|The New York Times
October 14, 2007
Sniffles That Precede a Recession
By ROBERT J. SHILLER
A RECESSION has much the same pattern as the flu — starting with vague
feelings of malaise and quickly building in misery until a patient's
activities are drastically curtailed. Then, all too gradually, comes
an extended period of recovery, accompanied by lingering symptoms of
With the unemployment rate up to 4.7 percent in September from 4.4
percent in March, the economy is feeling a chill. Is it descending
Most economists seem to be concluding that the current unpleasantness
is a false alarm. They point to some good vital signs: the stock
market is up, the dollar is cheap, the rest of the world is strong and
the Fed is ready to respond.
But there are worrisome symptoms, and they bear close watching. The
most important is a creeping sense of malaise that could turn into a
general loss of confidence. The downturn in the housing market and the
repercussions in financial markets are critical factors.
There have been only two domestic recessions in the last
quarter-century — both of them also global recessions. According to
the National Bureau of Economic Research dating committee, the first
began in July 1990, the second in March 2001.
There were familiar warning signs for both of them — an initial sharp
rise in unemployment, followed by slower increases that continued for
a couple of years. In each case, as often happens with recessions,
there was no agreement that a recession was under way until months
after it started.
Diagnosis of a recession is hard because no single virus causes it.
Instead, a recession seems to be a result of a confluence of many
hard-to-measure factors. A decline in investment spending is typically
one of them, and a recession is generally one of those rare events
when residential and nonresidential investment both happen to decline
In some respects, the current situation looks a lot like the period
leading up to the 1990 recession. We were coming out of a housing boom
then, and the economy was emerging from an associated lending crisis —
the savings-and-loan debacle. Now we are dealing with the subprime
mortgage "crisis," but so far, we have not seen the decline in
nonresidential investment that occurred in 1990.
There are also some similarities to the 2001 recession, which likewise
followed a huge speculative boom. The bursting of the Internet bubble
brought a huge decline in corporate investment, and the 2001 recession
helped to cleanse investors of their exaggerated hopes for the stock
market, particularly for technology and the dot-coms. A similar
cleansing of thinking appears under way regarding the housing market.
But residential investment is not as big a component of gross domestic
product as nonresidential investment; the decline in the housing
market has apparently not yet been enough to push us into recession
Consumer confidence indexes have not yet fallen as they did at the
onset of the last two recessions. But confidence is a delicate
psychological state, not easily quantified. It is related to the
stories that people are talking about at the moment, narratives that
put emotional color into otherwise dry economic statistics.
[standard measures of consumer confidence also aren't supposed to be
good predictors of anything, either.]
In August 1990, for example, a series of events in the Persian Gulf
severely damaged business confidence, and that sequence seems to
explain the timing of the 1990 recession. Saddam Hussein started his
surprise invasion of Kuwait on Aug. 2, 1990, and the United States
began sending jet planes to Saudi Arabia shortly thereafter; the Gulf
War abruptly became a virtual certainty. Mr. Hussein asked Muslims
around the world to join in a jihad against the forces opposing him.
In the United States, people started canceling business trips. August
was also the month when intense public conversation began about the
economy's weakness. In a sense, that was when the recession started,
not the July date given by the bureau committee.
It is clear that salient, emotion-arousing narratives — those that
capture the popular imagination and damage public confidence — are
central to the etiology of recessions. As these stories gain currency,
they impel people to curtail their spending, both in business and
their personal lives.
IS this happening now? A disturbing narrative began to unfold in the
last couple of months. People began talking of failed institutions —
of the possibility that savings socked away in a money market account
might actually be invested in subprime loans and so be lost. There has
been fear of locked credit markets, of possible bank failures and runs
Some of these tales have faded — bank runs no longer seem a risk. But
confidence in the economy remains fragile. More shocks are likely as
an era of huge real estate speculation apparently ends, with the
possibility of further surges in foreclosures and failures of
The narrative is still unfolding, and the extent of its virulence is
not yet known.
Robert J. Shiller is professor of economics and finance at Yale and
co-founder and chief economist of MacroMarkets LLC.
Copyright 2007 The New York Times Company