What is the Real Unemployment Rate?

What is the Real Unemployment Rate?

25%? Much Higher than 25% If you think the real unemployment rate is even close to being as high as 25%, please read this because the situation isn’t anywhere close to being as bad as you think.

We hear a lot about how the “real” unemployment rate is much higher than the “official” unemployment rate. These comments are often accompanied by a specific number that is supposed to accurately reflect the “real” unemployment rate. For purposes of understanding the state of the economy, the concept that the official unemployment rate understates reality is valid. However, coming up with a specific number that reflects a “real” unemployment rate is not as simple as it sounds. There are many different equally logical approaches to defining a “real” unemployment rate that would yield different results.

Let’s take a look at who is and isn’t included in the “official” unemployment statistics. Let’s begin with the formula used:

Unemployment rate = number of unemployed persons divided by the total civilian labor force

The unemployment rate is calculated from data provided by a monthly household survey conducted for the Bureau of Labor Statistics (BLS). It is published, along with other employment data, in a monthly news release.

The total civilian labor force is defined by BLS as those persons classified as employed plus those persons classified as unemployed.

Employed persons: Quoting from the BLS Technical Note accompanying the monthly news release:

People are classified as employed if they did any work at all as paid employees during the reference week; worked in their own business, profession, or on their own farm; or worked without pay at least 15 hours in a family business or farm. People are also counted as employed if they were temporarily absent from their jobs because of illness, bad weather, vacation, labor-management disputes, or personal reasons.

Unemployed persons: Again quoting from the BLS Technical Note:

People are classified as unemployed if they meet all of the following criteria: they had no employment during the reference week; they were available for work at that time; and they made specific efforts to find employment sometime during the 4-week period ending with the reference week. Persons laid off from a job and expecting recall need not be looking for work to be counted as unemployed. The unemployment data derived from the household survey in no way depend upon the eligibility for or receipt of unemployment insurance benefits.

For reference and as an example for what follows, the unemployment rate for March 2014 was 6.7%, calculated from approximately 10.5 million unemployed persons divided by a total civilian labor force of a little over 156 million.

People who don’t meet the above criteria for either employed persons or unemployed persons are not considered to be part of the labor force, and are excluded from the calculations for the unemployment rate. The reason is that the unemployment rate is designed to measure how many of the people who want jobs are able to get one, taking into consideration that those who are not actively looking for work are not likely to get a job. With this concept in mind, those who are not working, and not actively looking for work, are categorized as not in the labor force. They are not part of the unemployment rate calculation, but that doesn’t mean that they are not counted.

The question becomes, if you conclude that many of these people don’t have jobs because of economic conditions, and therefore their non-employed status reflects a hidden unemployment, then which ones should be counted in order to arrive at a “real” unemployment rate? The answer may not be as simple as it first appears.

Suppose, for example, that a person meets all of the criteria for being classified as unemployed, except that they have not looked for work in the past 4 weeks. They want to work, they are available for work, they have looked for work in the past, but their current situation indicates that job search at this particular point in time would be futile, so they accept their plight and don’t look for work. People in this category are not classified as either employed or unemployed. They are classified as not in the work force, and are part of the hidden unemployed. There is even a name for people in this category: discouraged workers. They are not in the “official” unemployment statistics, but they are counted. In March 2014, their numbers totaled approximately 700,000. This number was reported in the same news release that the official unemployment rate was reported, but didn’t make the headlines to the same degree. I would have to assume that most people searching for a “real” unemployment rate would include these people as unemployed. If we want to include them in the March 2014 calculations, then we would add that number to both the total number of unemployed persons and the total civilian labor force. That would give us an unemployment rate of 11.2 million divided by 157 million, or 7.1%. Adding discouraged workers to the official unemployment calculation would raise the unemployment rate from 6.7% to 7.1% for March 2014.

But I see reports of many people claiming that the “real” unemployment rate is much higher than this. There must be other categories that they include in their calculations. What if the person who wants to work, but hasn’t looked for work in the past 4 weeks, has quit looking for work for a different reason other than just giving up? What if that person gave up, and then later committed to doing something else, something that would prevent that person from taking a job if one were offered? Suppose, for example, that person enrolled in college and has committed a certain time frame for education, perhaps a semester or an entire degree program? Should that person still be counted as part of the “real” unemployment rate even if he wouldn’t take a job if one were offered? What if a person wanted a job, but illness or injury prevented him from looking for employment for an extended period of time? Should that person be in the same statistics? What if a person was forced to take care of a sick family member and couldn’t take a job if one were offered, even though that person would prefer to take a paid job? What if a person couldn’t look for work due to transportation problems?

Believe it or not, these people are counted in the BLS data. They just are not part of the official unemployment statistic. How many of these people do you think rightly deserve to be part of the “real” unemployment rate? I suppose the answer depends on what point you are trying to make. If you are counting underproduction in the economy, perhaps most or all of them should be counted. If you were trying to make the point that these people would be working if the economy were performing better, then perhaps you would have to count some but not count others. For the sake of argument, let’s count all of them. They are lumped together by the BLS in a category called “marginally attached to the labor force”. Let’s add them all back in and see how that changes the unemployment rate. This category already includes the discouraged workers that we added in above, so we won’t double those calculations. We’ll go back to the original 6.7% and add in all of the marginally attached persons. That number for March 2014 was a little over 2 million. If we add that number to both the number of unemployed persons and the total labor force, we get an unemployment rate of about 12.5 million divided by about 158 million, or 7.9%. Adding in all of the marginally attached persons to the ranks of the unemployed, even if doing so is somewhat questionable, would raise the unemployment rate from 6.7% to 7.9%. This is a sizable jump, but the numbers fall well short of some numbers that I see quoted as the “real” unemployment rate.

