Nobody From Nowhere (@i8dc)

Occasional Common Sense

Checking the Fact Checker – Job Creation Edition

with one comment

The Post Fact Checker Gless Kessler writes yesterday: “By the standard definition of job creation during a presidency, [Obama] is on track to be the first president to have negative job growth in the modern era.”

This is all arbitrary and stupid, since boom/bust periods and presidential administrations don’t magically align their cycles like Smith College roommates.  But two questions leap to mind: (1) what is the “standard definition of job creation during a presidency” and is it fair, and (2) is this true?

The “standard definition” cited by Kessler is a link to this WSJ 2009 analysis, showing that job creation under George W. Bush was the worst under any president since WWII.  I have to question whether this is a “standard definition,” rather than this one journalist’s assessment, since the Journal piece doesn’t claim to be using a standard. And since it was published hours after the first preliminary December 2008 unemployment figures were released by BLS, I think it’s more likely that the author was simply using the latest data and making all prior presidencies end in December to make an apples-to-apples comparison.

This wasn’t a big deal, since there is so rarely any very significant change in employment from month to month, and never has there been a very large change in the month a new president was inaugurated.

Until, of course, January 2009.

In the four transitions before 2001, job growth from December-January ranged between 0.24% and 0.3%.  In 2001, it was close to zero.  But during January 2009, the U.S. economy shed more jobs – 820,000– than in any month since WWII, and 0.61% of all jobs.  Additionally, the preliminary numbers used in the Journal piece were revised down several times as the severity of the nosedive became clear, so that GWB’s record is significantly worse than first reported.

So what should be the “standard” baseline for judging job creation?  Certainly it can’t be the December number.  Not only has a new president not had time to implement policy, he hasn’t even taken office; he won’t until more than halfway through January.  The very earliest you could make a baseline would be the January data, though there are compelling reasons why it should actually be several months later – employment as a lagging indicator and the time it takes to actually implement policy come to mind.  However, seems to me that picking a baseline later than January, which appropriate, would be filled with subjective choices and therefore impossible.  January it is, even if it screws Obama (raise your hand if you think Obama’s policies were responsible for 726,000 job losses in February 2009).

So the “standard definition” Kessler cites is fundamentally incorrect because of the dates used.

It’s also fundamentally flawed because it doesn’t take into account population growth. For example, during the Eisenhower administration jobs increased by 7.1%.  But the civilian non-institutional population grew by 10.8%.  Assigning credit for job creation that merely keeps up with population growth isn’t fair; job growth that mirrors population growth ain’t growth at all – it’s maintaining average living standards, nothing more.

A fair assessment of job creation has to incorporate population growth.

A much better assessment of job creation is the change in employment as a share of the population, the employment ratio.  Using that measure, half of the previous 10 presidents saw the employment situation decline during their tenures.  This should not come as a surprise; hearing that Obama would be the first president in history with bad job creation numbers should raise a red flag about the qualify of the analysis.

Number crunching time.  Here are the data for Eisenhower on, starting with the data for the month each took office:

President

UNRATE

CE16OV

CNP16OV

CLF16OV

CIVPART

EMRATIO

Eisenhower

2.9%

61,600

106,594

63,439

59.5%

57.8%

Kennedy

6.6%

65,776

118,155

70,447

59.6%

55.7%

Johnson

5.7%

68,267

123,192

72,418

58.8%

55.4%

Nixon

3.4%

76,805

133,324

79,523

59.6%

57.6%

Ford

5.5%

87,037

150,493

92,059

61.2%

57.8%

Carter

7.5%

89,928

157,688

97,208

61.6%

57.0%

Reagan

7.5%

99,955

169,104

108,026

63.9%

59.1%

Bush

5.4%

116,708

185,644

123,390

66.5%

62.9%

Clinton

7.3%

119,075

193,962

128,400

66.2%

61.4%

Bush

4.2%

137,778

213,888

143,800

67.2%

64.4%

Obama

7.8%

142,201

234,739

154,185

65.7%

60.6%

July 2011

9.1%

139,296

239,671

153,228

63.9%

58.1%

UNRATE: unemployment rate CLF16OV: labor force
CE16OV: number of employed people, 16 years+ CIVPART: participation rate (16+ labor force/16+ population)
CNP16OV: non-institutional population, 16 years+ EMRATIO: employees/16+ population

Now here are some metrics derived from the data:

President

ΔUNRATE

ΔCE16OV

ΔCNP16OV

ΔCLF16OV

ΔCIVPART

ΔEMRATIO

ΔEMRATIO/year

Eisenhower

 3.7%

  6.8%

10.8%

11.0%

 0.1%

-2.1%

-0.46%

Kennedy

-0.9%

  3.8%

 4.3%

  2.8%

-0.8%

-0.3%

-0.22%

Johnson

-2.3%

12.5%

 8.2%

  9.8%

 0.9%

 2.2%

 0.82%

Nixon

 2.1%

13.3%

12.9%

15.8%

 1.5%

 0.2%

 0.08%

Ford

 2.0%

  3.3%

 4.8%

  5.6%

 0.5%

-0.8%

-0.57%

Carter

 0.0%

11.2%

 7.2%

11.1%

 2.2%

 2.1%

 0.90%

Reagan

-2.1%

16.8%

 9.8%

14.2%

 2.6%

 3.8%

 0.77%

Bush

 1.9%

  2.0%

 4.5%

  4.1%

-0.3%

-1.5%

-0.59%

Clinton

-3.1%

15.7%

10.3%

12.0%

 1.0%

 3.0%

 0.60%

Bush

 3.6%

  3.2%

 9.7%

  7.2%

-1.5%

-3.8%

-0.76%

Obama

 1.3%

 -2.0%

 2.1%

 -0.6%

-1.8%

-2.5%

-1.89%

While we hear most often about the first column, the change in the unemployment rate, the most important data here is in the last column. This is annual change in employment as a percentage of the working age population (adjusted for partial years where necessary). Clearly, Obama’s record is the worst on the list, though as I said the data are arbitrary because of the lags in policy effects and the nature of the business cycle (and incidentally, if just February 2009 is assigned to Bush instead of Obama, the numbers change significantly).

But the assertion that Obama’s on track to be the first president with negative job creation is specious, as it doesn’t take into account changes in the workforce.  When those are included, 5 of the prior 10 presidents have seen the employment situation decline.

—–

Update, August 22:

I should also have noted that in that Kessler’s statement that ” [Obama] is on track to be the first president to have negative job growth in the modern era” isn’t true.  Comparing the presidents by change in payroll employment between their first month in office and their 31st, which is the latest Obama administration data, there were two prior presidents with job losses at this point in their terms: George W. Bush and Ronald Reagan.  The EMRATIO also declined during the first 31 months of the administrations of Eisenhower, Kennedy, Nixon, and George H.W. Bush, and through the 27-month term of Ford.

Advertisements

Written by David Clayton

August 20, 2011 at 12:21 pm

Posted in Debunkery, Punditry

One Response

Subscribe to comments with RSS.

  1. Very interesting. This is what you get when all the numbers related to the issue and all the “assumptions” are checked and balanced. One issue mentioned is the lag time from policy development to implementing and then the time it takes for the policy to work its way into the system. Too many people today want policy today, results tomorrow and then either beat their chest if the numbers are in their favor or the tend to ignore the issue because nothing has registered on/in any form of reliable measure. Fact Checker is a great resource and this makes it even better that it allows others to provide input. Thanks.

    barkingdog48

    September 5, 2011 at 6:34 pm


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: