January 06, 2004

The Slime Machine at Work Again

Daniel Drezner screams and leaps, fangs bared, for Paul Krugman's jugular. However, he trips over a tree root and falls off a cliff:

Daniel Drezner: CORRECTING KRUGMAN.... Krugman's assertion here is that the number of discouraged workers ("those who have given up looking for work") plus the number of part-time workers who wish they were full-time ("only marginally employed") are unusually high by historical standards.... [But] the percentage of discouraged workers... was much higher a decade ago.... [T]he percentage of Americans who are part-time workers but would prefer full-time... was higher a decade ago.... Krugman is either wrong or has a different definition of "unusual" than the rest of the English-speaking world. Distortions like this one...

There are, of course, two big problems with Drezner's "argument." When Krugman writes "an unusually large number of people have given up looking for work" he is tracking the flow of people who used to be employed into out-of-the-labor force status, and is referring to a much larger category of people who have dropped out of the labor force over the past three years than just the Bureau of Labor Statistics's "Discouraged Workers" category. When Krugman writes "many of those who say they have jobs seem to be only marginally employed" he is referring to a large group that has nothing at all to do with those who are working part-time for economic reasons. He is referring to those who tell the BLS household survey interviewers that they are working but for whom there is no corresponding employer telling the BLS payroll survey that they have somebody working for them

Does Krugman say that those who have "given up looking for work" are in the BLS "discouraged worker" category? No. Does Krugman say the words "discouraged workers" at all? No. Does Krugman say that those "marginally attached" are in the BLS "part-time for economic reasons" category? No. Does Krugman say the words "part-time for economic reasons" at all? No.

It is true that anybody who has been watching the labor market over the past three years--and seen the remarkably large fall in employment coupled with the remarkably small rise in the unemployment rate--will know that what Paul Krugman wrote was completely correct: this recession looks small as measured by the rise in unemployment, but it looks large as measured by the fall in employment as a share of the population or the duration of unemployment. Anybody who has been watching will know that Daniel Drezner's fangs-bared attack is fake and loony.

But the numbers of those who watch the flow of data out of the BLS are small. And the numbers of those who will read Drezner, and conclude that Krugman has written something wrong or questionable, are large.

Misrepresent somebody as saying something they did not say. Attack them for it. And then accuse them of "distortions." Way to go, Dan: you're now at the loony hack level. You ought to at least try to be better than that.

What Krugman did write:

Paul Krugman: An aside: how weak is the labor market? The measured unemployment rate of 5.9 percent isn't that high by historical standards, but there's something funny about that number. An unusually large number of people have given up looking for work, so they are no longer counted as unemployed, and many of those who say they have jobs seem to be only marginally employed. Such measures as the length of time it takes laid-off workers to get new jobs continue to indicate the worst job market in 20 years...

Me, from last December:

Look at what has happened to the U.S. employment-to-population ratio--estimated from the BLS household survey--over the past half century:

The employment-to-population ratio falls in each recessionary period.* Back in the old days, the rule of thumb was that the rise in the unemployment rate (in percentage points) was about five-thirds as large as the fall in the employment-to-population ratio (in percentage points). Thus the 1973-1945 recession saw the unemployment rate rise by 4.4% while the employment-to-population ratio fell by 2.4%. The 1979-1983 recessionary period saw the unemployment rate rise by 5.2% while the employment-to-population ratio fell by 3.1%.

But in the most recent 2000-2003 recessionary period, the employment-to-population ratio has fallen by 2.7% while the unemployment rate has only risen by 2.1%. The old pattern would have led us to expect such a fall in the employment-to-population ratio to have been accompanied by a rise in the unemployment rate of not 2.1% but 4.5%. More than half of the additional people who would have reported themselves as unemployed in a previous big recessionary period... aren't. They're reporting themselves as out of the labor force instead.

Why? What's happened to change the relationship between changes in employment and changes in the labor force? And what does it mean? (It's not the self-employed: this is from the household survey.)

It might be the sheer length of the downturn: a longer downturn may induce more people to give up looking, and produce more discouraged workers out of the labor force. But the 1979-1983 period was also prolonged, and although I remember Larry Summers and Olivier Blanchard worrying about how prolonged employment declines might discourage workers and produce a version of the European structural employment disease, it didn't.

It is a mystery to me.

*The employment-to-population ratio also rises over time as discrimination against women is severely reduced, and women find jobs outside the home in amazing numbers in the 1970s, the 1980s, and the 1990s.

UPDATE: Daniel Drezner states that when he wrote that "[Paul] Krugman is either wrong or has a different definition of "unusual" than the rest of the English-speaking world. Distortions like this one could explain parodies like this one..." he did not intend to:

  1. Provide an unfavorable description of Krugman
  2. Impugn Krugman's motives
  3. Convey the impression that Krugman lied
  4. Or ascribe any form of malign intent to Krugman

But intended only to state that he believed Krugman to be wrong.

Posted by DeLong at January 6, 2004 07:01 AM | TrackBack


Thank you, I just wanted to give a greeting and tell you I like your website very much.

Posted by: mark on January 6, 2004 08:27 AM


It was probably part of the guest-blogger agreement that Drezner must make at least one silly and inane attack on Krugman while Sullivan was out of the office. It wouldn't be "The Daily Dish" without some good Krugman bashing. Gotta keep the readers happy...

Posted by: bryan on January 6, 2004 08:42 AM


You really do miss the point of Daniel Drezner's piece when you state that he should at least "try to be better".

His piece has nothing to do with credibility, accountabllity, accuracy, honesty, or any other positive attribute.

It is written solely in attempt to discredit Krugman by any means possible - and lying, misrepresentation, and distortion are clearly on the table.

Drezner doesn't need to "try to be better". He achieved his goal of deception quite well!

Posted by: Mark-NC on January 6, 2004 08:46 AM


Are there any other words of wisdom that can be imparted to quantitatively minded laymen like myself (I'm a physicist.) who saw Drezner's link to the BLS's "Table A-12. Alternative measures of labor underutilization", plotted out all the time series over the last 20 years and saw that, as claimed, each measure looked worse in both the 1982-3 and 1992-3 recessions (provided the data went back that far, or else for the more recent series, looked worse in Jan 94) than it did at the peak of this current recession?

In short, what unemployment statistics are most interesting and in what circumstances?

(In Drezner's defense, this large and seemingly pertinent mound of data may have convineced him that Luskin was actually onto something and not just making another transparently specious argument. I concede, of course, that Drezner--and I--should have known better.)

Posted by: Bill on January 6, 2004 08:59 AM


Hacktackular! Get that man a job at Slate!

Posted by: Kaus Hackula on January 6, 2004 09:00 AM


>Misrepresent somebody as saying something they did not say. Attack them for it. And then accuse them of "distortions." Way to go, Dan: you're now at the loony hack level. You ought to at least try to be better than that.

Hey, he can now join Glenn Reynolds and Charles Krauthammer in GOP Hack Hell.

Posted by: Phoenix Woman on January 6, 2004 09:05 AM


Is it possible that the lower employment rate has not turned into a higher unemployment rate because there are a larger number of older workers now who have chosen to retire rather than seek to start over at a new job?

The population is aging, so if the laid off include a disproportionate number 57 year olds compared to 1980, they may decide to leave the job market altogether.

Posted by: Rv. Agnos on January 6, 2004 09:16 AM


When Luskin accused Krugman of just getting the numbers wrong on his assessment of the cost of job creation under the Bush program, Krugman first pointed out that his own figures were indeed correct. He then went on to note that, given his facility with math, casually accusing him of egregious math errors is stupid. That argument remains sound, as far as I can see from the occassional attacks on Krugman's claims about data. He gets the data right. He gets arguments about data right. But don't expect is math-and-integrity-challenged detractors to let that slow them down.

Posted by: K Harris on January 6, 2004 09:28 AM


You're supposed to say something nasty about Krugman ( guaranteed to make the Salem Witch hunts look like a weekend barbacue, and I mean you will be served.) or praise him. ( not getting nearly the attention.)
But to address the issue without even mentioning PK!! Could be precedent setting...I'll give it a go:
So the retiring or nearly-retiring boomers are thinking 'what the h..let's pack it in?' The ones that can afford it maybe, but it seems to me there are many more that can't. And rather than suffer the big drop in their standard of living, they are going to suffer some sort of underemployment which the government statistics are going to capture about as well as they do for inflation...productivity...

Posted by: calmo on January 6, 2004 09:58 AM


Once you go Hack, you'll never go back!

Posted by: Mikov Kauskov on January 6, 2004 10:07 AM


Ken Hirsch the premise of your second paragraph answeres the question raised in your first.

Even if we accept that certain categories of social and economic displacement were worse in 1993 than in 2003 what is this saying? On many levels, the country was in bad shape then.

Think about the state of Cities in 1993 for example. LA had just rioted. New York, Baltimore and D.C. were exporting the Crack wars to the Midwest. Philly was taken over by the state of Pennslyvania.

During the last decade a lot of the social problems that caused "non"-employment/under-employment and "informal" (i.e. illegal) employment were brought under control.

To say in 2003 that we are a bit better off. Particuarly after huge intervening expansion in the size of our economy is simply to accept stagnation with a smile.

