August 11, 2004

Why Oh Why Can't We Have a Better Press Corps? (Yet Another Mickey Kaus Edition)

*Sigh*. Mit der Dummheit kaempfen Goetter selbst vergebens:

Mickey Kaus: This John Crudele column from Thursday's N.Y. Post is now looking mighty prescient. It also offers a relatively benign, statistical explanation for the weak July job growth numbers released today...

The Crudele "explanation":

New York Post Online Edition: seven: ...the Labor Department's so-called Net Birth/Death Adjustment tabulates July as one of only two months in which there are more companies dying and taking jobs away than creating new jobs. The other month is January, when Labor takes out a massive number of jobs because it assumes a large number of companies die off after Christmas. I assume the White House doesn't understand these assumptions because the small jobs gain in January — helped by these assumptions — set off a massive amount of administration bellyaching.

Now let's look at what the Bureau of Labor Statistics says about its numbers:

Employment Situation Summary : Table A. Major indicators of labor market activity, seasonally adjusted (Numbers in thousands)

As an exercise for the reader, explain that one would not expect a predictable and regular seasonal pattern in the adjustments the BLS makes to the raw payroll data (to account for not-yet-known changes in the number of companies) to have any effect on seasonally adjusted employment estimates. Seasonally adjusted estimates are, by definition, seasonally adjusted: predictable and regular movements in them due to the standard seasonal pattern of economic activity are removed.


Kaus then gets his second wind:

The other glaring flaw in the BLS "payroll" survey, from which the net job number is generated, is that (as the BLS puts it) "Data exclude proprietors, self-employed ... farm workers, and domestic workers." The exclusion of the self-employed seems particularly troubling, given the economy's apparent increased reliance on freelancers and consultants....

[...]

Alert reader T.G. emails with another plausible reason for the steady rise in jobs in the government's "household" survey (which includes the self-employed) but not the "establishment" survey (which doesn't).

[I]t isn't just that we have so many more free-lancers in our economy, it is also that during the internet bubble, big bucks offers for most any warm body were drawing down the "normal" pool of self employed [as they were hired by establishments]; now we are merely returning to the norm.

...P.S.: I think T.G.'s point is not that there is some huge boom now going on in the self-employment sector (which doesn't feel like what is happening). It's that the payroll job growth of the late '90s was not quite as good as it seemed: The payrolls weren't always adding new net jobs. They were also sucking people from the self-employment sector (in which they weren't counted in the payroll survey even if they were working) into the corporate sector (in which they were counted). For the past few years those people have been moving back from (counted) payrolls into (uncounted) self-employment. Even though they may still be working, this has had the effect of lowering the payroll number and making it look worse than it actually is.

As an exercise for the reader, explain that a 139 thousand increase in the number of self employed (from 9374 to 9513) since the last employment peak in March 2001 is an order of magnitude to small to erase the 1.23 million drop in payroll employment (which is even more striking given that the long-run trend is for payroll employment to grow at about 1.4 million per year).

For extra credit, readers can explain Alan Greenspan's view of the household and payroll surveys:

Greenspan Testimony Report: 'I wish I could say the household survey were the more accurate,' Alan Greenspan, the Fed chairman, said in congressional testimony on Feb. 11. 'Everything we've looked at suggests that it's the payroll data which are the series which you have to follow.'... The Fed's conclusion was that the household survey's results had been inflated by overestimates of population growth.... If the population estimate is too high, the estimated number of jobs will also be too high. The bureau bases its population estimate on the 2000 census, but it then updates that estimate yearly with data on births, deaths and immigration. But immigration numbers are largely guesswork, because so much immigration is illegal. Fed officials suspect the immigration estimate is inflated because it fails to reflect tighter immigration controls since Sept. 11, 2001, as well as declines caused by the economic slowdown...


