The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

« Higher unemployment: Labor force effects versus job loss effects | Main | Good news in income growth? »

December 23, 2008

Are modern recessions different?

In case you are hungering for something to while away the time as the holiday approaches—since apparently you are not out shopping—Barbara Kiviat provides some food for thought at The Curious Capitalist:

“Let me leave you with something to toss into the conversation when your family starts talking about the economy…

“Since WWII, recessions have lasted 10 months on average. We hear that a lot. Our current recession started in December 2007. Even though this go-around seems worse [than] usual, that's still a little cause for optimism.

“Or maybe not. This week Merrill Lynch economist David Rosenberg put out a report—thanks for the tip, John!—questioning why we'd only look at recessions over the past 60 years.”

She then quotes Rosenberg…

“This recession has more in common with the pre-WWII era. However, most of the data we have and most of the analysis still being conducted is done within the context of post-World War II cycles. That will not work, as this is a balance sheet recession and not just within the confines of the financial sector, but within the broad US household sector. This involves debt repayment and asset liquidation, and for the first time in recorded history, the entire $70 trillion household balance sheet is in the process of shrinking.”

… and goes on to provide the punchline:

“He then goes back to 1855 (a better timeframe for true historical context, he figures), and finds that recessions last an average 18 months. If his logic is right, then we're not due for sunnier skies until mid-2009.”

Interestingly enough, Berkeley’s Christina Romer—the next chair of the president’s Council of Economic Advisers—has in the past had a fair amount to say about comparing recessions past and present. First, the statement of an obvious problem:

“The obvious series to compare over time are standard macroeconomic indicators such as real GNP, industrial production, and unemployment. Such comparisons, however, are complicated by the fact that contemporaneous data on these quantities have only been collected for part of the 20th century. For example, the Federal Reserve Board index of industrial production begins in 1919, the Commerce Department GNP series begins in 1929, and the Bureau of Labor Statistics unemployment rate series begins in 1940. Furthermore, because World War II marked a radical change in the data collection efforts of the U.S. government, many of these series are only available on a truly consistent basis after 1947.”

In the paper Professor Romer goes on to recap her efforts to reconcile the older (pre-World War II) data with that of the modern era. The details reveal the considerable novelty for which she is duly admired among economists—part of the solution is to forgo the impossible task of making bad data good and instead induce comparability by making the good data bad—but I will leave those details to you and get right to the point:

“The first finding is that recessions have not become noticeably shorter over time. The average length of recessions is actually one month longer in the post-World War II era than in the pre-World War I era. There is also no obvious change in the distribution of the length of recessions between the prewar and postwar eras. Most recessions lasted from 6 to 12 months in both eras. Recessions were somewhat longer in the interwar era. However, an average for this period is virtually impossible to interpret since it includes the Great Depression, where 34 months elapsed between the peak and the trough. Probably the most sensible conclusion to draw for the interwar period echoes that from the previous section: the 1920s and 1930s were simply very peculiar.”

So, if you are inclined to buy the argument that comparisons of the current downturn to pre-WWII era contractions is a legitimate exercise, there’s some comfort for these cold winter nights. Happy holidays.

By David Altig, senior vice president and research director of the Federal Reserve Bank of Atlanta

Because of the Christmas holiday, today’s posting will be the only macroblog posting for this week.

December 23, 2008 in Business Cycles | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Are modern recessions different? :


By trying to look for comparisons with past recessions, people are just trying to predict the future.

Unfortunately, every recession occurs under a specific set of circumstances that are mostly not comparable with one another.

However, the one factor that seems to be present in most of the recent recessions is the price of oil and energy in general. That commodity has such a multiplier effect that it can break almost any growth economy.

Posted by: Alan | December 29, 2008 at 09:56 AM

One thing one never should forget: Averages as just that, they are _not_ maximums.

And judging from the graph i suspect that the length of the average recession with a "below average" length is 10 months, while the average length of the 7 longest recessions is about 35 months.

Posted by: Uwe Ohse | December 29, 2008 at 09:56 AM

So, if we are in as bad of shape as some of the statistics suggest this downturn could last until at least October of 2010. Add the fact that our FED and Treasury are throwing everything, and the kitchen sink at the problem, which is only slowing down the proccess and this contraction could last much, much longer. We are going to come out a different people on the other end of this period of adjustment.

Posted by: ConcernedCitizen | December 29, 2008 at 09:56 AM

I read an IMF study that said that recessions based on monetary problems compared to inventory and stock aberrations last four times as long and is twice as deep. I believe the logic is that distortions in the monetary area are much more damaging than distortions in inventory and stock prices. The loss of capital is so great and so diffused in this downturn that it will take much longer to adjust and come back to equilibrium

Posted by: Michael A. Cuttler | December 29, 2008 at 09:57 AM

The comfort from comparisons to pre-WWII era contractions may be colder than the current weather. By 1884, the beginning date for Professor Romer's data set, the U.S. was the largest national economy in the world and its future prospects were not subject to question. While the U.S. is still the largest sovereign economy, its national credit and the status of the dollar as a reserve currency are now genuinely in doubt. Therefore, it may be more instructive for economists to look to the last time that the U.S. had to restore its national credit, the period following what is called The Panic of 1873. Those five plus years were - by conventional measure - the longest recession in our history.

Posted by: LetUsHavePeace | December 29, 2008 at 09:57 AM

Since the list of recessions is short, can you provide the length of each recession acording to Prof. Romer?

Posted by: tyaresun | December 29, 2008 at 09:57 AM

A primer on "quantitative easing" in layman's terms:


Posted by: dl | December 29, 2008 at 09:57 AM

Are these recessions as defined by the NBER or as defined by two consecutive quarters of negative economic growth?

Posted by: MW | December 29, 2008 at 09:57 AM

First, the length of pre-World War II recessions is inapposite to the current recession because of the activities of central banks. The Federal Reserve system and the vigorous application of monetary policy to recessions is a modern phenomena that makes comparison to recessions of nineteeth century perilous at best. In effect, the Fed itself is a "new" exogenous factor and hence comparisons are not genuinely possible to recessions of a pre-central bank era.

Second, the lack of reliable 19th century data also makes comparison perilous. The paucity of data is somewhat analogous to the climate change debate because of the lack of information available about not only the 19th Century, but also earlier eras.

Last, this recession differs chiefly because of the interrelated, world-wide economies that are driven by an increasing ease of doing business across national lines and oceans. In short, the electronic transfer of funds and the spread of market economies, combined with the availability of information through the web, have made this genuinely a "new era." Hence, analogies to prior periods will be weak.

The good news is economics itself, which posits policies and provides tools. The kind of mistakes the Exchequer made to deal with the recession in Ireland in 1845 are quite unlikely to be repeated anywhere.

Posted by: Mike Egan | December 29, 2008 at 09:58 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad