Sunday, June 7, 2009

Permanent Versus Temporary Output Loss: Policy Alternatives

In my post on April trade data I mentioned the potential for the economy moving to a different but lower growth path. While I'm not up to speed on business cycle theory, the genesis for my contention is this 2003 IMF paper* which investigates the response of economic growth to the 1997-98 crisis in a number of countries. The finding was that these economies suffered not just a temporary loss in output, but a permanent loss as well.

This matters because the policy response required is different with respect to the type of recession being faced. But to illustrate the different concepts I'm talking about, the figure below is reproduced from the paper:

To use the terminology of the paper for Malaysia's growth experience, the 1997-98 and 1984-85 recessions were Hamilton recessions, while the 1975-76 and 2000-2001 recessions were Friedman recessions. My guestimate is that the current downturn will turn out to be the former. What's the real difference here?

The normal, Friedman recession is one which is associated with an increase in the output gap, which is the difference between potential output based on labour and capital capacity to produce. The Hamilton recession is where potential output itself is dropping, which signals a permanent loss in output.

While specific policy prescriptions would differ based on the causes of a downturn,
in broad terms a Friedman recession only requires countercyclical monetary policy and the automatic stabilizers built into fiscal policy. The fall in growth will be superseded by above-trend growth, which returns the economy back to the former growth path.

The Hamilton recession however requires much more drastic measures, if output loss is not to become permanent. This would necessarily involve more fiscal spending measures, particularly investment which would be needed to replace the capacity loss. Attracting FDI would also help, which can be seen in the experience after the mid-1980s recession (RGDP volume index 1970-2007; log scale):

The trend line indicates the Malaysian economy had a trend growth of 6.44% over the last 35-odd years. However, the slope of RGDP is a little steeper before the 1997-98 recession, which indicates that not only did the Malaysian economy fail to fully recover the output capacity lost at that time, but also failed to attract enough investment to maintain its previous rate of growth. But that's hardly news to anyone.

But perhaps this graph might underscore the seriousness of the current situation, and the necessity of more intervention in the economy (as if anyone needs more ammunition):

This graph shows the weighted average of RGDP per capita against Malaysia's trade partners (2000=100). It's clear from this that Malaysia has continued to lose ground against our trade partners in terms of increasing the welfare and income of our people. And that something more drastic needs to be done to maintain our economic standing in the years to come.

*Cerra, Valerie & Saxena, Sweta Chaman, "Did Output Recover from the Asian Crisis?", IMF Working Paper No. 03/48

Technical Notes:
RGDP Volume Index from IMF International Financial Statistics Database.


  1. The Granger Causality Test is indeed very widely used. But it gives rise to this infuriating disclaimer by econometricians when asked to explain their empirical findings using the test - "Coexistence does not imply causation."

    I agree Granger's contribution to statistical theory is significant.

  2. Hi HishamH,

    This post is great..
    Are there any underlying causes to the difference of these 2 types of recessions ?

  3. Not that I know of - although some recent research (also at the IMF) suggests that deeper recessions tend to be caused by banking crises.