I thought I would provide a bit of a primer on the economic data that will be released tomorrow for non-farm payrolls and the unemployment rate in an attempt to reduce the confusion many on this blog seem to have with the actual data.
First, let's start off by linking to the Bureau of Labor Statistics (bls) page where the report will be released tomorrow at 8:30am (est). The employment situation summary report will detail the non-farm payroll gain/decline and the unemployment rate in the first paragraph. It is important to note that the non-farm payroll number and the unemployment rate are not from the same survey and thus it is possible to have one number go up and the other down.
The next important place to look in the employment situation summary is the information provided under the "household survey data" heading. This area will tell us the total number of unemployed (as counted under U-3, which excludes the discouraged, marginally attached, and employed part-time for economic reasons), which is the numerator in the unemployment rate equation. The survey will also tell us what the labor force participation rate is, which is important, as this is the denominator (well the total labor force is, the participation rate is simply a percentage out of the total possible population ages 15-65). These numbers are important to note because the unemployment rate can go down (like it did last month) simply from the labor force shrinking instead of actual job growth. The labor force shrinks when people take themselves out of the labor market (which is different from being a discouraged worker in U-6) by claiming they are now a student, retired, housewife, etc., which means they are no longer considered part of the labor force (even if the reason they left was discouragement or long-term unemployment). The final number to watch from the "household survey data" section is the employment-population ration, as this gives a better impression of the overall labor market by showing what percentage of the total working age population that is employed (ages 15-64). This ratio was at 59.4 last month (at its lowest level since the mid-80s). Using either the labor participation rate or employment-population rate for long-term comparisons is difficult, as the movement of women into the workforce that began in earnest in the 60's skews much of the pre-90's data.
Next, we have the "establishment survey data" section, which is where the job loss/gain number is reported and broken down. The key points to look at in this section are towards the end and reference hours worked. The workweek of non-supervisory and production workers is closely watched, as it is believed this number needs to move up before jobs will be created (it was at 33.1 hours last month) and the manufacturing workweek is important because it is a factor in the leading economic indicators that are released each month (it was at 39.8 hours last month).
Finally, we have the alternative measures of employment (ie where U-6 is). These measures add in additional workers that are otherwise excluded from the U-3 (the one reported everywhere and referenced in the official release) and can be useful in determining what the unemployment rate is when we start adding back in various classes of people. I would personally argue that while U-6 is a very good statistic to watch, the employment-population ration simply provides a much better view into the percentage of the working age population that is employed.
One final note. The birth/death adjustment is very controversial and is included in the establishment survey's non-farm payroll numbers. It is a statistical method that add or subtracts (but has so far on a yearly basis only added) jobs based on a formula devised by the BLS (under direction from the Bush administration) to more accurately estimate jobs created or lost from the birth and death of businesses. You can see the numbers and their breakdowns by sector here.
Update: The unemployment rate has nothing to do with unemployment claims or unemployment insurance.