Discussion Paper

No. 2009-6 | January 07, 2009
The Universal Shape of Economic Recession and Recovery after a Shock


We show that a simple and intuitive three-parameter equation fits remarkably well the evolution of the gross domestic product (GDP) in current and constant dollars of many countries during the times of recession and recovery. We then argue that it can be used to detect shocks and discuss its predictive power. Finally, a two-sector theoretical model of recession and recovery illustrates how the severity and length of recession depends on the dynamics of transfer rate between the growing and failing parts of the economy.

Paper submitted to the special issue “Reconstructing Macroeconomics”

JEL Classification

C32 O23 O41

Cite As

Damien Challet, Sorin Solomon, and Gur Yaari (2009). The Universal Shape of Economic Recession and Recovery after a Shock. Economics Discussion Papers, No 2009-6, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2009-6


Comments and Questions

Anonymous - Referee Report
January 12, 2009 - 10:05

The universal shape of economic recession and recovery after a shock
Damien Challet, Sorin Solomon and Gur Yaari

They report that the superposition of two exponential functions with
three parameters (1) fits the evolution of the GDP during the times of
recession and recovery. This ...[more]

... is confirmed by employing the GDP data of 23 coutries. The data are examined from various angles. They also explain the rationale behind Eq. (1). The logic flow is clear.

Finally policy making is discussed by using the model.
I think this paper is very interesting and should be accepted.
However, I have noticed the following:

1) The references of figures are confusing.
2) The notations after Eq. (4) in third paragraph of section 3 and Eq.
(5) are confusing.

I think the authors should check them again.

Damien Challet - From the authors
January 13, 2009 - 13:01

We could not understand precisely in what respect the references of the figures are confusing: does the referee means the figure captions, in which case we can expand them, or the references inside the captions, or the references to the figures, in which case we can amend the core of ...[more]

... the text?

Similarly, we checked our notations at the places indicated by the referee and could not detect undefined or unexplained quantities.

We would be more than pleased to make our paper easier to read but would need more precisions from the referee, if possible.

Vladimir Popov - Reader Comment
February 23, 2009 - 08:35

see attached file

Damien Challet - From the authors
March 06, 2009 - 21:30

Prof. Popov's report is very instructive. We are most grateful for the time he took reading our manuscript and also for writing a kind crash course of Economics' view on the mechanisms underlying various types of recessions.

He raises two issues. The first one is that our claim ...[more]

... about the universality of the shape of recessions may not hold for some types and sub-types of recessions whose mechanism he explains in details. He then suggests that the shape of the GDP may be something else, such as a parabola.

Taking the example of an economy whose all sectors shrink at the same rate and then recover for some reason, he argues that the first phase of such kind of recessions corresponds to setting both rates \lambda_+ and \lambda_- to the same value, hence that the fitting equation is not adequate for this case. It may be, and we would be grateful if such an real-life example can be pointed out tu us in order to study this case in details. But the wording 'sectors' should not be taken literaly: one can always try to fit this peculiar kind of recession with the proposed function. But once again, one would first to find a few examples of such events and fit them. It is possible in passing that the examples included in our submission do contain at least in part such mechanisms.

Similarly, if the transfer rate is indeed the essential limiting factor of the speed to recover, the fitting equation does not assume the existence of such mechanisms. We wish to point out once more that the fitting equation that we propose works because it respects the auto-catalytic hence exponential nature of (both positive and negative) economic growth.

We also wish to say that the ongoing worldwide recession shows that our model does not take well into account the very first stages of recessions, as it implicitely assumes that the GDP is a convex function, whereas the GDP of the UK ( http://www.statistics.gov.uk/cci/nugget.asp?id=192 ) is clearly a concave function during the first few quarters. This may be due to the fact that the numbers of sectors affected by the slowdown increases as a function of time, that is, that f is not a constant at the beginnning of the crisis. We are of course willing to amend our article to emphasize this point.

The second issue concerns the theoretical model that we analyze in the second part of our submission. He rightly points out that we need to be more specific at places and to expand the discussion so as to take into account his suggestions, in particular with respect to the optimal transfer rate.

In short, we understand better the need to adjust the way we present our main ideas and our results so as to be more precise and to communicate better to the audience of the journal.

Paul Ormerod - Comments
March 05, 2009 - 08:46

see attached file

Damien Challet - From the authors
March 09, 2009 - 09:55

see attached file