How is recession determined? The common definition is two consecutive quarters of falling output. But this is not a definition used by economists.
Governments typically
do not formally announce when the economy is in recession, in part because such
announcement may not be credible. For example the government in power may be
tempted to declare, before an election, that the economy is booming. Instead,
the dating of recessions is left to independent economists. In Canada, the C.D.
Howe Institute, an independent research organization, has recently undertakenthe task of systematically dating Canadian recessions. The C.D. Howe Business
Cycle Council determined, for example, that there was an expansion from April
1992 to October 2008 (the longest since the beginning of the data, in 1926),
and a recession from October 2008 to May 2009 (one of the shortest). In the
United States, most economists accept the dates for business cycle recessions
and expansions determined by the National Bureau of Economic Research (NBER).
For example, NBER determined that the business-cycle expansion began in
November 2001 and ended in December 2007, and it determined that the following
recession ended in June 2009, when the next expansion began.
How does the C.D. Howe Institute’s Business
Cycle Council determine when a recession begins and ends? There are no simple,
clear rules. The Council defines a recession as a pronounced, pervasive, and
persistent decline in aggregate economic activity. The Council looks both at
the length and size of the decline in output and employment and the sectors of
the economy that are affected. To call a recession, the Council requires that
output falls for at least one quarter. The decline in output needs not be
large, but it has to be persistent: A 0.1% drop in GDP in one quarter when in
adjacent quarters the economy is growing strongly would not be considered a
recession. The Council also looks at the proportion of sectors with falling
output, particularly in manufacturing and construction. These two sectors vary the
most over the business cycle. A general decline accompanied by robust growth in
manufacturing and construction is unlikely to be a recession.
The recession timing as determined by the C.D. Howe’s Business Cycle
Council is as follows:
C.D. Howe Recession dates
|
|
Beginning, quarter
|
End, quarter
|
1929:Q2
|
1933:Q1
|
1937:Q3
|
1938:Q2
|
1947:Q2
|
1948:Q1
|
1951:Q1
|
1951:Q4
|
1953:Q2
|
1954:Q2
|
1960:Q1
|
1961:Q1
|
1974:Q4
|
1975:Q1
|
1979:Q4
|
1980:Q2
|
1981:Q2
|
1982:Q4
|
1990:Q1
|
1992:Q2
|
2008:Q3
|
2009:Q2
|
The Great Recession, as you can see from the table, was short - it lasted only 3 quarters, the 1990-1992 recession lasted almost 4 times longer. The term Great Recession is imported. It was the worst recession in the U.S. since the Greta Depression, but in Canada it was not as bad.
The economic data used by the C.D. Howe are collected by Statistics Canada, the Canadian federal statistical agency. Statistics Canada is generally considered to be one of the the best statistical
offices in the world. It provides estimates of macroeconomic variables fairly quickly
and the estimates are quite accurate: subsequent revisions are small.
Nonetheless, the information is retrospective: first, the data measure economic
activity during a period that has already passed; second, the collection and
processing of data takes time. As a result, the start of the recession is
determined several months after the recession has begun. The C.D. Howe Business
Cycle Council list of recessions was published more than two years after the
end of the last recession, and we do not know how long its delay in announcing
future recessions and expansions will be. In the United States the delays are also
substantial. For example, the announcement that the most recent recession began
in December 2007 was made by NBER twelve months later, in December 2008.
No comments:
Post a Comment