiscal governance.
3.2
Fiscal Rules
Following Kopits and Symansky (1998), a numerical fiscal rule is "a permanent
constraint on fiscal policy, expressed in terms of a summary indicator of fiscal
performance". The rationale behind the introduction of rules in a fiscal domain is
that they can act as effective instruments in containing budgetary imbalances,
provided that they are equipped with appropriate characteristics within the
framework of budgetary policy they adhere to (Anderson et al., 2006). Recent
research suggests that numerical fiscal rules exert a strong influence on budgetary
outcomes (Debrun et al. 2008, EC 2009, [4]).
However, the extent of the influence of formal rules varies according to the
nature of rules and their characteristics (Bohn and Inman, 1996). Moreover, through
the estimation of fiscal reaction functions, some authors (Ayuso-i-Casals et al. 2007)
support the argument that numerical rules also exert a disciplinary impact on fiscal
policy.
Fiscal rules address various objectives, although the primary emphasis is on
numerical rules aiming at fiscal sustainability. Although rules should not be
considered as a substitute for political commitment to fiscal discipline, if
appropriately designed and implemented, they can serve the purpose of fiscal
sustainability. Rules for that purpose are the following:
- Budget balance rules (e.g. cyclically adjusted balance targets, ‘golden rule’).
- Debt rules (e.g. debt ceilings, debt-repayment capacity limits, such as debt service
156
DECEMBER 2010
to revenues ratio).
- Expenditure rules (expenditures control through binding spending ceilings).
- Revenues rules (ceilings on tax burden, constraints on tax revenue developments).
It must be noted that more than one third of the fiscal rules in place in the EU
are budget balance rules, whereas expenditure and debt rules account for about one
fourth of the total in both cases. Revenue rules account for less than ten percent (EC,
2009 [8]). Existing fiscal rules, by design, refer to the short-to-medium run horizons.
Nonetheless ensuring compliance with the short/medium run does not necessarily
ensure compliance with the long run as well. In this respect, existing fiscal rules in
the EU miss an important time dimension.
3.3
Independent Fiscal Institutions
Another basic component of a fiscal framework is fiscal institutions. National fiscal
institutions are non-partisan, independent public entities, other than the central bank,
government or parliament that prepare macroeconomic forecasts for the budget,
monitor fiscal performance and/or advise the government on fiscal policy matters.
These institutions which are primarily publicly funded but functionally independent
do not have a mandate to formulate or conduct fiscal policy, despite suggestions in
favor of such an option (Wyplosz 2005).
Independent fiscal institutions are in place in many EMU countries with the
mandate to provide independent analysis and forecasts of fiscal developments and
estimates of the cost implications of various budgetary initiatives. In addition, they
provide normative assessments which include even the appropriateness of fiscal
policy stance and consistency between announced fiscal actions and budgetary
outcomes. Independent fiscal institutions complement fiscal rules and it is believed
that when such institutions are embedded in domestic fiscal frameworks, they
contribute to improved fiscal performance by influencing budgetary developments
(Steclebout et al. 2007, Jonung et al. 2006, Debrun 2007, EU 2010, [2]).
3.4
Budgetary Procedures
Typically, domestic budgetary procedures encompass all the procedural rules laid
down in law covering the planning, approval and execution of the budget process.
According to the literature, seven budgetary dimensions are conducive to the quality
of the budget process (EC 2007 [8], Von Hagen et al. 1999). These are:
transparency, realistic economic assumptions, multiannual budget planning,
budgetary centralization at the planning and approval stages, budgetary
centralization at the implementation stage, top-down budgeting and performance
budgeting. Nonetheless, countries follow different procedures regarding drafting,
approval and budget execution.
3.5
Medium-Term Budgetary Frameworks
According to EC 2007 [5], medium-term budgetary frameworks (MTBFs) can be
considered as policy instruments that allow for an extension of the horizon over
157
which fiscal policy is conducted. This horizon usually extends beyond the annual
budgetary calendar. Barring three Member States (EL, LU and PT), all the other
EMU countries examined in this paper, declared to have an MTBF in place in 2008
(EC 2009, [4]). MTBFs usually cover a period of three to four years and the most
common institutional sector targeted is the general government. All MTBFs are
rolling and time-flexible frameworks. However, MTBFs are not without
shortcomings, for they do not necessarily incorporate binding targets.
Overall, the experience so far regarding the constituent elements of fiscal
frameworks suggest that these institutional arrangements can be useful devices in
improving the conduct and quality of fiscal policy in terms of better budgetary
results and less pro-cyclicality (IMF (2009), [1]). And although rules and institutions
cannot always reconcile fiscal soundness and budgetary flexibility, emphasis should
be placed on their appropriate design for they have to take into account countryspecific circumstances.
Even though numerical fiscal rules are of increasing importance, insufficient
independent monitoring and weak enforcement mechanisms remain the achilles’ heel
of current fiscal rules. Fiscal institutions and MTBFs continue to be wide-spread in
the EMU, although progress has been limited lately and some revision – especially
for MTBFs – is warranted. Poor monitoring mechanisms and lack of predefined
measures in case budgetary developments depart from medium-term budgetary
objectives, remain principal weaknesses in most member states MTBFs. Finally,
well-designed domestic fiscal frameworks can enhance policymakers’ commitment
to a lasting fiscal consolidation and sustainable budgetary policies.
4.
Fiscal Rules and Budget Performance
4.1
Fiscal Rules Strength Index12
Fiscal Rules is the most practical, and thus important component of fiscal
frameworks and evidence indicates that the discretionary or cyclically adjusted
budget balance “…on average improved in the years following the introduction of
numerical fiscal rules” (EC 2006, [6]). To corroborate this argument, we relate the
coverage and strength of a fiscal rule index to the budget performance in the twelve
EMU countries in our sample. On one hand, we use the numerical version of the
fiscal rule index (FRI) taken from the ‘Fiscal Rules database‟ (EC-DG Ecofin)13,
using as a benchmark the numerical average value of FRI over the period 2003-2008.
On the other hand, we use a measure of budget performance (BP), in terms of
Cyclically Adjusted Balances (CAB) taken as average over the period 2003-2010.
The strength of the index of fiscal rules is calculated by taking into account
five criteria (EC 2006, [6], EC 2009 [4]). These are: (a) the statutory base of the rule,
(b) the nature of the body in charge to monitoring respect of the rule, (c) the nature
of the body in charge of enforcement of the rule, (d) the enforcement mechanisms of
the rule and (e) the media visibility of the rule.
158
DECEMBER 2010
4.2
Fiscal Rules Strength Index and Budget Performance
Using the numerical version of the fiscal rule strength index (FRI), we assign to
countries:
‘Strong’ FRI S  : if FRI  0.5
‘Weak’ FRI W  : if FRI  0.5
Concerning the budget performance ( BP ) and taking into account the
Maastricht Treaty deficit/GDP reference value of -3%, we assign to countries:
‘High’ performance (H): if the average Cyclically Adjusted Balance, CAB  3% ,
‘Low’ performance (L): if the average Cyclically Adjusted Balance, CAB  3% .
