Recessions Often Begin With Positive GDP Data

After the Advanced GDP came out last week at +0.6%, I was surprised to read a variety of commentary about the economy that was factually incorrect. Several pundits and economists had concluded that since GDP was positive, we therefore could not possibly be in a recession

The meme "Positive GDP = No Recession!" is demonstrably false, as we show in the proceeding pages.

It took only a brief look at historical GDP data to unequivocally prove this to be the case. We used publicly available GDP data from the Bureau of Economic Analysis and from the Federal Reserve Bank of Philadelphia. The dating of recessions was as per the official tables kept by the Business Cycle Dating Committee of the National Bureau of Economic

The data so overwhelmingly proves that Recession can and often do begin with positive GDP, that one suspects the people making opposite arguments must never have actually reviewed any GDP data beyond the most recent headline. I have no other explanation for why so many people got this so wrong.

Before we go to the actual data, briefly consider just what a recession is. As formally defined by the NBER, it is the "Peak to Trough decrease in business activity" during an economic cycle. The peak marks the end of the expansion phase and the beginning of a recession. During the other phase of the cycle, between trough and peak, the economy is in an expansion. This is described as the economy’s "normal state."

Given that the NBER dates the beginning of a Recession from the economic peak in business activity, one would expect that GDP during that quarter would be mostly positive — not negative. And in fact, that is what the historical data often shows.

1. Many Recessions begin with a Positive GDP

Let’s look at a the beginning of several post-WWII recessions:

• The 1980 contraction was officially dated from January 1980 through July 1980. GDP for the first quarter of 1980 was +1.09%. This contraction lasted only 6 months.

Note the 1980-82 period can be called a "double dip recession, with the next contraction beginning exactly 12 months later — July 1981 — and running another 16 months to November 1982. 

• The deeper 1973 recession ran for 16 months, from November 1973 – March 1975. That first quarter GDP was a positive +1.34%.

• The 1957 recession began with a GDP reading of +1.78%. It ended 8 months later in April 1958.

• GDP in the fourth quarter of 1948 was +3.61%. That 11 month recession was dated from November 1948 to October 1949.

• Lastly, its also worth noting that the 1960 and 1969 recessions began almost flat — they had a marginally negative GDP number of -0.05% and -0.33% respectively.

Hence, the historical data shows that recessions do not always begin with negative GDP numbers,. Of the 11 post WWII recessions, 4 started with positive numbers, two were flattish.

1980:01    1.09%
1973:04    1.34%

1957:03    1.78%

1948:04    3.61%

1969:04    -0.05%
1960:02    -0.33%


Leading Quarter of 6 Post WWII Recessions, GDP


Data source Bureau of Economic Analysis, Federal Reserve Bank of Philadelphia
(Note: I will update this chart with the 1969 recession)


2. Advanced GDP that was Positive at beginning of recessions can be revised to Negative.

In the two most recessions, the Advanced GDP data was reported as
positive, only to be revised to a negative Final number as more data
became available.

Indeed, we frequently see revisions in GDP data — in both directions — as we move through each subsequent release from Advanced to Preliminary to Final.

The business contraction of 2001, as well as the consumer led
recession of 1990, each began with a positive Advanced GDP release.
However, the subsequent revisions took the GDP data in the negative
column.   In fact, there were actually two quarters of positive GDP
data during the 2001  recession which subsequently were revised to


Advanced GDP Data, Final GDP, and Revisions
2001 Q1     2.0%    -0.6%
2001 Q2     *0.7%    -1.6%
1990 Q3     1.8%  -1.6%

* 2004 Revision  


Data source Bureau of Economic Analysis, Federal Reserve Bank of Philadelphia

Merrill Lynch’s North American Economist, David Rosenberg noted that
this is "why the NBER, unlike the media, don’t place as much emphasis
on GDP when it makes the recession call."

3. Some recessions never have two consecutive quarters of negative GDP

Lastly, consider the traditional measure of recessions as two
consecutive quarters of negative GDP growth. That never occurred during
the 2000-2001 recession:

3Q00: – 0.5%
4Q00: +2.1%
1Q01: – 0.5%
2Q01: +1.2%
3Q01: – 1.4%

Although we saw three quarters of falling GDP in 2000 and 2001, none were consecutive.

