December 01, 2008
Recession Dating: Some People Are Going to Be Surprised
"The data are pretty clear that we are not in a recession," Council of Economic Advisers Chairman Edward Lazear told a meeting of editors and reporters from the Wall Street Journal and Dow Jones Newswires.
"I would be very surprised if the NBER, looking back at this period, would date this as a recession," Mr. Lazear said. There are even indications that revised first-quarter estimates would be slightly stronger than 0.6%. "The optimists seem to have been closer to right on that than the pessimists," he said.
Just to reiterate, that quote is from May 2008.
Figure 1: Gross domestic product (blue), and gross domestic income (red), in Ch.2000$, SAAR. Real GDI calculated as nominal GDI deflated by GDP deflator. NBER defined recession shaded gray, and dashed line at NBER peak. Source: BEA GDP release of 25 November 2008, and NBER.
I thought at the time Ed Lazear would regret those remarks. But Lazear's views were not unique. Here are some additional quotes of interest, hoisted from the comments sections in Econbrowser:
From March 2008, in response to my worries about 2008H2 growth:
While the financial market turmoil and dysfunctional credit markets are significant wild cards, interest rates are so low (and could be 2.25% this week!) that you have to think that growth later this year and into '09 should be positive, if not strong.
The Macroeconomic Advisers monthly real GDP index rose to a record high in December. It fell in October but rebounded in November and December.
Preliminary indicators suggest it rose again in January.
I get a sense that some academic economists are actually rooting for a recession, a way of punishing the country for having elected Bush.
In other words, looking at data too close to the candidate turning point can easily lead to subsequently embarrassing remarks. Nonetheless, the specific dating of the recession to December 2007 has spurred lots of queries about whether it matters, when (i) the recession is so much in evidence, and (ii) whether any greater importance should be accorded to a declaration by the NBER as opposed to any other group or individuals. Here are some thoughts on both questions, augmented with some retrospective views.
First, to the question of whether it matters that there's a declaration. Here, I turn to Jeffrey Frankel, who is on the NBER BCDC, but is speaking on his own behalf:
We sometimes hear the question "Who needs the NBER Committee anyway?" This question most often comes in one of two forms:
(a) Everyone in the real world has known that the economy has been in a recession for some time. In past cycles, media reports have sometimes taken the line "Ivy Tower Eggheads Finally Figure Out What Everybody Else Has Known All Along." The implicit critique is that the committee takes too long after the event -- typically almost a year -- to make its declaration. One short answer is that our job is to be definitive, authoritative, but not fast. We don’t want to have to revise our dating of the peaks and troughs later, in part because it would sow confusion among those who rely on them (from econometric researchers to political speechwriters). GDP and other official statistics are often revised after the fact, for example. We leave it to others -- pundits, forecasters, consulting companies, financial newsletters, and so on -- to try to get there first. We deliberately get there last.
(b) The other form taken by the question "Who needs the NBER committee?" runs as follows: "The rule of thumb is simple: two consecutive negative quarters of GDP growth. Why complicate things?" The Frequently Asked Questions segment of the BCDC announcement answers this in detail. For now, observe simply that questions (a) and (b) are inconsistent with each other. As of December 1, 2008, the US economy has not yet experienced two consecutive negative quarters. So an argument that we should wait for two consecutive quarters (critique b) is the opposite of the critique that we should have acknowledged a recession before now (critique a).
Taking Frankel's points into account, we can then move to the second question, whether it matters that NBER BCDC makes the determination. Think back to the two-quarter rule of thumb. Why is it a "rule of thumb"? Because the data are revised over time -- most importantly GDP -- the two quarter (sometimes defined as a "technical recession") is heavily reliant upon the series that is most subject to revision. I've discussed the importance of revisions numerous times in this context: , , , , .
