The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Monday, June 14,2010
Phil Angelides
Chairman
Via E-mail and FedEx
Professor Pierre-Olivier Gourinchas
691A Evans Hall, #3880
Economics Department
University of California, Berkeley
Berkeley, CA 94720
pog(a{berkeley.edu
Hon. Bill Thomas
Vice Chairman
Re:
Follow-up to the Financial Crisis Inquiry Commission Forum
Dear Dr. Gourinchas:
Brooksley Born
Commissioller
Byron S. Georgiou
Commissioner
Senator Bob Graham
Commissioner
Keith Hennessey
The Financial Crisis Inquiry Commission thanks you once again for your
participation in the "Forum to Explore the Causes of the Financial Crisis" on
February 26 and 27, 2010.
Enclosed are follow-up questions which were posed by the Commissioners
during the forum, as well as additional questions which have arisen over the
course of our investigation which we would like your assistance in answering.
Please respond to the questions by Friday, July 2,2010. If you have any
questions, or would like more information, please contact Scott Ganz at
sganz@fcic.gov.
Commissioller
Douglas Holtz-Eakin
Commissioner
Heather H. Murren, CFA
Commissioner
1. What were the mechanisms by which the global and safe asset imbalances
affected the home mortgage market in particular? By which they affected the
market for other assets?
2. In the 1970s the United States had a period of influx of petrodollars that
led to extensive sovereign lending and subsequent loan defaults. Is there a pattern
of global inflows and safe asset imbalances that can be seen in various countries
at various times in history? Did bubbles often result?
John W. Thompson
Commissioner
Peter J. Wallison
3. Please provide us with any data that you think would be of use to the
Commission regarding the quantity and nature of international capital flows, both
in gross and net form.
Commissioner
4. You stated that a growing belief in self-regulatory financial markets,
which you call the "Greenspan doctrine," has been a casualty ofthe current
financial crisis. Please explain how that belief contributed to the financial crisis
and how the belief developed and influenced policy making and market oversight.
Wendy Edelberg
Executive Director
1717 Pennsylvania Avenue, NW, Suite 800 • Washington, DC 20006-4614
202.292.2799 • 202.632.1604 Fax
5. One topic of great interest to the Commission is the extent to which the crisis was an
international crisis. Can you provide the Commission information regarding the differential
impact of the crisis on various countries, these countries' responses to the crisis, and the
effectiveness of these responses?
Sincerely,
Wendy Edelberg
U.S. Monetary Policy, ‘Imbalances’ and the Financial Crisis
Answers to follow-up questions, sent June 14, 2010
Pierre-Olivier Gourinchas
UC Berkeley, NBER and CEPR
1. What are the mechanisms by which the global and safe asset imbalances affected the
home mortgage market in particular? By which they affected the market for other
assets?
The U.S. mortgage market responded to the growing global demand for safe liquid debt
instruments through two main channels. First, as argued in my prepared remarks, a substantial
share of the global demand for safe assets was channeled to the U.S. financial markets, seen as
offering unparalleled liquidity and security (U.S. government and agency securities, triple-A
corporate bonds). This surge in demand pushed down the yields and increased the price of U.S.
safe assets. As the price of safe assets rose, financial market participants reallocated their
portfolio away from these assets –perceived as too expensive or offering too low returns - and
towards riskier assets. In turn, this reallocation of demand towards risky assets increased the
price of risky assets and reduced their expected return. Under unchanged market perceptions of
risk, the decline in the return on risky assets would be similar in size to the decline in return on
safe assets. The risk premium would have remained constant and the entire yield curve would
have shifted down. 1
In other words the increased global demand for U.S. safe assets triggered an increase in the
demand for risky assets (U.S. or otherwise). The associated decline in broad market rates had a
direct and positive impact on the demand for U.S. mortgages. Figure 1 illustrates this effect. It
reports the spread between the 30-year fixed rate mortgage and the 10-year Treasuries. Between
1
This is a good characterization of market developments after 2004, when the demand for U.S. safe assets is partly
fuelled by the growing savings of China and oil producers (see my prepared remarks p.20).
