Non-polictical (I hope): Unemployment--Interesting opinion piece by Krugman

Interesting article by Krugman (NY Times) on unemployment #s. Krugman is usually pretty left, and has been a bit of a shill over the past five years, but occasionally he returns to his strict economist roots. This piece brought up something that has been bugging me since before the election as the unemployment #s never seem to jibe with common sense; it is generally accepted that between 150-250k jobs need to be created every month to keep force with population growth (# of people coming into the labor force); how can unemployment as a % keep dropping when not that many jobs have been created? I don’t want this to be about Bush, but if you take a look at the #s something doesn’t make sense–assume that during 2000-2004, that it was breakeven–that is that during the first 3 years 2 million or whatever lost jobs & then during the last year 2 million found work. These would be the official #s (if memory serves it wasn’t until November or December that the #s crossed the plane where he could be the first POTUS since Hoover to avoid negative growth during term). However, when you take into consideration the 150-250k that are supposedly coming into the workforce, you are looking at a monster #/% of people w/o work. There is also the separate household study that always shows the actual unemployment # as quite a bit smaller than the employer study. Something just doesn’t jibe here. It can’t be the # retiring strictly due to age–population growth growth takes this into consideration. Any (non-conspiracy/non-political) thoughts on this?

    • The Dropout Puzzle

Many seemingly authoritative figures, not all of them partisan shills, say that the American economy has fully recovered from the recession that began in 2001. They point to the unemployment rate, which has fallen from a peak of 6.3 percent in 2003 to 5 percent last month. That’s not quite as low as the 4.2 percent unemployment rate in February 2001, when the recession began, but it’s fairly low by historical standards.

For some reason, however, the public isn’t feeling prosperous. Gallup tells us that only 3 percent of Americans describe the economy as “excellent,” and only 33 percent describe it as “good.”

Maybe people are just ungrateful. Maybe they’ve been misled by negative media reports. Maybe they’re grumpy about their paychecks: adjusted for inflation, average weekly earnings have been flat for the past five years.

Or maybe the figures on unemployment are giving a false signal.

Economists who argue that there’s something wrong with the unemployment numbers are buzzing about a new study by Katharine Bradbury, an economist at the Federal Reserve Bank of Boston, which suggests that millions of Americans who should be in the labor force aren’t. “The addition of these hypothetical participants,” she writes, “would raise the unemployment rate by one to three-plus percentage points.”

Some background: the unemployment rate is only one of several numbers economists use to assess the jobs picture. When the economy is generating an abundance of jobs, economists expect to see strong growth in the payrolls reported by employers and in the number of people who say they have jobs, together with a rise in the length of the average workweek. They also expect to see wage gains well in excess of inflation, as employers compete to attract workers.

In fact, we see none of these things. As Berkeley’s J. Bradford DeLong writes on his influential economics blog, “We have four of five indicators telling us that the state of the job market is not that good and only one - the unemployment rate - reading green.”

In particular, even the most favorable measures show that employment growth has lagged well behind population growth over the past four years. Yet the measured unemployment rate isn’t much higher than it was in early 2001. How is that possible?

The answer, according to the survey used to estimate the unemployment rate, is a decline in labor force participation. Nonworking Americans aren’t considered unemployed unless they are actively looking for work, and hence counted as part of the labor force. And a large number of people have, for some reason, dropped out of the official labor force.

Those with a downbeat view of the jobs picture argue that the low reported unemployment rate is a statistical illusion, that there are millions of Americans who would be looking for jobs if more jobs were available. Those with an upbeat view argue that labor force participation has fallen for reasons that have nothing to do with job availability - for example, young adults, recognizing the importance of education, may have chosen to stay in school longer.

That’s where Dr. Bradbury’s study comes in. She shows that the upbeat view doesn’t hold up in the face of a careful examination of the numbers. In fact, because older Americans, especially older women, are more likely to work than in the past, labor force participation should have risen, not fallen, over the past four years. As a result, she suggests that there may be “considerable slack in the U.S. labor market”: there are at least 1.6 million and possibly as many as 5.1 million people who aren’t counted as unemployed but would take jobs if they were available.