Listen, we have to pay for the retirement of the Baby Boomers in TEN YEARS! Stagnant employment **AS A SHARE OF POPULATION** is waste of precious time. We can not afford to be marginally better of in terms of discouraged workers than in 1993 because we are ten years closer to doomsday.

We had the right to expect that after the decade that just passed things would be much better.

Posted by: Michael Carroll on January 6, 2004 10:10 AM


Just remember: Dan Drezner and Tacitus are among the few remaining supposed examples of the genus "rational conservative". If we lose these two, what do we have left?

The rational Conservative: extinct or mythical? Who knows?

Latham and MacArthur at Harpers strike me as rational conbservatives due to their pessimistic, anti-populist elitism, but in America they count as liberals. The conservatives of today are quite comfortable with know-nothing demagogic populism and millenarian superstition, and are mostly willing to follow the Republican Party over a cliff.

Posted by: Zizka on January 6, 2004 10:25 AM


Perhaps an increasing proportion of the population, and not only illegal immigrants, are working under-the-table for ready cash without deductions or benefits.

Posted by: Zizka on January 6, 2004 10:27 AM


Just remember: Dan Drezner and Tacitus are among the few remaining supposed examples of the genus "rational conservative". If we lose these two, what do we have left?

The rational Conservative: extinct or mythical? Who knows?

Latham and MacArthur at Harpers strike me as rational conbservatives due to their pessimistic, anti-populist elitism, but in America they count as liberals. The conservatives of today are quite comfortable with know-nothing demagogic populism and millenarian superstition, and are mostly willing to follow the Republican Party over a cliff.

Posted by: Zizka on January 6, 2004 10:32 AM


Perhaps an increasing proportion of the population, and not only illegal immigrants, are working under-the-table for ready cash without deductions or benefits.

Posted by: Zizka on January 6, 2004 10:33 AM


Drezner is such a lying whore.

Not too long ago on Slate he wrote a piece in which he tried to rebut the Center for Responsive Politics data that showed that the companies that got the plum contract in Iraq were big contributors to the GOP campaign.

He again used a "straw man" argument, distorting the data into saying something that it never in fact said, or no one even suggested it said, and then he attacked the data for it.

Specifically, he tried to set up a thesis that the data argued for a direct proportional relationship between the size of GOP campaign contributions and the size of the Iraq contracts. He then said that since GE was the largest contributor to the GOP, but didn't get the largest rebuilding contracts (I think Halliburton did), that that shows that the data
from the Center was erroneous.

I e-mailed him and pointed out his straw man. I also said that his argument conveniently ignored the fact that GE gets a lot more from the Administration than just Iraq rebuilding contracts (e.g., military and weapons system contracts), so that it was unfair to look at just the Iraq contracts as evidence of whether a quid pro quo existed.

I also indicated that by setting up this straw man, he was conveniently ignoring the overall point of the data--that no matter what the precise size of the contribution or the construction contract, the entities that received the major contracts were ALL major GOP contibutors.

Needless to say, I didn't get a response from him.

He's another whore lying son-of-a-bitch that has no integrity.

He's getting paid off by someone, somewhere.

Posted by: Phillip G. on January 6, 2004 10:39 AM


Echoing Ken Hirsch'a comment, does Prof. DeLong have a basis for knowing what Paul Krugman precisely meant? Has he communicated with Dr. Krugman? If Prof. DeLong is just making a judgment, then he ought to say so. If he doesn't know for sure, then his charge against Drezner is as unfair as Drezner's charge against Krugman.

It's also conceivable that Paul Krugman may not have had precise definitions in mind. At the very least, he's guilty of making ambiguous statements which are incorrect under one plausible interpretation.

Prof. DeLong worried abou the large number of people who will read Drezner. However, the readership of any blog is dwarfed by the readership of the New York Times. Millions of people now "know" that an unusually large number of Americans have given up looking for work, even if they don't know what that phrase is intended to mean.

Posted by: David on January 6, 2004 11:20 AM


I wonder what ever became of Luskin's alleged letter to the NYTimes asking if Dr. Krugman was using a different definition than the one Luskin devised AFTER Luskin decided to attack Krugman's oped? And I also wonder if Luskin ever took the time to notice that even his narrow definition has shown an upward trend over the past couple of years?

Posted by: Harold McClure on January 6, 2004 11:30 AM


>>does Prof. DeLong have a basis for knowing what Paul Krugman precisely meant?<<


Posted by: Brad DeLong on January 6, 2004 11:33 AM


Just an amusing (to me anyway) FYI: Donald Luskin is criticizing this same Krugman column over at NRO, and links to a chart here on Brad's site to support his argument.

Who cares?

"Krugman lied!" "No, Luskin lied! And Bush lied too!" "Nuh huh!" "Yeah huh" "Nuh huh to infinity!" "Mommy! He pushed me!" "He pushed me first!" and so on.