1
2004 and the onset of the financial crisis, the spread remains very stable, around 157 bp. In terms
of levels (reported in Figure 3 of my prepared remarks), 30-year fixed rate mortgage rates
remained quite low, between 5.8 percent and 6.7 percent, and this, despite the sharp increase in
policy interest rates over the same period. This first channel affects all asset classes, including
but not limited to mortgage assets.
The second channel comes from the endogenous response of the U.S. financial system to the
surge in the global demand for safe assets: it actively sought to increase their supply. How? The
main vehicle was the securitization of physical assets (e.g. cash CDOs) as well as the creation of
synthetic assets (e.g. synthetic CDOs) with well-defined risk profiles. The triple-A rate tranches
of structured credit instruments was considered as safe as U.S. Treasuries. Churning out large
quantities of triple-A rated securities was a way to increase the supply of “quasi” safe U.S.
assets. But this process initially required large quantities of the underlying physical assets that
could be systematically securitized. The demand for the underlying credit instruments surged,
especially for mortgage assets given the importance of mortgage-backed-securities in the overall
market for structured credit products. 2 By repackaging risky assets into supposedly “safe”
bundles, this process directly increased the demand for the underlying mortgages instrument,
pushing further down mortgages rates (i.e. reducing the risk premium) and fueling house price
increases. As documented elsewhere, lax regulation, irresponsible underwriting practices, the
broad failure of rating agencies, as well as incentive problems, especially on the part of the
mortgage servicers and originators, made it possible to manufacture a large quantity of these
2
In his presentation to the FCIC roundtable, Chris Mayer reports that the annual issuance of non-agency MBS
jumped from 400 billion of USD in 2003 to more than 800 billion in 2005.
2
“quasi” safe assets in a relatively short period of time. 3 This second channel also had an
important impact on other markets, as triple-A securitized assets became the dominant form of
collateral on the repo market, and playing an important role in the shadow banking system. 4 As
the amount of collateral increased, the overall market became more liquid. With the onset of the
crisis, the value of that collateral came into question, forcing a run on the repo market.
The first channel was sufficient to get a housing boom going, as evidenced by the fact that many
countries experienced a residential housing boom over that period, irrespective of the degree of
securitization of their underlying mortgage market. 5 The second channel fanned the flames of the
U.S. housing boom, spread ‘toxic’ assets through the U.S. and global financial system and
created the conditions for a generalized financial crisis to occur.
3
See C. Mayer’s prepared remarks for the FCIC forum.
4
See G. Gorton’s prepared remarks to the FCIC forum.
5
See C. Mayer’s prepared remarks for the FCIC forum, figure 1.
3
2. In the 1970s, the United States had a period of influx of petrodollars that led to
extensive sovereign lending and subsequent loan defaults. Is there a pattern of
global inflows and safe asset imbalances that can be seen in various countries at
various times in history? Did bubbles often result?
The short answer is a qualified Yes. The increase in the price of oil following the 1973 and 1979
oil shocks transferred vast resources to oil-producing countries. It was undesirable --and in any
case difficult in the short run-- for these countries to let their internal demand adjust to the surge
in oil revenues. Instead, they started to run large external (current account) surpluses. The
corresponding capital outflows (the so-called “petrodollars”) were invested in the US and other
major financial markets. That part of the story is similar to the recent crisis.
However, unlike the recent crisis, monetary instability played a key role in the 1970s. That
period was marked by significant inflation in the US and other parts of the world. This high
inflation was partly the consequence of the oil shocks themselves, but also reflected the
widespread –and largely correct-- perception that most monetary authorities were unwilling to
tackle the rise in the cost of living. Inflation and lax monetary policy combined to generate very
low –even negative—global real interest rates during that period. It is this factor –an extended
period of low real interest rates-- that accounts for the period of instability that followed. Low
real interest rates often lull investors –private and public-- into borrowing excessive amounts and
can fuel financial bubbles and instability. The crisis occurred when the U.S. Federal Reserve –
under chairman Volcker— tightened dramatically monetary policy, increasing real interest rates
in the U.S. and elsewhere.
4
Hence, the root causes of the Latin American sovereign debt crisis of the early 1980s are quite
different from the current situation, but some elements are common. In particular, that episode is
also characterized by low real interest rates. Other recent episodes marked by low real interest
rates include:
-The Greek/euro crisis of 2010, where cheap financing became available in many peripheral euro
zone countries (Spain, Portugal, Ireland, Greece) following the adoption of the Euro in 19992001.
-The Argentinean crisis of 2001 where cheap financing became available following the adoption
the convertibility law in 1991 fixing the value of the Argentinian peso in terms of the US dollar
-The Mexican crisis of 1994 where cheap financing became available after the adoption of a
crawling peg between the Mexican peso and the US dollar, and later the adoption of NAFTA.
Many more examples are documented in the masterful recent of work Reinhart and Rogoff
(2010). 6 From a policy perspective, the body of empirical evidence suggests that policymakers
should monitor with extreme caution periods of cheap financing. Regardless of their source,
these episodes tend to end badly.
6
Reinhart and Rogoff, This Time is Different, Princeton University Press, 2010
5
3. Please provide us with any data that you think would be of use to the Commission
regarding the quantity and the nature of international capital flows both in net and
gross form
The spreadsheets attached to this document provide publicly available data on net and gross
capital flows into the U.S. Below is a short description of each spreadsheet.
File Name
Source
Remarks
Used for figure:
(R: prepared remarks; P:
presentation; F: follow-up)
ITA_table.xlsx
Bureau of Economic
Analysis International
Transaction. Tables 1,
8a and 8b.
Glob_imbalances.xlsx
World
Development
Indicators,
World
Economic
Outlook,
Statistical Yearbook of
the Republic of China,
Deutsche
Bank,
International Financial
Statistics and OECD
Economic Outlook.
Treasury_survey_2010.xlsx US Treasury Survey.
6
Gross and Net US P9, P10
capital flows.
1990-2008, annual.
Current
account P1.
balances by region as a
fraction of world GDP.
Quarterly, 1980-2009
Foreign
Portfolio
Holdings
of
US
Securities. Historical
data, 1974-2009
4. You stated that a growing belief in self-regulatory financial markets, which you call
the “Greenspan doctrine” has been a casualty of the current financial crisis. Please
explain how that belief contributed to the financial crisis and how the belief
developed and influenced policy making and market oversight.
This is a complex question and I will not pretend to give it here a comprehensive treatment.
What I will do is offer some elements of reflection that illustrate how periods of relative
economic prosperity and stability can lead to a weakening of regulatory standards.
The “Greenspan doctrine” is associated with the view –expounded on numerous occasions by
former Federal Reserve chairman Alan Greenspan—that markets have a natural tendency to
self-correct and regulate and that government regulation is likely to be detrimental. An
illustration of these views is contained in remarks delivered on April 12, 1997 at the Annual
Conference of the Association of Private Enterprise Education on the evolution of banking in
a market economy. There, chairman Greenspan stated:
“It is most important to recognize that no market is ever truly unregulated in that the selfinterest of participants generates private market regulation. Counterparties thoroughly
scrutinize each other, often requiring collateral and special legal protections; self-regulated
clearing houses and exchanges set margins and capital requirements to protect the interests of
the members. Thus, the real question is not whether a market should be regulated. Rather, it
is whether government intervention strengthens or weakens private regulation, and at what
cost.”
Later on in that same speech, Greenspan adds:
“To a significant degree, attitudes toward banking regulation have been shaped by a
perception of the history of American banking as plagued by repeated market failures that
ended only with the enactment of comprehensive federal regulation. The historical record,
however, is currently undergoing a healthy reevaluation. In my remarks this evening I shall
7
touch on the evolution of the American banking system, focusing especially on the pre-Civil
War period, when government regulation was less comprehensive and less intrusive and
interfered less with the operation of market forces. A recent growing body of research
supports the view that during that period market forces were fairly effective in assuring that
individual banks constrained risktaking to prudent endeavors. Nonetheless, the then nascent
system as a whole proved quite vulnerable to various macroeconomic shocks essentially
unrelated to the degree of banking regulation.”
In a concise form, these two paragraphs delineate the contours of a regulation framework.
First, there is a distinction between systemic and non-systemic events: Market forces should
tackle idiosyncratic or non-systemic risks; government intervention should deal exclusively
with “macroeconomic” or systemic risks. This is largely non-controversial.
