June 23, 2003

It Depends on What a Recession Is

The NBER's Business Cycle Dating Committee continues to hesitate. But it is not the case that more time will make the state of the U.S. economy between the start of 2002 and today clearer. The state of the U.S. economy is clear: slow growth in demand and output accompanied by rapid underlying productivity growth and so declining employment. What is unclear is what the NBER thinks a "recession" is. More time may be needed, but not because more time would shed better light on the state of the economy--rather, more time seems to be needed for the Business Cycle Dating Committee to think through what its system of categories really is.

The NBER's Recession Dating Procedure: According to the most recent data, the U.S. economy continues to experience growth in income and output but employment continues to decline. Because of the divergent behavior of various indicators, the NBER's Business Cycle Dating Committee believes that additional time is needed before interpreting the movements of the economy over the past two years...

Posted by DeLong at June 23, 2003 10:35 AM | TrackBack


Absurd. Is the idea that an economist must really be useless to anyone other than an economic historian and possibly not then? Two quarters of negative GDP growth: recession. Enough.

Posted by: lise on June 23, 2003 11:06 AM

Please do explain -

Brad DeLong noted that low interest rates make a given present value stream of earnings worth more, so price earning ratios can be higher in a low interest rate environment.

Paul Krugman just noted the high price earning ratios for stocks but was doubtful that low interest rates made such high priced stocks a good buy.

Please do explain.

Posted by: Emma on June 23, 2003 11:31 AM

They're no contradictory opinions, actually. Low interest rates affect the WACC (weighted cost of capital) used to discount the future cash flows generated by corporations. Hence, the equity value of the company rises relative to its current net income (earnings). Krugman's observation is simply that interest rates may indeed be lower leading to a lower WACC but that doesn't mean that stocks are necessarily a good buy.

Posted by: JT on June 23, 2003 11:41 AM

Lise you are surely correct but how do you describe the condition that the economy is in at the moment which may not be recession but where something is clearly out of whack? This looks like Goodhart's law for economic definitions.

Posted by: Jack on June 23, 2003 11:47 AM

The problem now is an economy that is growing but growing too slowly to get industry to create enough jobs even to keep up with labor force growth. We have to get GDP growth high enough to absorb natural labor force growth and cut the 3 million jobs lost from the onset of the recession.

We are in a labor market recession. This recession and slow growth period has been unlike any since 1945.

Why should industry begin to add significantly to job creation? Consumer demand has been fairly high these 2 years, but not high enough to cause capacity use to climb past 75%. Productivity growth has stayed high enough to allow industry to meet production needs with fewer workers.

Will the tax cut add enough to demand to cause industry to create several million jobs in the oming year? I am worried!

Posted by: anne on June 23, 2003 12:24 PM

Anne, the answer to your question is "no." There is no chance that the Bush promises of jobs from the most recent tax cut will hold up to scrutiny in november, 2004, although whether the so-called liberal media will notice and report is another matter.

Emma, to continue on JT's point, what Krugman notes is that if we start growing faster, the odds are that interest rates go up and the PEs make less sense than they do now. In short, we could have profits improving and PEs falling simultaneously.

Brad, what would you call the current conditions? Personally, i call them slow-growth, not recession, but that's just because i think the old definitions are as good as any....

Posted by: howard on June 23, 2003 12:35 PM

I agree with Anne. The recession had the character of a cycle before 1945. Perhaps the cycle ending the explosion of railroad growth after the Civil War had a bit of this character. The problem is employment. We need to create 80,000 jobs a month to keep up with labor force growth, and we have to grow fast enough to again absorb those who have lost jobs.

Economists have to make the problems more readily comprehensible.

Posted by: lise on June 23, 2003 12:45 PM

Maybe we need a third category to go with recession and expansion that captures more of the nuances.

How about 'muddle along"?

Posted by: bakho on June 23, 2003 01:03 PM

So Lise, a quarter of say 0.5% growth, followed by one of negative 2%, followed by one of 0.5% growth, followed by one of negative 1% growth, is not a recession? Stylised, but not so dissimilar from Japan over the last 10 years.

Posted by: Matthew on June 23, 2003 01:06 PM

So Lise, a quarter of say 0.5% growth, followed by one of negative 1%, followed by one of 0.5% growth, followed by one of negative 1% growth, is not a recession? Stylised, but not so dissimilar from Japan over the last 10 years.

Posted by: Matthew on June 23, 2003 01:06 PM

A single quarter of negative growth, is a single poor quarter. Two negative is a recession. Slow growth and job erosion in the midst of a bond market boom is quite a serious problem. Deficit growth, with scant job creation, even more of a problem.

Posted by: lise on June 23, 2003 01:13 PM

I am much more concerned about what the unemployment rate represents. The unemployment rate is not seriously high from a historical perspective and yet talk to anyone who has been in the market lately and you know things are bad. Where is the employment weakness being represented in the data that is commonly reported? If we have the answer to that question we don't have to ponder the definition of "recession".

