December 05, 2003

Disappointing Labor Market News

*Sigh.* Unexpected disappointing labor market news. Employment growth in October below what was expected, and what is needed to keep payrolls growing as rapidly as the labor force:

WSJ Online: ...Nonfarm business payrolls grew by a net 57,000 last month, raising the total of job gains since July to 328,000. Economists had expected a heftier payroll gain of 150,000, and for the unemployment rate to hold at 6%, according to a survey by Dow Jones Newswires and CNBC.

Over the last three years, employers have cut more than three million private-sector jobs. In order to replace those jobs and sustain the recovery, most economists are looking for monthly payroll gains in the neighborhood of 200,000 to 300,000.

Ed McKelvey, an economist with Goldman Sachs & Co. in New York, puts that number at more like 400,000. That far exceeds the average of the late-1990s economic boom.

In a speech last month, Federal Reserve Vice Chairman Roger Ferguson said, "Even if the growth is sufficient to make meaningful progress in reducing the slack in our nation's labor and capital resources, that pool of underutilized resources is large and it will take some time to be worked off completely." Most forecasters, as a result, don't expect the central bank to begin raising interest rates at least until the middle of 2004.

In its report Friday, the Labor Department revised its estimates of job growth for September and October, saying employers added 137,000 jobs in October -- 11,000 more than previously thought. But the job gains in September were smaller than first thought -- an increase of just 99,000 that was smaller than the first estimate by 26,000.

Posted by DeLong at December 5, 2003 06:08 AM | TrackBack


Rigorously, it's unexpected; but is it intuitively unexpected? Everything I've been reading here (and elsewhere) for the last year tells me that the employment situation isn't going to improve any time soon.

I have to admit that I'm frustrated that economists can't tell a coherent story about what's happening. Clearly, something isn't being accounted for. What is it?

Posted by: Keith M Ellis on December 5, 2003 07:25 AM


Aren't we supposed to gloat about this, being America-hating liberals and all?

Posted by: Zizka on December 5, 2003 07:28 AM


I go through moods of being irritated by all this management of expectations. Can't we just say "the handbasket is 38.2% of the way to Hell this month" or something equivalent?

Damn, now I know how Harry Truman felt when he demanded a one-handed economist.

Posted by: David Lloyd-Jones on December 5, 2003 07:30 AM


Well, when I saw the revised productivity numbers, i certainly didn't expect much in the way of job growth this month.

Posted by: howard on December 5, 2003 07:43 AM


I know that the employment figures from the CPS (used to calculate the unemployment rate) are not supposed to be as accurate as the payroll numbers, but a quick glance at the employment figures shows 1 million more people working in October and November. That's a big time divergence between the payroll and employment figures. Very strange.

Posted by: Art Woolf on December 5, 2003 07:44 AM


Interesting. The report on NPR this morning was far more positive, noting that the unemployment rate had fallen to 5.9% and that new jobs had been created. The report did note that factory jobs were still disappearing, though.

Posted by: PaulB on December 5, 2003 07:46 AM


How can we have falling unemployment despite lower-than-expected job creation? One reason is that a single month's results can be anomolous. Those who are chortling at the month's low job creation numbers might bear in mind that one month's results isn't much to of a base for predicting a trend.

Posted by: David on December 5, 2003 07:48 AM


I know that the employment figures from the CPS (used to calculate the unemployment rate) are not supposed to be as accurate as the payroll numbers, but a quick glance at the employment figures shows 1 million more people working in October and November. That's a big time divergence between the payroll and employment figures. Very strange.

Posted by: Art Woolf on December 5, 2003 07:49 AM


How can we have falling unemployment despite lower-than-expected job creation? One reason is that a single month's results can be anomolous. Those who are chortling at the month's low job creation numbers might bear in mind that one month's results isn't much to of a base for predicting a trend.

Posted by: David on December 5, 2003 07:53 AM


"Aren't we supposed to gloat about this, being America-hating liberals and all? — Zizka"

On balance, I'm pleased with this news. A big fear I have is that the economy will improve sufficiently enough to ensure Bush's reelection.

I said this in another comments thread, but it is perfectly rational to believe that continued employment doldrums through '04 is the lesser of two evils, the greater evil being Bush's reelection. Polite people don't crow about bad news in public, but polite people also don't dishonestly claim outrage at the suggestion that bad news is good news for the opposition party. It is.

