Unemployment numbers still point to partial recovery

By Edward Harrison On Oct 2, 2009

In July, I blogged on an interesting take on how employment affects equity returns during cyclical recoveries by Van Hoisington and Lacy Hunt. Their thesis was that a recovery in which employment lags the overall upturn significantly is bearish for stocks. Since then, employment has indeed lagged other economic indicators. Witness the most recent employment situation summary released earlier today.

Nonfarm payroll employment continued to decline in September (-263,000), and the unemployment rate (9.8 percent) continued to trend up, the U.S. Bureau of Labor Statistics reported today. The largest job losses were in construction, manufacturing, retail trade, and government.

Household Survey Data

Since the start of the recession in December 2007, the number of unemployed persons has increased by 7.6 million to 15.1 million, and the unemployment rate has doubled to 9.8 percent…

The civilian labor force participation rate declined by 0.3 percentage point in September to 65.2 percent. The employment-population ratio, at 58.8 percent, also declined over the month and has decreased by 3.9 percentage points since the recession began in December 2007.

Establishment Survey Data

Total nonfarm payroll employment declined by 263,000 in September. From May through September, job losses averaged 307,000 per month, compared with losses averaging 645,000 per month from November 2008 to April. Since the start of the recession in December 2007, payroll employment has fallen by 7.2 million.

In September, construction employment declined by 64,000. Monthly job losses averaged 66,000 from May through September, compared with an average of 117,000 per month from November to April. September job cuts were concentrated in the industry’s nonresidential components (-39,000) and in heavy construction (-12,000). Since December 2007, employment in construction has fallen by 1.5 million.

Employment in manufacturing fell by 51,000 in September. Over the past 3 months, job losses have averaged 53,000 per month, compared with an average monthly loss of 161,000 from October to June. Employment in manufacturing has contracted by 2.1 million since the onset of the recession.

In the service-providing sector, the number of jobs in retail trade fell by 39,000 in September. From April through September, retail employment has fallen by an average of 29,000 per month, compared with an average monthly loss of 68,000 for the prior 6-month period.

Government employment was down by 53,000 in September, with the largest decline occurring in the non-education component of local government (-24,000)…

In September, the average workweek for production and nonsupervisory workers on private nonfarm payrolls edged down by 0.1 hour to 33.0 hours. Both the manufacturing workweek and factory overtime decreased by 0.1 hour over the month, to 39.8 and 2.8 hours, respectively. (See table B-2.)

In September, average hourly earnings of production and nonsupervisory workers on private nonfarm payrolls edged up by 1 cent, or 0.1 percent, to $18.67. Over the past 12 months, average hourly earnings have risen by 2.5 percent, while average weekly earnings have risen by only 0.7 percent due to declines in the average workweek. (See table B-3.)

The change in total nonfarm payroll employment for July was revised from -276,000 to -304,000, and the change for August was revised from -216,000 to -201,000.

A few points here:

Losing over 250,000 jobs per month nearly two years into recession is an indication of a still weak employment market. This is 400,000 jobs per month below where we want to be.
Why is the labor force participation rate still falling? This is a sign of a deteriorating, not improving labor market.
Manufacturing is still shedding workers even though industrial production is rising. That demonstrates a weakness whose source is record low capacity utilization.
Local and state governments are cutting workforce and countering the stimulus provided by the Federal Government as I indicated in January they would.
The workweek is a record low and this is crimping earnings power.

Conclusion: the labor market is still weak, weaker than it should be at this point in a cyclical recovery. Unless this changes in the fall and winter, a double dip recession is going to be more likely. While the preceding points stress the negative, I should point out that my baseline view is for job losses to continue to diminish, albeit at a slow pace. I would anticipate job gains to appear by the end of the year or early in 2010.

That gets me back to Hunt and Hoisington and partial recovery. Even if we see job gains by Q1 2010, this will be a full 6 months after the manufacturing sector turned up. This must limit consumption because spending can only increase through higher employment and income or increased debt and leverage. As most of the cost-cutting and productivity gains inherent in those cuts is now behind us, the heavy lifting begins. Earnings growth is likely to be weak in this environment.

How a fully priced equity and corporate bond market continues to rally in the face of these factors is beyond me. I see government bonds as a better bet than either corporates or equities for the medium-term.

Update: I failed to mention the rather large (over 800,000 jobs) benchmark revision of prior unemployment data. It’s this sort of thing which makes people not trust the numbers. But, revisions are always necessary if you are going to do month-to-month measurements in an economy as large as the United States.

Here’s what the BLS said:

Preliminary Estimates of Benchmark Revisions to the Establishment Survey

In accordance with usual practice, the U.S. Bureau of Labor Statistics is announcing its preliminary estimates of the upcoming annual benchmark revision to the establishment survey employment series. The final benchmark revision will be issued on February 5, 2010, with the publication of the January 2010 Employment Situation news release.

Each year, the Current Employment Statistics (CES) survey employment estimates are benchmarked to comprehensive counts of employment for the month of March. These counts are derived from state unemployment insurance tax records that nearly all employers are required to file. For national CES employment series, the annual benchmark revisions over the last 10 years have averaged plus or minus two-tenths of one percent of total nonfarm employment. The preliminary estimate of the benchmark revision indicates a downward adjustment to March 2009 total nonfarm employment of 824,000 (0.6 percent).