Asymmetrical Information

This makes no sence. The jobs data is at least a month old. Unless you're suggesting that the interest rate hike has some magical retroactive effect?

Yes, the decrease in job creation seems more of a fallout over the higher oil price which in turn is jacked up by at least an $8 a barrel terror premium. Oil prices are coming down as will terrorism, so economic progress will continue.

Rule of thumb: prices do not always go up smoothly. So, there is no need to panic.

not all kerry supporters cheer bad news. i don't "take heart" in poor job creation numbers; i am sorry that more people can't find work.

i do take heart that undecided voters might consider reports like this and swing to Kerry. but that's a secondary effect.

i'm just tired of being accused of enjoying bad news. i don't.

francis

That would be less waiter,cook and maid positions.

I wasn't saying that the rate hike caused the jobs fall; only that the jobs fall makes it less likely that Greenspan will further raise rates.

Uh, Jane . . .

"Jobs fall"????

Yes, I'm nit-picking. However, there IS a substantial difference between something declining and somethign rising (e.g., number of jobs) less than was expected.

Judson continues the lie. He must get his news from Krugman.

Another interesting tidbit, from the WSJ:

"In addition, employers also added 35,000 fewer jobs than previously thought in April and May."

Got that? We had b.s. high numbers for April and May, and June caught everyone off guard because it was lower than expected.

It wasn't quite BS numbers -- the numbers from earlier months were revised upward by even more in successive months, often as much as 50K. And even taking out those 35K jobs, that still leaves us with over 500K jobs created in April and May.

The Iowa market reacted accordingly?

Looking at the graph, doesn't seem like much change to me.

Keeping accurate employment statistics is a difficult job that the government doesn't do particularly well. The question of the moment is whether the economy will roll over, or will continue to grow. My guess is that it will continue to grow.

From an economic perspective, it probably doesn't make a great deal of difference whether Kerry or Bush gets elected. Presidents are figureheads who generally do little of any significance. Iraq will probably be a better issue for Kerry than the economy, though DNC rhetoric continues to compare Bush with Hoover.

Clinton got elected in 1996 when the unemployment rate was 5.6% and the media thought that rate was amazingly low.

What can’t Bush get elected when the unemployment rate is 5.6%? Could it be that the media is saying that rate is amazingly high?

"Yes, I'm nit-picking. However, there IS a substantial difference between something declining and somethign rising (e.g., number of jobs) less than was expected."

Given that there need to be ~140,000 new jobs every month just to keep up with population growth, an increase of 115,000 really is a net decline.

Jon H:

Uh, no.

A decline is when something becomes numerically smaller. What you are referring to is a change in the rate of increase.

Source, please, for your assertion of the need for 140,000 new jobs per month to keep pace with population growth?

Hey, Hondo: I've noticed the same statistic used like a club in the past two years. First of all, I don't know where it came from and I never heard it being used during either Bush 41 or Clinton times. Second of all, Krugman uses it, but the number is 150,000.

Who knows?

Hondo, Klug:

It due to the growth of the labor force.

Jane,

I don't think the Iowa markets have a very good record.

"It due to the growth of the labor force."

Yes, obviously. The questions is how much does the labor force typically increase every month, and what is the source for the number?

"I don't think the Iowa markets have a very good record."
IIRC the Iowa market had a perfect record until 2000, and even then I believe it was pretty close. The problem with the Iowa market is that its pretty much useless until right before the election, just like other polls.

Jon H, GT:

Anyone can make up a number. The assertion that the correct number is 140,000 may or may not be accurate.

Again: an AUTHORITATIVE source, please, (e.g., government, academic, or widely accepted private organization/think tank) for the 140,000 figure for monthly growth of the US labor force?

If you can't give an AUTHORITATIVE source, you should wonder if you are correct. It's possible your are wrong, or are being misled.

The Iowa markets were not close at all in 2000. And their previous record is bad as well. I used to follow them but not anymore. They simply follow what the polls say so I might as well check the polls.

Hondo,

Just go over to the BLS website where you can check the numbers for yourself.

Uh, GT . . . .

In case you haven't noticed, the BLS web site has literally thousands of documents. Telling someone to "go check out the BLS website" is about as helpful as telling them "go look it up in your local library."

Can you be a little more specific - say, a URL maybe?

This makes no sence. The jobs data is at least a month old. Unless you're suggesting that the interest rate hike has some magical retroactive effect?

I think the rate hike was widely anticipated. Hence it could in fact have a "retroactive," effect without the need for any magic.

Hondo,

You can start here:

http://data.bls.gov/labjava/outside.jsp?survey=ln

GT:

Thanks for the cite. I like to check facts; an amazing number of "well known facts" - aren't.

I looked at the cited Java applet. While it does give figures for adjused and/or raw numbers of unemployed by month - and does give the unemployment rate by month - it doesn't clearly identify the need for job creation by month to maintain a given rate.