There’s both good news and bad news in that assessment. The good news is that the economy probably has plenty of room to expand before inflation becomes a problem (which implies that the Fed’s decision to start raising interest rates was premature).

The bad news is that it’s hard to see where further expansion will come from. We’ve already had four years of extremely loose fiscal and monetary policy. Tax cuts have pushed the federal budget deep into the red. Low interest rates have helped generate a housing bubble that has lifted real estate prices to ludicrous heights in major parts of the country.

If all that wasn’t enough to give us a full economic recovery, what will?

I have believed for the past several years that the reported payroll numbers do not reflect the changes in our economy. There are so many people like myself that don’t show up in employment reports. We are business people, but we don’t pay ourselves a salary.

I don’t know how you run your wine business, but if you don’t report wage income, you would not be included in payroll numbers even if you do work 100 hours a week.

I doubt many of the people on my construction site show up in payroll reports. They are working, believe me.

Having said this, I understand Greenspan has rejected this line of reasoning for the most part. I seldom disagree with him, and I only do it at my own peril, but on this matter I think he is just wrong.

What else is Krugman going to write about? If he thought the economy was great, unemployment was way down, we had the best economy by far of any industrialized nation, and tax receipts were going through the roof, you don’t actually think he would write about that, do you?

I don’t doubt that there are quite a few who like myself, started our own companies right around 2000–the dotcoms did in a way pave the way, showing many of us that with a bit of hard work, ingenuity and luck the brass ring is obtainable.

I also acknowledge that California (and from what you posted Florida) have a huge cash economy when it comes to day laborers–however for my part, these are part-time workers (once or twice a week) & are mostly undocumented individuals anyway that don’t show up on any government counts, except estimates of #s of illegal aliens :-). Everyone of my employees + myself shows up on federal & state reports–I pay the required taxes so I’m sure we are getting counted somehow.

I have more than a few friends who did very well during the late 90s and just haven’t gone back to work, but again, them + us just doesn’t measure up to the #s that we are talking about.

How difficult can it be to actually track these #s?

I take Krugman with more than a grain of salt, but this issue (for me anyway) has been knocking around in my head for a while now–his article just freshened it a bit.

Anecdotally related maybe, but one of the Sunday night “news magazine” shows last night (can’t remember which) reported that the percentage of women opting to stay at home and raise kids is up dramatically in the past few years (after dropping for much of the 60’s through 90’s).

So, these women are opting out of the work force and leaving jobs available for others–which jibes with your article’s thesis.

I have always felt that the unemployment statistics sampling process leaves much to be desired. I think it misses the mark in several ways, but the errors probably do cancel out to some degree.

In my case, I am really not talking about the day laborers. They actually get cycled through a temporary company for insurance purposes. They would be counted, for the most part. Probably not for their full hours though, since their pay is augmented with cash, mostly to cut down on insurance expenses, but also to help them avoid taxes.

I am really talking about the guy who runs the landscapping company, the plumbing company, the electrical contractor. I don’t know how they manage their books, but I doubt they report much in the way of wages to themselves and step into the resulting employment taxes.

I am able to structure my gains as long term capital gains for the most part. Why would I step into the taxes resulting from paying myself a salary?

I saw that piece too. But I don’t think they were reporting a dramatic rise in women leaving the workforce who were below upper middle class (I definitely could be wrong). The most surprising numbers were the amount of the super rich (based on the study of that quasi-feminist type lady) who were staying at home, the quote “future Senators and CEOs”.

However, based on what I was hearing last night, we should assume that most of these women will re-enter the workforce.

The sampling style will change when it is convenient for the next President to do so. That person will change it so they can show employment improvements under their term/s.

I could be wrong because I haven’t taken an Econ class since getting my MBA in 1984 (I’m old!), but, it is my understanding the sampling process hasn’t changed significantly since the '30s.

With the internet and the huge cash economy (I can virtually get any kind of service I need by paying someone cash that I am sure never gets recorded as income or employment) I would think some better means of estimation should be available.

