Ed Dolan's Econ Blog

August Job Numbers are a Disappointment for Democrats

The August job numbers released today by the BLS will come as a disappointment, especially to Democrats who were hoping for a strong headline number during their convention week. The U.S. economy created just 96,000 new payroll jobs last month. An increase of 103,000 private-sector jobs was partly offset by a loss of 7,000 government jobs. Private-sector goods-producing jobs actually fell by 16,000. To make matters worse, the June jobs number was revised downward by 19,000 to a very weak 45,000, and the July jobs gain was revised downward by 22,000 to a more modest 141,000.

The unemployment rate decreased from 8.3 percent to 8.1 percent, but there was not as much good news in that number as it might seem. For one thing, the unemployment rate had already hit 8.1 percent once before, in April, so the August number did not break new ground. Also, much of the drop in the unemployment rate was due to a decrease in the size of the labor force. The number of employed persons, according to the household survey on which the unemployment rate is based, actually fell by 119,000 in August. The unemployment rate is based on a survey of household that is entirely separate from the establishment survey on which the payroll jobs report is based. The two surveys can produce numbers on job loss or gain that differ substantially, partly for methodological reasons and partly because the household survey, unlike the payroll survey, includes farm workers and self-employed persons.

August also saw a decrease in the broad unemployment measure, U-6, which fell from 15 percent to 14.7 percent. U-6 includes discouraged workers, other marginally attached workers, and involuntary part-time workers as well as those that meet the official definition of unemployed. The decrease in U-6 was welcome, of course, but it did not bring the broad unemployment rate back to the 14.5 percent it registered in March and April.

As always, you can find at least one good data point if you dig for it. Probably the most encouraging number this month was a decrease to 40 percent in the percentage of the unemployed who were out work for 27 weeks or more. That is the lowest number since November 2011. However, even that good news was offset by an increase in both the mean and median duration of unemployment.

All in all, it makes a hard set of data for Democratic spinmasters to work with. They are being forced to fall back on the argument that the previous administration left the economy in such a mess that even four years of heroic efforts have not been enough to get things back to normal. That argument may be factually true, but it is not ideally suited to crafting snappy TV spots.

If Republicans only had a convincing economic strategy of their own, they ought to be able to win the election in a landslide. Instead, all they seem to have on offer is more tax cuts for ‘job creators’ (unconvincing when after-tax profits are already running at record highs) and fiscal austerity (not exactly a resounding success when tried in Europe). Each party seems to be left hoping that voters will look mainly at the weakness of the other party’s claims to be the better manager of the economy.

Follow this link to view or download a slideshow with charts of all the latest jobs data.


6 Responses to “August Job Numbers are a Disappointment for Democrats”

ThomasGrennesSeptember 7th, 2012 at 4:40 pm

If voters are sensitive to month-to-month unemployment data, and candidates know this, government officials have little incentive to act today to deal with the fiscal cliff approaching at the end of 2012. They have even weaker incentives to act now to deal with the long-term debt problem and the viability of social security, medicare, and medicaid. What can one expect from Washington besides election promises and procrastination?

rjsigmundSeptember 7th, 2012 at 5:44 pm

if you took a poll, what percentage of americans do you think could tell you the unemployment rate, much less the number of jobs added in august?

Ed Dolan EdDolanSeptember 7th, 2012 at 11:31 pm

You raise an interesting point. I was just listening to an item on NPR that said that the state-by-state unemployment rate had more influence on presidential voting than the national rate. That is odd if you think of it. There is little enough that the president can do about the national unemployment rate, and for all practical purposes, nothing that he or she can do about the state-by-state rates.

What I make of it is that the actual numbers are not what matters. I suppose in a national quiz you would not get a high percentage of right answers. What probably matters is a feeling people get from their personal experience and that of their friends, plus the "buzz" they get from the media (without remembering the numbers) of whether things are getting better or worse.

Ben LeetSeptember 8th, 2012 at 2:26 pm

The employment to population ratio dropped to 58.3%, while in Jan. 2000 it was 64.6% its record high. The difference, 6.3% of the working age population, is 15.3 million workers who were employed back in 2000 who are not employed today, transferring the 6.3% to today's labor market. (One can google "data bls, employment to population" to see their data and chart) In Nov. 2010 the rate was 58.2%, and then not since 1983, 29 years back, had it been that low. For 20 years, 1988 to 2008 it had been over 62%. A look at the National Jobs for All Coalition,, shows 27.5 million under- and unemployed, or 17.0%. "In addition, millions more were working full-time, year-round, yet earned less than the official poverty level for a family of four. In 2010, the latest year available, that number was 16.8 million, 17.0 percent of full-time, full-year workers (estimated from Current Population Survey, Bur. of the Census, 9/2011)." And there are 7 under- or un-employed workers for every job opening.
17% under-and-unemployed
11% working full-time full-year for below poverty rate
28% total,
something like 10 million college graduates in jobs that require only a high school education
7 workers for every job opening
I agree with FDR that full employment should be a government goal and a personal right, part of a new Bill of Rights. If the jobs were created, the employees are there to fill them. This is very controversial, but we do have child labor laws, minimum wage, and the EITC which approaches $60 billion a year now. We have a demand problem in this economy, not a shortage of supply of labor, capital or industrial capacity. All of those we have in excess. Dan Alpert says this often at EconoMonitor. A demand problem is also a problem of grossly unequal distribution of income and wealth. There are not enough customers, just like the Great Depression, weakness and fragility reigns. This week Bill Clinton reminded people in his speech to the Democratic Convention that Obama's September 2011 jobs bill of $447 billion would have added another 1.6 million employed workers. But there are other plans out there to add 12 million and they are not more expensive than $500 billion. Half of the U.S. working labor force, 75 million workers, take in as wage income only 7% of the total personal income of the country. That's the inequality I mentioned. The economy grew by 68% per capita over 30 years and workers annual private sector income at the median increased by 5.7% according to the Economic Policy Institute's president Lawrence Mishel. My blog is

rjsigmundSeptember 8th, 2012 at 6:25 pm

just one more point, re "private-sector goods-producing jobs actually fell by 16,000"

the loss 15,000 manufacturing jobs has been widely cited, but that was largely due to a seasonal adjustment which reversed the manufacturing "jobs gained" when auto industry July retooling shutdowns were less than normal, which inflated July's manufacturing jobs by 25,000, SA…