"Brief & Rare:"

The NBER notes that "Expansion is the normal
state of the economy; most recessions are brief and they have been rare
in recent decades

The recent comments from Messrs. Mankiw, Wesbury, Kudlow and
Pethokoukis that an Advance GDP Release of 0.6% precludes the
possibility of a recession were not only factually incorrect, they were
quite puzzling. To be blunt, these are people that ought to know
better. Chalk it up to Cognitive Dissonance.

I cannot understand why so many people fight desperately against
any mention of what has come to be known euphemistically as the R word. Its not only silly — its bad economic


Special thanks to Michael Donnelly, Chief Economist, PBP for all his assistance in the preparation of these comments. You can see his most recent commentary on this subject at CEO Economic Update.


Congratulations! Its a Recession! (April 30, 2008)

Challenge for Economists: Positive GDP Recessions (May 6 2008)

Business Cycle Expansions and Contractions 
National Bureau of Economic Research

Recession 2001: Business Cycle Dating Committee
National Bureau of Economic Research 
NBER, November 26, 2001

Economic Research

Bureau of Economic Analysis

Confidence or complacency   
David Rosenberg
Merrill Lynch 05 May 2008

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. Matt commented on May 7

    You don’t think that our pols and the economists that prepare their data have an incentive to avoid the R word? Admitting recession depresses consumer confidence which depresses consumer spending which is 70% of GDP. Seems logical.

  2. DL commented on May 7

    Much of this debate has more political significance than investing significance.
    To the investor, there is not much difference between GDP growth of +0.5% or -0.5%. But from a political perspective, there is a big difference between GDP growth of +0.1% versus “growth” of -0.1%. Which raises the possibility of political pressures on the NBER and/or BEA to massage the data one way or another.

  3. The Financial Philosopher commented on May 7

    I believe it is more useful to define the economy as “strengthening” or “weakening” rather than fighting the world over the definition of recession.

    Whether we define current economic status as recession or not, most would agree that the economy is “weakening;” therefore, the conditions for corporate profits going forward are weakening as well.

    Let’s not get caught in “the smart-talk trap” of attempting to use knowledge to reason where reason is present…

    “I prefer the company of peasants because they have not been educated sufficiently to reason incorrectly.” ~ Michel de Montaigne

  4. Bob commented on May 7

    The recent debate reminds me of the old joke: “Economists have predicted 9 of the last 5 recessions.”

    Recessions are historical events by the time the official start and end is declared by the NBER. The excessive focus on recession and various bubbles and crises can also become a self-fulfilling prophecy.

    Most standard economics textbooks use the tradtional rule of thumb definition of 2 consecutive quarters of declining GDP. It would be more accurate to use real and per capita figures.

    I am faced with a dilemma. I teach high school economics and have to use the textbook definition since it is the definition that will be used on standardized tests. I also try to bring as much current events into the classroom as possible. This current debate is useful, yet it is confusing to many students as well.

    A lot of the dissenters about the recession are educated people who have only been exposed to the textbook definition. A big problem for students is that they equate recession as being totally bad. When it is not the case.

    So how do you inform about the basics and also get them to understand the current reality?


    Tough call — educate them, or teach to the test.

    BTW, I don’t know the origin of that joke, but as a group, economists on Wall Street have forecast exactly ZERO of the past 11 recessions.

  5. Stephen Keith commented on May 7

    Recessions in the aggregate (i.e., not per capita) start when the derivative of the economic growth rate turns negative, i.e., the change in “y” as “x” changes has peaked when the derivative of the growth rate is zero, after that, you’re in negative territory, so far as the slope of the growth-rate line goes. This is what I think you and the NBER are getting at–a declining rate of growth, no matter whether the growth remains positive throughout its decline, can be considered a recession.

    But really, aggregate GDP is lousy for measuring welfare. Per capita GDP is what matters, and by that metric we’ve been in a recession, by dint of actually negative growth rates, for two quarters now.

  6. dave54 commented on May 7

    REAGANOMICS: Falling interest rates, commodity prices, wages, taxes and regulation. Higher budget deficits, home prices, stock buy-backs and equity prices.

    ALZHIEMERS: Rising oil prices, commodity, prices, budget deficits, taxes, regulation (on CO2), interest rates and wages…Lower home prices, stock prices and economic growth rates.

  7. Alan Greenspan commented on May 7

    Recession….hmmmm…I suppose everyone noticed the revision DOWNWARD of last month’s housing numbers when today’s were released….and also note that even with the Fed’s gyrations the long term bonds have been going up in yield recently….I imagine this is due to both US and foreign investors becoming more and more skeptical that the Fed is on the right path….

    Frankly, it would be newsworthy ONLY if we had avoided a recession…

  8. kio commented on May 7

    Your efforts are tramendous to prove recession. It looks a bit biased, however. Sometimes it is worth to get at a more balanced point. It helps me feel how far I am one-sided when I find three to five arguments against current view.


    BR: Kio,

    I am not trying to prove there is a recession — I am attempting to demonstrate that those folks who claim that “+0.6% = No Recession” are incorrect.

    The data shows their conclusions are incorrect.

  9. John commented on May 7


    Regarding your teaching dilemma, when I teach adults investing and economics, I present the textbook material and then explain what happens in the real world. Admittedly I don’t have to worry about standardized tests, so that makes your job harder in presenting alternatives that might confuse high school students. Good luck!

  10. Estragon commented on May 7


    One problem with looking at the big picture is forgetting it’s made up of pixels.

    Defining a recession is a bit like trying to find an average American with exactly 2.3 children. Both the recession and the average American are abstractions of reality, not reality itself.

    In reality, the expansion/recession phases of the business cycle occur at the level of individuals, and these cycles interact with each other in different ways depending on the degree to which they’re in phase. Most residential construction has been in recession for at least a year, financials might be six months in, and commercial construction is likely just starting. These recessions have been out of phase with other individual cycles (eg. segments of tech, energy, commodities, etc.) so far. If and to the extent the individual recessions line up in phase, we’ll have what everyone describes as “a” recession.

  11. michael schumacher commented on May 7

    the only thing missing in the WSJ piece today is the “mission accomplished” banner across the front page.


  12. Winston Munn commented on May 7

    There is still so much damage yet to be accounted – no end in sight to the fall in home prices, the overbuilt-overleveraged commercial real estate being finished this year, and the big drop in governments revenues – that the argument about recession may well be mute, and we’ll be seeing in the not-too-distant future definitions of depression, instead.

    Almost 50% of the Fed’s balance sheet is MBS and now they are to accept student loans and credit card debt, too?

    When the Fed Swap-o-Meter hits 100%, what then?

  13. Winston Munn commented on May 7

    There is still so much damage yet to be accounted – no end in sight to the fall in home prices, the overbuilt-overleveraged commercial real estate being finished this year, and the big drop in governments revenues – that the argument about recession may well be mute, and we’ll be seeing in the not-too-distant future definitions of depression, instead.

    Almost 50% of the Fed’s balance sheet is MBS and now they are to accept student loans and credit card debt, too?

    When the Fed Swap-o-Meter hits 100%, what then?

  14. DL commented on May 7

    “The recent comments from Messrs. Mankiw, Wesbury, Kudlow and Pethokoukis that an Advance GDP Release of 0.6% precludes the possibility of a recession were not only factually incorrect … Chalk it up to Cognitive Dissonance”.

    All four of these guys are gunning for McCain. And Brian Wesbury wants a Fed funds rate above 4%.

  15. paul commented on May 7

    Barry, What part of “This is an election year” don’t you get?

  16. AGG commented on May 7

    They pretend to tell us the truth and we pretend we believe them.

  17. Bruce commented on May 7

    Can anyone tell me what per centage of the massive increase in consumer credit released today was credit card debt?