Was it just me, or did you find that feminist “philosopher” more than a little condescending? I mean, if women want to stay home and raise kids, who is she to judge? And she seemed to think raising kids wasn’t a very difficult or challenging job. Probably doesn’t have any!

Oh man did my girlfriend just LOVE that lady (my g/f wants to be a stay at home Mom). At first she was ok with the lady because she had some interesting insights and her research was so unique, but then the lady definitely spilled into the “Women who don’t work are ruining it for women everywhere” stuff. I was surprised her research did lead her to be a little more open-minded and to take a “I can see how this would work for different people in different situations”, though CBS did a good job of stating before the nutso went off that she is biased, and clearly she was.

But she actually has a daughter that she raised on her own while continuing her career, which is at least a major reason why she holds the opinion/s she does. To each their own, at least her research is interesting.

True, sampling techniques really haven’t changed much over time, but the way in which the employment stats are defined have been changed. Almost every administration gets the methods for calculating unemployment changed, and it goes to whatever suits them best. Bush changed them when he came into office and it caused a momentary stink, Clinton did the same thing, it’s nothing new.

I do remember one thing which changed recently and that was counting people in the military. Seems like a no brainer to me, I mean they are part of the work force as much as any other government “employee” and they are employed. Statistically, it only made like 1/10 of 1 percent difference, though…

For some reason, however, the public isn’t feeling prosperous.

Possibly because there’s more to “feeling prosperous” than the national employment rate. You also need to consider things like pay raises, promotions, opportunity to find a better job, etc etc, and my own feeling is that these issues are largely contributing to the perception that the economy is stagnant.

Krugman’s right, but he fails to mention that the unemployment number has always left those people out. I agree that 5.0% is not the exact number of people who would like a job and don’t have one, but for comparison purposes it works OK. Comparing today’s 5.0 to 4.2 and 6.3 at different times in the past is valid, IMO.

FWIW, I also agree that he’s a shill who has thrown quite a bit of economic knowledge out the window.

In large measure the public feels how the media tells them to feel. In the run up to the last election, we had endless articles about how rotten the economy was and how awful the job situation was. Of course lots of people will be persuaded by such a non stop tidal wave of bad news.

The mainstream media kills me these days. They run negative stories over and over again about whatever, then they run a poll about whatever and report as “news” that more people than ever think whatever.

The economy is OK, but weird: It’s not abysmal, its not great. But its never been great for most people. The late 90s were a brief fantasy that evaporated for 90 % of us … There’s so much that goes into this. Many people have quit looking for work, but many have a cushion: family, maybe, or assets or credit (not good) … And there is the phenomenon of the day laborers, etc … Five percent unemployment is pretty good, IMO, but 30-40 % of people without health care, well, that’s what will just get us in more and more trouble in the near future … Bigger crisis than debt, home prices, SS, anything.

The whole feeling prosperous bit is interesting: The majority of Americans are quite spoiled and we want more and more crap: New vehicles, bigger homes, more gadgets than we could ever use. Remember a 1,000 square foot home and a car were all most people aspired to 40-50 years ago. Now that won’t do for the middle class. I am also amazed at my unemployed friends who would rather live on credit than work a crap job while looking for a good job … But, also homes, health care and college tuition cost next-to-nothing 50 years ago.

Actually, while that number in general has always left certain people out, you can make a pretty persuasive argument that in certain economic times, such as those with extended recessions where large numbers fall off the charts, that unemployment statistics are much more likely to be understated than at say, peak times, or somewhat more gentle recessions. So I don’t think making comparisons across times is without some peril. That is exactly his point - the unemployment number without some other economic statistics to provide context is not by itself all that useful. And that’s why traders always watch a bunch of different numbers when they get issued.

The metric JBDelong uses is Total Males Employed / Total Population (25-64).

I think he does this to address some of the bias mentioned above.

Permalink here:

http://delong.typepad.com/sdj/2005/07/four_out_of_fiv.html

This metric would overstate the real unemployment problem. Lots of people retire before age 65–policemen, firemen, military, civil service, Art Franke, etc. Counting them in the available workforce, when they really may have no intention of working, would be misleading.