Based on this taxonomy, in terms of FRI
classification for the countries in our sample:
we obtain the following
Strong S  : DE, ES, LU, NL, FI
Weak W  : BE, IE, EL, FR, IT, AT, PT
whereas, in terms of BP we classify countries as:
High performance H  : BE, DE, ES, LU, NL, AT, FI
Low performance L  : IE, EL, FR, IT, PT.
FRI
Table 8. Rules and Performance
BP
H
DE
ES
S
LU
NL
FI
W
BE
AT
L
none
IE
FR
EL
IT
PT
Table 8 confirms the widely accepted notion that countries with strong fiscal
rules have, on average better fiscal performance 14, i.e. lower deficits, while countries
with weak (or no) fiscal rules in place have, on average worse fiscal performance,
i.e. higher deficits, although two countries (BE, AT) had weak rules but high
performance. Out of the twelve EMU countries under examination, five are placed in
the S-H territory (DE, ES, LU, NL, FI), another five in the W-L territory (IE, EL,
FR, IT, PT), while two countries (BE, AT) are positioned in the W-H domain. This
symmetric distribution of the countries in the sample based on their budget
performance vis-à-vis the strength of fiscal rules in place, provides further support to
159
the argument about the positive correlation between a strong FRI and high
budgetary performance.
Based on the EMU experience so far, the above exercise illustrates clearly
the point that carefully designed fiscal frameworks with effective numerical fiscal
rules, if in place, can contribute to the improvement of budgetary outcomes.
5. Concluding Remarks
In the aftermath of the economic crisis that hit the economies the world over, many
countries are left with taunting economic policy challenges. For fiscal policy in the
EMU in particular, it is important to identify existing structural weaknesses and
institutional vulnerabilities and take decisive and prompt action. Coordination
enhancing institutional solutions (not necessarily through complex constitutional or
Treaty amendments) is a possible avenue, though much depends on political will and
determination to pursue ‘common ground’ policies to uncommon (at least in
magnitude) fiscal challenges. An example could be the mandatory adoption of fiscal
frameworks in all EMU countries with a number of „least common structural
elements‟.
Apart from fostering sound budgetary policies and consolidation, instituting
such a mechanism - addressing domestic requirements - would not only reinforce
public governance, but it would also firmly anchor public expectations about the
course of future fiscal rectitude. Moreover, it would reduce discretion in the conduct
of fiscal policy and would enhance governments’ commitment and accountability.
One the other hand, since the root causes of fiscal problems in the EMU
countries are diverse, responding to the wide variety of fiscal challenges facing
Eurozone in a uniform way does not guarantee an effective and sustainable solution
to country specific fiscal setbacks.
A unified policy framework can only provide general guidelines for setting
policy targets and for steering the conduct of fiscal policy. A uniform strategy seems
inappropriate to confront the heterogeneity of diverse and growing fiscal challenges
the EMU member states are facing, especially with a view to reduce observed
asymmetries. Solutions tailored-made to countries’ idiosyncratic fiscal problems on
the other hand, could allow for temporary deviations from the SGP, as long as these
deviations do not threaten its validity or undermine future progress. Initial conditions
and available fiscal space, are also important parameters in determining future
economic performance.
After a decade of experience, euro-zone is now confronted with a profound
gap in its fiscal architecture. Specifically:
(a)
Fiscal asymmetries seem to be the rule rather than the exception: asymmetries
in fiscal performance (budget deficits and public debt), asymmetries in budget
outcomes across time (pre/post crisis) and across countries, and asymmetries in fiscal
stance, stemming among other from the uneven contribution of components to
160
DECEMBER 2010
budget outcomes. In addition, there are growing sustainability concerns: the
accumulation of sizeable public debts in several EMU countries and their projected
increase, represent an acute challenge and pose a serious threat to the sustainability
of public finances for the euro-zone as a whole. In view of these developments,
EMU member states have to engage in serious and possibly long-lasting
consolidation efforts, in order to put their fiscal finances on a sound footing. Failing
to do so, the repercussions will be immeasurable.
(b)
In this connection, the time path and magnitude of any debt reduction effort
should reflect country specific circumstances. A „one-size fits all‟ EMU strategy may
be sub-optimal for individual country fiscal sustainability objectives. Countries with
weak initial (structural) conditions are required to make sizeable adjustments in
primary balances to reduce the debt/GDP ratios to target levels. The required fiscal
adjustment has, nevertheless, to be reasonable/attainable, otherwise the purpose of
fiscal consolidation is not served.
(c)
Acknowledging the positive correlation between fiscal rules and budget
performance implies that the adoption of sound domestic fiscal frameworks in all
member states is a precondition for improved fiscal outcomes, including among
other (if not amplified by additional rules to address) the observed target gaps (i.e.
the distance between current levels of budget deficits and public debts from set
targets). The mandatory establishment of sound domestic fiscal frameworks in all
EMU member states can be considered as a positive factor in assisting governments’
efforts to reduce the observed asymmetries, and thus in promoting fiscal
convergence with the ultimate aim to minimize intra-EMU member states’ target
gaps, thereby serving the purpose of consolidation.
Although the crisis has had asymmetric effects across member states, a
common legacy has been the shadow cast on public finances and potential growth.
The European economies are still facing strong headwinds from the global economic
and financial turbulence which revealed well concealed structural and institutional
vulnerabilities. Thus far, the transition to a new steady-state has not yet found a solid
gait. Differentiation of policy responses of member states are expected to continue to
exist, reflecting marked differences in the scale and nature of structural and
budgetary adjustment challenges they face. Ongoing rebalancing will continue and
consolidation in the euro-area might prove a long process. Under current
circumstances the road to convergence seems quite long, unless a concerted effort is
carefully orchestrated and implemented.
161
Annex A. Decomposing changes in the overall balances into automatic and
discretionary effects
Starting from the nominal budget we derive the formulas for the estimation of the
Cyclical Balance (CB) and Cyclically Adjusted Balance (CAB) respectively. While
the CB captures the effects of the cycle on the budget, the CAB gives an indication of
the balance that would prevail in the economy if actual and potential GDPs were
equal. CAB is unobservable and it is usually obtained as a residual. Moreover, it is
commonly used as an indicator of discretionary fiscal policy. Total Fiscal Balance
(TFB) is the sum of Cyclical (CB) and Cyclically Adjusted Balance (CAB):
and
TFB  CB  CAB 15
(1)
TFB  CB  CAB
(2)
The change in CB is also defined as ‘Automatic Stabilizers’ (AS), so that:
AS  CB  TFB  CAB
(3)
In order to decompose TFB to its constituent parts, a calculation of the output gap
( y gap ) is required first, along with the elasticities of revenues (R) and expenditures
(G) with respect to variations in output (GDP).
Deviations of actual GDP (𝑌) from trend or potential GDP ( Y P ) as a share of
potential GDP ( Y P ) provides the formula for the output gap ( y gap ). Formally:
 Y Y P
y gap  
P
 Y




(4)
The change in CAB can be calculated from the cyclically adjusted revenues ( R Ad )
and expenditures ( G Ad ) defined as:
Y P
R Ad  R
 Y




er
e
(5)
Y P  g

(6)
G  G 

Y


where R is nominal revenues and G is nominal expenditures respectively, and e r and
eg are the elasticities of revenues and expenditures with respect to the output gap.
Ad
162
DECEMBER 2010
The Cyclically Adjusted Balance (CAB) is then given by:
e
e
Y P  r
Y P  g
  G

(7)
CAB  R

 Y 
 Y 


If there is a perfect correlation between revenues and the cycle (i.e. e r  1 ), while
expenditures remain insensitive to cyclical swings (i.e. e g  0 ), then the CAB
becomes:
Y P 
G
CAB  R

 Y 
(8)
By substituting eq. (8) in (1) the Cyclical Balance (CB) becomes:
 YP
CB  TFB  CAB  R1 
Y

 R P
  Y y gap
 Y

(9)
In this case, the automatic stabilizers AS , as a ratio to the potential GDP 16 ( as ) can
be calculated from eq. (9) as follows:
R

CB   Y P y gap 
Y

CB
 R P

 P  Y y gap 
YP
Y Y

CB
Y
and since
P

AS
YP
AS
YP
 as

  ry gap

(10)
(11)
(12)
(13)
it follows that the automatic stabilizers (as a share of potential GDP) can be
approximated by:
as  ry gap
(14)
The above approach is commonly used for its computational simplicity, nonetheless,
the CAB and CB are not without shortcomings (see for instance, Fedelino et al 2009,
EC – DG EC 2010 [1] and 2009 [4]), a discussion of which lies beyond the scope of
the present exposition.
163
Annex B. An analysis of the debt dynamics
The basic dynamics of debt accumulation can be computed as follows:
Gt  rDt 1  Tt  Dt  Dt 1   S t  S t 1 
Dt  Dt 1  rDt 1  Tt  Gt 
Dt  (1  r ) Dt 1  PBt
(1)
(2)
(3) 17
where at the end of a given period t,
𝐷 = is the country’s gross public debt stock
𝑟 = is the real interest rate on debt outstanding
𝐺 = is government expenditure
𝑇 = is government revenues
S = is Seignorage.
PB is primary balance (interest payments on debt excluded), with PB  T  G  0
denoting the primary surplus.
Expressing the above identity in percentage of GDP terms we get:
Dt
D Y
PB t
 1  r  t 1 t 1 
Yt
Yt 1 Yt
Yt
(4)
Manipulating (4) and setting real GDP rate of growth: g t 
Yt  Yt 1
, we get the
Yt 1
fiscal sustainability identity 18 :
 1  rt
d t  
 1  gt

d t 1  pbt

(5)
where small letters now denote the ratios of initial variables to GDP.
From (5) we note that debt/GDP is determined by three factors: the debt/GDP ratio
in the previous period, the interest rate/growth rate differential to GDP ratio and the
primary balance/GDP ratio. This formula can be extended to the long run by
systematically substituting the debt/GDP ratio up to the final T period (the starting
reference time is t  1  0 ). Based on the assumption of constant r and g rates and
without any loss of validity of arguments, we simplify calculations that yield:
 1  g 
 1  g  
 1  g  
do  
 pb1     1  r   pbT   1  r   d T


1

r






T
T
(6)
164
DECEMBER 2010
For an infinite time horizon eq. (6) becomes:
d0 

 1  g 
 1  g 
pb j  limT  
 dT
j 1  1  r  


 1  r  

j
T
(7)
Imposing the ‘no-Ponzi’ condition, the last term in (7) which implies that the
discounted value of the public debt must tend to zero, we get:
d0 

 1  g 
pb j
j 1  1  r  



j
(8)
This is the intertemporal budget constraint (i.e. the government’s solvency
condition), which shows that the discounted value of the sum of future primary
balances must equal the current value of public debt.
Debt-stabilizing Primary Balance
Under given macroeconomic assumptions regarding r and g , (as exogenously
determined variables), we calculate the primary balance necessary to stabilize the
stock of debt/GDP at current (or set) levels. Solving eq. (5) the required primary
balance ( pb * ) necessary to stabilize the debt/GDP ratio depends on the difference
between ( r  g ) and the debt level d 0 prevailing in year t  0 :
rg
d 0
pb *  
 1 g 
(9)
If r  g  0 , i.e. r  g , then from (5) we obtain: d t   pbt .
In this case pb will be the primary surplus equal with the change in debt d . It is
apparent that the greater the r  g difference is, the greater, ceteris paribus, the
required primary surplus has to be in order to stabilize the debt/GDP ratio.
Nonetheless, r  g has historically proven to be the case in our sample (with some
exceptions for some EMU countries during the early accession years in the EMU).
Debt-Target Analysis
Simply stabilizing the debt/GDP ratio does not satisfy the government’s solvency
condition (8). Producing a constant primary surplus to merely satisfy the solvency
condition on the other hand, cannot guarantee elimination of the debt/GDP ratio
down to zero. In order to reduce the debt/GDP ratio to a desired (target) level d * ,
over a period of T years, the required primary surplus pb ** is calculated by first
solving recursively eq. (5) for d which yields:
t
 1 r 
 d 0  pbt
d t  
 1 g 

 1 r 


j 0  1  g 


t 1
165
j
(10)
and then solving eq. (10) for pb , we get the primary surplus required to reduce the
debt/GDP to a specific debt-target level d * in T periods:
T
pb ** 
 1 r 
  d *
d 0 
1