Another motivation is to recall that the NBER is a research group, and not a policy group, trying to identify for historical purposes the ebb and flow of activity. In any case, we want to place the responsibility for identifying peaks and troughs in economic activity in the hands of a disinterested group, and not in -- say -- the government. (I'll add that it's an auspicious group: "Robert Hall, Stanford University (chair); Martin Feldstein, Harvard University and NBER President Emeritus; Jeffrey Frankel, Harvard University; Robert Gordon, Northwestern University; James Poterba, MIT and NBER President; David Romer, University of California, Berkeley; and Victor Zarnowitz, the Conference Board. Christina Romer of the University of California, Berkeley, resigned from the committee on November 25, 2008, and did not participate in its deliberations of November 28.")
(For those wondering why NBER gets to declare the recession dates, one hint is that it was NBER researchers Burns and Mitchell (1946) who developed the lens through which "cycles" experienced in industrial economies were perceived. For the inquisitive, here is a link to the volume which summarized that research: Burns, Arthur F. and Wesley C. Mitchell, Measuring Business Cycles (NBER, 1946).)
This is why I think so much of the commentary, saying that this was all obvious, not only demonstrates a distressing level of ignorance, but also a juvenile approach to discourse (see e.g., here). Yes, it was pretty obvious that we were in a recession, but when it began is something that could have been placed at any number of dates. We will have a similar debate when we starting thinking about dating the trough.
I think it's also of interest to look back at whether we should be so surprised. I'll focus on the debate that occupied our attention months ago: whether the yield spread had predictive power for future economic activity. Well, now we can conclude the answer was yes. The inversion in August 2006 predates the recession's beginning by about 16 months.
Now, what we don't know is how well the prediction holds cross-country. This graph from a March 2007 post suggests it will work for Europe (we'll have to see what CEPR says, although current forecasts seem pretty definitive).
Figure 2: Ten year - three month term premium, daily averages. For US (blue), ten year rate is constant maturity, three month rate is for T-bills in secondary market; for Euro area (red), the ten year rate is the GDP-weighted rate of on-the-run benchmark bond yield, three month rate is for interbank money rate; for Germany (green), the short rate is the daily interbank rate. NBER recession dates shaded gray, CEPR recession dates (converted from quarterly dates) shaded light blue. Squares are data for March 9 (where Euro ten year rate proxied by arithmetic average of benchmark bid yields for Germany, France, Italy, Netherlands and Spain). Sources: St. Louis Fed FREDII; IMF, International Financial Statistics; Financial Times; NBER; CEPR; and author's calculations.
Here's a cross-country snapshot from an October 2007 post.
Figure 3: Ten year benchmark bond yield minus three month yield spreads, from Economist, Oct. 12, 2007 and Oct. 11, 2006 issues, and author's calculations.
Australia and the UK seemed to be conforming to the prediction from the yield curve. Canada to date has failed to experience negative growth, and so might be thought to also conform to the prediction of yield curve -- at least so far.
Posted by Menzie Chinn at December 1, 2008 09:27 PMdigg this | reddit
So the recession started in December 2007 and since many feel it will not be "that bad" (say less than a year). Then it's already over! All these layoffs and stock market losses are just trailing indicators.
Posted by: oc bear at December 2, 2008 12:21 AM
Seems like a lot of economists could use a sympathy card.
Posted by: DiscreditedEconomist at December 2, 2008 02:20 AM
hm, the comments on the wsj page you linked to fit its nutty editorial pages quite nicely. no surprise here... ;-)
Posted by: zz at December 2, 2008 03:37 AM
I think you are reading too much into what was a very close call.
Posted by: Rich Berger at December 2, 2008 06:53 AM
I still don't think that it is obvious that it started in December of '07. October of '08, yes, that's obvious, but not all the way back to December of LAST YEAR? I don't see it.
Maybe my mind is warped because of the industry I work for and the area I live in, but even now, it is not all that obvious that we're in anything more than a mild recession. The parking lots of the stores are full. Restaraunts are full. This ain't the Great Depression!
Also, it is hard to understand why commodities spiked if the US was in recession.
Posted by: Buzzcut at December 2, 2008 07:12 AM
This illustrates the challenges of attempting to affix binary labels to continuous and complex phenomena.
And it's not as if affixing the label "recession" to a period provides us special benefits, so one wonders about the energy devoted to it...
