sheep giving birthEvery 6 months the BLS performs an exercise in hindsight. It looks over its surveys from the prior half-year and adjusts its extrapolation of the total number of employed. Then it goes back to making birth-death adjustments based on its estimate of how many people were employed by new businesses and how many were laid off by businesses that had failed in the prior month. In January of this year the BLS removed nearly half a million jobs that it had added in 2nd half of 2009. It decided that its cumulative birth-death adjustments for August through December overestimated actual employment during the period by 427,000. Starting in February the BLS began adding jobs again; its birth-death adjustments for the first half-year of 2010 totaled 728,000. June's gain in private jobs - 83,000 - depended on one of those monthly adjustments; without it, the gain in June would have been a loss of 64,000 jobs. The unemployment rate for June improved through a similar sleight of hand. In May the BLS survey showed 14,973,000 Unemployed out of a total Labor Force of 154,393.000 Labor Force; that produced the UE Rate of 9.7% For June the numbers were 14,623,000 Unemployed out of a total Labor Force of 153,741,000 (UE Rate = 9.5%). But for the magical disappearance of roughly 650,000 people (where, one wonders, did they go?) the Unemployment Rate would have remained the same as it was in May.

Rudy Hauser comments:

The payroll series after those adjustments are made is a very accurate survey as it is based on payroll reports which all employers, even if you only employ one person, have to file at least quarterly. That full information is not available on a timely basis but is reflected when the numbers are revised to incorporate that information. Those reports are used to determine what employers have to pay for unemployment insurance and as such it is illegal not to file them as required. The immediate numbers are based on voluntary reporting to the BLS by firms. New firms are not likely to be included and it is not always known why a firm stops reporting, hence the need for the initial birth/death of firm adjustment.

The household survey on which the unemployment rate is based is just that- a survey of households. A decline in the labor force could just reflect the usual variation for any survey, even a large one such as the household survey or it could reflect an actual decline of people actively looking for work or some combination of the two.





Speak your mind


Resources & Links