Nevertheless, analysts appeared perplexed when attempting to explain the sudden drop in unemployment – given how slowly the economy has grown and given that only 114,000 new jobs were created.
Rather than piling on the criticism of Welch, it is more productive to use this as a teaching moment. The decline in the unemployment rate seemed perplexing because observers failed to recognize the following:
- Recently, the economy has grown much faster than most analysts realize.
- The unemployment rate and the number of new jobs created are generated from two different surveys, the Household Survey, and the Payroll Survey. The former is almost never revised; the latter is almost always revised.
- The estimate of new jobs created (for September it was 114,000) is not final until two or more months after it is first released. In fact, the August and July estimates were recently adjusted upward by 48% and 28% respectively.
How the Numbers are Collected.
On the first Friday of each month at precisely 8:30 am, the Labor Department releases the latest Monthly Employment Situation Report (commonly referred to as the Jobs Report). It measures the unemployment rate and number of new jobs created for the previous month. Thus, the report released on Friday November 5, 2012 measured the employment situation for October.
The November Report showed the unemployment rate declined from 8.1% in August to 7.8% in September. This drop caught Republicans and Democrats off guard. Conservatives charged conspiracy! Republican Representative Allen West of Florida even went so far as to assert the data was so manipulated that it was reminiscent of an Orwellian plot. Democrats fired back quickly, but were at a loss to provide a rational explanation.
The conspiracy theories have quieted, and Welch has suggested his tweeted comment should have been in the form of a question. Therefore, we will answer Welch’s question by explaining why the unemployment report seemed so surprising.
There are two reasons why observers scratched their heads when the unemployment rate dropped from 8.1% to 7.8%.
- Reason #1: The rate has remained above 8% since the recession started, and the economy is growing so slowly it does not appear capable of creating enough jobs to reduce the unemployment rate so significantly.
- Reason #2: The employment report showed only 114,000 new jobs were created in September – a number that appears inconsistent with a drop in the unemployment rate from 8.1% to 7.8%. The report also showed that 873,000 more people were employed during September than in August, and 456,000 fewer people were unemployed.
The first point: The “unemployment rate” comes from a different source of data than does the number of new jobs created. The rate is based on information collected by census workers in the field who talk to a random sample of 60,000 households monthly. This is called the “Household Survey” and its results are rarely revised in subsequent months.
The number of new jobs created, the “Payroll Survey,” comes from a monthly survey of 140,000 businesses and government agencies. The time lag within which these businesses submit their data means current estimates of new jobs created is almost always revised in subsequent months.
This month, the August payroll survey was revised upward by 48%, from 96,000 to 142,000. Furthermore, the July survey was revised up by 28%, from 141,000 to 181,000.
The problem is that analysts look at the number of new jobs created as if it is gospel; when, in fact, it is one of the most variable numbers released monthly by DOL.
The second point: It is important that analyst look more deeply at what is happening in the economy. If they did so, they would notice many positive developments occurring. For example, last month’s developments that signaled a stronger economic recovery were as follows. New claims for unemployment compensation decreased, oil prices declined and the following indicators increased: labor productivity, personal spending, retail sales, auto and truck sales, housing starts, consumer confidence, manufacturing purchases and service purchases.
Given all of the positive developments, it is easy to see why the drop in unemployment rate occurred and more importantly, why it should not be a mystery.
Last modified: June 20, 2017