Posted by: Nick on January 6, 2004 11:43 AM


Perhaps an increasing proportion of the population, and not only illegal immigrants, are working under-the-table for ready cash without deductions or benefits.

It's the possibility of this effect that's made me think that we need a larger chunk of the lower-income segment to be tax-free in all regards. Perhaps $25k or so should be the cutoff point- however I am not an economist, so I dangle my feet over the pit of doom with such a statement. But I have personally seen this effect among many of my friends who are nominally "unemployed" or "students" with 6 credit hours a semester.

For example, a guy I know who subsists by fixing computers around his area. He keeps all the money (his rates are deadly dirt cheap, but who knows what you get...) and is officially living off his wife. Anecdotal, I know- but I see a great deal of it.

Posted by: a lesser mongbat on January 6, 2004 11:49 AM


>>does Prof. DeLong have a basis for knowing what Paul Krugman precisely meant?>Yes

Which needn't be shared.

But Brad is making the second post in a week's time that the ratio of employed to population is dropping, for no discernable reason. (Discernable using conventional stats, anyhow.)

But I don't see how to infer, from that, that
"the flow of people who used to be employed into out-of-the-labor force status ... people who have dropped out of the labor force over the past three years " represents a PROBLEM.

Let's consider a non-problem hypothesis.

A lot of workforce reduction is not layoffs or firings, but early retirement. In fact, some companies pay incentives to get 50-sometimes to retire ahead of schedule. Early boomers, born 1948-1953, represent a significant part of the total population. If a significant part of this significant part all twitches one way at one time, the national stats will reflect it. Won't they?

But how do these early-retirees see themselves? More important to this discussion, how do they REPORT themselves? Out-of-work? Looking-for-work? I don't see, from a psychological viewpoint, why they would do so -- and certainly I can't see them hustling resumes around to agencies and headhunters looking for a new job. On the other hand, one sees a lot of latemiddle aged early boomer people selling hand-tied fishing lures and wooden toys at various flea markets. Some may be cleaning out the attic via e-bay. Some may be doing part time work as "Greeters" at WalMart or the information desk at the mall. They do things they enjoy, and eke out more than enough income to cover their expenses. Will they report themselves unemployed?

Surely their former employers are proud to report these not-elderly workers off the payroll. No question. But I'm not sure the early-boomers are equally proud about declaring themselves "retired", (as in old, out-of-the game, and especially "not with it".) A pride discrepancy may account for a lot ...

This is a conjecture -- certainly not a theory, not even a hypothesis yet. But it is consistant with the other problems "Boomers" have inflicted on national trends. It seems to be testable, falsifiable, and worth some thought.

How big a sample of folks 50-60 would it take to start measuring a valid trend in move from traditional payroll to a "self-employed"/ "semi-retired" status?

Posted by: Pouncer on January 6, 2004 11:58 AM


Phillip G. -

Er, what sort of rhetoric did you employ in your emails to Drezner? It's possible that he understood the gravamen of your complaint but chose not to reply for reasons that have nothing to do with the cogency & merits of your arguments....

(expectantly dons flamesuit)

Posted by: Bruce Cleaver on January 6, 2004 11:59 AM




People who are 70 years old in 2015 by almost anyones Crystal Ball will not have to work. Even if they do want to work they will not find many well paid jobs -- but they still will get to vote. And, they will vote to keep their social security and medicare benefits for 10 or 20 or even 30 years until they meet their end. AND they will vote for people who have jobs to pay for this.

If the working age population grows in the next ten years it will be due to new immigration and to people forgetting to die.

Now people forgetting to die will not generate much cash to pay for 20 years expanding entitlements and illegal immigrants will not pay social security taxes. Legal immigrants will not make up the difference either.

Take it like this.

1) the population is growing moderately.
2) the **PROPORTION** of the population reaching retirement is growing and will continue to for 10 more years. = More entitlement demands
3) the **PROPRTION** of the working age population legally employed has been shrinking in the short term = fewer resources to pay for entitlements.
4) the deficit is growing. = less ability to defer the problem.
5) short term free trade growth is studdering = less ability to grow out of the problem.
6) the dollar is weakening = less ability to get invested out of the problem from abroad.

Yes a lot of this is short term but any one who is forecasting has to be thinking about the perfect storm that may be brewing out there. In my opinion hubris got us into this situation. but hubris will certainly not get us out 2004 may be our last good chance to get this situation under control and I am not confident that Conservative's are contributing anything here.

Is any of this sinking in Pouncer?

Posted by: Michael Carroll on January 6, 2004 12:23 PM


I am one of those early retirees. I hoped to work about half time doing about the same work at about 60% of my previous hourly remuneration (benefits considered). I'm working less than half time at about 35%. Unemployment here is the worst in the country -- ~7% I think -- and people who have jobs are keeping them. As I should have.