Now none of this is rocket science: Greenspan's views are not hidden under a bushel basket; the meaning of "seasonal adjustment" is not obscure; the Bureau of Labor Statistics does not conceal the fact that its household-survey estimates of self-employment have risen by only 139 thousand since March 2001.

Posted by DeLong at August 11, 2004 12:44 PM | TrackBack | | Other weblogs commenting on this post
Comments

Crudele is either being dishonest, or he has failed to pick up the phone and call a BLS economist, who would gladly have pointed out to him, as one did to me many months ago, that the birth/death figures are not seasonally adjusted, while the final payroll figures are. Either way, he's failing in his journalistic duties. Insert New York Post joke here.

Posted by: CaptWillard on August 11, 2004 12:53 PM

____

Before anyone start using the birth-death model to argue that Voodoo economics is working they better look at how it impacted employment growth from Feb to May when private payrolls increased 973,000. Over these 4 months the birth/death model added 800,000 of these 973,000 jobs.

Now do the Bush defenders really want to make that adjustment to the great Bush job growth record?

Posted by: spencer on August 11, 2004 12:55 PM

____

Its still a mystery to me why so many apparently smart people both to read Kaus, let alone take his rants seriously.

Posted by: Danf on August 11, 2004 01:06 PM

____

Its still a mystery to me why so many apparently smart people bother to read Kaus, let alone take his rants seriously. Was he good in the day? Or does he get disproportionate respect for being an early adopter of the blog thing?

Posted by: Danf on August 11, 2004 01:08 PM

____

Kaus is kinda like a male Maureen Dowd: a lot of surface wit and flash, but nothing underneath.

And any defender of this Administration who has a problem with the net birth/death adjustment had better wake up and smell the coffee, because according to the BLS, 1.35M new jobs (net) have been created since the last birth/death adjustment benchmarking in March 2003. 1.2M of them are attributable to the birth/death adjustment; without the adjustment, the establishment survey would show a gain of only 0.15M jobs - 153,000, actually - since March of 2003.

If they want to run on *those* numbers, they can be my guest.

Posted by: RT on August 11, 2004 01:15 PM

____

Danf: Kaus is A) paid and B) highlighted regularly on a medium-sized media site, so he naturally rates more curiosity seekers and trainwreck-watchers than the typical blog.

Posted by: V from VJ on August 11, 2004 01:16 PM

____

A male Maureen Dowd is exactly what Kaus want's to be, but unfortunately his rightful spot on the NYT editorial page was stolen from him by Paul Krugman. Such an injustice.

Posted by: Kaus Hackula on August 11, 2004 01:30 PM

____

What follows is my rant on my little hobby horse. And a repeat post from one below. My apologies, but I want to know what people think of this.

********
Here is an item that is the first hard evidence, that I know of, that any kind that cash disbursements to middle and lower income classes are signficant factors in driving the above and below trend bumps in employment changes. Apparently these folks have a model for retail sales that includes things like child tax credit checks.

I know very little about this kind of forecasting, but it reinforces my idea that we will see good employment numbers mostly when the Bush admin is forced to, or the law says, you give some cash to regular folks. So I think with no more rebates coming, and the income tax refunds spent, we will see more numbers like June and July. Let's see what happens the next few months.

http://story.news.yahoo.com/news?tmpl=story&cid=568&ncid=568&e=14&u=/nm/20040810/bs_nm/economy_retail_redbook_dc_1

Chain Store Sales Growth Slows
Tue Aug 10, 9:00 AM ET
NEW YORK (Reuters) - The absence of child tax credit checks hurt the pace of U.S. chain store sales growth in the first week of August, a report said on Tuesday.

Posted by: jml on August 11, 2004 01:41 PM

____

pop quiz: who's a bigger hack, kaus or easterbrook?

Posted by: wag on August 11, 2004 01:45 PM

____

Kaus just doesn't understand economics. Period. That's all there is to it.

If he'd just admit it, it wouldn't be so bad. But instead, he fakes his way through, in order to reach predetermined conclusions, which is what makes him a worthless hack.

