April 4, 2008

The March Employment Report

Filed under: Economy,Taxes & Government — Tom @ 8:22 am

SEE UPDATES BELOW — Raw-data reports from BLS show large increases in employment, though nowhere near as large as in March 2006 or March 2007.

________________________________________

Advance info:

  • ADP, in its monthly report, says that nonfarm private employment grew 8,000.
  • John Crudele at the New York Post says that “Wall Street thinks” that the economy lost 50,000 jobs.
  • More specifically, per Ken Sweet at Fox Business, “Economists interviewed by Thomson Financial expect the U.S. economy lost 50,000 jobs in March, bringing the nation’s unemployment rate up two notches to 5%.”
  • To show how spread out the guesses are, one economist notes that “In a survey by Bloomberg.com, economist forecast for NFP were in a range of -150k to 65k. The range in the unemployment rate forecasts was 4.7% to 5.1%. The wide range supports the view that the data remains relatively uncertain.” No kidding.

The actual report:

It’s from Uncle Sam’s Bureau of Labor Statistics, and arrives at 8:30. Here’s where it is.

8:30 a.m. — Here’s the first para:

The unemployment rate rose from 4.8 to 5.1 percent in March, and nonfarm payroll employment continued to trend down (-80,000), the Bureau of Labor Statistics of the U.S. Department of Labor reported today. Over the past 3 months, payroll employment has declined by 232,000. In March, employment continued to fall in construction, manufacturing, and employment services, while health care, food services, and mining added jobs. Average hourly earnings rose by 5 cents, or 0.3 percent, over the month.

This is basically a bad report from top to bottom, and leads me to believe that first quarter GDP, instead of being a couple of tenths of a percent positive, will might go a couple of tenths negative. (See Update below for some contrarian positivity.)

Digging into the detail:

  • In the Household Survey (the basis for the unemployment rate), the number of unemployed zoomed by 434,000 after dropping by 284,000 the previous two months. It appears to have occurred largely because a lot more people (410,000) entered the labor force (i.e., started looking for work, and didn’t get it).
  • The Household Survey only shows a net loss of 24,000 jobs in the economy during the month.
  • The net increase in the number of unemployed during the first quarter was 150,000. The quarter-long number isn’t that large, but the one-month increase of 434,000 will likely be the one that will get media attention.
  • It implies more precision than might exist, but I’m going to calculate it anyway — Carried to three decimal points, the unemployment rate was 4.975% in December, 4.925% in January, 4.812% in February, and 5.085% in March.
  • African-American unemployment, which had dropped from 9.2% to 8.3% in February, went up to 9.0% in March; teen unemployment, which was 18.0% in January, went down for the second straight month, from 16.6% to 15.8%.
  • In the Establishment Survey (the basis for the number of jobs gained/lost in the economy) — Prior-month revisions — February’s -63,000 was revised to -76,000 (a -13,000 net change), while January’s -22,000 (after last month’s revision) was revised to -76,000 (a -54,000 net change).
  • So the number of estimated number of Americans working at the end of March was 147,000 fewer (-80-13-54) than it was thought to be at the end of February.

I may update for the expected Old Media pile-on later today.

_______________________________________

UPDATE: I understand seasonal adjustments and the importance of reporting on that basis, but sometimes things in the raw data jump out. Here’s an example from Table A-1 supporting today’s Employment Situation Report:

UnempTableA1_0308

So the real increase in the number of people working, according to the Household Survey, was 558,000 jobs.

Now before getting too excited, March 2007′s analogous number was an increase of 844,000, and March 2006′s was 776,000. But I’m still left thinking that this year’s raw increase indicates that the job difficulties of the past few months might have bottomed out.

I think March’s large raw increase in the number of people employed goes a long way towards explaining why the stock market, which is down only 0.15% about 15 minutes after the opening bell, has essentially shrugged off today’s employment report. It would be nice to think the politicians and Old Media might react similarly, but of course they won’t.

UPDATE 2: The raw data in Table B-1 of the Establishment Survey (the basis for reporting the number of jobs gained/lost) is even more stunning:

UnempTableB1_0308

There were 1.1 million more people who were actually working in March than there were in January.

As above, the two-month results in the two previous years were better (2007 — 1.65 million [1.58 after revisions]; 2006 — 1.81 million [1.91 after revisions]). But again, the still-big increase this year supports the idea that the bottom-out may have already happened.

Share

3 Comments

  1. Should be fun to watch the doomsday predictions on the nightly news tonight

    Comment by Ben Keeler — April 4, 2008 @ 12:39 pm

  2. What’s interesting is the Not in Labor Force:
    Feb 79,436 March 79,211 -225,(000)
    So what’s the explanation of the drop here when in fact unemployment increased??? Which is just the opposite of what happened in February’s report where the unemployment rate went down but the Not In Labor force # increased. Are we seeing evidence of some kind of slosh factor between the two due to the way people ask the survey questions or interpret the questions?

    Given the unadjusted Employed # of +558,000, maybe the scenerio occured one of these ways:
    1. 225,000 people who were long term unemployed or retired got jobs, and a net 80,000 people lost jobs.
    2. 225,000 illegals went home and a net 80,000 people lost jobs.
    3. 225,000 represents a combination of illegals going home, seniors coming back to the work force for part time jobs and long term unemployed getting jobs. Since employers fire and lay off workers for various and sundry reasons, a net of 80,000 people lost their jobs in any event even though a sizable number of others gained employment.

    Also given the recent Bear Stearns fiasco, I would think lots of people in the financial services sector would be getting laid off.

    Comment by dscott — April 4, 2008 @ 12:59 pm

  3. dscott, you need to stay consistently with the raw data for a rundown like the one you did. In this case that means you’d stick with the data in Update 1, which does have some anomalous numbers:

    Adult population – up 186
    Labor force – up 632
    Employed – up 558 (as noted)
    Unemployed – up 74
    Not in labor force – down 446
    Currently want a job (but apparently not actively looking, otherwise they’d be considered unemployed) – down 197

    My take (net) – a whole lot of people came out of the woodwork to start looking for work in March (apparently happens every year, at least it has the last two). 88% of them (net) found work.

    Not sure how I get immigration into the narrative, except in the sense that the adult pop is not going up much at all.

    Adult pop:
    - March 2005 — 225,236
    - March 2006 — 227,975; up 2739
    - March 2007 — 231,034; up 3059
    - March 2008 — 232,995; up 1961

    The 12-month change from March 2007-2008 is the lowest since 1995-1996.

    The adult population has been adjusted for changes in population controls every January for the past six consecutive years — something that apparently wasn’t necessary except for one time before that. Every adjustment except one (2004, -348) increased the adult population from the previous December, presumably based on Census input (2000 – +578; 2003 – +256; 2005 – +197; 2006 – +128; 2007 – +552).

    2008′s adjustment was DOWN -540.

    Something’s going on in the population that the Census and BLS folks don’t have a handle on; otherwise, such a large adjustment wouldn’t have been necessary.

    What I also don’t get is that employment in financial services at Table B-1 hasn’t dropped all that much in the past two months, or in the past year (-112,000).

    Comment by TBlumer — April 4, 2008 @ 2:23 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.