To my mind, the deciding principle regarding the moral acceptability of this sort of supposed shadenfreude is whether its petty and personal or principled and rational. I think that the Bush administration is doing enormous damage to the US, in many ways that the general public does not comprehend, and that tangible examples of this that reach the voters' consciousness are *good things*. The best thing that can happen to this country is not to get another four years of Bush and co.

At this point, I'm willing to stomach Dean's (or another's) protectionism as, ahem, the lesser of two evils.

Posted by: Keith M Ellis on December 5, 2003 07:53 AM


Yeah, looks like expectations probably got ahead of where the job market is at right now. I kind of suspected as much. There has been improvement, but no hiring boom is in evidence. Stepping back a little from the November report, it appears clear that we have had a transition from job loss to moderate job growth reflected in the Sept-Nov payroll reports. Weekly initial unemployment claims now appear consistently below 400,000 and by some margin at this point (let's say around 370,000). I think we build from here and that in the Spring we start clocking 200,000 net new jobs per month in the payroll report.

Posted by: Joe Blog on December 5, 2003 07:59 AM


Bad news. In the heartland, we are still shedding mfg jobs.

Overall, Bush policy is bad. By my calculation, his tax cuts have been offset by the need for the government to borrow an additional $600 billion. The trade deficit is collecting money from Asia that needs to be spent on American mfg goods. States are still cutting projects/salaries/workers in an effort to meet budget shortfalls. Bush policy has created a giant sucking sound that is sucking the stimulus out of the economy.

The bright spot was housing and refinancing. However, we have a bubble and now new mortgages are down due to higher interest rates. Where is the demand for US goods? If we don't see that on the horizon, then this economy is flat or heading for a double dip.

I totally disagree with Keith Ellis. I would settle for a change in policy from Bush to correct the situation sooner rather than later. Unfortunately, he does not have it in him, so we will go into 2004 elections with a weak economy.

Posted by: bakho on December 5, 2003 08:12 AM


One reason interest rates are so low (considering the scheduled fiscal train wreck) is that the bond market doesn't believe the economy is all that healthy, so real rates decrease (compare the yield on 30 year TIPS today with past rates).

So far today, rates are down

Posted by: richard on December 5, 2003 08:17 AM


Other measures of job growth are improving.
The diffusion index just rose over 50% for the first time in three years. Overtime hours also expanded for the second month.

Joe Blog seems to be on target for now, but why would it be any different in the spring..

It looks like the quarter to quarter growth in
hours worked will be about 3% (ar). But this would imply that productivity growth is slowing sharply.

If we are in a new era of higher productivity growth it means that a given rate of economic growth will generate less employment growth.
4% real GDP and 3% productivity means 1% employment.

This is reflected in the GDP Gap. Using the official CBO estimate of potential real GDP
the GDP gap is only 0.5% of actual real GDP --
implying that the economy is operating at near full capacity. But if you assume that potential real GDP growth has been 4% since 1995 ( 3% productivity & 1% labor force growth) , the GDP gap is 7.6% of actual real GDP , about the same level as it was in the early 1980s. This is the GDP gap that would be in line with the manufacturing capacity utilization data.
But the ISM vendor performance index -- percent experiencing slower deliveries --just jumped to 56%. Historically, this measure had a tight fit with capacity utilization. But now vendor performance and commodity prices are rising despite the very low level of cap use.

We really need to take a serious look at what are the normal economic relationship in a high productivity environment and where manufacturing output growth is dominated by High tech -- over the past year HT ouput rose 20% while all other manufacturing output fell 0.5%.

For example, chemical firms are announcing job cuts. In a poor pricing and slow growth environment the only way to get earnings growth is through productivity gains. But that means they must cut jobs to get productivity.

Posted by: spencer on December 5, 2003 08:22 AM


"Joe Blog seems to be on target for now, but why would it be any different in the spring.. "

Well I am not an economist, but let me say why that is my read on the situation. Business profitability has improved dramatically over the last year from it's bottom which occurred roughly around mid 2002. As an illustration, the rise in S&P 500 core earnings is pretty staggering, roughly doubled from then to now.

I think this really takes some pressure off business employment, no longer a need to cut as many people as possible or not hire anyone no matter how great a specific need is. And increased economic activity provides growth opportunites for some companies. Definitely some industries, like chemicals, will continue to cut and we aren't going see any sort of mass binge hiring but I detect a change in the mood and hiring activity. In the second half of this year the immediate effect has been a decline in job loss and some selective hiring. A number of my previously displayed friends have found jobs that can broadly be categorized as in the areas of professional and technical services. I think we will see the hiring side kick in a little more when we come back from the holidays and some healthier payroll reports as a result.

The weak dollar is also a factor and will provide additional opportunities for some industries like agriculture, tourism, manufacturing. That takes a while to kick in as well but I think it is now.

Posted by: Joe Blog on December 5, 2003 08:50 AM


On the discrepancy between CPS and CES employment, there is a recent BLS paper:
that extensively discusses various potential contributing issues. A few things bear noting:

-- The CPS-CES disagreement conundrum that the BLS paper covers is that CPS employment growth significantly *lagged* CES nonfarm payroll employment growth during the mid-to-late 90s.

-- CPS and CES use significantly different concepts of employment. In particular, you don't have to be very employed at all (or necessarily even paid) to be counted as employed in CPS. It seems to me that this could make CPS employment growth look rosier now, though why it wouldn't have had a similar effect over the boom years isn't obvious.

-- CPS employment is sensitive to the underlying population controls. The last change to the controls added 576,000 jobs to CPS employment in January.

I'll also put a plug in for last weekend's op-ed in the Times by Austan Goolsbee if anyone missed it.
A friend recently forwarded the Feb. '03 QJE paper by Autor and Duggan referenced therein on the relationship between the disability rolls and unemployment, and it's interesting stuff too.

Posted by: Tom Bozzo on December 5, 2003 08:53 AM


I've got a graph up at my blog (click on my name) that I've been updating monthly showing our "progress" toward the 1.4 million new jobs Bush promised when he pushed through the increase in the birth tax (well, he called it a "tax cut", but it's really raising taxes on the future).

Posted by: Mike Jones on December 5, 2003 09:32 AM


I missed the chortling, but maybe I just wasn’t listening hard enough. The reason the jobless rate can go down when the headline (establishments survey) figure comes in below estimates is that they have nothing to do with each other.

I agree with Keith on the balance of evils, but since wishing won’t affect outcomes, I am more interested in seeing that the Plame affair, lies about Iraq, the truth about Medicare and the inheritance tax, the risk of the big deficits, undermining of environmental regulation and such are kept in the headlines.

As regards the data, kick in another 63,300 for California grocery store strikers. That takes the tally up to 120k, but most forecasters were aware of the strike, so that doesn’t get us any closer to what was expected. It just means that November was more like October than we thought. Tepid holiday sales are so far associated with 35k new (seasonally adjusted) retail jobs, ex-strikers, which seems a bit aggressive. Factory job losses continue, but the pace is way down. No durables sector jobs were lost in November, for the first time since October of 2000. As Spencer noted, the diffusion index is above 50 – the number of industries hiring has been above the number laying off for 3 months now, and November was the best of the lot. Temp hiring remains firm, and hour worked are up just a wee bit. Those are all signs of a tightening labor market. No visible impact on workers, however. Hourly wages rose just 0.1% in November. At best, that keeps up with inflation (probably not, as the hedonic mumbo-jumbo is left out of nominal data). The duration of average unemployment rose, as well. The steady rise in temp employment is an effort to avoid translating a tightening labor market into better pay or jobs.

The argument heard recently that the discrepancy between household and establishments’ job counts is due to a shift to entrepreneurship looks a bit shaky, based on household data. Of the 589k new jobs households reported in November, only 118k were accounted for by self-employment. Similarly, the 441k new jobs household reported in October included just 70k new self-employed. We’re left with a discrepancy of 842k jobs not accounted for between the household and establishment surveys in just 2 months. Were they all hired by the other 178k? When you consider that at least some of the self-employed include newly converted house painters, auto mechanics and nannies, entrepreneurship in the glowing sense that some economists have recently offered is just not born out.

Posted by: K Harris on December 5, 2003 09:43 AM


Darn, beaten to the tables. It looks, though, like much of the remaining November increase is accounted for in part-time employment categories, +340K part-time for non-economic reasons (i.e., people who report want to work part-time), +97K part-time for economic reasons.

Posted by: Tom Bozzo on December 5, 2003 10:32 AM


doesn't it seem likely that the continued lack of job growth is due to substantial perceived uncertainty in the future business climate, presumably as a result of the current administration's long term economic policies?

Posted by: anonymous on December 5, 2003 11:35 AM


For years I have done a chart of payroll employment growth vs household survey growth.
The survey growth almost always leads payroll
at both peaks and bottoms. The payroll data has always had trouble capturing employment in small
firms and startups, but they probably are quicker to respond to a changing economy than large firms.

At economic bottoms and the first year or two of the recovery the growth in the survey data is usually 1% to 2% stronger than the payroll data.
Late in the cycle and at peaks this reverses.