Posted by: Zizka on January 6, 2004 12:40 PM


>>does Prof. DeLong have a basis for knowing what Paul Krugman precisely meant?>Yes.

Which you are under no obligation to share, of course.

But, of course, we have no particular grounds to find your concern over this "problem" statistic compelling.

Because I am not an economist, or a newpaper columnist, I can airily wave away the problem with a couple of unsuported assertions. The ratio of employed to population is falling because :

1) Downsizing is creating an unprecedented pool of "early retirees" -- early boomers of age 50-something.

2) Women have wised up to the fact that given high FICA for not much additional Social Security benefit (over their entitlement to their husbands') their "second" income is a much less lucrative deal than first presented.

3) More people are working from home doing e-bay.

4) Drug dealers and Andrew Sullivan style bloggers report themselves gainfully employed, despite the opinions of others.

5-99) Whatever strikes my fancy ...

I can make such unsubstantiated claims because I'm not trying to persuade moderate voters that Bush is a great guy responsible for a booming economy. I'm just a loudmouthed (loud-typing-fingered?) commenter on a webpage.

Krugman, on the other hand, IS trying to persuade us there is a huge problem. It seems to me that using MY "airy hand-waving" technique might be a less-than-professional and less-than-effective approach.

Coming back to hypothesis (1) -- why shouldn't we try to falsify this? The boomers traditionally swing demographics... if a significant fraction of boomers all twitch the same way, the whole economy shifts. EmployERS who downsize workers aged 50-60 off the payroll are proud to report their leaving. But former employEEs in this age range, it would seem to me, might not yet be willing to report themselves "over the hill, out of the swing, not with it", or the like, anymore. They might be doing the greeter at WalMart thing, or the selling-crafts at the flea market thing, or any number of things that modestly boost morale and chump-change disposable income -- and report themselves "fully employed".

Another falsifiable hypothesis for an economics researcher -- WOMEN joined the workforce in the 70's to make the house payment. (Interest rates were high, women's rights were newsworthy, and the stagflation problem was real.) So, 30 years later, the mortgage is paid off, and that worker can justifiably come home to stay. SO we might poll 50-60 year old WOMEN specifically to see if they, preferentially, are leaving the current job market. (Women have a strong and justifiable incentive to report themselves "working" when they consider the mountains of laundry still to do...)

Given reasonable NON-problematic hypothesis about the drop in the workforce, economists telling me the sky is falling have to work harder explaining why such ideas are amatuerish.

Posted by: Pouncer on January 6, 2004 12:49 PM


The question that has been asked repeatedly: Why is the firm survey more reliable than the household survey? A householder knows if they are working or not - it is a simple distinction. A firm has full-time and part-time employees, contractors, consultants, and business partners, only the first of which count. The firm survey is known to undercount smaller businesses (which form and fail "under the radar" to a great extent) and new businesses.

Besides, let's say you are right, and that more people have exited and fewer are looking than normal. If you take the 2.7% drop in the proportion of the population in the market, and add the 1.3% of the of the Population which is now out of work (LFPR*delta(unemployment)) you get about 4.0% reduction in workers. Your other two recessions, submitted to the same analysis: 79-83: 5.2%*.61+3.1% = 6.0%
75: 4.4%*.58+2.4% = 5.0%
These were 25-50% worse than the current one.
89-92: 2.8*.62+2.0 = 3.7%

So what we see is that, accounting for Brad's finding, this recession is just slightly worse than Bush I's, which is pretty good considering the absurdly low base from which unemployment moved this time. Which is what we would suspect from circumstances. We also see that the trend toward more exit and less unemployment was already underway in 89-92 - the ratio was 1.4 to one, as opposed to 1.83 in '75 and 1.67 in '82. (arguably, the data suggests that this downward trend has been in motion since at least the mid '70s.)

If the labor force participation rate growth rate is leveling off, shouldn't that mean that recessions will show up as larger in it than back when the LFPR was growing quickly? In other words, in '75, the LFPR should have been 1%+ higher than in '72, but by the 90's, LFPR growth was down to less than 1% per decade peak to peak. As we age, it may start dropping long term.

For the pessimists among us, the drop in labor force participation has masked an increase in labor force participation amongst the 65+ set in this recession. In prior recessions, older LFPR has gone down. Probably reduced retirement funds. (LFPR amongs the 65+ set has been trending down for decades, but turned upward in the early-mid '90s.

Posted by: rvman on January 6, 2004 12:57 PM


I _hate_ the way Brad's site processes comments.

I thought I'd lost one, now I see I've posted twice. Sort of. Oh well.