Would you have a better cite (e.g., one that directly gives the figures for monthly labor force growth needed to maintain a given unemployment rate, preferably with some explanation as to how they are derived), or will I need to find an hour or two, download the data, and set up a spreadsheet to figure things out for myself?

Hondo:

I believe that the number is derived as follows:

(((US Pop.) * (pop growth rate)) * ((employed + unemployed)/total pop))/mos in year

or ((total new cits) * (% of US employed))/12

SomeCallMeTim:

To a first approximation that is likely correct. It seems common-sensical. However, I'm not an economist, so I don't know if it's correct or not. Something along those lines is likely what I would have used.

However, given the various things professional economists do to include/exclude/adjust employment figures, I believe your formulae might be a bit simplistic. One variant refers to "Total US Population", the other "total new cits". Do we count employed and unemployed illegal immigrants or not? Do we count those who have left the labor market, either temporarily or permanently (e.g., retirees, those - of either sex - who have chosen to stay at home and raise family vice work, those who can't work due to transient or permanent health conditions, etc . . . )?

Since each of these groups is fairly large, including or excluding them from the calculation will have a proportionally large effect. I'm not sure that "Total US Population" is the relevant number. I think that "Total Seeking Employment" is likely a better number. Then again, I don't KNOW - and would like to.

Since I'm not a pro in this field, I'd appreciate a pointer to some form of authoritative document - if one exists - that discusses how such figures as "monthly new job requirements" is calculated.

Hondo:

I was sloppy above, so I'll restate (though your points continue in force). The arguement is that there are X new people in the US every year, and we assume (a) that they add to the population at the same number per month for that year, and (b) that they want to work at the same rate as the rest of the US population. So the formula is:

(new US pop per month) * (% of US pop that wants to work).

IIRC, B DeLong confirmed the formula on his website, so you could check there if you want further evidence. (I would, but he has a lot of posts on Bush's misuse of the unemployment statistics, and going through them is more work than I'm willing to do for something I'm pretty sure is right).

I agree that this is a quick and dirty number. For example, you might expect that we need more jobs per month than this predicts, because the baby-boomers' kids are coming on-line and people are starting to work longer. But I'm not sure that much more than a quick and dirty number is needed. It's just a tracking number; it isn't as if (politically) the government could order up new jobs on a month by month basis if there was a shortfall.

SomeCallMeTim:

Thanks. I think we agree on this one.

Any idea on where to get authoritative (or even decent) numbers, say for the past 12 months or so, on:

a. The monthly new US population, and
b. The proportion of the US population desiring to work?

I'm not trying to be obtuse - but when I hear a figure used and no one can seem to give a clear reference as to how it was derived, I start to wonder if it was made up out of thin air.

I'm not suggesting that is the case here - the 140,000 number seems plausible - but I WOULD like to see how the number was obtained, and what assumptions were made in obtaining it.

Hey, Hondo:
I'm a grad student in organic chemistry, so I don't have a lot of immediate references. However, my free copy of the 2nd edition of "Macroeconomics" by Blanchard seems to say that there are 350,000 people entering the market per month and 200,000 retiring per month. (This in the "labor markets" chapter.)

That seems to jibe with the 140,000 to 150,000 jobs factoid, but it still doesn't mention where these numbers come from.

Tim and Hondo:

I actually did check Brad DeLong's website for all references of "150,000" (all six pages), since that's the number that I've heard again and again.

DeLong himself referenced it once as I recall, but without derivation. The mentions were mostly in the comments, again mostly in the vein of "you need 150,000 just to keep pace, etc., etc." It seems to be a consensus figure among economists that has turned into a handy hammer to beat the Bush administration with.

"It seems to be a consensus figure among economists that has turned into a handy hammer to beat the Bush administration with."

It's a non-partisan hammer. Unfortunately for Bush he's President right now. If the 150,000 job number seems like a new issue, it's because Clinton didn't have much trouble adding more than 150,000 jobs a month during his 8 years in office, so there was no reason to complain, for a very long time.

If Kerry wins, and the economy has many

But after the first six months to a year, any sub-150k months will be all Kerry's, and he'll take heat for them.

Hm. What happened to that sentence.

I meant to say that if Kerry is elected, and he has many sub-150k months, he'll take heat from the left. He'll get a break for the first 6-12 months, in that low hiring months will be blamed on lingering effects of Bush policies. (Just as the 2001 recession is blamed on the Clinton administration.)

etc.

Folks:

I'm really not trying to be obtuse here - but can someone PLEASE point me at a decent source that can tell me where the damn number comes from, how it is calculated, and what assumptions are made in calculating it?

For a commonly-accepted and important number, this doesn't seem too much to ask.

Comments are Closed.