g


T 1  1  r




j 0  1  g 


j
(11)
Notes
1
Panagiotis Korliras is Professor at the Athens University of Economics and
Business and Scientific Director of the Centre of Planning and Economic Research,
Athens, Greece, [email protected] and Yannis Monogios is Research Fellow at the
Centre of Planning and Economic Research, Greece, [email protected]. An earlier
version of this work has been presented at the Deutsche Bundesbank Conference on
July 29-31, 2010 in Frankfurt, Germany on the occasion of the 10 th Biennial APF
International Conference on ‘Regulatory Responses to the Financial Crisis’. The
usual disclaimer applies.
2
The sample consists of BE, DE, IE, EL, ES, FR, IT, LU, NL, AT, PT and FI. The
rest of the EMU countries namely CY, MT, SI, SK and recently EE have been
excluded from the sample due to their short history in the EMU.
3
COM (2008) 800 final, 26/11/2008, 'A European Economic Recovery Plan', see:
http://ec.europa.eu/commission_barroso/president/pdf/Comm_20081126.pdf
and
also in Progress report on the implementation of the European Economic Recovery
Plan - June 2009" and ditto December 2009, which is available at
http://ec.europa.eu/financial-crisis/documentation/index_en.htm.
4
For a technical exposition see Annex A.
5
Admittedly CAB is not without methodological shortcomings (see for instance
Larch and Turrini, 2009).
6
In our exposition TFB includes debt-interest payments.
7
See for instance, Noord (2010).
8
For a technical exposition see Annex B.
9
Note that in the following analysis the stock of debt does not include government’s
contingent liabilities. By the same token, no adjustments have been made for the
government’s assets.
10
LU has been excluded from the exercise since the debt/GDP ratio lies at very low
levels (see also Table 3).
166
DECEMBER 2010
11
The combinations of r and g were adopted from various sources (Deutsche Bank
2010, OECD database, IMF (2009)) and estimates by the authors.
12
See EC 2009, [4] and EC 2006, [6].
13
See http://ec.europa.eu/economy_finance/db_indicators/fiscal_governance/fiscal_rules/index_en.htm
Similarly, the IMF has constructed an FRI using a principal component analysis, see
IMF (2009), [1].
14
In Ayuso-i-Casals et al. (2007) higher values of the FRI are associated to an
improvement in the cyclically-adjusted primary balance. Our results corroborate this
argument.
15
If interest payments are excluded from TFB, then one arrives at the Primary
Balance (PB) net of interest : TFB  PB  INT  CB  CAPB  INT and thus
TFB  PB  INT  CB  CAPB  INT which is a more accurate way of
decomposing overall balances. However, including or not interest payments in the
decomposition of overall balances requires further adjustments (see Fedelino et al.
2009, Bouthevillain and Quinet, 1999).
16
When scaling the choice of either potential or nominal GDP plays a role in the
computation of the automatic stabilizers.
17
S which denotes Seignorage is excluded from the basic relationships for
expositional simplicity without any loss of generality, as it is not an instrument in a
Member State’s Central Bank monetary policy toolbox. In addition this analysis
does not deal with stock-flow adjustments.
18
See for instance Escolano (2010), Ley (2009).
References
Anderson B. and Minarik J.J. (2006), ‘Design Choices for Fiscal Policy Rules’,
OECD Journal on Budgeting, Volume 5 – No. 4.
Ayuso-i-Casals, J., Gonzalez Hernandez D., Moulin L. and Turrini A. (2007),
'Beyond the SGP: Features and effects of EU National-level fiscal rules,' in:
Deroose, S., E. Flores and A. Turrini (eds.), The role of fiscal rules and
institutions in shaping budgetary outcomes. Proceedings from the ECFIN
workshop held in Brussels 24 November 2006, pp. 191-242.
Balassone F., Cunha J., Langenus G., Manzke B., Pavot J., Prammer D. and
Tommasino P. (2009), ‘Fiscal sustainability and policy implications for the euro
area’ ECB Working Papers Series, No. 994, Jan. 2009.
Balassone, F., and Franco, D. (2000). ‘Assessing Fiscal Sustainability: a Review of
Methods with a View to EMU’, Banca d‟ Italia, Fiscal Sustainability, Roma.
Banca d’ Italia: „Current Issues and Challenges‟. Papers presented at the Bank of
Italy workshop held in Perugia, 29-31 March 2007.
167
Banca d’ Italia: „Fiscal Rules‟. Papers presented at the Bank of Italy workshop held
in Perugia, 1-3 February 2001.
Becker S., Deuber G. and Stankiewicz S., (2010), 'Public Debt in 2020. A
Sustainability Analysis for DM and EM Economies’, Deutsche Bank Research,
March 24, 2010.
Bohn, H., and Inman P. (1996), 'Balanced budget rules and public deficit: Evidence
from the U.S. States', National Bureau of Economic Research, Working Paper,
No. 5533.
Bouthevillain C. and Quinet A. (1999) ‘The Relevance of Cyclically Adjusted Public
Balance Indicators - the French Case’ in Indicators of Structural Budget
Balances Proceedings of the Bank of Italy Public Finance Workshop (Perugia,
November 26-29, 1998).
Debrun, X., L. Moulin, A. Turrini, J. Ayuso-i-Casals and M. S. Kumar (2008), 'Tied
to the mast? The role of national fiscal rules in the European Union', Economic
Policy, April 2008.
Debrun X., and Kumar M. S. (2007), ‘Fiscal rules, fiscal councils and all that:
commitment devices, signaling tools or smokescreens? Banca d’ Italia: „Current
Issues and Challenges‟. Papers presented at the Bank of Italy workshop held in
Perugia, 29-31 March 2007.
Escolano J. (2010), ‘A Practical Guide to Public Debt Dynamics, Fiscal
Sustainability and Cyclical Adjustment of Budgetary Aggregates’, International
Monetary Fund.
European Commission (2010), [1], ‘Cyclical Adjustment of Budget Balances’,
Autumn 2010.
European Commission (2010), [2], Directorate General for Economic and Financial
Affairs, ‘European Economic Forecast – Autumn 2010', European Economy No.
7/2010.
Affairs, 'Public Finances in EMU – 2010', European Economy No. 4/2010.
European Commission (2007), [5], Directorate-General for Economic and Financial
Affairs, ‘Sustainability Report – 2009', European Economy No. 9/2009.
European Commission (2009), [8], ‘Domestic Fiscal Frameworks. How National
fiscal governance can contribute to Budgetary consolidation over the mediumterm’, Note for the attention of the Economic Policy Committee, October 2009.
European Commission (2009), [9], ‘Economic Crisis in Europe: Causes,
Consequences and Responses’, European Economy, No. 7/ 2009.
168
DECEMBER 2010
Affairs, ‘Impact of the current economic and financial crisis on potential
output’, European Economy Occasional Paper 49, June.
Fedelino, A., Ivanova, A., and Horton M. (2009), ‘Computing Cyclically Adjusted
Balances and Automatic Stabilizers’, Fiscal Affairs Department, November
2009, International Monetary Fund.
Greiner, A., Kollery, U., and Semmler, W. (2007), ‘Debt sustainability in the
European Monetary Union: Theory and empirical evidence for selected
countries’, Oxford Economic Papers 59, pp. 194–218.
Hallerberg, M., Strauch R. and Hagen J. von (2007), ‘The Design of Fiscal Rules and
Forms of Fiscal Governance in European Union Countries’, European Journal
of Political Economy, Vol. 23, No. 2, pp.338-59.
Hagen, J. von (2006), ‘Fiscal rules and fiscal performance in the European Union
and Japan’, Monetary and Economic Studies, Vol. 24:1, pp. 25-60.
Hagen, J. von and Wolff, G. (2006), ‘What do deficits tell us about debt? Empirical
evidence on creative accounting with fiscal rules in the EU’, Journal of Banking
and Finance, Vol. 30:12, pp. 3259-3279.
Hagen, J. von (2002), ‘Fiscal rules, fiscal institutions and fiscal performance’, The
Economic and Social Review, Vol.33, No. 3.
Hagen, J. von and J. Poterba (1999), 'Fiscal institutions and fiscal performance',
National Bureau of Economic Research and University of Chicago Press.
IMF (2009), [1], ‘Fiscal Rules - Anchoring Expectations for Sustainable Public
Finances’, Fiscal Affairs Department, December 2009, International Monetary
Fund.
IMF (2009), [2], ‘The State of Public Finances. Cross-Country Fiscal Monitor:
November 2009’, Staff Position Note, Fiscal Affairs Department, November
Jonung, L. and Larch M. (2006), ‘Fiscal policy in the EU – Are official forecasts
biased?’ Economic Policy, July 2006.
Kopits, G. and Symansky S. (1998), 'Fiscal Policy Rules,' IMF Occasional Paper
Larch, M. and Turrini A. (2009) ‘The cyclically-adjusted budget balance in EU fiscal
policy making: A love at first sight turned into a mature relationship’, European
Economy Economic Papers, 374.
Ley, E. (2009). Fiscal (and External) Sustainability, MPRA Paper No. 23956.
Monogios, Yannis (1998), ‘Debt, Savings and Aggregate Profits. Does Fiscal Policy
Matter?’, PhD Thesis, University of Cambridge.
Noord, P. van den (2010), ‘Turning the page? EU fiscal consolidation in the wake of
the crisis’, address given at the conference: The Aftermath of the Crisis, on
November 5-6, 2009 at the Österreichische Nationalbank.
OECD (2010) ‘Economic Outlook‟, N. 87, May 2010.
169
Schuknecht, L. (2004), ‘EU Fiscal Rules. Issues and Lessons from Political
Economy’, ECB Working Paper Series no. 421/December 2004, European
Central Bank.
Stéclebout, E. and Hallerberg M. (2007), 'Who provides signals to voters about
government competence on fiscal matters? The importance of independent
watchdogs', European Economy Economic Papers, European Commission,
number 275.
Sturzenegger, F. (2002), ‘Toolkit for the Analysis of Debt Problems'. Universidad
Torcuato Di Tella.
Wyplosz, C. (2005), ‘Fiscal Policy: Institutions versus Rules’, National Institute
Economic Review, No. 191, pp. 70-84.
Who Is To Blame for the Great Recession?
Herbert Grubel
Simon Fraser University and The Fraser Institute1
Abstract: This paper uses empirical information and economic theory to show that
the primary causes of the Great Recession of 2008 were the non-market policies of
China and energy producing countries, which resulted in the current account
imbalances that existed before the recession began. The savings of these countries
did not have the normal beneficial effects on global interest rates and investment
because they were used to buy only US debt instruments and none of other
