The Great Depression and the Great
Recession: What Have We Learnt?* Michael D. Bordo
Introduction
The Financial Crisis of 2007-2008 and the Great
Recession of 2007-2009 are now in the past although
the US economy is still recovering but at an abnormally
slow pace, and Europe is in recession again following
the debt crisis of 2010-2011. During the worst of the
recent financial crisis/Great Recession many observers
made comparisons between that event and the Great
Depression. In this lecture I re-evaluate the experience
of the two events. I raise and answer five questions:
1. What is similar between now and then?
2. What is different?
3. What were the monetary policy experiences that
came out of the 1930s experience?
4. Which of them were of value in dealing with the
recent crisis?
5. What do we need to learn from the recent
experience?
2. Some Comparisons Between Now and
Then: The Similarities
2.1 The Downturn
The first point to note is that in terms of the
decline of the real economy as measured by real GDP
or Industrial Production or Unemployment, the Great
Recession was a relatively minor event. Between 1929
and 1933 real GDP fell in the US by close to 30 per
cent, whereas between 2007-2009 it fell by a little
over 5 per cent (Chart 1). Unemployment in the US
peaked at 25 per cent in 1933 versus a little above 10
per cent in 2009 (Chart 2). Other advanced countries
experiences were similar. For a comparison between world industrial production in the 1930s and recently
see Chart 3.
|
Secondly, both the recent financial crisis and the
Great Depression were global financial crises. Bordo
and Landon Lane (2010a) demarcated global financial crises for a large panel of countries from 1880 to the
present using cluster analysis. We identified 7 such
events of which 2007-2008 was one of the mildest.
Its global incidence was considerably less than the
1930s (Chart 4). Moreover, we found that the mean
weighted cumulative per centage loss of real output
for countries with financial crises in 2007-2008 which
was -2.95 per cent was one third of the comparable
measure for the early 1930s which was -9.35 per cent.
These data, which suggest that the recent
experience wasn’t that bad, doesn’t answer the question – ‘what would have happened if the Federal
Reserve and other central banks didn’t follow the
aggressive policies that they did?’ Nor if there hadn’t
been in place other elements of the financial safety
net, such as deposit insurance and various automatic
stabilisers.
|
A third point of similarity is the nature of the
recoveries after the trough of the recession. Both
episodes had sluggish recoveries in terms of the real
economy expanding after the business cycle trough at a
slower pace than the downturn. The recovery after 1933
was very rapid but not quite sufficient to completely
reverse the preceding downturn. One plausible
explanation for the incomplete recovery in the 1930s is
that it was impeded by the New Deal National Industrial
Recovery Act (NIRA) policy which attempted to cartelise
both goods and product markets (Cole and Ohanian,
2004). The recent recovery is also sluggish and has been
labeled a ‘jobless recovery’ as was the case for the two
preceding recessions (Charts 5 and 6).
There are several explanations for the sluggish
recent recovery including Reinhart and Rogoff (2009)
who see this pattern as typical of countries that have
had big banking crises. My recent work with Joseph
Haubrich (2011) finds that financial stringency does
not explain the difference between this recovery and
previous recoveries that had financial crises. We find
that the collapse of residential investment which proxies for the collapse of the housing sector is the
key determinant of the shortfall.
A fourth similarity between now and then is that
both episodes were preceded by asset price booms
and busts. There was a housing boom and bust in
the 1920s and the Wall Street boom and crash in
1929 (Charts 7 and 8). In the recent crisis, it was the
subprime mortgage related housing boom that burst
in 2006 that triggered the crisis. Unlike the 1920s,
the tech boom of the early 2000s did not precipitate a
financial crisis (Charts 9 to 11).
2.2 The Key Difference between the Two
Crisis Episodes was in the Nature of the
Banking Crisis
The 1930s episode was an old-fashioned liquidity
based banking crisis brought on by the Fed’s failure to
serve as lender of last resort. By contrast, the recent
crisis was more of a late twentieth century solvency
driven banking crisis. Although Gary Gorton (2010)
has argued quite forcefully that there was an oldfashioned
panic in the repo market and other aspects
of the shadow banking system.
Federal Reserve tightening to stem the Wall
Street boom beginning in early 1928 led to the
downturn which began in August 1929 followed by
the stock market crash in October. However, there is
considerable evidence to refute the view that the Wall
Street crash caused the Great Depression.
The real problem arose with a series of banking
crises that began in October 1930 and ended with
the Banking Holiday of March 1933. Milton Friedman
and Anna Schwartz (1963) posited that the panics by
reducing the deposit currency and deposit reserve ratios reduced the money multiplier and hence the
money supply (Charts 12 to 16).
