Calling recessions in real time

James D. Hamilton, 18 July 2010

a

A

Is the world economy about to experience a "double-dip" recession, going back into a new downturn before the recovery from the previous recession is even complete? Or, if there is a subsequent downturn, should it be described as a new, separate recession? The answer requires an objective characterisation of what we mean when we say the economy is in a recession.

What’s in a word?

In the US, the commonly used dates for economic recessions come from the judgemental conclusions of the Business Cycle Dating Committee of the National Bureau of Economic Research. This may be the best approach available, but it would be nice to be able to supplement it with purely objective statistical inference. In addition to clarifying what exactly one means by the statement that the economy is in recession, such an approach might be able to provide more timely economic assessments that one can be sure are immune from any political influence.

Harvard Professor James Stock and Princeton Professor Mark Watson proposed one algorithm for identifying business cycle turning points in 1988 (Stock and Watson 1989), and provided widely publicised predictions from the model up until 2003. Their press releases highlighted the more ambitious component of their approach, which offered the promise of being able to predict turning points in advance of their occurrence. This particular goal proved to be elusive for both the 1990 and the 2001 economic downturns. But their lead in suggesting that researchers should establish a real-time, out-of-sample track record is one well worth following, and the immediate publication possibilities of the internet allow many researchers today to do just that.

In a new research paper (Hamilton 2010), I survey the methods and conclusions of a number of these real-time dating methods currently in use by various researchers. For the last five years, I have been using Econbrowser as a forum for publicly reporting the real-time output of a mechanical GDP-based algorithm for dating US business cycle turning points. Figure 1 below plots those inferences in the form of a GDP-based recession indicator index. Values to the left of the vertical line are what I would describe as "simulated real time" – each observation plotted was constructed solely from a vintage data set as it would have been reported at the indicated date. To the right of the vertical line are "actual real time" numbers – each observation plotted was publicly announced at the indicated date.

Figure 1. GDP-based recession indicator index

Graph plots for each quarter t the value for Prob(St|Yt+1) where Yt+1 is the history of observed GDP growth rates as reported as of date t+1.

Based on the procedure detailed in my earlier paper with Marcelle Chauvet (Chauvet and Hamilton 2006), the algorithm determined that the most recent recession began in the fourth quarter of 2007 and ended in the second quarter of 2009. A subsequent economic downturn would be viewed by this procedure as the beginning of a new recession rather than a continuation of the previous one. Figure 2 shows the simulated real time and actual real time record of the algorithm.

Figure 2. Start and end dates of recessions

Start of recessions
Peak as
Date
Recession start
Date
Algorithm
determined
NBER made
as determined
algorithm made
announcement lead (-)
by NBER
declaration
by algorithm
declaration
or lag (+) in months
1969:Q4
N.A.
1969:Q2
May 1970*
N.A.
1973:Q4
N.A.
1973:Q4
May 1974*
N.A.
1980:Q1
Jun 1980
1979:Q2
Nov 1979*
-7
1981:Q3
Jan 1982
1981:Q2
Feb 1982*
+1
1990:Q3
Apr 1991
1989:Q4
Feb 1991*
-2
2001:Q1
Nov 2001
2001:Q1
Feb 2002*
+3
2007:Q4
Dec 2008
2007:Q4
Jan 2009
+1
 
Start of expansions
Trough as
Date
Recession end
Date
Algorithm
determined
NBER made
as determined
algorithm made
announcement lead (-)
by NBER
declaration
by algorithm
declaration
or lag (+) in months
1970:Q4
N.A.
1970:Q4
Aug 1971*
N.A.
1975:Q1
N.A.
1975:Q1
Feb 1976*
N.A.
1980:Q3
Jul 1981
1980:Q2
May 1981*
-2
1982:Q4
Jul 1983
1982:Q4
Aug 1983*
+1
1991:Q1
Dec 1992
1991:Q4
Feb 1993*
+2
2001:Q4
Jul 2003
2001:Q3
Aug 2002*
-12
N.A.
N.A.
2009:Q2
Apr 2010
N.A.

Business cycle turning points and dates at which announcements were issued by NBER and the GDP-based algorithm. Starred entries denote simulated real-time declarations, unstarred are actual real-time declarations. N.A. indicates information is not available.

