<http://www.iht.com/articles/7768.html>
The Unpredictable Economy: Experts Missed Last 9 Recessions
Steven Pearlstein Washington Post Service Wednesday, January 17, 2001
WASHINGTON In presenting his annual economic outlook last Thursday, the
chairman of President Bill Clinton's Council of Economic Advisers was having
nothing to do with all the recession talk going around.
.
"Let me be clear," Martin Baily said. "We don't think that we're going into
recession."
.
The same message was delivered the next day by Mr. Clinton in a Rose Garden
economic valedictory. Citing the predictions of 50 private forecasters known
as the Blue Chip Consensus - "the experts who make a living doing this," as
he put it - Mr. Clinton assured Americans that the economy would continue to
grow this year at an annual rate of 2 percent to 3 percent.
.
What the president and his adviser failed to mention was that "the experts"
did not predict any of the nine recessions since the end of World War II.
.
That is true of the members of the Blue Chip Consensus as well as the
Council of Economic Advisers, the forecasting staffs of the Federal Reserve
Board and the Congressional Budget Office.
.
And if, as a few renegades have begun to predict, the U.S. economy is
heading into a mild recession this year, that would mean one more forecast
added to that dismal record.
.
"We really aren't very good at calling the turning points of the economy in
either direction," said Murray Weidenbaum, the top economic adviser in the
Reagan administration.
.
Allen Sinai of Primark Decision Economics Inc., a respected private
forecaster, agreed. "It's probably only fair for forecasters to admit at
times like this that we're simply not well equipped to predict turning
points," he said. "A recession, by its nature, is a speculative call."
.
At first blush, such humility may seem at odds with the aura surrounding
modern-day forecasters.
.
Using high-speed computers and sophisticated models of the U.S. economy,
they constantly revise their two-year predictions for everything from
unemployment to business investment to long-term interest rates, expressed
numerically to the first decimal place.
.
But according to the forecasters themselves, what may appear to be a precise
science is really a black art, one that is constantly confounded by the
changing structure of the economy and the refusal of investors, consumers
and business executives to behave as rationally and predictably in real life
as they do in economic models.
.
"The reason we have trouble calling recessions is that all recessions are
anomalies," said Joel Prakken, president of Macroeconomic Advisers of St.
Louis, one of the leading U.S. forecasting firms.
.
According to Mr. Prakken, every modern recession has been caused by a
combination of overly aggressive interest rate increases by the Federal
Reserve Board, which weakens the economy, and some "external shock" that
pushes it over the edge.
.
Because such shocks - the 1973 oil embargo, for example, or the 1990 Iraqi
invasion of Kuwait - are random and, by their nature, unpredictable, Mr.
Prakken argues that it is virtually impossible to predict when a slowing
economy will turn into a shrinking one.
.
For the moment, Mr. Prakken, like most other forecasters, is confidently
predicting that the economy will skirt the edge of recession early this year
and then pick up its pace, growing at the annual rate of 2.6 percent.
.
A hearty band of optimists in the forecasting community still says it should
be possible to come up with leading indicators for the economy that could
state, with some reliability, that a recession is just over the horizon. It
is just that such indicators have not been found yet.
.
James Stock of Harvard University's Kennedy School of Government said there
had been any number of early warnings that, for 20 to 30 years, were highly
correlated with a recession: sharp declines in the stock market, a slump in
housing starts, a rapid expansion of the money supply and an unusual
alignment of interest rates in which long-term bonds had lower yields than
short-term securities did.
.
The problem is that just when there are enough occurrences to prove an
indicator reliable, the economy changes in some fundamental way, and the
model loses its predictive power, Mr. Stock said.
.
Victor Zarnowitz, a longtime student of the business cycle, said much of the
problem was that forecasters had become prisoners of an economic ideology
that assumed too much rationality on the part of the economic players or
factors.
.
Thus, they spend too much time looking for recession indicators in official
economic data, which often arrive too late to be of any use in forecasting,
and not enough at the anecdotal evidence around them.
.
Mr. Zarnowitz starts from the assumption that there are natural ups and
downs to the economy, dictated by the ability of businesses to find good
things in which to invest.
.
At the end of a recession and at the beginning of a new expansion, there are
plenty of low-risk, high-payoff opportunities and a rush of new investments,
sparking a cycle of higher profits and greater investment that feeds on
itself and pushes the economy higher.
.
At some point, however, the supply of "good" investments diminishes, Mr.
Zarnowitz said. Businesses and investors then become overconfident and start
paying too much for assets, assuming risks that are too great and taking on
too much debt.
.
Finally, a tipping point is reached, and overconfidence gives way to an
equally excessive pessimism, leading the economy to unwind quickly.
.
After a recession cleanses the economy of its excesses and imbalances,
growth resumes.
.
On casual observation, this theory of investment booms and busts squares
nicely with the run-up in the stock market that preceded the 1973 recession
and the real-estate boom and bust before the 1990-91 recession.
.
But the problem for forecasters is having to acknowledge a high level of
irrationality on the part of business executives and investors and consumers
that, by its nature, follows no neat pattern or rational timetable.
.
"It's easy to say that the bubble will burst sooner or later but very
difficult to say exactly when," said Mr. Zarnowitz, who now is a consultant
at the Conference Board in New York.
.
Difficult but not impossible, said Edward Leamer, director of the UCLA
Anderson Business Forecast.
.
Starting with about the same model of the U.S. economy as Mr. Zarnowitz lays
out, Mr. Leamer last year went looking for conditions that appeared to be
uniquely present at the end of expansions.
.
In each case, he found that investment spending was growing faster than
profit. He also found that interest rates on 10 year, medium-risk corporate
bonds fell sharply in relation to short-term notes. Finally, there was in
each instance a noticeable reduction in the hours worked by the average
employee each week.
.
Based on these three indicators and a heavy dollop of gut instinct, Mr.
Leamer last month became the first Blue Chip Consensus forecaster to predict
a mild recession in the United States this year.
For Related Topics See:
Business
Front Page
< < Back to Start of Article WASHINGTON In presenting his annual economic
outlook last Thursday, the chairman of President Bill Clinton's Council of
Economic Advisers was having nothing to do with all the recession talk going
around.
.
"Let me be clear," Martin Baily said. "We don't think that we're going into
recession."
.
The same message was delivered the next day by Mr. Clinton in a Rose Garden
economic valedictory. Citing the predictions of 50 private forecasters known
as the Blue Chip Consensus - "the experts who make a living doing this," as
he put it - Mr. Clinton assured Americans that the economy would continue to
grow this year at an annual rate of 2 percent to 3 percent.
.
What the president and his adviser failed to mention was that "the experts"
did not predict any of the nine recessions since the end of World War II.
.
That is true of the members of the Blue Chip Consensus as well as the
Council of Economic Advisers, the forecasting staffs of the Federal Reserve
Board and the Congressional Budget Office.
.
And if, as a few renegades have begun to predict, the U.S. economy is
heading into a mild recession this year, that would mean one more forecast
added to that dismal record.
.
"We really aren't very good at calling the turning points of the economy in
either direction," said Murray Weidenbaum, the top economic adviser in the
Reagan administration.
.
Allen Sinai of Primark Decision Economics Inc., a respected private
forecaster, agreed. "It's probably only fair for forecasters to admit at
times like this that we're simply not well equipped to predict turning
points," he said. "A recession, by its nature, is a speculative call."
.
At first blush, such humility may seem at odds with the aura surrounding
modern-day forecasters.
.
Using high-speed computers and sophisticated models of the U.S. economy,
they constantly revise their two-year predictions for everything from
unemployment to business investment to long-term interest rates, expressed
numerically to the first decimal place.
.
But according to the forecasters themselves, what may appear to be a precise
science is really a black art, one that is constantly confounded by the
changing structure of the economy and the refusal of investors, consumers
and business executives to behave as rationally and predictably in real life
as they do in economic models.
.
"The reason we have trouble calling recessions is that all recessions are
anomalies," said Joel Prakken, president of Macroeconomic Advisers of St.
Louis, one of the leading U.S. forecasting firms.
.
According to Mr. Prakken, every modern recession has been caused by a
combination of overly aggressive interest rate increases by the Federal
Reserve Board, which weakens the economy, and some "external shock" that
pushes it over the edge.
.
Because such shocks - the 1973 oil embargo, for example, or the 1990 Iraqi
invasion of Kuwait - are random and, by their nature, unpredictable, Mr.
Prakken argues that it is virtually impossible to predict when a slowing
economy will turn into a shrinking one.
.
For the moment, Mr. Prakken, like most other forecasters, is confidently
predicting that the economy will skirt the edge of recession early this year
and then pick up its pace, growing at the annual rate of 2.6 percent.
.
A hearty band of optimists in the forecasting community still says it should
be possible to come up with leading indicators for the economy that could
state, with some reliability, that a recession is just over the horizon. It
is just that such indicators have not been found yet.
.
James Stock of Harvard University's Kennedy School of Government said there
had been any number of early warnings that, for 20 to 30 years, were highly
correlated with a recession: sharp declines in the stock market, a slump in
housing starts, a rapid expansion of the money supply and an unusual
alignment of interest rates in which long-term bonds had lower yields than
short-term securities did.
.
The problem is that just when there are enough occurrences to prove an
indicator reliable, the economy changes in some fundamental way, and the
model loses its predictive power, Mr. Stock said.
.
Victor Zarnowitz, a longtime student of the business cycle, said much of the
problem was that forecasters had become prisoners of an economic ideology
that assumed too much rationality on the part of the economic players or
factors.
.
Thus, they spend too much time looking for recession indicators in official
economic data, which often arrive too late to be of any use in forecasting,
and not enough at the anecdotal evidence around them.
.
Mr. Zarnowitz starts from the assumption that there are natural ups and
downs to the economy, dictated by the ability of businesses to find good
things in which to invest.
.
At the end of a recession and at the beginning of a new expansion, there are
plenty of low-risk, high-payoff opportunities and a rush of new investments,
sparking a cycle of higher profits and greater investment that feeds on
itself and pushes the economy higher.
.
At some point, however, the supply of "good" investments diminishes, Mr.
Zarnowitz said. Businesses and investors then become overconfident and start
paying too much for assets, assuming risks that are too great and taking on
too much debt.
.
Finally, a tipping point is reached, and overconfidence gives way to an
equally excessive pessimism, leading the economy to unwind quickly.
.
After a recession cleanses the economy of its excesses and imbalances,
growth resumes.
.
On casual observation, this theory of investment booms and busts squares
nicely with the run-up in the stock market that preceded the 1973 recession
and the real-estate boom and bust before the 1990-91 recession.
.
But the problem for forecasters is having to acknowledge a high level of
irrationality on the part of business executives and investors and consumers
that, by its nature, follows no neat pattern or rational timetable.
.
"It's easy to say that the bubble will burst sooner or later but very
difficult to say exactly when," said Mr. Zarnowitz, who now is a consultant
at the Conference Board in New York.
.
Difficult but not impossible, said Edward Leamer, director of the UCLA
Anderson Business Forecast.
.
Starting with about the same model of the U.S. economy as Mr. Zarnowitz lays
out, Mr. Leamer last year went looking for conditions that appeared to be
uniquely present at the end of expansions.
.
In each case, he found that investment spending was growing faster than
profit. He also found that interest rates on 10 year, medium-risk corporate
bonds fell sharply in relation to short-term notes. Finally, there was in
each instance a noticeable reduction in the hours worked by the average
employee each week.
.
Based on these three indicators and a heavy dollop of gut instinct, Mr.
Leamer last month became the first Blue Chip Consensus forecaster to predict
a mild recession in the United States this year.