Posted by: Dan on June 23, 2003 01:22 PM

The problem with using the two quarters rule is that "recession" has become a political word as much as an economic word. To real people in the real world - the US is still in recession because it's shedding jobs.

One of the things that disgusted me in the early 90's was hearing about the jobless recovery - to me, if jobs weren't being created, it wasn't a revovery.

So yeah, I know jobs are a lagging indicator (though I didn't back then), but when someone hears "we're out of the recession" they don't think "so in 6 months to a couple years the economy might start adding jobs".

GDP and productivity are both rather troublesome indicators of economic health. Because it ain't just how big the economy is, and it ain't just how much profit is being rung out of it - it's how the economic pie is being sliced. And right now, it ain't being sliced in a way that benefits the poor, the working class or the middle class.

Bottom line: call it whatever you want - for normal people, the economy still stinks.

Posted by: Ian Welsh on June 23, 2003 03:06 PM


Earnings themselves can be affected by interest rates. If rates are lower, companies are willing to make investments which bring lower rates of return because the alternative uses of cash are not so good. At least this should happen in competitive industries, maybe not so where competition is scarce. I'm not really sure about that part.

Many people do not consider P/E ratios to be a primary measure of company valuation for the general reason that changes in competition can change earnings potential. Do a search on "Tobin's Q" to get an alternative point of view to this debate over P/E ratios.

Posted by: snsterling on June 23, 2003 04:23 PM

Dan writes "I am much more concerned about what the unemployment rate represents."

It doesn't help that the definition of 'unemployed' hinges on whether the government deigns to grant you unemployment benefits. If you've run out, you aren't unemployed, even if you don't have a job yet, and are looking.

They cutely say that such people have 'stopped looking', but the lie in this is shown in that there's no mechanism by which a person who "starts looking" will be counted in the unemployment numbers.

Recently saw an article where the author said that if all those people "started looking for jobs again" the unemployment rate would rise to 9.x%. Lucky for Mr. Bush, that can't happen. Even if they all were not looking for work, and all started looking for work en masse, it wouldn't show up in the statistics.

Posted by: Jon H on June 23, 2003 04:27 PM

Here's a naive question, in the spirit of trying to understand the real economy we're living in: we've been hearing from many quarters, including several posts here, that productivity is growing. At the same time we hear that capital expenditure figures are dismal. So where does that productivity growth come from? Could it simply be that people are being cut and not being replaced?

I ask in part because I've heard stories recently about book publishers that do business by fax and only take phone messages to be returned when people can get to them. This must mean that they have a few overworked people trying to manage their contacts as best they can.

That isn't good for customer service and can't be good in the long term for the health of the business, and it would point to an eventual breakdown, or at least increasingly inefficient performance, as time goes on unless business volume falls. Or am I wrong in this?

Posted by: Altoid on June 23, 2003 05:50 PM

altoid, in a giant economy like that of the U.S., all kinds of things are happening.

Improved information flow, for instance, enables productivity gains, but as you say, so, undoubtedly, does the simple fact that fewer people are doing the same work, often out of fear of losing their own jobs.

The question that we are in no position to answer is how much of the productivity gains are sustainable - meaning, essentially, that we really are (dare i say it) working smarter and not harder (couldn't help it, sorry) - and are not sustainable (meaning arising from what used to, in assembly-line and union-organizing days, be called the "speedup").

Posted by: howard on June 23, 2003 07:37 PM

Jon H
It is fine to say the unemployment rate should be 9.x%, but the miscalculation you site is the same miscalculation in years past. In other words, in the early 90's when the official unemployment rate was much higher the same "no longer looking" people were not counted. So relatively speaking, why is the unemployment rate, consistent miscalculations and all, so low compared to other periods of poor employment?

Posted by: Dan on June 23, 2003 08:35 PM

The thread's probably dead, but Dan's is a very good question.

One answer which has been suggested is that the massive increase in the imprisonment rate takes up a lot of the unemployment slack.

If true, that would push the unemployment rate up by at least a percent.

Posted by: MFB on June 24, 2003 03:32 AM

Recession and unemployment become important factors at election time because they influence how people vote. Politically, it does not matter how these are defined. Voters choose based on their personal perception. You can change the way unemployment is calculated, but that does not change the employment scene as experienced by the voter. Politicians believe they can convince people that the economy is better or worse than it really is, but they really cannot. In the end, the politics muddies the economic picture, but does not change it.

Posted by: bakho on June 24, 2003 07:55 AM

There are a fair number of people out there in the market place "looking for jobs" who are seeking, say, a computer programming job paying $60,000/yr. If the market rate for such a job is $35,000/yr, I'm not willing to qualify those people as "looking".