Ziska, my sympathies about your under-employment. But you seem to confirm my point about the unreliable nature of the current statistics. If Professor Krugman is going to assert your (and your peers) plight is the Shrub's fault, it seems to me he is going to be working to develop some brand new statistics, first.

Mr Carroll, you correctly identify a collection of problems. The problem of younger workers "carrying" the retiree population is valid. What proposals are on the table to address your issues? I know of two: (1) A social security "Lock Box" which hides the "Budget Surplus" under a mattress against next decades' obligations. (2) Privatization of some portion of worker's investment in their own retirements.
Both raised in 2000, neither advanced since.

The problem of immigrant workers is real -- yet bringing workers to the jobs here may be better than sending the jobs overseas, yes? Here, at least, we have some chance of closing down a sweatshop or two, and maybe inspecting a few machines for safety-regulation compliance.

The optimistic point to be extracted from this is that the early-boomers who are now off the corporate payroll have an "unusally high" combination of experience, energy, and leisure time. American hobbyists historically have tended to produce amazingly useful inventions, widgets, and advances -- in their garages, despite the advice of their betters, and sometimes in defiance of local law. I would not be completely surprised to see some of these folks develop entirely new industries in the next decade. (Virgina Postel and her tale of the nail salon industry comes to mind...) There are certainly enough boomer/geezers here in North Texas experimenting with emus, ostriches, llamas, catfish farms, and new products made from the humble "black eyed pea". These old guys are BUSY -- ain't none of 'em making a dime in real profit yet, far as I can tell. But not too many would claim to be unemployed, nor yet are any of them starving or otherwise dying-of-poverty.

It's an interesting time.

Posted by: Pouncer on January 6, 2004 01:12 PM


"The problem of younger workers "carrying" the retiree population is valid. What proposals are on the table to address your issues? I know of two: (1) A social security "Lock Box" which hides the "Budget Surplus" under a mattress against next decades' obligations. (2) Privatization of some portion of worker's investment in their own retirements.
Both raised in 2000, neither advanced since.

Here is another solution to the problem that Krugman and various democrats have proposed:


If we can keep down the red ink and avoid a meltdown in the dollar, technical solutions like the ones you mentioned stand a lot better chance.

"It's an interesting time. "

What is that Ancient Chinese warning about interesting times? Lets hope the times don't get anymore interesting.

Posted by: Michael Carroll on January 6, 2004 01:41 PM


To me the telling factors relating to Prof. K's
words are the inability of his detractors to begin
to achieve any traction with their "wrong in any
case mindset". luskin and his ilk serve the power
and they've held their asses up for service for so
long they've got the view upside down. Complex
factors and mutable results continue to befuddle the critics but still may be routine to the Prof.
As the facts are clear...our present resident at
Pa. Ave. has devalued discourse and ruined the
foreseeable future by sheer theft and cronyism
and some of his protectors must feel that they
will get a piece of the pie...grab it now for after Nov04 things will be very different and the days of law are near at hand...

Posted by: xaxx on January 6, 2004 01:48 PM


Oh, and sound monetary policy wouldn't hurt either.


Someone correct me if I'm wrong but are there any meaningful upward pressures on the Dollar besides the peg to the Yuan?

Posted by: Michael Carroll on January 6, 2004 02:17 PM


There is an interesting analysis of changes in labour force participation by demographics comparing the 1990-1992 and 2001-2003 periods on http://www.bls.gov/opub/ils/opbilshm.htm : "Labor Force Participation During Recent Labor Market Downturns".

Check out the table "Persons who did not work or look for work by age, sex, and reason for not working in 1991 and 2001". The percentage who say their reason for not participating is "could not find work" is very low, much lower than in the 1990-1992 recession. More men and women are dropping out through early retirement before 55, more are going to school, and more are ill or disabled (could be disguised unemployment I guess).

Now I'm just pointing out what I genuinely understand, in my own simple-minded way, to be a discrepancy between the published data and Mr. Krugman's opinion. I do hope that doesn't get me called a fangs-bared slime machine loony fake, because that would be most uncivil.

Posted by: dc on January 6, 2004 02:20 PM


Pouncer, many people here have a pretty good understanding of economics, and those who don't (such as myself) tend to trust DeLong over people such as yourself who seem to be reinventing economics from scratch. Your string of conjectures is mildly interesting, but that's all they are. It's up to you to convince us that things really are that way. All you're giving us is what-if, what-if.

As for my own case, I don't know where I fit into Krugman's statistics, but I do fit into the general picture of people who are working less and for less pay primarily because of high unemployment. I do NOT fit the theory you pulled out of your butt about early-retirees, even though I am one.

