QandO: 32k? A measly 32k?

August 06, 2004

32k? A measly 32k?
Posted by Dale Franks

As I wrote yesterday, I was nervous about today's employment release. I thought I was being pessimistic, but even I wasn't this pessimistic. Yesterday, I said anything over 200k would be a blessing. I should've said 50,000. The ISM survey's employment index really jolted me, when I saw the employment index decline to 50 from 57, as 50 essentially means no job growth at all. 50 is the neutral number on the ISM index. And, neutral job growth is precisely what we got.

32,000 new jobs in July, by the way. That was the actual number. Even worse, the anemic 112k from last month was revised sharply downward to 78k. And May was revised downward again to 208k. I'm kind of wondering where all the economists were getting 250k new jobs last week from.

Oddly, the household survey picture is far different. I say "oddly" because you expect a rough congruence between the household and establishment surveys, although, the past several months, there's been quite a divergence between the two. This month is no different. The unemployment rate actually fell to 5.5%, a post-recession low, fueled mainly by a huge 629k rise in the household survey as the number of unemployed fell by 52k. Oh, and I imagine some previously discouraged job seekers re-entered the labor force, too. For all the good it did them.

Anyway, that's the good news, and that's about all the good news there is. Oh, I guess you can point to the 10k rise in manufacturing jobs, and say, "See, 1/3 of all new jobs last month were in manufacturing!" but, really, you're just whistling past the graveyard on that one.

I'm looking at the report for signs of increased labor demand, but I'm really not seeing 'em. The average workweek rose by 0.1 hours per week, but, of course, that follows a 0.2 decline in June, so, big deal. We haven't broken above a 33.8-hour workweek for two years. So it's not like workers are putting in a lot of overtime. And, with only a 0.3% rise in hourly earnings, which is only 1.9% year over year, there's no sign that labor market concerns are figuring in to employment decisions.

Politically, this is a big problem for Bush. Democrats will imply hammer away at the job losses we've had over the past few years, and blame him for not creating new jobs at the rate he promised. And, frankly, the rate of job creation Bush promised hasn't materialized. Of course, since presidents don't actually create jobs, making promises on something you have no power to deliver is unwise.

You gotta wonder, though, how Kerry's economic plan, assuming he could get it through Congress, would help create more jobs, since tax increases—but only on the richest Americans, i.e., the ones who create jobs—are the centerpiece of his plan.

But, Kerry's not in the center seat now, and Bush is, and I expect he's hoping for a damned good August and September employment report.

TrackBack

Comments

Kerry is going to increase taxes on those making over $200,000. Unfortunately, small businesses make a higher percentage of that income range than they do of lower ranges. Raising taxes on those businesses will mean more people laid off.

I know the payroll survey is the one we should be looking at for job creation. But this month, there was a huge disparity between the payroll and household surveys. Payroll said 32,000 jobs created...household said 649,000. I don't recall ever seeing a gap that large before this month.

Posted by: Steverino at August 6, 2004 11:05 AM

Should he be elected, John Kerry will create jobs the same way Bill Clinton did: by doing absolutely nothing and taking credit for the hard work of others.

Posted by: Tom Ault at August 6, 2004 11:38 AM

11 months of numbers, each better than the last, and this is a matter for gloom and doom? Keep in mind, this is summer. We're dealing with vacations and whatnot.

Posted by: Bithead at August 6, 2004 03:05 PM

What if any impact does this have on your assesment?

what if any significance does this have on your assesment.

"Total employment rose by 629,000 to 139.7 million in July, ...."

http://www.bls.gov/news.release/empsit.nr0.htm

does this data even matter?

Posted by: james at August 6, 2004 10:27 PM

Not much. The Household survey is pretty quirky, and there's a lot of information it leaves out. In general, the establishment survey is regarded as the more accurate. Also, a lot of those new entrants to the work force are, as I pointed out, formerly discouraged job seekers who have renentered the market. Maybe they'll leave the job market next month.

The household survey is often intriguing, but you have to have the gifts of Nostradamus to figure ou what it actually means.

Posted by: Dale franks at August 7, 2004 02:21 AM

All the "horrible" job data didn't seem to hurt Clinton's reelection. The average unemployment rate during his first term was 6.0%, a half point more than the current rate.

Posted by: Lance Jonn Romanoff at August 8, 2004 12:39 AM

Yeah, but that was because the rate was 7.5% under Bush, so he looked good by comparison. After 4 years of 4.5% or less, 6% doesn't seem like it's so hot any more.

At the time, though, it seemed pretty good

Posted by: Dale Franks at August 8, 2004 02:30 AM

Dale, I do not recall anyone ever comparing the unemployment rate under Clinton to the unemployment rate under Bush in 1996. By that time the George HW Bush presidency was ancient history. For that matter, I do not hear any Democrats now specifically comparing the current rate to the second Clinton term, presumably because they would then have to come up with some argument regarding how they would achieve a 4.5% rate. Instead we are getting vague, statistically shaky arguments about how many jobs have been "lost."