What about those who work part time? People get counted as employed if they have a job. It doesn’t matter if they work part-time or full-time, or if they have more than one job. If they are employed, they simply get counted once as employed. Many of these people want to work full-time, but due to economic conditions they are only offered part-time work. These people are classified as being underemployed; but in the official statistics, they are simply “employed”. What would happen if we added them to the ranks of “unemployed” for the purpose of coming up with what we consider to be an accurate “real” unemployment rate? Doing so would probably distort reality, because it would discount all of the production that is already being done by these part time workers. But since they want to work full-time and are only working part-time, let’s count them all as unemployed instead of their current classification of employed. We can’t count all part time workers in this category, but we can count the ones who would rather work full-time. They are already being counted, and are included in the data released by the BLS alongside the official unemployment rate. The BLS data has them listed as “part-time for economic reasons” and include not only people who could only find part-time work, but people who have been reduced to part-time due to slow business conditions that perhaps are temporary. Let’s count them all; that number is approximately 7.4 million people. That’s a lot of people, and it will make a huge difference in the unemployment rate. Let’s go back to our last adjustment, which counts all of the marginally attached persons, and change 7.4 million from employed to unemployed. That would change our unemployment rate calculation to about 20 million divided by about 158 million, or 12.7%. That is higher than any annual unemployment rate since official records have been kept, beginning shortly after WWII. The unofficial unemployment rate during the Great Depression was more than twice as high.

This brings up an important point about comparing the unemployment rate from one point in time to another point in time. We have used some very questionable calculation adjustments to get from the official rate of 6.7% to a “real” rate of 12.7%. It is not logical to take that adjusted figure, 12.7%, and compare it to an official figure of an earlier point in time. If you make adjustments, rational or questionable, to one point of comparison, then you need to make the same adjustments to the other point of comparison. Otherwise, the comparison makes absolutely no sense. So the reference above about the historical unemployment rates compared to our adjusted rate of 12.7%? That was going beyond the questionable adjustments we made to raise the current rate to 12.7%. That was adding an irrational comparison.

What is the “real” unemployment rate? I don’t have an answer. If you want to make the point that it is different from the official unemployment rate, then any number that you come up with must be based on data that is consistent with the point that you are trying to make. We didn’t do that in the calculations above, we just added in whatever real numbers we could justify in order to make the unemployment rate look as large as possible. But what kinds of points would we be making that would justify using a real unemployment rate that is higher than the official unemployment rate?


  1. Are we trying to say that the economy has not recovered as much as the official numbers show?
  2. Are we trying to compare today with a specific point in the past, perhaps for political reasons?
  3. Are we trying to say that the economy is operating at far below capacity?
  4. Are we trying to say that there are more people who are hurting for lack of jobs than what the official numbers indicate?
  5. Are we trying to say that too few workers are supporting too many free-loaders?

Perhaps each one of these points would require making different judgments as to which of the types of situations we would be justified in making adjustments for in the calculations. Do we count part-timers as unemployed instead of employed? Which categories of marginally attached individuals should be included, instead of blindly including all of them like we did? Do we really want to count full-time students, temporarily disabled people, stay at home moms, people taking care of sick parents, and the like as unemployed people? Perhaps for some arguments and purposes, but we certainly cannot be justified in doing so across the board.

The only reason anybody even discusses a difference between the official unemployment rate and a “real” unemployment rate is the fact that different people are trying to make different points as to the meaning of the numbers. Certainly, different numbers can be justified for making different points. As long as those insisting on using a certain number or a certain method of calculation clarify which argument they are making, and limit the use of their version of a “real” unemployment rate to that argument only, then there really isn’t much of a reason for disagreement. But we do have disagreements which only serve to confuse the issue and help to keep the public misinformed.

And what about using a higher “real” unemployment rate to point out that too few workers are supporting too many others? We have already counted every single marginally attached person; we have reclassified part time workers for economic reasons from employed to unemployed, and we have made these adjustments without regard to appropriateness. We just counted them all. With everything counted, we nearly doubled the unemployment rate from 6.7% to 12.7%. Yet I continue to hear people arguing that the “real” unemployment rate is perhaps double that, to over 25%. How can that be, if we already added in everybody that we could to the ranks of the unemployed? There is no rational approach to the data that would justify a “real” unemployment rate as large as nearly 25% as being reasonably close to reality, unless the only point being made is to compare the “makers” to the “takers”. The only way you could come up with a number that high is to count a lot of people who are not even marginally attached to the labor force, and call them all unemployed. If you are trying to make that point, be very careful about the conclusions that you make. If you want to make the point that the “real” unemployment rate is significantly higher than what we calculated here, then you will need to point out that you are counting as unemployed: significant numbers of children, disabled people, retired people, full-time students, caretakers, homemakers, and/or people who live in households that already have breadwinners to support them.

A version of this essay is included as a chapter in the book Common Misconceptions of Economic Policy by Jerry Wyant. You can purchase this book in paperback form from Amazon and other online book distributors. The list price is $12.99 (only $9.99 using discount code TA9GTK7E when ordering, depending on the distribution channel). Or if you prefer, you can download a digital version on your device (Kindle, Nook, etc.) for $4.99.

Paperback version from Amazon
Kindle version from Amazon
All eBook formats from Smashwords

Jerry Wyant