Household vs. Establishment
Our friends at the Bureau of Labor Statistics (motto: "Hold still while I count") conduct two employment surveys. The headline number -- the one flashed on the screen that produces much ululating and hand-waving among those who are still trading bond futures via open outcry -- is derived from a mail survey of 400,000 business establishments. A separate telephone survey of 60,000 households is conducted as well. The two surveys can differ markedly, and as is usually the case, we can learn something from the differences.
If we display the establishment survey's total as a percentage of the generally larger household survey, a pattern emerges. Periods of strong labor demand, such as during the Korean War in the early 1950s, the Vietnam War in the late 1960s and the late 1990s technology bubble, see an increase in the ratio of establishment to household labor totals. The percentage falls once labor demand falls.
This suggests that incremental labor demand comes not from sources such as self-employment and entrepreneurial activity, both of which are represented in the household survey, but rather from business establishments. When hiring gets large, the large get hiring: Business establishments can raise wage levels and bid people away from self-employment and the like.
|Will the Trend Line Be Unbroken?