Posted by: diz at December 2, 2008 07:43 AM
Clearly the data have indicated that the economy has been weak for some time. I suspect when Chairman Lazear made the statements he was basing them off the data series the NBER claims they use to date recessions. Given his read of the data, the early part of the year did not look like the economy fit into what is typically classified as a recession by the NBER. Edward Leamer's recent paper on "is this a business cycle" provides some rules of thumb to identify past calls on recessions (that do not have false positives). Presumably the intent of this paper is help provide guidance when the next recession may get called in 'real time'. According to his metric the early part of the year did not look like a recession to him either.
On another note, do we actually care about the nominal identification of a recession? We typically do not wipe or brow and say the economy is doing well when GDP growth is 0.1% and become hysterical when GDP growth is -0.1%. Everyone has known the economy has been soft for some time, and I suspect people making policy decisions are as concerned about -.1% GDP Growth and rising unemployment as if GDP growth was .1% with similarly rising unemployment. Economists provide valuable services; dating recessions 10 months after the fact is not one of them.
Posted by: Newton's_second_law at December 2, 2008 07:58 AM
Also, it is hard to understand why commodities spiked if the US was in recession.
One word: China
Posted by: Imelda Blahnik at December 2, 2008 08:04 AM
diz and Newton: I discussed that issue in my post last June on Is this a recession and do we care?
Posted by: JDH at December 2, 2008 08:08 AM
NBER-as a research group they sure seem to enjoy all the attention they get when they influence policy.
Posted by: L Moore at December 2, 2008 08:31 AM
I think the more important question is whether a second--and far more serious-- dip of the recession is ahead. If this were the bottom of the recession, we'd call it relatively mild. But it does not feel like the bottom of the recession to me.
And here's a related question: we are creating so much money out of thin air, how does that affect the GDP statistics?
Posted by: Charles at December 2, 2008 09:33 AM
Jim Lindgren is claiming that NBER said that the recession started in January 2008. His analysis seems okay, but it means that a bunch of other people can't read. Is he right?
Posted by: Dan Weber at December 2, 2008 09:53 AM
" not all that obvious that we're in anything more than a mild recession. The parking lots of the stores are full. Restaraunts are full"
Obviously you are no where near Atlanta. Maybe you have also failed to note the increase in unemployment.
Posted by: me at December 2, 2008 10:00 AM
Prof. Frankel's comments got my mind whirling. So the NBER is there to map out the bad things, to recognize that "yup, things have gone too far to the downside," and now we are seeing negative growth, now if any part of the fleet is not yet there, get it going.
In turn, if we are now starting to come to the other side of the general acceptability of academics and researchers to 'see' and chart bubbles, would it be worthwhile for something similar to the recession dating committee to establish a Bubble Dating Committee? Would this not give us an early warning system, and do so in an institutionalized manner?
Posted by: Mr. L at December 2, 2008 11:00 AM
Obviously you are no where near Atlanta. Maybe you have also failed to note the increase in unemployment.
So the unemployed are going out to eat and stopping off at Best Buy to pick up a sweet 1080p Plasma afterwards?
Somebody is at the stores. Black Friday sales are up.
Who are these people, don't they know that we're in the worst economy since the Great Depression?!?
Posted by: Buzzcut at December 2, 2008 12:18 PM
People of low quantitatitve intelligence use subjective anecdotes to confirm simplistic views. Examples include "The parking lots are FULL" or "Restaurants/movie theaters are FULL".
Nevermind that :
1) A parking lot being full means that people are unemployed, bored, and spend hours browsing in stores, but NOT BUYING ANYTHING. How do you whether the owner of a parked car spend $0 or $2000?
Parking lots are full at the library too, BTW.
2) Movie theaters and cheaper restaurants have a disproportionate amount of young people who are either in college or working low-wage jobs that are easy to get. Expensive restaurants may be a better indicator, but even then, Salespeople have to entertain their clients at company expense, which still has to continue in a downturn in order to generate much-needed revenue.
3) Even in a recession, 90%+ of people are employed. They may be underemployed, but still continue to live their lives (go to movies, window-shop, etc.). Pray tell, are you able to see people pass in the mall, and judge exactly who is fully employed, underemployed, or unemployed? Their shopping bags may still have only $10 of items, you know.