Posted by: Zizka on January 6, 2004 02:50 PM


As far as the Luskin Truth Squad, its sole mission seems to be feeble attempts at finding alleged lies by Paul Krugman. I have seen lots of lies in Luskin's writings but alas the lies were statements made by Luskin, which he tries to pass off as truth. Then again, maybe Luskin really believes the nonsense he writes given his total lack of knowledge about economics. But then why on earth would Rick Lowry allow such pathetic writing at NRO? Take a look at Lowry's 16 questions for Howard Dean as a clue. One put forth the amazing claim that education spending has risen by 65% over the past 3 years. Maybe Federal spending has risen that much but is the editor of NRO now saying state public spending and private spending don't count to the quality of education?

Posted by: Harold McClure on January 6, 2004 03:02 PM


I have no problem trusting Brad. When, that is, he offers an opinion and a quick and dirty layman's version of why he holds it.

But in this case, he's said, "you think I'd would know. I think I SHOULD know. But I don't know."

So, in the spirit of helpful consideration, I'm offering some conjecture.

Now, I have less reason to respect Professor Krugman. At least, I have no more reason to respect PK in his role as a newspaper columnist than I do, say, Thomas Sowell. As a professor offering a class in his discipline or as author of a textbook, maybe. But an airy wave of the rhetorical device in a well-crafted paragraph? Anybody can do that. Me too.

As for how well your personal situation applies to the population at large, I'm not sure either of us in a position to say. Mr Carroll, above, cites BLS stats that more people ARE in fact taking "early retirement." What that means for the overall economy is still an open question. Some people no doubt like having the "leisure" time and some may do surprising and useful things with that time. Others suffer. I regret that some, like yourself, seem to be suffering. But how big a problem it is helps determine what actions to take. So, it's too bad those helping shape the course of action don't share whatever facts and data and statistics they use to try and influence opinion.

Posted by: Pouncer on January 6, 2004 03:07 PM


With respect, dc, I fail to see the "discrepancy" that you see. The BLS table you cite states that, in 2001, only a very small percentage of non-participants said the reason they did not particpate in the work force was because they could not find work. Interesting and surprising, yes. But inconsistent with PK's opinion? Why? Why is this fact about the state of the labor force in 2001 inconsistent with PK's opinion about the state of the labor force in 2004?

Posted by: joe on January 6, 2004 03:22 PM


Pouncer wrote, "I _hate_ the way Brad's site processes comments."

I do, too.

I seem to have found a foolproof way of dealing with it:
(1) Hit post.
(2) After waiting for as little as a second, hit "preview." (DON'T hit post again.)

Your post will probably show up as posted, and if not, you have your (precious) text in the preview area.

Posted by: Stephen J Fromm on January 6, 2004 03:30 PM


1) Early retirement by the well-off
2) Women leaving the workforce - possibly cultural change
3) Many more people on disability pay
4) An expanding shadow economy

No. 4 is a taxation problem.

Did anyone ever consider the structural similarity between our modern, apparently booming shadow economies and the Depression Era-tendency to go back to/hold on to subsistence farming (which may, however, have been a European phenomenon only - I sure donīt know about the U.S. and its very different circumstances at the time)? Transition delay on a Kondratieff-cycle scale?

Posted by: Joerg Wenck on January 6, 2004 04:02 PM


Unemployment vrs working Stats.
I humbly suggest that there has been a cultural
shift in the 2 adult "family working unit" of the 80's baseline. ( a 50% divorce rate and a 70% second marriage divorce rate) has taken a toll of support out of the work force. Certain minorities still maintain family working units
The majority of this culture is disolving into a sometimes partially employed workforce.

Posted by: DERIK B on January 6, 2004 05:41 PM


Brad, you were telling us just yesterday that you don't know why the payroll/household employment ratio changes over time. If Krugman has been telling you that its recent reversion to what appear to be (eyeballing the graph from your earlier post) 1996 levels is such good evidence of weakness in the labor market that the failure of either the BLS discouraged-worker stat or the BLS involuntary-part-timer stat to confirm it presents no problem-- well, presumably he gave you some convincing reason for believing this. Is it the sort of reason the rest of us can be trusted to hear?

Posted by: Paul Zrimsek on January 6, 2004 06:44 PM


Pouncer, the reason why Brad believes he knows what Krugman is saying is probably mostly that Krugman was using ordinary economic terminology in an ordinary way. The reason he gave you such a short answer is probably that he judged you to be a game-playing lightweight.

I have no way of knowing what Brad actually thought. Perhaps I'm projecting my own opinion of you onto Brad.

Posted by: Zizka on January 6, 2004 07:00 PM


Phillip G,

Thanks for calling attention to the GE/administration weapons deals. I had suspected for some time that my local Organic Foods Co-op was getting screwed out of contracts. Now it can be told.