developed countries. This asymmetric effect was the result of the fixed exchange
rate, which the surplus countries maintained against the dollar and not against other
currencies that otherwise might have shared the burden of absorbing the high levels
of savings.
JEL Classification: F32, F34
Keywords: Great Recession, Chinese non-market policies, current account
imbalances, fixed dollar rate, energy-producing countries
1. Introduction
Economists who consider the basic cause of the global recession that started at the
end of 2007 (hereafter referred to as the Great Recession) are divided into three main
camps. The first, discussed in part one of this paper blames the excessively easy
monetary policy of the US Federal Reserve between 2002 and the middle of 2005.
The second part considers the argument that the accumulation of large fiscal and
trade surpluses by sovereign wealth funds, central banks and government pension
plans are at fault. Part three contains a theoretical model that demonstrates the interdependency of these two main causes of the crisis. Part four examines the arguments
made by the third camp that blames the US Congress for legislation that resulted in a
housing bubble and the use of innovative but flawed financial practices.
2. The Fed Is To Blame
The most prominent members of the blame-the-Fed camp are John Taylor (2009)
Allan Meltzer (2002) and Anna Schwartz (2009). They argue that the Fed policies of
monetary ease were maintained for too long a period after the 2000 recession had
ended. This mistaken policy led to the housing bubble and other manifestations of
171
172
DECEMBER 2010
excess aggregate demand, which in turn prompted the Fed to raise rates sharply and
caused the recession that started at the end of 2007.
This proposition is backed by Figure 1, where the solid line shows the Federal
Funds Rate between 2000 and 2009. As can be seen clearly, the rate fell sharply
during all of 2001, reflecting the recession-fighting ease of monetary policy in the
wake of the bursting of the high-tech bubble. The graph also shows that between
2002 and the middle of 2004 the easing continued until this policy stance was
reversed in the middle of 2004, when the rate rose until it reached a peak early in
2006.
Figure 1
Source: for the federal funds rates:
http://www.federalreserve.gov/datadownload/Download.aspx?rel=H15&series
Taylor (1993) presents the results of his simulation models to suggest what
interest rates set by the Fed would stabilize both employment and inflation. He
specified a rule in the form of an equation2 that was applied to the evaluation of
actual US monetary policies in later years and was seen to have tracked them very
well during the period 1987 to 2001 when the US and world economy experienced
stable prices and high rates of economic growth. 3
Taylor’s rule was used to calculate what the Federal Funds rate should have
been, given US inflation and the output gap in the years after 2001. The results of
this exercise are shown as the dotted line in Figure 1. As can be seen, if the Fed had
followed the rule, at the end of 2001, when economic recovery was well under way,
VOL.7 NO.2
GRUBEL: WHO IS TO BLAME FOR THE GREAT RECESSION?
173
the Fed should have raised the federal funds rate and continued the increases until
early in 2005 and then kept the rate at this level until the end of 2006.4
This counter-factual application of the Taylor rule demonstrates clearly that
the Fed’s monetary policy was too easy for about 3.5 years and ultimately resulted in
monetary tightening, which in turn led to the recession.
While Taylor’s analysis relied on the development of the federal funds rate,
Meltzer and Schwartz used data on the growth of monetary aggregates and interest
rates and, not surprisingly also concluded that Fed policies in the wake of the 200001 recession were excessively expansionary for too long. 5
2. Foreign Governments are at Fault
A search of the financial and economic literature reveals many exponents of the view
that “global imbalances” and a resultant “savings glut” have been the ultimate causes
of the US recession and global economic crisis of 2008-09. Ben Bernanke (2005),
the Chairman of the Federal Reserve and Hank Paulson (2008), then the Secretary of
the Treasury are two of the most prominent persons in public office who expressed
this view. Martin Wolf (2006) is one of the well-known members of the financial
media who has written many articles supporting this position.
The argument in support of the role of global imbalances in the creation of the
recession is summarized by Paulson (2008) in a section of his speech headed
“Genesis of Financial Turmoil: Global Imbalances”
“The world was awash in money looking for higher return, and much of this
money was invested in U.S. assets. The combination of a huge amount of
capital and low interest rates stoked greater risk taking, financial innovation
and complexity…” 6
What caused these global imbalances and savings glut? They were due to the
efforts of government agents in different countries to accumulate foreign exchange
reserves for use in the event of international payments imbalances and the resultant
need to prevent the depreciation of currencies. China was also motivated by its
desire to maximize the growth of manufacturing through the creation of large
surpluses in its trade balance that resulted in work for underemployed workers in the
rural sector. Table 1 shows the total of foreign exchange reserves held by the
world’s central banks at the end of 2008: $7.4 trillion in total, with China and Japan
holding the largest sums of $2.2 trillion and $1.0 trillion, respectively.
Several countries’ policies were driven by the desire to accumulate funds for
future inter-generational equalization of income, as for example sovereign wealth
funds of the producers of energy and the investments of national pension systems.
All of these efforts clearly involved interference with free and efficient allocation of
resources and instead served the economic and social engineering efforts of the
relevant countries’ governments.
174
DECEMBER 2010
Table 1
Official Foreign Exchange Reserves
(End of 2008)
Country
China
Japan
Russia
India
South Korea
Brazil
Hong Kong
Singapore
Algeria
Germany
Others
Total
US$ billion
2,243
1,031
387
248
201
201
183
166
138
133
2,469
7,400
Source: International Financial Services London (IFSL) (2009), Sovereign Wealth
Funds 2009, found at:
http://www.google.com/search?hl=en&q=ifsl+swf&aq=0p&oq=ifsl&aqi=g%3Ap2g
3g%3As3g2, table 5, page 6
Table 2 shows the amount of money held by Sovereign Wealth Funds to meet
the needs of future generations. The second column gives the value of the
investments under management by the funds identified in column one.
As can be seen, the total value of the funds summed to $3.9 trillion at the end
of 2008, having grown from $3.3 trillion at the end of 2007, in spite of the decline in
world commodity prices and stock market averages. The largest of the top seven
funds were owned by energy exporting governments, with the United Arab Emirates
heading the list with $875 billion.
The last column of the table indicates whether the funds dominantly came
from profits from the sale of energy or from the accumulation of fiscal surpluses
created through taxation or mandated premiums on social insurance programs. As
can be seen, the funds financed through premiums numbered eight, with the other 12
holding funds derived from energy sales.
VOL.7 NO.2
175
Table 2
Sovereign Wealth Funds Assets under Management
(End of 2008)
US$ billion Source
875
Commodities
433
Commodities
330
Taxation
312
Taxation
301
Commodities
265
Commodities
225
Commodities
200
Taxation
173
Taxation
134
Taxation
82
Commodities
74
Taxation
60
Commodities
50
Commodities
47
Commodities
44
Taxation
38
Commodities
30
Commodities
30
Taxation
29
Commodities
168
3,900
Total
Source: International Financial Services London (IFSL) (2009), Sovereign Wealth
Funds 2009, found at
http://www.google.com/search?hl=en&q=ifsl+swf&aq=0p&oq=ifsl&aqi=g%3Ap2g
3g%3As3g2, table 3, page 3
Country
Abu Dhabi Investment Authority (UAE)
SAMA Foreign Holdings (Saudi Arabia)
Government of Singapore Investment Corp
SAFE Investment Company (China)
Government Pension Fund of Norway
Kuwait Investment Authority
National Welfare Fund (Russia)
China Investment Corporation
Hong Kong Monetary Authority Invest. Portfolio
Temasek Holdings (Singapore)
Investment Corporation of Dubai
National Social Security Fund (China)
Qatar Investment Authority
Libyan Investment Authority
Revenue Regulation Fund (Algeria)
Australian Future Fund
Kazakhstan National Fund
Brunei Investment Agency
Korea Investment Corporation
Alaska Permanent Fund
The two tables presented here show that by the end of 2008 governments held
$11.3 trillion in the portfolios of their central banks and wealth funds. This figure
can be put into perspective by considering that at the end of 2008 the US federal debt
accumulated since the founding of the country in 1776 was $11.5 trillion, of which
$6.1 trillion was held by the public.
While the preceding two tables present the stock of foreign assets held by
different countries, it is clear that these stocks were accumulated through annual
purchases7, the size of which is presented in Figure 2 and expressed as a percent of
the World’s GDP for the period 1997-2008, for groups of countries that are of
particular analytical interest: the United States, the Euro-area, Japan, the emerging
countries of Asia and the oil exporting countries. 8
176
DECEMBER 2010
The graph shows that the US deficits were matched almost completely every
year by the surpluses of the other countries. During the crucial years 2002-04, when
according to Taylor US monetary policy was too easy, the three-year total US
current account deficit was 4.3 percent of world GDP while the surplus of the other
countries over the same period came to 4.1 percent. This near equality is as expected
since the world as a whole cannot run imbalances and the surpluses of some