Second, and this is much more controversial, is the notion that self-regulation to avoid
idiosyncratic failures –e.g. the establishment of self-regulated clearing houses, margins and
capital requirements etc…-- can reduce the exposure to systemic risk, by “assuring that
individual banks constrained risktaking to prudent endeavors”. Under this view, there is little
need for ex-ante regulation of the financial system. On the contrary, all such regulations are
likely to reduce the efficiency of the financial system, and lower standards of living.
Moreover, such regulation is unlikely to reduce further systemic risk, hence the statement
that “the nascent system as a whole proved quite vulnerable to various macroeconomic
shocks essentially unrelated to the degree of banking regulation.” If systemic risk is unrelated
to the degree of banking regulation, while market forces reduce systemic risk, the solution is
to let markets provide as much self-regulation as they want, nothing more. If a systemic crisis
nonetheless occurs, the Federal Reserve should provide ex-post assistance. In other words,
8
the regulatory framework should provide light regulation during the upswing phase of the
bubble, followed by a “mopping up” after the collapse.
In light of the crisis, it is this second notion -- that financial innovations directed at reducing
idiosyncratic risks will also reduce systemic risk-- that has been seriously damaged. Consider
one such innovation: securitization. It is obvious that securitization allowed financial
institutions to better control and manage their risk profile, thereby reducing idiosyncratic
risks. But it would be more difficult to argue that it also reduced systemic risk. On the
contrary, when the crisis came, the complexity and opacity of structured credit instruments
vastly exacerbated counterparty risk. As chairman Greenspan himself stated in his remarks to
the Economic Club of New York in February 2009,
“But in August 2007, the risk management structure cracked. All of the sophisticated
mathematics and computer wizardry essentially rested on one central premise: that
enlightened self interest of owners and managers of financial institutions would lead them to
maintain a sufficient buffer against insolvency by actively monitoring and managing their
firms’ capital and risk positions. When in the summer of 2007 that premise failed, I was
deeply dismayed.”
The view that market forces could act in a self-correcting way was reinforced in the years
preceding the crisis by two observations. First, there was the perception that the pace of
financial innovation rendered many older regulations obsolete. To quote again from
Greenspan’s 1997 speech,
“[…] I should like to emphasize that the rapidly changing technology that is rendering much
government bank regulation irrelevant also bids fair to undercut regulatory efforts in a much
wider segment of our economy. The reason is that such regulation is inherently conservative.
[…] With technological change clearly accelerating, existing regulatory structures are being
bypassed, freeing market forces to enhance wealth creation and economic growth. In finance,
regulatory restraints against interstate banking and combinations of investment and
commercial banking are being swept away under the pressures of technological change. […]
9
As we move into a new century, the market-stabilizing private regulatory forces should
gradually displace many cumbersome, increasingly ineffective government structures. This is
a likely outcome since governments, by their nature, cannot adjust sufficiently quickly to a
changing environment, which too often veers in unforeseen directions.”
The pace of financial innovation was seen as an argument in favor of less, not more,
regulation. We now know that some of the financial innovations were re-creating the
financial risks of another age. For instance, G. Gorton’s testimony to the FCIC illustrates
how the repo market became a modern version of the unregulated fractional banking system
that the US used to have before the establishment of the Federal Reserve Act. If anything,
faster financial innovation requires more not less vigilance.
Second, the period between 1990 and 2007 saw the U.S. and the world economy weather
successfully a string of “crisis” such as the 1994 Mexican crisis or the 1997-98 Asian
financial crisis, or the 1998 collapse of Long Term Capital Management. None perhaps, was
as important as the 2001 implosion of the dotcom bubble. The relatively mild impact on
economic activity convinced the US monetary authorities that it had the capacity to handle
and prevent –through traditional monetary policy-- a systemic crisis.
Accordingly, the benign outlook on financial regulation was reinforced both by the speed of
financial innovation –which should have been a source of added vigilance—and the fact that
the world and US economies seem to be moving along without major disruptions.
In closing, I would like to re-emphasize that while better financial regulation could have
strengthened the US financial sector, it is not clear how/whether the crisis would have been
10
avoided altogether. The reason is that the excess demand for safe assets and the resulting
pressures on world interest rates would have been present even with a better regulated
financial system. The conditions for a financial bubble to emerge were ripe. Perhaps the best
one could hope for is to insulate the core of the financial system, so as to avoid a complete
meltdown.
5. One topic of great interest to the Commission is the extent to which the crisis was an
international crisis. Can you provide the Commission information regarding the
differential impact of the crisis on various countries, these countries’ responses to
the crisis and the effectiveness of these responses
One can think of two main channels of transmission of the crisis from the US to the rest of the
world: financial and trade channels. 7 The first channel of financial transmission was exposure to
U.S. assets backed by sub-prime mortgages. Foreign financial institutions, especially in Ireland,
France, Germany, Switzerland and the United Kingdom, were substantially exposed. When
markets became unable to price these assets, these foreign financial institutions suffered great
losses that endangered these countries’ financial system and required prompt action.
7
There is an active empirical literature looking at the transmission of the crisis to the rest of the world. My answers
to this question rely on the work of P. Lane and G.M. Milesi-Ferretti “The Cross-Country Incidence of the Global
Crisis”, forthcoming IMF Economic Review (2010), O. Blanchard, M. Das and H. Faruqee, “The Initial Impact of
the Crisis on Emerging Market Countries”, Brookings Papers on Economic Activity (2010) and D. Giannone, M.
Lenza and L. Reichlin, “Market Freedom and the Global Recession”, forthcoming IMF Economic Review (2010).
11
A second channel of financial transmission is through global deleveraging. As investors around
the world pulled back, many foreign financial institutions found themselves unable to obtain
funding, especially dollar funding. This dollar shortage was especially acute for European
financial institutions with short term exposure to dollar markets. The retrenchment of global
investors also had a devastating impact on emerging equity markets.
Finally, as economic activity slowed down markedly in industrial countries, demand for durable
and investment goods collapsed.
Export-oriented countries (Singapore, China, Japan,
Germany….) suffered dramatic collapses in trade. The global downturn also reduced demand for
commodities. The associated decline in commodity prices affected the economy of commodity
producing countries.
The empirical evidence available so far suggests the following: everything else equal, the crisis
had a larger impact on more advanced economies, countries with more liberalized credit markets
or more rapid growth of credit prior to the crisis, countries with larger external deficits, and more
open economies. There does not seem to be a sizeable effect of having a larger stock of official
reserves, or of having a fixed exchange rate.
The policy responses to the crisis fall broadly into three areas: financial policy, monetary policy
and fiscal policy. On the financial front, most governments acted quickly to provide liquidity to
their financial system and recapitalize or take over insolvent financial institutions. This was
coupled with aggressive and often coordinated monetary policy that dramatically lowered policy
rates in many countries. With many advanced countries approaching the zero nominal bound (i.e.
12
the fact that it is policy rates have to remain positive), traditional monetary policy was
supplemented by non-traditional policies that often involve making use of the central bank
balance sheet. In some countries, these interventions took the form of quantitative easing,
whereby the central bank increases the money supply and expands its balance sheet by buying
government securities and other assets (e.g. the U.K.). In others such as the U.S., the Federal
Reserve acquires certain types of assets (such as MBS) where private markets are impaired in
order to reduce spreads and speed up a return to normalcy. In addition, many countries, in
coordination with the IMF and the G-20, agreed to implement sizable fiscal stimulus plans to
supplement monetary policy. The consensus is that the combination of forceful and coordinated
monetary and fiscal policies stabilized the world economy and allowed some return to normalcy
to occur. The outcome is far from uniform, however. At one end of the spectrum, countries like
China have recovered vigorously and are now trying to slow down their economy. At the other
end, the generalized re-pricing of risk in financial markets is still going on, and some Southern
European countries (Greece, Spain, Portugal) find themselves in a very difficult financial
situation. While this broad assessment is widely shared, it is still too early to assess precisely
which policies worked and where.
13
3.5
3
Percent
2.5
2
1.5
1
0.5
AAA‐10yr treas
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
0
30 yr mtg‐10yr treas
Figure 1 Spread between AAA-corporate bonds and 10-year Treasuries and between 30 year fixed rate mortgage rate and
10 year Treasuries. Source: Federal Reserve, series H.15.
14
@FedFRASER