The Bureau of Labor Statistics (BLS) produces a number of monthly and quarterly reports that we regularly review. Here are some brief descriptions of their important monthly reports:
Report on the Employment Situation
Arguably among the most important domestic data releases, the monthly Employment Report comprises two major series: the Current Employment Survey (CES), aka the Establishment Survey, or non-farm payroll (NFP); and the Current Population Survey (CPS), aka the Household Survey.
There are major differences between these two series. For the CES, the BLS surveys employing establishments to get a count of the number of workers who received paychecks during the pay-period including the 12th of the month, the “reference week,” or survey week in common parlance. Since the CES surveys businesses, it by definition excludes the self-employed. It also excludes agricultural workers because seasonal variation in that work is so high.
The CES survey covers about 554,000 business establishments, approximately one-third of total non-farm employment. The over-the-month change has to be at least 94,600 to be statistically significant. The headline employment number in the monthly employment report comes from the CES, as do details on hours and earnings, including aggregated hours and payrolls.
The Current Population Survey surveys a much smaller 60,000 households each month and is so noisy that it has to move +-436,000 a month to be statistically significant. The CPS is never benchmarked, so we have no way to check its accuracy, a primary reason the BLS considers the CES more reliable, as do we.
The six unemployment rates come from the CPS. There is a lot of confusion surrounding the unemployment rates, and why they are collected as they are. We’re here to remedy that. The unemployment rate can function both as a measure of wage-rate pressure, and as an indicator of how much pain is out there, and the six different rates are designed to capture both. The U-3, the official unemployment rate, includes only the unemployed who are ready, willing and able to work, and have actively looked for work in the prior four weeks. Since the U-3 functions as a measure of wage pressure it makes sense that it includes only those whoa re actively looking for work: it takes a while to get discouraged workers back into the force.
But it would not make sense to use the narrow U-3 as a measure of the health of the labor force: it’s obviously not good to have a large number of workers sitting out because they don’t think there are any jobs available to them, or working part-time when they would rather have full-time jobs, which is why we have the broader measures up the the U-6, which includes the officially unemployed, plus those marginally attached to the force, and unwilling part-timers.
Demographic details, and information on multiple job-holders and part-time workers also come from the CPS.
Job Openings and Labor Turnover Survey (JOLTS)
JOLTS, released with a one-month delay, details job-market churn. Every month there are millions of separations, both voluntary and involuntary, millions of hires, and millions of jobs up for grabs.
There are two important details in the JOLTS series. The Job Openings Rate, the number of job openings expressed as a percentage of total employment plus job openings, is plotted against the CPS’s unemployment to produce the Beveridge Curve, a measure of how well the labor market is matching the unemployed with available jobs. The curve gets its name from UK economist William Beveridge, and is fodder for both heated debate and speculation about the relative importance of structural and cyclical forces on the unemployment rate, including the labor mismatch theory popular in recent years.
The other important component of JOLTS is the quit rate, which measures the percentage of voluntary separations and is a measure of worker confidence that the Federal Open Market Committee (FOMC) tracks carefully. As you can see from this graph, like the Beveridge Curve, the quit rate is currently recovering slowly, but remains well below its normal range.
And here is some information about their quarterly reports:
Business Employment Dynamics (BED)
Job creation depends heavily on the rate of business start-ups. Although mature firms are responsible for most employment, their staffing levels tend to be stable; the bulk of new jobs are generated by young firms. The quarterly BED reports provide information on jobs created and destroyed at expanding and contracting firms, and jobs created or lost in newly born or dying firms. Also detailed are rates of establishment births and deaths. Establishment births fell off the chart in the Great Recession, and have managed only a staggering recovery since, a primary reason job creation has been so weak. The following graph shows our progress out the depths of the recession, and the distance we have yet to travel.
Quarterly Census of Employment and Wages
The Quarterly Census of Employment and Wages, the QCEW, is the last word in employment statistics. Taken from the reports employers must file with unemployment insurance payments, the QCEW covers close to the entire non-farm employment universe, and is the basis of the annual benchmarking process for the mighty Non-farm Payroll. It also provides detail by sector and county for the entire country, which we employ in our regional reports.