countries must always match the deficits of the rest of the world, except for
measurement errors.9
However, this fact does not help establish the merit of the arguments by the
two camps blaming either the Fed or the countries accumulating large savings. To
shed light on the elements of truth in both camps, it is useful to entertain the thought
experiments contained in the next section, which attempts to explain why markets
did not channel any of the demand for financial assets by the surplus countries into
countries other than the United States.
Figure 2
Current Account Balances
2.5
2.0
Emerging
Asia
Percent of World GDP
1.5
Oil
Exporters
1.0
Japan
0.5
0.0
-0.5
-1.0
Euro Area
United
States
-1.5
-2.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Source: Milesi Ferretti (2009), data provided and updated by the author and Gian
Maria
3. A Model for Evaluating the Arguments
Consider a world that consists of two countries only, C and M. Assume that the
government of C controls the heights of the economy through monetary and fiscal
VOL.7 NO.2
177
policies, but it also uses exchange controls, labour market policies and major
publicly owned enterprises and banks to determine wages, profits, exports, imports,
savings, investments and international capital flows. In particular, it uses these
instruments to maintain a fixed exchange rate against the currency of country M.
Country M is assumed to have a market economy with free international
capital flows and with aggregate savings and investments influenced almost
completely by monetary and fiscal policies. The country’s exchange rate is free to
float in principle, but in practice is fixed through C’s exchange rate policy.
Assume that initially both countries enjoy full employment, price stability and
balanced trade at the existing fixed exchange rate. Now consider that the
government of C decides to use its control over the economy to increase exports and
limit imports to create a large trade surplus, mainly by keeping wages low and
channelling capital to the export and import competing sectors. The labour needed
to generate this surplus is drawn from the rural sector.
The government of C has three main reasons for the use of these policies.
First, it seeks to raise total national output by shifting labour from the rural sector,
where it has low productivity, into manufacturing where its productivity is higher.
Second, it wishes to accumulate international reserves for future use in foreign
exchange market interventions. Third, it wishes to accumulate investments as
backing for its future public pension obligations.
Of central interest for the present analysis is the impact, which the trade
surplus created by C has on the economy of country M. Standard economic analysis
implies that M faces a reduction in aggregate demand and higher unemployment
because the money spent by consumers on imports leaves the country and is not used
to purchase output in the domestic economy. In the face of these influences on the
economy of M, what options does its government have in dealing with the resultant
problems?
The first set of available policies can be aimed at eliminating the trade deficit.
This requires either a depreciation of the currency or a lowering of the domestic
price level and costs of production. The first alternative is not available since
country C keeps the exchange rate fixed by its unilateral actions.
The second alternative requires a tightening of monetary and fiscal policies.
One problem with these policies is that they are costly in terms of high
unemployment and lost output over a prolonged period. The second problem is the
absence of any guarantee that country C will allow trade to become balanced since
its desire to run surpluses is undiminished and it has all the instruments of control
needed to make it happen.
Under these conditions, country M has no option but to live with the trade
deficits, use easy monetary and fiscal policies to maintain full employment and
engage in diplomatic efforts to persuade C to change its fixed exchange rate policies
and increase domestic consumption, investment and imports.
The easy monetary policy induces domestic agents to borrow for consumption
and investment, which increases the supply of stocks and bonds in capital markets.
178
DECEMBER 2010
The fiscal deficits of governments also add to this supply. However, since agents in
C buy these securities, the decrease in their prices and higher market yields that
would result otherwise do not take place.
Moreover, this process was enabled by the fact that this deficit spending by
the private and public sector did not lead to overall inflationary pressures because the
goods were imported from C at constant prices and virtually unlimited quantities
because the labour markets in that country were able to draw labour from the rural
into the industrial sector without having to pay higher wages.
The imbalances created by policies in C cannot continue without limits
because, while C is able to accumulate assets indefinitely, agents in M issuing debt
face constraints as their debt to income ratio rises to levels where capital markets
doubt the ability of the issuers to service them and the agents begin to be unwilling
to commit more of their future income to debt payments, especially so if they are
consumers and did not use the borrowed funds for productive investment.
It is obvious that in the preceding model country M characterizes the United
States with its market economy. Country C stands for the countries with the
surpluses noted above, most notable China and the energy exporting countries, in
which governments have much control over the economy.
Thus, China uses
controls to determine most of the composition and level of international trade. The
energy exporting countries are characterized by institutional arrangements that
funnel trade surpluses into accounts owned by ruling families or government
agencies that are immune from democratic, legislative pressures to use the bulk of
the funds for domestic purposes and imports.
In contrast, in the US market economy the downward pressures on aggregate
demand caused by the trade deficits were limited by the only policy instruments
available, easy monetary policy and government deficits.
The accumulation of US financial assets by the controlled economies and the
resultant growth in financial obligations generated in the US kept the world economy
humming for a number of years. However, this process had to end at some time
since the indebtedness of the private sector became so large as to raise doubts about
its ability to service it and debt-service problems began to appear. At that time, the
Fed tightened monetary policy to prevent further deterioration in these conditions,
which in turn aggravated the debt-service problems, increased financial distress of
borrowers that pulled down financial intermediaries and marked the onset of the
Great Depression.
3.1 Some Perspectives
First, the model, reasoning and data presented here support the proposition that the
Great Recession is due to the policies of countries that do not have free markets and
legislatures that respond to electorates, in particular China and the energy producing
countries.
However, the model is also consistent with the view that tighter US monetary
policies could have avoided the Great Recession. But it is essential to note that such
policies would have caused a global recession earlier in the decade, which would
VOL.7 NO.2
179
have continued until the imbalances were eliminated or shared with other countries
than only the United States. This recession might have lasted a long time since the
surplus countries were very committed to their policy goals and were not reluctant to
use non-market controls to achieve them.
Second, there is the question why Britain and the Euro countries did not
suffer from the same problems as the United States. The answer is found in the
important fact that the exchange rates of the surplus countries against these
currencies were not fixed. Trade deficits of Britain and the Euro countries led to
currency depreciation large enough to eliminate the cost advantage of China’s
exports. As a result, these countries did not suffer from aggregate demand
deficiencies and had no need to ease monetary and fiscal policy. These countries did
not generate a surplus of financial assets that China wanted to acquire.
Third, basic economic principles imply that increases in savings in the world
lead to lower interest rates, more investment and more rapid growth in productivity,
world output and income. These beneficial effects of higher savings were not
realized in recent years because the resultant demand for financial assets was
concentrated almost exclusively on the US economy and overwhelmed its capacity to
absorb them. In the end, debt to income ratios for private US deficit spenders
reached critical levels.
4. The Role of Housing Policies and Financial Institutions
There remains the need to discuss the argument that the root causes of the Great
Recession were distortions in the US housing market and the greedy and
irresponsible behaviour of financial institutions in the United States and elsewhere.
A full presentation of these arguments is not possible here, allowing only the
following, short summary.
4.1 Housing Market:
The US Congress has long attempted to use legislation to gain favour with the public
by promoting universal home ownership. For this reason, the interest costs on home
mortgages have been tax-deductible for a long time. More recently, policies were
enacted that increased the flow of mortgages to borrowers that previously had been
judged ineligible for mortgages on the basis of demographic and income
characteristics.
The increased flow of funds to these borrowers was facilitated by the policies
of the quasi-government Federal Home Loan Mortgage Corporation, known as
Freddie Mac and the Federal National Mortgage Association, known as Fannie Mae,
which buy mortgages issued by private institutions using funds obtained in private
financial market at preferential rates because they are backed by government