Marcelle Chauvet and Jeremy Piger (2008) have suggested an approach that uses growth rates of industrial production, personal income, sales, and employment. Professor Piger has been regularly updating these inferences on his web page since August 2006. There is a trade-off in these methods between trying to make an inference using the most recent available data about where the economy is at the moment (called the "filter probabilities"), or waiting for data to be revised and the trend to become clearer before trying to form an inference about where the economy was n months previously (called the "n-month smoothed probabilities"). Piger's actual real-time filter probabilities for each month are plotted in the top panel of Figure 3. This series first moved above 50% in Piger's November 1, 2008 report which was based on data describing August. The probability moved to 99.2% when the September data became available but fell back to 16.1% with the next month's data which showed industrial production and real personal income less transfers to be growing again in October.

Chauvet and Piger (2008) had recommended an inference rule of declaring a recession as soon as three consecutive values of recent smoothed probabilities were all above 80%. This threshold was passed with the release of the October data, for although the current month filter inference was only 16.1% based on t= October 2008 data, the 1-, 2-, and 3-month smoothed probabilities at that time were still all above 80% (see the bottom panel of Figure 3). Chauvet and Piger's (2008) announced rule would then date the start of the recession as the earliest n for which the smoothed probability was above 50%, which turned out to be February 2008. Thus their approach would have announced on 1 January 2009 that a recession had started in February 2008, although the filter probabilities available in January 2009 raised the possibility that the recession could already have been over by October. Subsequent data confirmed the downturn was ongoing (panels 2 and 3). The Chauvet-Piger rule of waiting for a reading of 3 consecutive months below 20% would have resulted in a declaration in January 2010 that a recovery likely began in July of 2009.

Figure 3.

Source: reproduced from Hamilton (2010); constructed from data supplied by Jeremy Piger.

My 2006 paper with Chauvet proposed a similar 4-indicator inference. One important difference from the specification of Chauvet and Piger is the reliance on the Bureau of Labor Statistics household survey for the measure of employment rather than the establishment payroll data. Figure 4 shows the sequence of smoothed probabilities associated with different vintages of data. This model would have sent a signal in August 2008 that a recession had begun in December 2007. Like Piger's calculations, it showed a temporary hope of an end when only data through October 2008 were available. Chauvet used this model to announce on her website in October of 2009 that the recession ended in June or July of 2009.

Figure 4. Recession probabilities

Source: Marcelle Chauvet.

One of the most promising approaches in this area may be that developed by Camacho et al. (2010). They show how one can use an unbalanced panel of mixed-frequency indicators to update an inference daily as each new datum gets released. Figure 5 reports the probabilities of a Eurozone recession that would have been calculated by their approach based on the actual data available as of each day during 2008 and 2009. The model yields a remarkably sharp and stable inference over this period, with the probability jumping from 3.8% on 24 June 2008 to 98.1% one month later. The probability remained above 69% until 23 April 2009, at which point it fell to 6%.

Figure 5. Recessions probabilities in the Eurozone

Source: Camacho, Perez-Quiros, and Poncela (2010).

Although using different data sets and different methods, all of these algorithms agree on one point. The most recent recession ended in the summer of 2009. If we do see a subsequent downturn, it should be classified as a separate economic downturn.

References

Camacho, Maximo, Gabriel Perez-Quiros, and Pilar Poncela (2010), “Green shoots in the Euro area: a real time measure”, working paper, Bank of Spain.

Chauvet, Marcelle and James D Hamilton (2006), “Dating business cycle turning points”, in Costas Milas, Philip Rothman, and Dick van Dijk (eds.), Nonlinear Time Series Analysis of Business Cycles, Elsevier.

Chauvet, Marcelle, and Jeremy Piger (2008), “A comparison of the real-time performance of business cycle dating methods”, Journal of Business Economics and Statistics, 26:42-49.

Hamilton, James D (2010), “Calling recessions in real time”, working paper, University of California, San Diego.

Stock, James H, and Mark W. Watson (1989), “New indexes of coincident and leading economic indicators”, in Olivier Jean Blanchard and Stanley Fischer (eds.), NBER Macroeconomics Annual 1989, MIT Press.

Stock, James H and Mark W Watson (1991), “A probability model of the coincident economic indicators”, in Kajal Lahiri and Geoffrey H Moore (eds.), Leading Economic Indicators: New Approaches and Forecasting Records, Cambridge University Press.

Stock, James H and Mark W Watson (1993), “A procedure for predicting recessions with leading indicators: econometric issues and recent experience”, in James H Stock and Mark W Watson (eds.), Business Cycles, Indicators, and Forecasting, University of Chicago Press.

Topics: Global crisis, Global economy
Tags: business cycle, economic forecasting, recessions

Professor of Economics, University of California, San Diego

Subscribe