The panics which began in 1930 and worsened
in 1931 when it became global, reflected a contagion
of fear as the public converted their deposits into
currency, i.e., they hoarded currency. The public staged
a series of runs on the banking system leading to
massive bank suspensions. In modern terms, there was
a big liquidity shock. The collapse of the money supply
led to a decline in nominal spending and, in the face of
sticky wages, a decline in employment and output.
The process was aggravated by banks dumping
their earning assets in a fire sale and by debt-deflation
as falling prices increased the real burden of debt
leading to insolvencies of banks with initially sound
balance sheets. Bernanke (1983) also regarded the
banking panics of the 1930s to be the key cause of
the Great Contraction. Bank failures crippled the
mechanism of financial intermediation. This effect
can be seen in the quality spread (the Baa less 10 year
US Treasury bond rate) (Chart 17).
There is considerable debate over whether the
clusters of bank failures in the early 1930s were really driven by contagious liquidity shocks as Friedman and
Schwartz (1963) argued, or whether the bank failures
reflected an endogenous response to the downturn
that was caused by other non-monetary forces, as
posited by Peter Temin (1976) and recently by Charles
Calomiris and Joseph Mason (2006).
Recent evidence by Gary Richardson (2007) using
newly unearthed Federal Reserve bank examiners’
analyses of all the bank failures that occurred in
the early 1930s, finds that illiquidity shocks largely
explain what happened in the banking panic windows
identified by Friedman and Schwartz (1963). Bordo and Lane (2010b) conducted an econometric study of
the banking panics of the 1930s to ascertain whether
it was illiquidity or other factors including insolvency
that largely explained the panics.
We estimated a structural VAR using data on
bank failures due to illiquidity, total bank failures,
M2, unemployment and the quality spread. Chart
18 shows the historical decompositions from the VAR
model. Historical decompositions are a way to engage
in counterfactual experiments by showing the impact
of one variable holding the others constant. The chart
shows that the liquidity shock mimics the actual data
quite well for all of the crisis windows.
The upshot of the banking panics according
to Friedman and Schwartz (1963), Meltzer (2003),
Bernanke (1983) and Wicker (1996), is that, they
represented a major Fed policy failure. The Fed which
was founded in 1913, in large part to be a lender of
last resort to the banking system, failed in its duty.
The Fed could have prevented the Great Contraction
by using expansionary open market policy, a tool
which was familiar to Fed officials.
3. The Recent Crisis
The crisis of 2007-2008, like 1929-33, started
with an asset price boom that later bust. The collapse
of the subprime mortgage market led to a panic in
the shadow banking system which was not regulated by the Fed nor covered by the financial safety net.
These institutions (e.g., Goldman Sachs, JP Morgan,
Bear Stearns, Lehman Brothers) which had expanded
after the repeal in 1999 of the Depression era Glass
Steagall Act which had separated investment banking
from commercial banking, had much more leverage
than traditional banks and were much more prone to
risk. When the crisis hit, they were forced to engage in
major deleveraging involving a fire sale of assets into
a falling market. This lowered the value of their assets
and those of other institutions. A similar negative
feedback loop had occurred in the 1930s. Gorton
(2010) posits that the crisis started in the repo market
which had been collateralised by opaque (subprime)
mortgage-backed securities by which the investment
banks and universal banks had been funded. The repo
crisis continued through 2008 and then morphed into
an investment/universal bank crisis after the failure
of Lehman Brothers in September 2008.
The crisis led to a credit crunch which led to a
serious recession. The effects of the credit crisis can
be seen in the spike of the quality spread in the fall
of 2008. It looks similar to what happened in 1931
(Chart 19). However the recent crisis was not a
contagious banking panic. There was no collapse
in money supply brought about by the collapse of
the deposit currency ratio as had occurred in the
1930s. M2 didn’t collapse. Indeed it rose, reflecting expansionary monetary policy (Chart 20). The depositcurrency
ratio rose (Chart 21). There was no run on the
commercial banks because depositors knew that their
deposits were protected by federal deposit insurance
which had been introduced in 1934 in reaction to
the bank runs of the 1930s. The deposit-reserve
ratio declined reflecting an expansionary monetary
policy-induced increase in the banks excess reserves
rather than a scramble for liquidity as had occurred in
the 1930s (Chart 22). The money multiplier declined
in the recent crisis (Chart 23) reflecting a massive expansion in the monetary base (Chart 24) seen in the
Fed’s doubling of its balance sheet in 2008. Moreover,
although a few banks failed between 2007-2010 the
numbers of failures and the size of the losses were
small relative to the 1930s. Thus, the recent financial
crisis was not driven by a Friedman and Schwartz type
banking panic. But there was a panic in the shadow
banking system and it was driven more by insolvency
than by contagious illiquidity considerations.