guarantees.
In 1991 Freddie Mac came under the supervision and thus influence of the
U.S. Department of Housing and Urban Development (HUD), which is subject to
political influence. In 1995 political motives resulted in the creation by Congress of
“affordable housing credits” that were used by Freddie Mac to purchase of
180
DECEMBER 2010
“subprime mortgage securities”. These new types of securities were developed in
the private financial sector and were backed by mortgages issued to borrowers with
low incomes and unfavourable demographic characteristics.
The purchase of these securities by Freddie Mac as well as other financial
institutions channelled large amounts of funds into the mortgage market, lowered
borrowing costs and induced more low-income earners to buy and mortgage homes.
This result was desired by politicians and many Americans. It also was responsible
for large increases in house prices, which in turn induced speculative buying and
further price increases in a cycle that characterizes all price bubbles. In this case, the
higher house prices further induced many Americans to use their new wealth to
borrow against the increased value of their homes. They used the funds to increase
the purchase of consumption, many of which came from China.
4.2 Financial Institutions:
Commercial and investment banks as well as other financial intermediaries are
constantly developing new institutions and products designed to increase their profits
but which typically also serve the public interest by making capital markets more
efficient as they reduce risks and the cost of making funds flow from ultimate
lenders to ultimate borrowers. However, occasionally these capital market
innovations are accompanied by negative externalities and can endanger the stability
of the entire financial sector.
At least since the early 1990s several financial innovations were adopted
widely, some of which have been considered to be the root causes of the financial
crisis associated with the Great Depression.
First, there was the practice of bundling mortgages, credit card obligations,
consumer and car loans, which served as the backing for new securities that were
liquid and offered high, risk-adjusted returns. The bundling of these credit
instruments was alleged to reduce default risk because of the law of large numbers
and the diversified nature of the borrowers. These securities were bought my many
banks and other financial intermediaries and increased the flow of funds into
mortgage and other credit markets.
Second, there was the growth of the so-called hedge funds and some
insurance companies like AIG, which were largely outside the regulatory framework
for commercial and investment banks. They used sophisticated analytical models to
develop financial derivatives, which promised high returns and increased the
efficiency of the capital market. However, these institutions also deliberately took
on high risks in the expectation of high returns for their firms. In doing so, they
increased their own profits as long as markets were stable, but lost large amounts of
money and threatened the stability of the entire financial system once extra-ordinary
developments took place (Black Swans appeared). Some of these firms took on
these risks in the belief that the government would consider them “too large to fail”
and bail them out to prevent a collapse of the entire financial system.
Some banks purchased mortgage backed securities and financial derivatives
or created affiliates to do so, in fact imitating the behaviour of the unregulated hedge
VOL.7 NO.2
181
funds and issuers of derivative. They too considered the risks of holding these assets
worth the expected returns and were emboldened to do so by the expectation that
they were considered too big to fail.
Third, firms that rate the riskiness of holding securities issued by private firms
and governments failed to meet their responsibilities satisfactorily. In retrospect,
they should have warned investors of the increased risks associated with holding the
securities backed by mortgages and consumer loans. The failure of these rating
agencies to meet what is essentially a fiduciary duty is deplorable, but may be
explained by the general euphoria about economic prosperity and capital markets
before 2007.
The patterns of behaviour by government and financial institutions just
described reflect causes of the Great Depression that are easy to explain and coincide
with populist notion about the shortcomings of big government and big banks. There
is also much truth to the story of how the bursting of the real-estate bubble, the
financial distress of intermediaries and the failure of rating agencies resulted in the
development of the vicious cycle characteristic of all recessions – banks cut back on
lending to sanitize balance sheets, firms without access to credit are forced to curtail
activities and investments and lay off workers, the unemployed reduce spending,
which in turn increases defaults, less lending and spending on investment and
consumer goods and further financial difficulties for banks and producers, and so on.
However, it is not correct to imply that the behaviour of government, banks
and other financial institutions have caused the development of this vicious cycle
and the Great Recession. In the absence of the large demand for the debt issued by
US borrowers coming from China and energy producing countries, capital markets
would have constrained the behaviour of governments, banks and other financial
institutions long before they became as large as they did. In fact, in can be argued
that the behaviour of the allegedly guilty US financial institutions helped the
efficient allocation of the savings that China and the energy producing countries
supplied and that the Fed policies accommodated.
In assessing the role played by financial intermediaries during the last decade,
it must be remembered that financial innovation is a continuing process that raises
productivity just as technical innovation in the manufacturing sector. But neither is
smooth and both lead to periodic crises during which failed innovations are shaken
out of the system. This shake-out took place in most recent times during the collapse
of the high-tech boom at the end of the last millennium and innumerable times
before. The recessions accompanying these events typically are short-lived and an
essential characteristic of free markets.
On the basis of this analysis it follows that the great challenge facing
governments in the wake of financial sector problems is to adopt policies that
eliminate failed innovations while preserving the sector’s incentives and ability to
continue its tradition to innovate and raise economic efficiency and productivity.
But it is also clear that the crisis in the financial markets in the US was not the root
182
DECEMBER 2010
cause of the Great Recession. The blame for that falls on the trade imbalances
discussed above.
At most, the excesses of the financial markets can be blamed for having made
the recession deeper and possibly longer than it would have been otherwise, though
no one can know how long and deep a global recession would have been if US
policies and institutions had not accommodated the flood of capital into the economy
caused by the non-market driven policies of China and the energy-producing
countries.
5. Summary and Policy Implications
This paper presents arguments and empirical data supporting the view that China and
energy producing countries have been responsible for the Great Recession. China
used non-market policies to generate export surpluses and the energy producing
countries accumulated large profits that the owners of the resources invested in
foreign assets to share them with future generations. Both of these accumulators of
financial instruments maintained fixed exchange rates against the dollar while the
value of other major currencies remained flexible. As a result, the entire burden of
supplying financial assets to the surplus countries fell upon the United States.
The easy monetary policies of the US Fed accommodated this demand for
foreign assets and in the process prevented a recession that the payments imbalances
of these countries otherwise would have caused. But the easy monetary policy
induced a real estate bubble and excessive consumer borrowing as well as risky
financial market innovations, all the while the inflow of low-priced consumer
products from China prevented inflation.
When it became obvious that the accumulation of debt by home owners and
consumers was approaching a critical debt to income ratio, the Fed tightened
monetary policy and set off the Great Recession with all of its many adverse effects
on economic well being.
In an important sense, the Great Depression has the same cause as the
recession that gripped the world in the late 1970s, when oil exporters accumulated
large surplus funds. At that time they deposited the funds with banks, which in turn
lent them at excessively easy terms to developing countries that eventually defaulted
on them and brought on the crisis. Important for the present discussion of policy
issues is the fact that the world never found a solution to the problems raised by the
recycling of petro dollars and thus sowed the seeds for the Great Recession 30 years
later.
For this reason it is important that the proper policy implications are drawn
and acted upon this time. Thus, the main policy implication of the preceding
analysis is that the world will again face the threat of recession if the trade surpluses
of China and other countries are not eliminated or reduced substantially. Changes in
government housing policies in the United States and increased regulations of
financial institutions with global reach cannot do the job.
VOL.7 NO.2
183
In the light of these conclusions, what can be done to prevent these problems?
The first-best solution is found in the prevention or at least significant reduction of
the trade surpluses of China and energy producing countries.