Posted by: JP on August 11, 2004 01:48 PM

____

http://www.nytimes.com/2004/08/11/business/worldbusiness/11CND-OIL.html?hp

Global Oil Demand Expected to Exceed Forecasts, Report Says
By JAD MOUAWAD

PARIS — Global oil demand is expected to be higher in 2004 and 2005 than initially forecast, increasing pressure on oil producers to boost their output at a time when rising oil prices may hurt a recovering world economy, according to a report by the Paris-based International Energy Agency.

Supply disruptions in Iraq, uncertainty over the fate of Russia's top producer Yukos, and rising demand in China have pushed oil prices to record highs recently. Still, the report published by the agency today tries to dampen concerns that oil is in short supply by pointing to rising production in Saudi Arabia and Russia.

'The market is tight, production and infrastructure capacity is less than desired and uncertainties continue to weight on the market,' the report said. 'But does this justify $45 oil? Current oil prices are a concern and are causing economic damage.'

Posted by: anne on August 11, 2004 01:50 PM

____

It's funny that these numbers have been coming out monthly for decades and we've always interpreted them in a straight-forward way. Now we discover that you actually need a secret decoder ring to understand what they REALLY mean.

Posted by: cailte on August 11, 2004 02:02 PM

____

It's funny that these numbers have been coming out monthly for decades and we've always interpreted them in a straight-forward way. Now we discover that you actually need a secret decoder ring to understand what they REALLY mean.

Posted by: cailte on August 11, 2004 02:07 PM

____

Is there a calculation of people who have dropped out of the labor pool because they have stopped looking for work, etc.?

If it exists, how does that look, especially given that job growth is basically flat, but unemployment went down slightly last month?

That would be an interesting number to publicize along with new employment and jobless numbers.

Posted by: Phil on August 11, 2004 03:08 PM

____

"Kaus just doesn't understand economics. Period. That's all there is to it."

Well, there's more to it. His screw ups on the meaning of polls are as bad as is humanly possible. He doesn't do numbers of any kind, as best I can make out -- he really has to be seated at a separate table from them.

Posted by: frankly0 on August 11, 2004 03:20 PM

____

The Labor Department does publish estimates of discouraged workers. In Jan. 2001, there were 301,000. At the end of the recession, there were 328,000. Now there are 504,000.
These are folk who tell the BLS that they would like a job but aren't looking right now because they don't think any job is available for them. More than 1 million more aren't looking for reasons other than discouragement... probably stuff like: I can't get a ride to work or I'm disabled or I can't find someone to watch my kids....

Posted by: rex on August 11, 2004 03:26 PM

____

What does everyone make of this arguument? It's from yesterday's NYT letters section. Seems bogus to me, but what do I know?
-----------------------------------------------
To the Editor:

The seeming contradiction between low job gain in the establishment job survey and declining unemployment rate in the household survey is connected to the increasing aging of the population.

Most voluntary retirements and many involuntary retirements among the upper age groups result in withdrawal from the labor force. Others in these groups, because of fundamental changes in the economy, continue work as independents (not included in the establishment survey). Most of the upper age groups don't go into unemployment.

A result is that the number of employed measured in the establishment survey goes up slowly (despite the increase in population), while the unemployment rate from the household survey goes down, and the number of independent workers, as well as the totals in the household survey, go up faster.

Peter Gutmann
New York, Aug. 7, 2004
The writer is a professor of economics at Baruch College, CUNY.

Posted by: mario on August 11, 2004 03:31 PM

____

What does everyone make of this arguument? It's from yesterday's NYT letters section. Seems bogus to me, but what do I know?
-----------------------------------------------
To the Editor:

The seeming contradiction between low job gain in the establishment job survey and declining unemployment rate in the household survey is connected to the increasing aging of the population.

Most voluntary retirements and many involuntary retirements among the upper age groups result in withdrawal from the labor force. Others in these groups, because of fundamental changes in the economy, continue work as independents (not included in the establishment survey). Most of the upper age groups don't go into unemployment.