  18. Alfred commented on May 7

    Barry is absolutely right in his assertion of positive GDPs at the beginning of past recessions and he is absolutely wrong in calling the current slowdown a recession. NBER looks at lasting, significant declines in real GDP, real income, employment, industrial production, and wholesale-retail sales. All those indicators have weakened but it is simply too early to call it significant and lasting. Barry’s guess is as good as Brian Wesbury’s.

  19. Estragon commented on May 7


    No, but the increase appears (on a seasonally adjusted basis) to be about equally split between revolving and non-revolving debt, skewed a bit towards revolving in percentage terms. Details here

  20. BG commented on May 7

    This thread is so indicative of the political/financial community when it comes to spinning things the PC way.

    Recession/No Recession? Who gives a rat’s ass? It is a MOOT POINT!! Things are bad out here and getting worse and anyone with half a brain in THIS world already knows it!

    The only thing that remains constant is the Government hand-wringing and blather; but when it comes to actually doing something. Oh…well…I mean, shit…ah,… we’ll be OK.

    Actually, I don’t think there is all that much hand-wringing any more. From my perspective, the current Administration doesn’t really give a diddly damn, OR they don’t want people to realize just how little influence they have left with people who can make a difference.

  21. Melancholy Korean commented on May 7

    “I cannot understand why so many people fight desperately against any mention of what has come to be known euphemistically as the R word.”


    Keep Iraq just stable enough until it’s someone else’s problem. Keep the economy stable enough through the election (it’s not a recession!) until it’s someone else’s problem. Tuesday’s Washington Post article re our dear Leader’s schedule was unbelievable. Two events for the entire day. One was a dinner party. The other was the postprandial entertainment.

  22. Ivo commented on May 7

    “I cannot understand why so many people fight desperately against any mention of what has come to be known euphemistically as the R word. Its not only silly — its bad economic analysis.”

    Well, a good question, I keep thinking of it all the time as well.

    Some obvious proposals:
    1. Politics
    2. Sell-side self-interest
    3. Dependence on consumer sentiment to boost corporate profits and lobbyst’s pay.
    4. I live in Germany – here most common people are not (heavily)invested in the equity markets and have savings and state pension and healthcare. So their income and well being is not that dependent on the stock market at least mid-term. In the USA the stock market seems to have become “THE common good” – everybody is invested and depends on its inflation and this resembles the state ownership and the planned economy of communism in a strange way at the end (common ownership). A recession is bad for stocks and bad for the common wealth. The stock market seems far detached from reality, it has become more of a false source of a well-being promise and propaganda, totally detached from reality and fundamentals.
    5. People like to think in black and white and prefer “some but now” than “more later”. People have not changed through history, only their toys, so they never learn.

    Just thoughs…

  23. Winston Munn commented on May 8

    “Can anyone tell me what per centage of the massive increase in consumer credit released today was credit card debt?”

    About 1200%, after securitization and being traded to the Fed for treasuries.

  24. SocraticGadfly commented on May 8

    “Positive GDP”? Please. As Kevin Phillips points out in his new book, those numbers have been finagled, including changing GNP to GDP, for more than 40 years.

    I don’t for a minute believe we actually had “positive GDP” last quarter.

  25. kio commented on May 8


    +0.6% is obviously +-1% (or even more for quartarly estimates) uncertain, as analysis of NIPA revisions demonstrate and as BEA people say in their papers available at BEA site.
    Even after comprehensive revisions one can not be sure about real GDP growth rate. So, +0.6% and -0.4% (recession?) is esentially the same.

  26. Vladislav Iordanov commented on May 8

    Please check your numbers. First you say that the 2001 Q2 final GDP reading was -1.6% but further down you say it is +1.2%. Both times you use it to show that your “theory” is correct. Do you fit the numbers to match your “theory” or your “theory” to the numbers? If you are arguing that we are in a recession, you should know that a reading of -20,000 non-farm payrolls change is not statistically significant from 0 which is unusual for the “worst recession since the Great Depression” as it is called by the left press. Usually you see numbers like -200,000 to -400,000 during recessions. If you go back to the 1950s, you can see monthly changes in seasonally adjusted employment of -1% which given the 137 million payrolls today will translate to 1.37 million lost jobs in a month. You have to look at the numbers as an aggregate and so far the GDP, non-farm payrolls, unemployment, payroll taxes, service industry growth doesn’t point to recession.
    Probably slowdown.