The problem stemmed from the difficulty
of pricing securities by a pool of assets where the quality of individual components of the pool varies
and unless each component is individually examined
and evaluated, no actual price can be determined. As
a result, the credit market, confronted by financial
firms whose portfolios were filled with securities of
uncertain value, derivatives that were so complex
the art of pricing them had not been mastered, was
plagued by the inability to determine which firms
were solvent and which were not. Lenders were
unwilling to extend loans when they couldn’t be sure
that a borrower was creditworthy. This was a serious
shortcoming of the securitisation process that was
responsible for the paralysis of the credit market.
4. The Response to the Crisis
The Federal Reserve had learned the Friedman
and Schwartz lesson from the banking panics of the
1930s of the importance of conducting expansionary
open market policy to meet all demands for liquidity
(Bernanke, 2002). In the recent crisis, the Fed
conducted highly expansionary policy in the fall of
2007 and from late 2008 to the present. Fed policy
was likely too cautious in the first three quarters of
2008 seen in high real interest rates and flat growth
in the monetary aggregates (Hetzel, 2009). However,
from the last quarter of 2008 to the present, Fed policy
has been highly expansionary as it pushed the federal
funds rate close to zero and embarked on an aggressive
policy of quantitative easing.
Also, based on Bernanke’s (1983) view that the
1930s banking collapse led to a failure of the credit
allocation mechanism, the Fed, in conjunction with
the Treasury, developed a plethora of extensions to
its discount window, referred to by Goodfriend (2011)
as credit policy, to encompass virtually every kind of
collateral in an attempt to unclog the credit markets.
Finally, another hallmark of the recent crisis
which was not present in the Great Contraction is that
the Fed and other U.S. monetary authorities engaged
in a series of bailouts of incipient and actual insolvent
firms that were deemed too connected to fail. These
included Bear Stearns in March 2008, the Government
Sponsored Enterprises (GSE’s) (Fannie Mae and
Freddie Mac) in July 2008 and American International
Group (AIG) in September. Lehman Brothers had been
allowed to fail in September on the grounds that it
was both insolvent and not as systemically important
as the others and as was stated well after the event
that the Fed did not have the legal authority to bail it
out.
Indeed, the deepest problem of the recent
crisis was not illiquidity as it was in the 1930s but
insolvency and especially the fear of insolvencies
of counterparties. This has echoes in the
correspondent-induced banking panic in November
1930 (Richardson 2006). But very different from
the 1930s, the too-big-to-fail doctrine which had
developed in the 1980s ensured that the monetary
authorities would bail out insolvent large financial
firms which were deemed to be too inter-connected
to fail. This is a dramatic departure from the original
Bagehot’s Rule prescription to provide liquidity to
illiquid but solvent banks. This new type of systemic
risk raises the specter of moral hazard and future
financial crises and bailouts.
Thus, the policy response of aggressive monetary
policy learned from the experience of the Great
Depression greatly helped attenuate the impact of the
financial crisis on the real economy. But the sources
of systemic risk differed considerably between the
two episodes: a contagious banking panic then, an
insolvency-driven counterparty risk problem recently.
The monetary authorities were slow to catch
on that insolvency was the key issue. Once they
did, they bailed out institutions deemed to be too
interconnected to fail. This doctrine was not followed
in the US in the 1930s. The insolvent Bank of United
States, which was one of the largest banks in the
country, was allowed to fail.
5. The Key Pressing Monetary Policy
Issues for the Future
The first issue is the exit strategy, i.e., when
should the Federal Reserve return from its current
stance where the federal funds rate is close to zero to
one consistent with long run growth and low inflation?
The risks facing monetary policy with respect to the
exit are two-fold: tightening too soon and creating
a double-dip recession as occurred in 1937-38; and
tightening too late leading to a run-up of inflationary
expectations. There are many examples of each type
of error.
My paper with John Landon Lane (2010c) on the
history of Federal Reserve exits since 1920 suggests
that in the post-World War II era, the Fed usually
began tightening once unemployment peaked and
often after inflation has resurfaced. We found that
tightening often occurs two quarters after the peak
in unemployment. It seems doubtful that this will
happen since unemployment is still above 8 per cent.