Based on the past history the current surge in the survey employment vs the payroll employment is a perfectly normal cyclical development and the some 1% to 2% difference is what happened in every cycle since 1970.

Correction: earlier I said hours worked were up 3%, actually it is more like 2.5%. Sorry.

Posted by: spencer on December 5, 2003 11:49 AM


In the early months of the Bush administration, one often heard that "the grown-ups are back in charge." But if being a grown-up means planning for the future — in fact, if it means anything beyond marital fidelity — then this is the least grown-up administration in American history. It governs like there's no tomorrow.

Nothing in our national experience prepared us for the spectacle of a government launching a war, increasing farm subsidies and establishing an expensive new Medicare entitlement — and not only failing to come up with a plan to pay for all this spending in the face of budget deficits, but cutting taxes at the same time.

Recent good economic news doesn't change the verdict. These aren't temporary measures aimed at getting the economy back on its feet; they're permanent drains on the budget. Serious estimates show a long-term budget gap, even with a recovery, of at least 25 percent of federal spending. That is, the federal government — including Medicare, which Mr. Bush has given new responsibilities without new resources — is nowhere near solvent.

Then there's international trade policy. Here's how the steel story looks from Europe: the administration imposed an illegal tariff for domestic political reasons, then changed its mind when threatened with retaliatory tariffs focused on likely swing states. So the U.S. has squandered its credibility: it is now seen as a nation that honors promises only when it's politically convenient.

What really makes me wonder whether this republic can be saved, however, is the downward spiral in governance, the hijacking of public policy by private interests....

Posted by: anne on December 5, 2003 12:28 PM


Mind the Gap
A new study of operating cash flow and earnings for the 87 nonfinancial companies in the Standard and Poor's 100 suggests that the recent upturn in corporate profits may not be sustainable. -Ronald Fink, CFO Magazine

This has a huge impact on how we think the economy is performing and what the job picture should be. If cash flow is being used to mfg profits, then that could affect productivity / profits / GDP and other indicators. Garbage In- Garbage Out. Have we slipped into a new accounting world that biases the data?

Posted by: bakho on December 5, 2003 12:55 PM


Well if productivity growth is greater than GDP growth, why should there be a spurt in employment? We have asked the question often, but we need to keep asking. Demand growth is not sufficient yet to foster enough job creation. The fiscal and monetary stimulus has helped demand, but not enough to cause an employment spurt. If productivity were less, there would be more need for workers.

Posted by: anne on December 5, 2003 02:23 PM


I agree about the continued IT revolution - There's no reason why the continued development of the economy into more efficient and interlinked production of goods and services suddenly stopped because the stock market bubble popped. if anything, the decline of the bubble refocussed companies on core competencies which meant heightened productivity gains throughout the shallow recession. our companies are probably the "leanest and meanest" the've ever been = they have very high operating leverage which means that output growth is going to continue to exceed labor growth for awhile to come (at least in established companies).

I think this is essentially the same point you are making about this business cycle being different. you are recognizing the high operating leverage and saying therefore employment growth will not rev up.

...that's where we part...

what I think you (and most headlines) fail to appreciate is that people - at least in the US - make their own work outside of the "establishment". the wsj ran an article yesterday about small business employment that discussed this anecdottally - but it's begginning to appear in more hard statistics in the household survey which continue to suggest far more employment growth than the "establishment" survey
the household survey suggest that over 1 million jobs have been created in the last 2 months while the establishment survey suggests that only 200,000 jobs have been created in the last 2 months. (which is another reason why the rate is down to 5.9% despite the supposed weak gains in the establishment survey)

and this explains why they are so different

That article on small business in WSJ again cited the supposed decline of 2.4 million jobs during the recession - this data is from the establishment survey of 160,000 existing businesses. But we know that since the bubble burst, existing businesses have been restructuring, _outsourcing_ and getting "lean and mean" so it makes a lot of sense that the largest established 160,000 business in the US have lower employment levels now that Nov. 2000.

but we also know that new business were still formed during the recession, and that the outsourcing was not just to India but to US based small businesses and that small business are quicker to respond to upticks in demand than large business. which is why the "Household" survey of actual individual people and not existing businesses shows that there are more jobs now than ever before (tick the first box and hit the Retrieve Data button at the bottom)

so you should stop your shameless lies about loss of employment under Bush:
(event / date / # of people employed)
Bush took office: Jan 2001 - 137,846,000
Bush signed tax cut # 1: May 2001 - 137,036,000
World Trade Center Bombing #2: Sept 2001 - 136,858,000
Lowest point during GWB: Jan 2002 - 135,791,000
Bush passed signed tax cut #2: May 2003 - 137,487,000
NOW: Nov 2003 - 138,603,000

so 1.6 million more people are employed since Bush first started implementing his economic policies and 2.8 million more people are employed since the end of the worst part of the economic dislocation from 9/11. And there would probably be another 1 to 2 million people employed if 9/11 hadn't happened.

