Analysts and
journalists that comment or report economic news, be they or not economists,
conclude, often and automatically, that there is a recession when an economy real
GDP contracts two consecutive quarters. Also, when politicians are asked: what
is the definition of a recession? Frequently, their staff recommends them to
answer: it is at least two consecutive quarters of real GDP contraction.
So, why a recession
in Canada lasted only one quarter (first quarter of 1975), and why the most
recent recession in the United States (December 2007 to June 2009) started
without having two consecutive quarters of real GDP decline during the first
half of 2008?
The answer: organizations
that have the responsibility of dating business cycles use not only real GDP
but also other macroeconomic indicators to make their decisions.
The National Bureau of Economic Research (NBER)
Business Cycle Dating Committee looks at real GDP, real income, employment,
industrial production, wholesale trade, retail sales and any other indicator
that, depending on the circumstances, could guide its judgment of the
occurrence of a recession in the United States.[1]
The Centre for Economic Policy Research (CEPR)
Business Cycle Dating Committee, which has responsibility for dating recessions
in the euro area, uses also a set of indicators related to production and
labor market, such as real GDP, employment, investment, industrial production
and consumption.[2]
The C. D. Howe Institute Business Cycle
Council retains real GDP and employment to identify recessions in Canada. The Institute
took this responsibility in 2012 after Statistics Canada decided to cease the
dating of the country’s business cycles.[3]
Those four organizations
insist on the importance of looking not only at the duration, but also at the amplitude
and the scope (degree of propagation
to the whole economy) of the downturn in economic activity to determine the
occurrence of a recession.
In Japan, the Economic and Social Research Institute (ESRI)
Business Cycle Indicators Committee analyses many coincident indicators, such
as industrial production, retail sales and wholesale trade. After that, ESRI’s
President renders a decision as to the occurrence of a recession.
The Conference Board does not play an
official role in dating recessions. But, it employs real GDP and its index of
coincident indicators to identify recessions for many countries. The coincident
variables are, most often, employment, industrial production and retail sales. The
Economic Cycle Research Institute examines
also business cycles for different countries, and it uses the NBER method to determine
recessions.
For Quebec’s economy,
Desjardins’ economists identified recessions by uniquely using real GDP. But in
a note to their analysis, they mentioned:
“…two or more consecutive quarters of declines by the real GDP do not
necessarily signify a recession. A major pullback in economic activity for the
period also has to be recorded. If the drop is weak in scope, it could denote a
slowdown phase.”[4]
For the world economy global business cycles, International Monetary Fund (IMF) economists use, as key indicator of
a recession, world real GDP per capita based on purchasing power parity (PPP)
weights. They also look at industrial production, total trade, capital flows, oil
consumption, unemployment, per capita consumption and per capita investment.[5]
Also, a weakness of the two consecutive quarter approach is that when
there are revisions of data, a small quarterly decrease can become a small
increase. What happens, after the revision, of the recession call made on the
basis of preliminary data? It is forgotten or ignored.
Even if the two quarter approach would have advantages, such as being
simple, easy to explain, automatic and popular, the business cycle experts
refute it. They prefer to exercise their judgment after having used a certain
number of indicators and criteria (duration, amplitude and scope).
Finally, where does come
from this idea or belief that two consecutive quarters of real GDP contraction
equals a recession?
It seems that it
originates from an erroneous interpretation of a NBER statistical observation, during
the 1960s, to the effect that recessions in the Unites States lasted at least
six months.[6]
To conclude, taking inspiration
from the NBER, CEPR and C.D. Howe Institute experts, wouldn’t it be more
appropriate to suggest to politicians, and their staff, the following answer to
a question on the definition of a recession:
-
It
is a significant decline of the activity, propagated to the whole economy, lasting
at least a few months.
N.B.: The French version of this article is available at: http://leblogdejpfsurlesindicateursavances.blogspot.ca/2014/09/pib-et-recession.html
[1] For more information, consult
the most recent decision of the NBER Business Cycle Dating Committee and its
Q&A section at:
[2] The CEPR
committee presents its decisions and methodology at:
http://www.cepr.org/content/euro-area-business-cycle-dating-committee
[3] Cross, Philip and Philippe
Bergevin. “Turning Points: Business Cycles in Canada since 1926”. C.D.
Howe Institute. October 2012. Available
at:
https://www.cdhowe.org/sites/default/files/attachments/research_papers/mixed/Commentary_366_0.pdf
[4]
Bégin, Hélène and Jonathan Créchet. “New Features for the Desjardins
Leading Index”. Desjardins Economic Studies. January 2013. Available at: http://www.desjardins.com/en/a_propos/etudes_economiques/actualites/point_vue_economique/pv011613a.pdf
[5] IMF, “World Economic Outlook”,
April 2009, Box 1.1 « Global Business Cycles », pages 11 to 14. Available
at: http://www.imf.org/external/pubs/ft/weo/2009/01/#ch1box
[6] Cross and Bergevin, page 4 :
“The
notion that a recession is defined by two or more consecutive quarterly
declines in GDP has become well entrenched in popular discussions. The origins
of the consecutive-declines guideline go back to a mistaken interpretation of a
simple statistical observation by the NBER that, in practice, recessions in the
United States lasted at least six months (Moore 1967). Lay people, anxious to
penetrate the byzantine process used at the time to assess cycles, quickly
jumped on this as a rule even though it was just a statistical artifact.
Indeed, the NBER itself has never used consecutive quarterly declines in GDP as
a definition of a recession.”