Keep up the good work, Doctor...

Posted by: Tommy G on January 6, 2004 07:03 PM


Welcome, mark. Not all discussions here get this, shall we say, heated up. Once in awhile, if you really get sick of the discussion, try a couple jokes or stories and a lot of people appreciate it, even if they are not the ones who write here.

OK. My comment. I believe this is something like the Christmas movie where if a bell rings, an angel gets its wings?

That's my theory anyway. People keep talking about Mr. Krugman getting a Nobel Prize and he hasn't gotten one yet, so every time someone trashes him, it means that he acquired one more Scandinavian joke or ate one more piece of lutefisk on his way to proper Scandinavian recognition.

Way to go Mr. Krugman! I hope Dan Drezner does you some more super big favors real soon! Don't count on any help from "The Big Guy" though. He'll just say "I have no idea what you are talking about" and will keep his jokes and lutefisk to himself.

Posted by: northernLights on January 6, 2004 07:31 PM


I've been scratching my head on this a bit and am wondering if one missing piece of data might help clarify things...what is the population age distribution now as opposed to, say ten years ago?

The boomer cohort is starting to hit retirement (albeit one possibly greased by lack of opportunities); I'd be fascinated by data that linked employment % to age...


Posted by: Armed Liberal on January 6, 2004 07:33 PM


Since Brad is unlikely to put it up himself:

here is the link to Dan's response:


After reading both, seems Brad is a bit unhinged. "Fangs bared" ? Read the original Dan post and his response to see a very civil dissection of the Krugman asserstions.

Posted by: lee kane on January 6, 2004 07:38 PM


The reality on the ground for me is that one third of the people [all middle-aged] I know are out of work here in Silicon Valley. I myself have been out of work for 2+ years despite having excellent I.T. qualifications. Others I know have been looking longer though, and without luck. Don't know why they don't show up in the statistics. But I'd guess it's because a bunch of people in power don't want the public to know how bad it is.

Posted by: russell s. on January 6, 2004 09:25 PM


Since this discussion concerns Krugman's language, I find it particularly useful to examine the language used by DeLong and Drezner in their analyses. What could be a mature and civil debate strains immediately under the heft of De Long's epithets ("slime machine" and "fangs bared"), and not to mention those of his supporters ("loony hack" and "lying whore"). If you disagree with Drezner's argument, let's hear about it, but the irony is acute when you're making charges about mischaracterizations with hyperbolic language of your own.

Posted by: Liam on January 7, 2004 07:02 AM


Help me--I'm confused. The 5.9% unemployment number doesn't come from the household survey, which is, we're told very often by Krugman et al, less reliable. The household survey suggests that the 5.9% number should be lower, not higher, but Krugman uses it to suggest that the firm survey--the source of the 5.9% number--understates, rather than overstates, the unemployment rate. How is that supposed to work?

Posted by: Thomas on January 7, 2004 08:18 AM


The 5.9% unemployment number *is* from the household survey...

Posted by: Brad DeLong on January 7, 2004 09:15 AM


Thomas: The problems of measuring unemployment is a difficult one, as many statistical measures that sound pretty straightforward superficially.

The underlying idea is to measure the degree of "labor force (under)utilization" as an important economic indicator. For this purpose (1) the notions of un/employment are defined, and (2) several methods are used to measure them, and (3) the raw data are subjected to adjustments to account for known blips in the data, like "calendar effects" -- reduced construction in rainy or cold months, increased salesforces during shopping seasons etc. Consider the BLS website:
and there specifically the "summary", "explanatory note" and table A-12. (It is also instructive to look at the breakdowns by age, race, and other characteristics.)

The officially reported rate is the seasonally-adjusted U-3 figure that excludes dicouraged, marginally-attached, and involuntary part-time workers (definitions are in the first two document, summary and explanatory note). I would rather go with the U-6 rate, not because it is higher (it is by definition as it includes more people), but it provides a better picture as it includes important categories of un/underemployed people.

The "reliability" problems of the survey are that (lay) people have to label themselves, and may misinterpret their situation, and also the survey coverage is low (60,000 households) and may be subject to bias. One thing that has been pointed out is that when extrapolating the data to the whole population, the household survey indicates 2 million more jobs than the establishment survey, which should be impossible.

On a related note, "self-employed" people between jobs are (apparently) not counted as unemployed, but I think most people would agree that they should. That you have your own (single person or family) company and can only find 1/2 as many clients as you would want to be fully engaged should be equivalent to somebody who wants to works fulltime, but can only find either (a) a 50% position, or (b) fulltime positions for a total of 6 months anually. U-3 includes (b) people, but not (a) people and the self-employed, which is inconsistent.

Posted by: cm on January 7, 2004 09:57 AM


Post a comment