The fed better start raising rates soon, growth is too strong and the currency is too weak because our interest rates are too low.

Posted by: alex on December 5, 2003 02:45 PM


alex, since there isn't an irony-signaling emoticon in your post, I'll just note that the only employment data mentioned in Elaine Chao's statement on today's news is the CES payroll growth:
If even the Bush administration isn't taking credit for those jobs from CPS, doesn't that tell you something?

Posted by: Tom Bozzo on December 5, 2003 07:54 PM


Steel restructuring removed layers of management and reduced workers. It is estimated that about 6 people do what previously required 10. A few of these productivity gains really bump the average.

Posted by: bakho on December 5, 2003 08:27 PM


Now then, folks, does Tijuana constitute a Hooverville?


Posted by: David Lloyd-Jones on December 6, 2003 03:10 AM


then why the dollar per euro rate of exchange has gone from $0.88 to $1.20 in te last two years?

In Euroland we have an economy that is always blamed as inefficient. Your claims do not make sense to me.


Posted by: Antoni Jaume on December 6, 2003 05:03 AM


The Euro is where the problems of the budget deficit is first appearing. It is really very simple, the supply of dollars, the current account defict ( the savings investment gap) is soaring so it takes higher interest rates spreads-
what drives demand for dollars -- to attract the needed increase in foreign capital inflow. But since rate spreads are not high enough the dollar is falling so that the given supply of Euros translates into a larger supply of dollars.
Partially because of the federal deficit we are now at the point where it takes a combo of higher rates and/or a weaker currency to attract
sufficient capital to finance the deficit.
The problem is we are just seeing the start of this process, before it ends it probably will
generate a recession.

For example, a year from now even though China will have a large trade surplus with the US, its
total current account will be in deficit.
Consequently, the Chinese will be selling T bonds and dollars to buy other countries goods rather than financing our deficit. I know, than someone else will buy the dollars so it does not matter.
But that line of thinking ignores the question of at what price will they buy the dollars.

Posted by: Spencer on December 6, 2003 05:27 AM


If China has a very big surplus of dollars, it would make sense to buy land in the USA and exporting Chineses there. China could easily export 200 millions of individuals there, and get rid of quite a few problems. They should put some 5 millions in 40 states, that would change instantly the political landscape...


Posted by: Antoni Jaume on December 6, 2003 06:45 AM


Alex - That is not the way it looks here in NYC - there are less jobs and more unemployed and the only new jobs are seasonal retail. Almost all of these numbers are crap and based on political agendas.

Posted by: Jim on December 6, 2003 07:32 AM


"The fed better start raising rates soon, growth is too strong and the currency is too weak because our interest rates are too low."

Utter nonsense.

Posted by: Ari on December 6, 2003 08:49 AM


Alan Krueger of Princeton has carefully explained the differences between the business and household employment surveys. The household survey when properly adjusted for sampling changes shows numbers close to the business survey, and both show that this is easily the poorest employment recovery from a recession since the 1930s.

Administration fiscal policy has plunged us to debt that we can not grow out of, while failing to produce enough demand stimulus to create a significant number of jobs. The employment weakness is indeed the worst for any Presidential term in office since Herbert Hoover. The reason is the awful fiscal policy.

Posted by: anne on December 6, 2003 09:43 AM



Medicare Plan For Drug Costs Bars Insurance

Medicare beneficiaries will not be allowed to buy insurance to cover their share of prescription drug costs under the new Medicare bill to be signed on Monday by President Bush.

Posted by: jd on December 6, 2003 10:29 AM


Just when you think you understand how bad the Medicare bill really was, you find out it was worse.

As Paul Krugman pointed out the pending energy bill has a provision for a Hooters in Louisiana. Love them "compassionate conservatives."

Should the federal government subsidie Hooters? Hummmmm.

Posted by: jd on December 6, 2003 10:38 AM


I criticize by creation -- not by finding fault.

Posted by: Serota Pamela on December 10, 2003 07:45 PM


A little nothing goes a long, long way.

Posted by: Maceiras Rachel on January 9, 2004 10:50 PM


Post a comment