Less government control over the economy in China would lead
automatically to higher incomes of workers and returns to capital, more consumer
spending on imports and the purchase of foreign assets. Setting free the exchange
rate against the dollar would complement these changes, which together would
reduce the trade surpluses and restore sustainable balances in international markets
for goods, services and assets. The governments and ruling families of energy
exporting countries can similarly increase domestic demand and allow exchange
rates to float.
These needed changes in government policies are in longer run interest of the
citizens of China and the energy producing countries, as well as the world as a
whole. It remains to be seen whether the surplus countries will adopt the policies,
but at least in China, signs are appearing that they may not have a choice.
In the middle of 2010 in China two major companies have been persuaded to
pay higher wages to their workers. One of them was the electronics giant Foxconn
Technology Group. It had experienced a wave of suicides among its workers, which
were attributed to low wages and excessive work requirements. In response to
public pressure, the company increased average wages by 70 percent. Another
company owned by the Honda Motor Company faced a strike that endangered its
entire, globally integrated production system. Honda settled the dispute by raising
the workers’ wages.
If these wage increases are allowed to spread through the entire economy of
China, domestic spending and imports will increase significantly, resulting in the
desirable reduction in the country’s trade surpluses. The continuation of the slow
increase in the value of the Chinese currency will assist these developments.
The surpluses generated by energy-exporting countries may also diminish
through time as domestic real investment is increased. This development already has
taken place in some Gulf countries, most notably Dubai, Abu Dhabi and Kazakhstan,
which may be accelerated if some of the ventures fail and need continuous support
from the sovereign wealth funds.
If increased domestic spending in the surplus countries does not reduce the
imbalances sufficiently, the world needs a collective effort to prevent a future
recession. This effort might be coordinated by the IMF, which would urge the
surplus countries to let their exchange rates be determined by market forces. The
IMF and OECD might also urge all developed countries into sharing the adjustments
needed to deal with the imbalances caused by the surplus countries by accepting
trade deficits and the sale of financial assets abroad. These adjustments would
require easy domestic monetary and fiscal policies.
The benefits of such policies lie not only in the prevention of a future global
recession but also in the opportunity to increase domestic investments in the capital
184
DECEMBER 2010
importing countries. The proper operation of capital markets will ensure the efficient
allocation of the capital imports between real investment and consumer borrowing.
In the longer run, we may expect that the trade surpluses will shrink
automatically as wages and consumption in China increase, stocks of oil and gas in
the energy producing countries will shrink and new technologies for the production
of energy in the developed countries will reduce the demand for imports.
Unfortunately, efforts to coordinate national macro-economic and exchange
rate policies in the past have met with limited success and are unlikely to do much
better this time. They involve a surrender of national sovereignty that most Western
countries, but especially the United States, are unwilling to engage in. The design of
optimum monetary policies is difficult enough without constraints from international
agreements. Even if the political will existed to have a collective sharing of the
burden imposed by the surplus countries, the practical implementation will be very
difficult.
When first- and second-best solutions to economic problems cannot be
adopted, the world tends to muddle through by adopting elements of both – until the
problems arise again in a previously unknown guise. Such periods of muddling
through are not necessarily bad. During the 30 years following the crisis of the
recycled petro-dollars in the 1970s the world experienced unprecedented prosperity
and growth in output that dwarfs the costs of a short-lived recession. Maybe
muddling through yet again will produce similarly good results.
Notes
1
The author is Professor of Economics (Emeritus) at Simon Fraser University and
Senior Fellow at the Fraser Institute. His thanks for comments on earlier versions of
this paper go to John Taylor, Allan Meltzer, Miranda Xafa, James Dean, Robert
Mundell, Jacob Frenkel and other participants at conferences organized by the
Athenian Policy Forum in Frankfurt, Germany; by the Alamos Alliance in Alamos,
Mexico; by Robert Mundell in Santa Colomba, Italy and at the III Astana Economic
Forum in Astana, Kazakhstan. email: [email protected]
2
The Taylor rule is: r = p + .5y + .5(p - 2) + 2 , where r is the federal funds rate, p is
the rate of inflation over the previous four quarters and y is the percentage deviation
of real GDP from a target.
3
See Taylor (2007a) and (2007b) for detailed information about and references to
studies that compared actual monetary policies to those implied by the Taylor rule.
4
The dotted line was derived by smoothing a set of numbers calculated by the
application of the Taylor rule to the actual data. It replicates the image found on
Taylor (2009), page 3. The graph reproduced there was published in The Economist,
October 18, 2007, which in turn drew on the image in Taylor (2007).
5
Figure 1 uses the latest available data on the Federal Funds Rate, which were not
available when the graph was first produced and later presented in Taylor (2009).
VOL.7 NO.2
185
The graph shows the typical reaction of the Fed to the development of a recession.
The rates dropped sharply to a historically unprecedented low target level of 0-.25
and an actual rate of .15 percent.
6
There is no page number in the document Paulson (2008).
7
The annual flows represent current account balances, which predominantly reflect
the net of exports minus imports of goods and services, the rest consisting of
relatively minor unilateral transfers.
8
The oil exporting countries in the study are: Algeria; Angola; Azerbaijan; Bahrain;
Congo, Republic of; Ecuador; Equatorial Guinea; Gabon; Iran, Islamic Republic of;
Kuwait; Libya; Nigeria; Norway; Oman; Qatar; Russia; Saudi Arabia; Syrian Arab
Republic; Turkmenistan; United Arab Emirates; Venezuela, Rep. Bol.; Yemen,
Republic of. The countries in the group “Emerging Asia” are: China,P.R.: Mainland;
Hong Kong; Indonesia; Korea; Malaysia; Philippines; Singapore; Taiwan; Thailand.
9
Taylor (2009) pp. 6-7 presents data produced by the IMF that show the equality of
global savings and investment between 1970 and 2004. While these data show small
discrepancies between the two magnitudes that are due to measurement errors, the
main goal of the IMF study was to discuss the decline of global savings and
investment as a percent of world GDP during the period. These data cannot be used
to refute the notion that there has been a glut of savings in one part of the world that
was matched by dissavings in another part of the world, the phenomenon that is
argued by Bernanke, Paulson and other mentioned in the text to have been the basic
cause of the current crisis.
References
Bernanke, Ben S. (2005), “The Global Saving Glut and the U.S. Current Account
Deficit”, presented as the Sandridge Lecture, Virginia Association of
Economics, Richmond, Virginia on March 10, 2005, found at
http://www.federalreserve.gov/boarddocs/speeches/2005/200503102/
Blanchard, Olivier (2009), “What is needed for a lasting recovery”, Financial Times,
June 18, found at http://www.ft.com/cms/s/0/e8bcc516-5c33-11de-aea300144feabdc0.html
International Financial Services London (IFSL) (2009), Sovereign Wealth Funds
2009, found at http://www.google.com/search?hl=en&q=ifsl+swf&aq=
0p&oq=ifsl&aqi=g%3Ap2g3g%3As3g2
Meltzer, Allan (2002), A History of the Federal Reserve, Volume 1, 1913-1951,
University of Chicago Press, January. Volume 2, forthcoming. This volume
includes a Postscript entitled “The Global Financial Crisis of 2007-9”, which
Meltzer presented as the keynote address at the meetings of the Alamos
Alliance in Alamos, Mexico in February 2009
186
DECEMBER 2010
Milesi-Ferretti, Gian Maria (2007), “IMF Offers Compromise Path on Imbalances”,
IMF Survey Magazine, August 7, p. 1
Paulson, Henry M (2008), Remarks at the Ronald Reagan Presidential Library,
November 20, US Department of the Treasury, HP-1285, found at
http://www.treas.gov/press/releases/hp1285.htm
Schwartz, Anna J., (2009), “Origins of the Financial Market Crisis of 2008”, Cato
Journal, Vol. 29, No. 1 (Winter), 19-23
Taylor, John (1993), Macro Economic Policy in a World Economy: From
Econometric Design to Practical Application, New York: W.W.Norton
----------------- (2007a), “Housing and Monetary Policy”, presented at a conference in
Jackson Hole, Wyoming in the summer of 2007. The paper can be found at
http://www.stanford.edu/~johntayl/Housing%20and%20Monetary%20Policy-Taylor--Jackson%20Hole%202007.pdf
--------------- (2007b), “The Explanatory Power of Monetary Policy Rules”, The
Adam Smith Lecture, Annual Meeting of the National Association of
Business Economics, September 10, San Francisco, found at
http://www.stanford.edu/~johntayl/Adam%20Smith%20Lecture--Taylor-2007.pdf
----------------- (2009), GETTING OFF TRACK: How Government Actions and
Interventions Caused, Prolonged, and Worsened the Financial Crisis,
Stanford, Calif.: Hoover Institution Press, Stanford University
Wolf, Martin (2006), “Do not believe everything you hear about global imbalances”,
Financial Times, March 20, found at http://www.ft.com/cms/s/0/c5960930bec0-11da-b10f-0000779e2340.html

JEA-Vol.7-No.2 - Athenian Policy Forum

Transcription

Similar documents

Acronyms Used