By the brain-dead 'parking lots are full' paradigm, we have never, ever had a recession in the last 60 years. The only thing that matters is the data, specifically 1) Employment data, and 2) stock market/earnings levels.
Posted by: GK at December 2, 2008 12:31 PM
Lazear might have been able to argue that we were not in recession in May of 2008 as GDP growth remained positive still then, if just barely. But there is no case whatsoever for putting off the start of it until October. Third quarter GDP growth was negative, and all those other elements had already gone south, with employment declining since the beginning of the year, that switch now coinciding with their dating. Lazear's remark was made at what was the last possible point that one could argue for any sort of ambiguity.
Posted by: Barkley Rosser at December 2, 2008 01:24 PM
Again, Buzz, your mumbling. In the early 80's the slasher film type trend was huge. Made 100's of millions. Oh, wait that was the early 80's inflaton blowoff recession. How could they afford that?
Buzz's problem is he doesn't understand the nature of people and recession. Oh yeah, Shirley Temple was huge during the 30's depression. How on earth did anybody care with so little money?
Oh yeah, Black Friday sales were DOWN. Get it straight.
Posted by: Sandman at December 2, 2008 01:38 PM
Lest we forget there are 2 geniuses at Calculated Risk, Bill McBride nailed the recession's origin to December 2007 as early as the next month, January 2008.
Posted by: egc at December 2, 2008 05:21 PM
I nailed the start date as being December 2007, on February 5, 2008. So I am second only to Bill McBride.
Posted by: GK at December 2, 2008 07:24 PM
The reason commodities spiked early this year is because the housing bubble had popped but not yet deflated. In other words, there was this all this liquidity generated by a massive asset bubble, but people were no longer buying houses. Also bonds were looking more and more risky, equities were starting to trail off, so what was left? Commodities my friend, we all piled into commodities. Once the bubble deflated all that liquidity was gone, pulling the rug out from under everything, including housing prices, bond yields, equities, commodities, everything.
Posted by: X Man at December 2, 2008 07:54 PM
Thanks for allowing me to be your whipping boy.
X Man, your explanation makes a ton of sense.
Posted by: Buzzcut at December 3, 2008 06:24 AM
I agree with your assessment, but is 2.8% growth 'just barely'? I think the size of that growth disqualifies Q1 and Q2 from being called a recession.
Posted by: Tom at December 3, 2008 11:19 AM
I agree that Ed Lazear was over confident in his assestment that the US was not in a recession in May of 08, but according to JDH's recession indictor:
The porbability that we were in a recession based on data available at the time was on the order of 20%. Besides, I don't think it is wise for members of the CEA to say publicly that we are in a recession... better to leave that to more indepedent groups like NBER or the FED.
Posted by: MikeR at December 3, 2008 11:28 AM
MikeR: All I'm saying is he could have been a bit more nuanced about his assessment, and saved himself some embarrassment (e.g., "the data we have at this time is not entirely consistent with recession.").
Posted by: Menzie Chinn at December 3, 2008 12:42 PM
Definitely agree with x-man. Thing with the commodities bubble is that it exacerbated the debt bubble. Wages stayed flat, expenses went up, risk went up.
Posted by: aaron at December 4, 2008 12:46 PM
Even with rising inflation, and falling home construction, real GDP contracted at only a tiny 0.2% annual rate in the fourth quarter of 2007, and grew in the first two quarters of 2008, including a quite healthy 2.8% growth rate in Q2.
Not since 1970 has the NBER called a recession for a period including such a strong quarter of real GDP growth (and remember that the 1970 data has been revised substantially in the intervening years). In fact, GDP data were much worse than in recent quarters during 2000-2001 and using them could have led the NBER to date the last recession to mid-2000, not March 2001, and it would have been called the Clinton Recession, not the Bush Recession.
Posted by: Buzzcut at December 4, 2008 02:15 PM
I saw the 2000 recession. When I graduated in december there were no jobs. People who had offers in the summer had them recinded. During the election I suggested it would be a good strategy for the democrats to let the republicans win and take the blame for the recession.
Posted by: aaron at December 4, 2008 02:46 PM