A result is that the number of employed measured in the establishment survey goes up slowly (despite the increase in population), while the unemployment rate from the household survey goes down, and the number of independent workers, as well as the totals in the household survey, go up faster.

Peter Gutmann
New York, Aug. 7, 2004
The writer is a professor of economics at Baruch College, CUNY.

Posted by: mario on August 11, 2004 03:32 PM

____

John Crudele? JOHN FREAKING CRUDELE?

I haven't heard that name since he was spreading "Vince Foster was murdered" rumors back in the Clinton days.

Do a Google search of "John Crudele" and "Vince Foster" and you'll see what I mean.

Posted by: Swopa on August 11, 2004 04:26 PM

____

Brad,

Let us assume that I know nothing whatever about you--a proposition is close enough for jazz.

I still know this: that you do not have time to read Mickey Kaus.

Posted by: Frank Wilhoit on August 11, 2004 04:51 PM

____

One other thing to remember about Kaus -- he claims to be supporting Kerry. His pathetic spinning of the job numbers is proof, if proof was needed, that this is just a pure pose to free him to snipe at Kerry without the implication that he's hitching his wagon to Bush. The only remaining mystery is why there are people out there (many of whom are New Republic readers) who need to believe that Kaus is some kind of liberal.

Posted by: P O'Neill on August 11, 2004 05:33 PM

____

That is pathetic. Voters don't care about the number. Voters only care that cousin Joe and Aunt Jane can get good jobs. Voters know the jobs are not there. We don't need statistics to tell us so. Having this moron try to make the employment situation is better than we know it is insults our intelligence. People like that make all economists look bad.

Posted by: bakho on August 11, 2004 06:29 PM

____

Straws in the wind, from Alan Abelson's column in that well-known "left-wing, comsymp, liberal smear machine" known as Barron's ("Rallying the Troops," Aug. 2, 2004, pp. 7-8):

"Investors, when they aren't basking in their natural state of euphoria, tend to be an edgy lot. What makes them more than a little antsy these days is a nagging doubt as to whether the recovery is the genuine article. No secret, we suspect it isn't. And skepticism, whether unarticulated like that of investors or outright like our own, wasn't exactly allayed Friday by
the preliminary reading on second-quarter GDP, which slowed sharply to a 3% annual rate from 4.5% in this year's opening three months. It also
contrasted with expectations among the cogniscenti of 3.7% to 4%.....

"Not only is the economy in a braking mode, but the increase in the GDP deflator, one of Mr. Greenspan's many favorite measures of inflation (having many measures enables him to choose, at any given time, the most convenient one, i.e., the one registering the least increase) was a brisk 3.2%. A slowing economy and accelerating inflation used to be called stagflation. Which is the economic equivalent of running harder and harder in place and enjoying it less and less.

"Agreed that it isn't an exceptionally elegant word, and these days it rarely slips from the lips of economists, especially those of the Wall
Street genus, who prefer "robust" when they can't say "boom." In any case, stagflation and the condition it describes is bad for people and hence bad for stocks."

Abelson also cited an interesting thought from Stephanie Pomboy of the consulting firm MacroMavens:

"Stephanie views the [Microsoft dividend] as not so much the beginning of a trend but rousing confirmation of one. And what that trend consists of is that, as we observed about Microsoft, corporations can't think of what to do
with their increasing trove of cash other than to pay dividends, buy back their stock and contentedly watch it pile up.

"By her reckoning, undistributed corporate profits now weigh in at a cool $550 billion, handily more than double their total a scant few years ago. In the good old days, she notes, those overflowing coffers would be put to productive use via capital investment. However, the times they've been a-changin' and corporate attitudes along with them.

"For, like Microsoft, most companies appear far from eager to spend their ever-growing reservoirs of cash on capital projects. "This earnings
season," Stephanie remarks, "has been noteworthy, if only for its raft of share-buybacks and dividend announcements." She tallies 938 companies buying back stock during this stretch, compared with 808 in the last quarter, while 410 companies have declared dividends this earnings season, compared with 184 last.

"Stephanie takes this as persuasive evidence, along with the wave of privatizations (whether or not they are, as claimed, due to the costs of
complying with Sarbanes-Oxley), THAT CORPORATIONS ARE CHOOSING TO SHRINK RATHER THAN EXPAND (emphasis added). Which opens to considerable doubt expectations that what she calls the standard recovery package of capital expenditures and employment gains have merely been delayed in delivery.

"For that matter, she sighs, the very fact that companies have been "amassing cash even as the Fed has been taxing them for doing so" via negative real short rates was a tip-off about the lack of compelling investment opportunities. "Underwhelmed and uninspired," she relates, "companies are now opting to give the money back."

"That isn't a favorable harbinger for capital spending, nor for jobs, since "no single force shapes the employment outlook like capital spending."

So in other words, the very people one would normally consider Dubya's natural constituency - CEOs, CFOs, and the like - have so little faith in this bogus economic program that they're not even supporting it with new investment spending. Kinda makes you wonder what's going through Greg Mankiw's mind, doesn't it?

Posted by: Uncle Jeffy on August 11, 2004 07:30 PM

____

The BLS exclusion of sole proprietors, self-employed SOHO's, seasonal farm workers, and domestic workers is actually crafty politics.

1 in 10 SOHO's makes it. I would argue 1 in 20.
That leaves 19 out of 20 bankrupt, representing a huge net deficit to the US economy. (Bankrupt spent life savings to stay open for business until they are spending someone else's life savings who foolishly gave credit to them.)

Read your local newpaper's bankrupcy notices. Huge. Ditto farm and domestic workers, grossly underpaid far below a living wage. All of the above are a net load on the dole system, and so a net structural deficit in the US economy.

Increased numbers of sole proprietors, self-employed SOHO's, seasonal farm workers, and domestic workers are *not* good employment stats, just as increasing numbers of OK'ys moving to California fruit-picking jobs during the Dust Bowl was not "good", and increasing numbers of "self-employed" WWI vet's lining the streets with their pencil and apple stands was not "good", or 50% inner city unemployment stats.

US is a permanent two-class Depression economy, just like post-industrial England. We just don't have an Australia to off-ledger poor off to, so BLS just ignores them like everyone else does.
The City of Seattle is waging wars in rural areas trying to set up tent cities for the poor. Television has so warped and rotted our values that anyone wearing sub-par Old Navy is Saddam.

To paraphrase Will Smith in 'Men in Black', BLS makes the SOHO's and seasonal's (net) look good.
Just one of many little Bushco vanishing acts.
Rewriting Modern Reality into a Reaganesque. Wait 'til he saws Condi in half. Wow!

Posted by: Tante Aime on August 11, 2004 08:50 PM

____

I at least sometimes laugh *with* Maureen Dowd on occasion. Same cannot be said for Kaus...

Posted by: EricT on August 11, 2004 09:57 PM

____

jml,

One thing to know about chain store results. They are usually reported as year-over-year changes. Thus a big rise last July makes a strong showing, in the July/July % change, unlikely. However, the implication that last July's conditions cut into this July's actual sales, which is how the piece you quote reads, is nonsense. There is pretty clearly a big difference between consumer demand in Q2 and early Q3 and in other recent quarters. To argue that prior good news is holding back sales is nutty, though. Tax rebate/refund checks boosted sales in prior quarters, but do not hurt sales now.

Your point, though, holds. In the absence of a suppliment to wages, consumer spending seems unlikely to growth very fast.

Posted by: kharris on August 12, 2004 04:53 AM

____

ä = alt + 0228. just a tip for ya, brad.

Posted by: Steve-O on August 12, 2004 12:13 PM

____

Post a comment
















__