I'm also curious why you are so quick to dismiss the household survey when the establishment survey systematically ignores all non-payroll workers (of which there are now at least fifteen million), doesn't count those who are employed but on unpaid leave, counts people with two jobs TWICE, and so forth. It seems each survey has some questionable methodology.

Posted by: Lance Jonn Romanoff at August 8, 2004 10:02 AM

As far as I can determine, not one of the jobs I have held, in over thirty years in the workforce, would have shown up in the payroll survey. That includes three created after Bush 43 took office.

Posted by: triticale at August 9, 2004 01:25 PM

I know the payroll survey is the one we should be looking at for job creation. But this month, there was a huge disparity between the payroll and household surveys. Payroll said 32,000 jobs created...household said 649,000. I don't recall ever seeing a gap that large before this month.

Posted by: Steverino at August 6, 2004 11:05 AM

Yea, but kerry is going to cut taxes for these companies, and lighten their burden, while closing the loophose the big companies use to get out of taxes altoghether. He has incentives to create jobs to get tax breaks, and all sorts of great ideas, check it out yourself:
www.johnkerry.com/plan

Posted by: Peace + Love at August 10, 2004 11:45 PM

Dale, I do not recall anyone ever comparing the unemployment rate under Clinton to the unemployment rate under Bush in 1996. By that time the George HW Bush presidency was ancient history. For that matter, I do not hear any Democrats now specifically comparing the current rate to the second Clinton term, presumably because they would then have to come up with some argument regarding how they would achieve a 4.5% rate. Instead we are getting vague, statistically shaky arguments about how many jobs have been "lost."

Posted by: Lance Jonn Romanoff at August 8, 2004 10:02 AM

The reson the unemploymment rate isn't a good judgement of actual unemployment is this... the unemployment rate is technically all those on unemployment. Over time you can get dropped off. and a new job is any part time or full time. and you don't get unemployment from a part time job.
Those numbers are always going to be screwy. Plus, why doesn't anyone thingk about this.... WE HAD LESS POPULATION BACK THEN. THERES A TTTTOOOOONNNN MORE PEOPLE WHO NEED JOB EVERY SINGLE MONTH

Posted by: Peace + Love at August 10, 2004 11:50 PM

As far as I can determine, not one of the jobs I have held, in over thirty years in the workforce, would have shown up in the payroll survey. That includes three created after Bush 43 took office.

Posted by: triticale at August 9, 2004 01:25 PM

you've had 3 different jobs since "dubya"?
so you get payed under the table then? explain to me why you wouldn't have shown up on the payroll survey.

Posted by: Peace + Love at August 10, 2004 11:53 PM

Dear Peace + Love;

Before you start commenting on the unemployment report, you should probably take the time to actually learn how the report is compiled. You comments so far indicate you dont have a clue.

The reson the unemploymment rate isn't a good judgement of actual unemployment is this... the unemployment rate is technically all those on unemployment. Over time you can get dropped off.

Uh, no. The unemployment rate is compiled through a survey given to individual households. Respondents are asked if they are employed. If they are not employed, they are asked if they are actively seeking work.

Those who are not actively seeking work are dropped from the numbers as being 'out of the labor force". Then they do the simple math to figure out the percentage of the unemployed who are actively seeking work. That then becomes the unemployment rate.

It has nothing whatsoever to do with those who are on unemployment.

you've had 3 different jobs since "dubya"? so you get payed under the table then? explain to me why you wouldn't have shown up on the payroll survey.

Because the payroll survey doesn't count people who are self-employed. It only counts the number of employees who are on actual employee payrolls, which is why it's called the "payroll survey". So, if you are a free-lance web designer, or a plumber, or any other type of independent contractor, you don't exist as far as the payroll survey is concerned.

It is, however, now clear why you think John Kerry has sound economic policy ideas.

Posted by: Dale Franks at August 11, 2004 12:19 AM

Dear Lance John Romanoff;

I'm also curious why you are so quick to dismiss the household survey when the establishment survey systematically ignores all non-payroll workers (of which there are now at least fifteen million), doesn't count those who are employed but on unpaid leave, counts people with two jobs TWICE, and so forth. It seems each survey has some questionable methodology.

Because, on balance, the establishment survey at least asks questions that have some factual, documentary basis. The household survey is so nebulous, it's hard to figure out exactly what it is telling us.

It tells us how many people consider themselves to be employed. Employed at what? We have no idea. It tells us how many people are unemployed, and, of those, how many are not actively seeking work. Why aren't they seeking work? Are they living off the trust fund dad left them? Are they beach bums? Are they unable to find work in their chosen field, but are unwilling to work at any job except those for which their degree in 9th Century Saxon poetry qualifies them? I don't know. And neither does anyone else.

Forthe last several months, we've seen really wierd disparities between the establishment survey and the household survey. There's millions of people who say they're employed, but they aren't on any payrolls we can find in the establishment survey.

It would be interesting to know why, but we certainly aren't gonna find out by perusing the household survey.

Posted by: Dale Franks at August 11, 2004 12:26 AM