    BR: For 2001 Q2 (#3 above) those numbers include the subsequent revisions in 2004 (fixed above). Hence, we get yet a different number from the Advanced and Final releases.

    But do not miss my point: I am using the data to show that merely because Advanced GDP is positive does not preclude the possibility of a recession.

    Additionally, do not assume that recessions must have 2 consecutive quarters of negative GDP, as 2001 showed.

    I am not responsible for what the media or other economists say or do. I am using the actual data to show that the factual declarations of Mankiw, Wesbury et. al were wrong.

  27. Ivo commented on May 8

    Hi Vlad,

    First, what Barry seems to argue is the fact that reported number is not negative, does not mean that it is not a recession.

    Second, you cannot compare data of 1950s with today’s – the accumulated distortions and manipulations are just too high. If you adjust for that bias the data may look even worse than any data you have seen since the 1950s.

  28. Mike Milken commented on May 8

    We believe the U.S. economy tipped into a mild recession in the first half of 2008. A severe correction in housing markets, oil prices that breached the one hundred dollars a barrel barrier, the weakening labor market, overextended consumers, and a continuing credit crunch have overwhelmed the underlying strength of export markets.

    The result will be a modest decline in overall economic activity. Our outlook calls for real GDP to decline at an annual rate of 0.6 percent in the first quarter of 2008, followed by a 0.9 percent drop in the second. We expect that the “advance” GDP report for the first quarter, released April 30 and estimating an increase of 0.6 percent, will be revised to show a decline.

    The good news is that Congress and the administration have reacted quickly, uncharacteristically so, to enact fiscal stimulus measures. The Federal Reserve, meanwhile, has aggressively cut interest rates and is supplying new liquidity to the financial system. These actions should mitigate the severity of the contraction and nudge the economy on an upward trajectory in the second half of the year, making this the mildest recession in the post-World War II period.

    However, the potential exists for other downside risks to precipitate a far more severe economic contraction. And by then it won’t matter to consumers, homeowners, investors, or businesses whether or not the National Bureau of Economic Research (NBER) defines the downturn as a recession—because it will certainly feel like one.

  29. drop73 commented on May 9

    This article addresses what has been the case in the past and attempts to find a similar pattern for today. GDP reports are backwards looking. What we should be looking for are signs of improvement or decline, such as whether there is a current increase or decrease in home prices from the mortgage crisis, or if banks have the current ability to balance their balance sheets with projected mortgage defaults to double in 2008 compared to 2007. We know the answer/outcome to these two questions, but somehow our most renowned economic advisors are telling us the problems don’t exist. No matter how long we keep our heads buried in the sand, the mortgage debt and bad securities we’ve created is not going to just go away. If you treat houses like a commodity, don’t be surprised when they behave like a commodity, and most people in the last 7 years bought high.

  30. BobC commented on May 10

    I’m surprised at that lack of logic displayed in many of the responses to this article. It’s pretty straight forward. The premise is that recessions have not always had negative growth at their inception. The author then proceeds to provide empirical data that proves his assertion. That’s it. Nothing more. He never actually states that we are in a recession or even that we will be in one soon. He only says that a +.6 growth in GDP does not prove we are NOT in recession.

  31. Barry Ritholtz commented on May 11

    Thanks BobC. I dislike when the comment thread gets polluted with incredibly silly or ignorant comments.

    I like when people disagree and challenge my reasoning and logic — but with data and intelligence, not idiocy.

    To maintain the intellectual integrity of the site, as well as to delete what looks like trolls, I will unpublish those comments that fail logic or econ 101.

  32. Scott commented on May 13

    Barry, my man. I believe I erred in making an ad hominim attack re Kudlow in what had been my first and only post. This was made prior to the “asshat” email. Guilty as charged. Consider me chastened. Please let me stay in the playground and I’ll be good. That said, as an old Merc guy, who came across Kudlow occassionally back in his Bear days: he was more fun in the 80s!! ;)

Posted Under