Indeed political pressure and the way the Fed has
interpreted its dual mandate will likely ensure that
the Fed doesn’t begin tightening before 2013 and
moreover the Fed has signalled that it will keep its
policy rate low until 2014. But the lessons from the
last two jobless recoveries is that the Fed kept rates
too low for too long leading, first, to the inflation
spike in 1994 and the Tequila crisis and, second, the
housing boom in the 2000s. Moreover to the extent
that the sluggish recovery reflects problems in the
housing market, it is not clear what more monetary
policy can do. Will history repeat itself? My guess is
that it will.
The second pressing issue coming from the
bailouts of 2008 is that in the future the too-big-tofail
doctrine will lead to excessive risk-taking by such firms and future crises and bailouts. This was a major
concern in the debate leading to the recent Dodd-
Frank Wall Street Reform and Protection Act, passed
in July 2010 and still to be fully implemented. The
Act attempted to address the too-big-to-fail problem
by establishing a Financial Stability Oversight Council
made up of members from the Federal Reserve Board,
the Treasury, the Securities and Exchange Commission
and a number of other financial agencies. The Council
was charged with identifying and responding to
emerging risks throughout the financial system. It
would make recommendations to the Federal Reserve
to impose increasingly strict rules for capital and
leverage and other requirements to prevent banks
from becoming too large and systemically connected.
It remains to be seen whether it will be effective in
preventing future crises.
The financial reforms of the 1930s (Federal
Deposit Insurance, Glass-Steagall Act and interest
rate ceilings) were designed to prevent the problems
perceived as having created the Great Depression.
They succeeded in creating a stable but repressed
financial system for over three decades. Beginning in
the late 1960s, rising inflation, financial globalisation
and financial innovation to circumvent the controls
made much of this regime obsolete. Along with
financial innovation and the proliferation and
expansion of financial markets and institutions, the
financial crisis problem has resurfaced. That historical
experience suggests that financial crises may not be so
easily abolished.
References
Bernanke, Ben (1983), ‘Nonmonetary Effects of
the Financial Crisis in the Propagation of the Great
Depression’, American Economic Review, vol. 73, pp.
257-76.
Bernanke, Ben (2002), ‘On Milton Friedman’s
Ninetieth Birthday’, Speech given at the University of
Chicago, November 8, 2002.
Bordo, Michael, and John Landon-Lane (2010a),
‘The Global Financial Crisis of 2007-2008: Is it
Unprecedented?’, NBER Working Paper No. 16589.
_____ (2010b), ‘The Lessons from the Banking Panics
in the United States in the 1930s for the Financial
Crisis of 2007-2008’, NBER Working Paper No. 16365.
_____ (2010c), ‘Exits from Recessions: The U.S.
Experience 1920-2007’, NBER Working paper No.
15731.
Bordo, Michael and Joseph Haubrich (2011), ‘Deep
Recessions, Fast Recoveries and Financial Crises:
Evidence from the American Record’, Federal Reserve
Bank of Cleveland (mimeo).
Calomiris, Charles and Joseph Mason (2003),
‘Fundamentals, Panics and Bank Distress during the
Depression’, American Economic Review, Vol. 93(5),
pp. 1615-47.
Cole, Hal and Lee Ohanian (2004), ‘New Deal Policies
and the Persistence of the Great Depression: A General
Equilibrium Analysis’, Journal of Political Economy,
Vol. 112, pp. 779-816.
Friedman, Milton and Anna J. Schwartz (1963), A
Monetary History of the United States. Princeton:
Princeton University Press.
Goodfriend, Marvin (2011), ‘Central Banking in the
Credit Turmoil: An Assessment of Federal Reserve Practice’, Journal of Monetary Economics, Vol. 58,
pp. 1-12.
Gorton, Gary (2010), Slapped in the Face By the
Invisible Hand: Banking and the Panic of 2007, Oxford
University Press.
Hetzel, Robert (2009), ‘Monetary Policy in the 2008-
2009 Recession’, Federal Reserve Bank of Richmond
Review, September.
Meltzer, Allan (2003), A History of the Federal Reserve,
Vol. I, University of Chicago Press.
Reinhart, Carmen and Kenneth Rogoff (2009), This
time is different : Eight Centuries of Financial Folly,
Princeton University Press.
Richardson, Gary (2006), ‘Correspondent Clearing and
the Banking Panics of the Great Depression’, NBER
Working Paper No. 12716.
_____ (2007), ‘Categories and Causes of Bank Distress
during the Great Depression, 1920-1935: the Illiquidity
versus Solvency Debate Revisited’, Explorations in
Economic History, Vol. 44, No 4. October, pp. 588-607.
Temin, Peter (1976), Did Monetary Forces Cause the
Great Depression? New York: W.W. Norton.
Wicker, Elmus (1996), The Banking Panics of the Great
Depression, New York: Cambridge University Press.
|