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.

« Are modern recessions different? | Main | Will tax stimulus stimulate investment? »

December 30, 2008

Good news in income growth?

One of my New Year’s resolutions is to be more consistent in responding to questions and comments from the loyal readers of macroblog. Though it remains the case that time constraints prohibit a response to all worthy queries, we’re still listening. Next year we’ll endeavor to give a shout back just a little more often.

In that spirit, I received an interesting inquiry from reader Robert Schumacher:

A cursory examination of the monthly trends in real disposable income in light of the NBER official business cycles suggests to me that a sustained rise in disposable personal income (at least three if not four months) signals the end of the recession is at hand. In that real disposable income rose in October and November how are we to interpret this amidst the dire economic forecasts for the coming year?

It does seem, as Robert suggests, that a sustained rise in real disposable income is characteristic of a typical recession’s end. Using a graphical device from a few posts back, here’s a look at the trajectory of disposable income up to and after December 2007 (the start date of the current recession according to the NBER Business Cycle Dating Committee), compared with the average experience of the previous seven recessions dating from 1960:


As in the previous post, “time 0” represents the peak of a business cycle, or the month a recession begins. The average length of US recessions from 1960 through 2001 was 10.7 months, so the line indicating 10 months from the peak roughly coincides to the end of the average recession over this period.

On average, Robert’s conjecture looks right on track. In the typical case, growth in real disposable income stalls and then begins to pick up three or four months before recession’s end. If you smooth through the spike associated with the stimulus package of late spring, income growth was roughly flat through August but has increased since (and at a reasonably good clip). That would seem to portend well for all of us—and I assume it is all of us—hoping for a sooner rather than later end to the current contraction.

The picture is equally encouraging if we look at the income series preferred by the Business Cycle Dating Committee, which subtracts out transfers (that is, payments made to the public by the government):


That’s all encouraging, but there is a caveat: Individual results may vary. Here are the comparisons for the long-lived (16-month) recessions of 1973–75 and 1981–82:



In these two recessions—which are arguably better benchmarks than the average at this point—income measures were not such reliable harbingers of expansion.

Still, in current circumstances a glimmer of hope is better than nothing.

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

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

December 30, 2008 in Business Cycles , Data Releases | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference Good news in income growth? :


Doesn't it bother you that a big chunk (at least) of this real disposable income growth is probably due to the decline in the CPI over the same period? Sticky wages in a deflationary period would be very worrisome.

Posted by: Aaron | December 31, 2008 at 09:09 AM

If you compare real income to real spending you find that during an expansion and a recession real income tends to be a concurrent indicator. But in recoveries real income( excluding transfers) tends to be a lagging indicator.

Posted by: spencer | December 31, 2008 at 11:48 AM

Aaron beats me to it...basically plummeting gas prices, but also hard to put a finger on any economic activity (other than bk filing) that might account for this income "surge". The stimulus package (a step function then as the checks went out all at once, yes?) of $600/adult appears to have lasted 3 months --longer than I would have imagined...and surprisingly more than the savings from cheaper gas prices --did consumers actually drive that much less?
Not to B even gloomier than Aaron, but I know of some who are working longer hours and/or for less...giving me the other impression that DPI is moving down, not up...in line with fewer jobs for more qualified applicants.

Posted by: calmo | December 31, 2008 at 11:48 AM

I have the feeling that the stimulus spike will be repeated and the future versions of the graph you show will come to resemble an electrocariogram after defibrillation treatments. Let's just hope the patient recovers.

Posted by: don | January 05, 2009 at 08:34 AM

This recession differs from 73-75 and 81-82 by virtue of price deflation, a phenomenon more associated with the Great Depression. A rise in real disposable income is hence a symptom of a deep causal problem, the lack of confidence in an impaired financial sector. However, effect can become cause. Or, put in slightly differently terms, when the price of gasoline drops dramatically, consumers are freed immediately to buy other goods. The price of gasoline works as a giant brake, or a drag on the American economy. The drag has been greatly ameliorated and thus a rise in real income can harbinger recovery, i.e. disposable income becomes a leading indicator.

The end of this recession is not at hand because of this data, but it is going to help end the recession, as effect morphs into cause.

Posted by: mme | January 05, 2009 at 08:35 AM

THE problem we face is that the principal cause of our GDP growth for five years was absurdly leveraged speculation. It's folly to propose reducing over-leveraged money center banks will in any way address the problems we face.

Posted by: bailey | January 05, 2009 at 08:35 AM

David - a nice dissection and response. Thanks. I too feel/conclude that the recent upblip in income (& real wages) btw are somewhat misleading because of the sudden drop in inflation. Let's "see" what happens as unemployment worsens. My view is that consumer spending - the engine - is based on real wages, jobs and wealth. We've just taken the biggest hit in wealth in the post-war world so the asset ATM is gone. I've found that the sum of YoY changes in Employment and Real Wages (hattip Joseph Ellis) is powerfully indicative of future consumption spending and it's still headed down. Paul Kasriel at Northern Trust gets nearly identical results looking at (my recollection is bad on this) the product of employment and wage changes. His forecasts are well worth reviewing.

Posted by: dblwyo | January 05, 2009 at 08:35 AM

Aaron -- You are correct that the November increase in real disposable income was a result of the monthly decline in prices. Here is the data series showing nominal DPI:

July -- $10765 billion
August -- $10649 billion
September -- $10659 billion
October -- $10676 billion
November -- $10664 billion

It's not clear exactly what to make of the data -- it's still real income that matters. But your and calmo's skepticism in light of other data seems justified. As always, we will have to wait and see.

Posted by: David Altig | January 05, 2009 at 01:51 PM

Susan Woodward and Robert Hall also have a blog post related to my last comment at their blog "Financial Crisis and Recession": http://woodwardhall.wordpress.com/2009/01/02/consumption-surprise/.

Posted by: Dave Altig | January 06, 2009 at 09:38 AM

I wonder how the same series would look using "core" inflation, which seems to have fallen from grace these days?

Posted by: John G | January 09, 2009 at 12:04 PM

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