Your Ultimate Cheat Sheet To Unemployment Numbers

Today the Bureau of Labor Statistics releases their newest set of data for the unemployment rate and related statistics in the US. Understanding the unemployment rate is crucial to a sound understanding of our economy.

Who Does the Government Consider “Unemployed”?

According to the government’s definition, the “unemployed” are people without jobs who are actively seeking work. It is important to note these people are seeking work because there are also unemployed people who are not looking for a new job for whatever reason.

The government has a hierarchy of unemployment categories:

U1 unemployment: Those who have been out of work for 15 weeks or more;

U2 unemployment: Those who have lost jobs or have only been able to find temporary positions;

U3 unemployment: Those without jobs that are available for work and actively seeking it. This is the official definition of unemployment — the one we read in the headlines;

U4 unemployment: U3 + “discouraged workers,” or those who have looked for jobs but feel they cannot find employment because of economic conditions;

U5 unemployment: U4 + “marginally attached workers,”  or those who would like to find jobs but have not looked recently;

U6 unemployment: U5 + part-time workers who cannot find full-time jobs for economic reasons. This is the widest definition of unemployment and gives the most accurate picture of the total number of under-employed people.

Who Counts the Unemployed People?

The first Friday of every month, the Bureau of Labor Statistics (AKA, the “BLS” in Wall Street talk) delivers a new report on the state of unemployment. The report is a constitution of labor statistics that covers the status of the labor force, employment, unemployment, persons not in the labor force, hours of work, earnings, and other demographic and labor force characteristics.

The information also takes into account all adjustments in unemployment by population change, seasonality, and renewed claims for benefits.

How is Unemployment Measured?

The unemployment rate is created from data collected in the Current Population Survey (CPS). This is a monthly survey of over 60,000 households (usually around 110,000 people) and is then weighted according to the demographics of the households.

Surveyors then determine the rate of employment by asking a series of questions on whether citizens have a job or not, whether they want a job and are available to work, and what they have done to look for work in the preceding 4 weeks.

People counted as unemployed are:

  • Not Self-Employed;
  • 16 Years of Age Or Older;
  • Without a job;
  • Currently available for work; and,
  • Have actively looked for work in the prior four weeks.

Moreover, unpaid workers putting in 15 hours or more of work in a family-owned enterprise (such as farmers) are considered part of the employed.

After these people are counted, the unemployment number is then taken as a percentage of the overall labor force. The “labor force” is defined as the number of people who are actively working jobs plus the number of people who are available for work.

Interestingly, the government drops people out of the “labor force” category for a variety of reasons including returning to school or when a person simply decides to stop looking for a job (i.e., they are unavailable for work). This adds a tricky component to the unemployment equation because U3 headline unemployment can actually go down even if the number of unemployed persons rises. This occurs when the “labor force” shrinks as previously included person are no longer considered “jobless” to be included in the final unemployment numbers.

Should We Trust The Numbers?

Before praising the BLS for stunning accuracy of America’s labor force, remember: the unemployment rate is known as a lagging indicator. This means the numbers are based on a previous period of time rather than the present. Therefore, things will have changed in the economy by the time the unemployment data is released.

The problem with lagging indicators is that they always look good at the beginning of a recession and bad in the middle of a recovery. Thus, unemployment numbers are not a direct representative of the health of the economy, but very close estimates. They are a very prominent indicator of the economy’s health because employment stimulates economic activity.

What are Seasonal Adjustments?

The BLS uses a statistic called Seasonal Adjustments to calculate and smooth out the numbers for the unemployment rate in the face of economic fluctuations. As new jobs become available for different seasons of the year, the BLS also places additional seasonally-adjusted statistics into their original figures to come up with a number that more accurately represents the state of employment in relative terms.

For example, jobs often come and go in the summer as well as during and around holidays. So, the BLS uses this additional statistic to assess the impact of seasonal fluctuations and the workforce. These seasonal adjustments allow the BLS to analyze the cyclical and underlying trends of unemployment without constraint of seasonal data.
However, seasonally adjusted information is used only for comparing month to month trends. The data used to calculate annual average estimates are taken from the data series which has not been seasonally adjusted.

What About Unemployment Insurance Benefits?

The BLS is not affiliated with any unemployment insurance programs. Those are conducted by the Employment and Training Administration (a division of the Department of Labor). These numbers are delivered in categories of either initial claims for workers beginning a period of unemployment or continued claims if the unemployed are then receiving benefits.

While some people think that the unemployment numbers are taken from the number of claims filed, this is a false notion. When benefits expire, some people still remain jobless and are still counted as unemployed.

While not directly related to the national unemployment rate, unemployment insurance numbers give better estimates for local and state numbers of unemployment. Numbers regarding unemployment insurance are released every week here. Also, Wall Street analysts and investors use initial claims data as a leading indicator of how the BLS unemployment data may change.

What is the Birth-Death Adjustment?

One of the more questionable aspects of BLS numbers is the Birth-Death Adjustment. The B-D Adjustment (as Wall Streeters call it) is an artificial adjustment regarding the number of employers who have recently formed new businesses, but who have not yet been added to the Unemployment Insurance tax files. Since new businesses form every month, failure to include these businesses within the model would overestimate the number of unemployed, so the BLS uses the the Birth-Death Adjustment model to estimate this part of the population. The BLS takes the number of non-farm payroll employees, adds the Birth-Death Adjustment, and then seasonally adjusts the numbers.

As many small firms continually find themselves opening for business and closing, the BLS finds a way for the two statistics to offset each other: a business birth accounts for a business death. However, firms that die usually do not report that they have gone out of business; instead, they no longer report anything.  Follow-ups with unrespondent businesses are then made to determine whether a company still remains in business.

Unfortunately, the Birth-Death Adjustment also fails to offer accurate numbers in the booms and busts of the business cycle. The Birth-Death Adjustment will catch more job creation in the beginning of the cycle, but at the end of the cycle underestimates the number of job losses in the economy. This skews the model and inflates the numbers.


Unemployment numbers are important gauges of economic health. In order for us to improve foresight as investors, we must keep an eye on these very useful tools.