I don't know if it is coincidence, but most of the people I know who have been out of work for a while are still out of work because they haven't adjusted their expectations of salary for the fact that they were wildly overpaid during the tech boom. Most computer jobs involve running queries on databases, simple coding, nothing too creative, nothing that requires more than, or in many cases even, a BA, largest physical risk is repetitive stress injury. In other words, $30K to start, $40K-50K with significant experience would be a good expectation of equilibrium long run wage. How many people here know computer people who wouldn't dream of accepting $30K to run a computer, because they were getting $100K or $60K plus options four years ago?

Even for the more creative jobs, why would the equilibrium long run wage for programmers be higher than that for, say, engineers or marketing research? Programmers typically have the same or less education, and require similar, or less, analytical skill.

Do you know non-computer types out of work? Are they truly looking? (That is, are they spending even 20 hours a week of their free-time seeking employment? Are they calling every temp agency in town looking for leads on positions?)

I'm not sure the "lack of jobs" is necessarily a demand side (of the labor market) phenomenon.

Posted by: rvman on June 24, 2003 08:31 AM

There have been changes to to the models and rules used in determining the unemployment rate since the last recession. Likewise the specific duration of insurance extensions has changed.

As far as the old "not looking" argument - the economy simply has less jobs. Period. It's like musical chairs - look as hard as you like, someone's going to be left without a chair. The labor force participation rate and the labor force itself has been dropping like a rock - that's what's keeping the unemployment rate down.

Posted by: Ian Welsh on June 24, 2003 11:16 AM

"The labor force participation rate and the labor force itself has been dropping like a rock - that's what's keeping the unemployment rate down."

Just so, sadly.

Posted by: dahl on June 24, 2003 11:51 AM


2.5 million total jobs and 3.o million private-sector jobs were lost between the beginning of the recession in March 2001 and April of this year. Both in terms of numbers and percentages, more private-sector jobs have been lost over this downturn than in any prior comparable period in post-WWII history.

Posted by: dahl on June 24, 2003 11:57 AM

I think I am detecting a bit of not seeing the forest for the trees - a bit of a disconnect. We're discussing here whether the economy is "improving" enough to say we are not in a recession.

Meanwhile the Fed is said to be about to cut interest rates to near zero, closing in on their "last bullet" in an attempt to keep the economy from tumbling over the steepest of edges. Japan is in deflation, and Germany appears to be entering the same. Consumers are still spending, but have taken on record levels of debt - many dipping into their home equity to maintain their lifestyle. Everyone acknowledges that the housing market is in a bubble, and it's a question of when it bursts, and how bad things will be as consumers lose their last bits of equity. The federal budget deficit has gone totally out of control, seriously threatening any possibility of paying Social Security to a working population that has already largely lost their pensions. Those higher income people who manage to have 401Ks and other retirement savings (most people don't) have lost much of those savings in the crash. Wealth is becoming ever more concentrated at the top, while for the first time even white collar and service jobs are being exported, dut to the capabilities of high-bandwidth internet.

And we're talking about whether we're in a recession or not?

Posted by: Dave Johnson on June 24, 2003 12:03 PM

If the computer programmer was getting $60,000 dollars and now can only get $35,000, there is certainly less of something about. If that won't make them work then the market is failing to use some capacity however you look at it.
Even if it did, something would have been lost in any case.
Essentially I think you are giving a moral rather than economic definition of unemployment.

Posted by: Jack on June 24, 2003 12:06 PM

No, we are not in a recession. GDP is growing. Yes, we have a serious employment problem. We need to create 80,000 jobs a month just to keep standing still in terms of labor force growth. We need far more job creation to employee the 3 million workers who have lost jobs since the recession began.

Posted by: dahl on June 24, 2003 12:07 PM

rvman writes: "There are a fair number of people out there in the market place "looking for jobs" who are seeking, say, a computer programming job paying $60,000/yr. If the market rate for such a job is $35,000/yr, I'm not willing to qualify those people as "looking"."

Chances are, if a $60k programmer applied for a $35k job, they'd probably decide she was overqualified and wouldn't consider her.

Also, if you're a programmer, and haven't worked in 6 months or so, nobody wants anything to do with you. This isn't really limited by pay either. Even a $10/hour part-time job at a tiny company is likely to be out of the question.

It doesn't matter if you look for jobs or not, you're past the sell-by date and recruiters treat you like stinky week old fish sticks.

Posted by: Jon H on June 24, 2003 04:39 PM

Sure, your 60K programmer can temp. Let's talk about temping, shall we? They'll be "employed," but with no or substandard health insurance (if the latter, they may be automatically placed on COBRA if they take time off), no paid vacation, holidays or sick leave and no 401K. Under the circs, your programmer would be far better off collecting unemployment and continuing to job-hunt.

Posted by: hesprynne on June 26, 2003 10:25 AM
Post a comment