Historical American Job Shortage Number Data, and How the AJSN Fits In

The American Job Shortage Number has varied considerably between mid-1994 and mid-2012, and since it uses unadjusted labor data to be consistent with population data, it can fluctuate more than other measures from month to month. However, using the same month each year can eliminate most of that illusion.  Here are the July numbers for each year since 1994, in graphic and chart form:

Year (July)	AJSN	Year (July)	AJSN	Year (July)	AJSN	Year (July)	AJSN
1994	15.7 million	1999	13.6 million	2004	16.0 million	2009	23.3 million
1995	15.0 million	2000	13.3 million	2005	15.6 million	2010	23.5 million
1996	14.9 million	2001	14.2 million	2006	15.4 million	2011	23.4 million
1997	14.1 million	2002	16.2 million	2007	15.3 million	2012	22.5 million
1998	13.9 million	2003	16.7 million	2008	17.3 million

The AJSN is made up of 11 components, of which one is based on the officially unemployed and 10 are based on other groups of nonworking Americans or American residents.  The shares of AJSN from those unemployed and from the other ten statuses together (non-unemployed) has varied as follows:

How does the AJSN relate to the Bureau of Labor Statistics measures?  When it is compared with the official unemployment rate, we get the following:

Note that the difference has been growing, especially over the past few years.  However, it not a completely fair appraisal, as we are comparing rates (unemployment) with absolute numbers (AJSN), which brings in distortion from population changes.  Comparing the AJSN, then, with the number of people unemployed gives us this:

At first glance, this chart may look like AJSN is only the unemployment level plus a near-constant.  However, when we look in more detail, the gap has not been staying the same.  Consider the following:

This chart shows how much AJSN has increased even with regard to the total number of people officially unemployed.  So what about population distortion, as again if the number of people is rising, the numbers within all of their categories may as well?  Here, then, is the AJSN as a per-capita metric, measured against the number of Americans aged 15 or older:

Higher numbers here are bad here, and we may be coming in on one additional absorbable job for each 10 working-age Americans. 

We know that the unemployment rate represents a share of people who ostensibly want to work, but how about the number of jobs that would quickly be taken by Americans not working but not officially jobless?  Here is a view of the per-capita job demand by that set of people:

The line is bowl-shaped, and noteworthy here is the last three years.  Although the July unemployment rate peaked in 2009 and 2010 with 9.7% apiece, the number of jobs others without them would work has continued to climb.  That is what we are seeing now, with job demand rising without regard to official unemployment.  I will follow these trends and we will see if they continue, as I expect they will.

----
Here is a chart of the estimated numbers of American citizens living elsewhere, which is an input to the AJSN.  All other data used is from the Bureau of Labor Statistics and the United States Census Bureau.
 

Year (July)	American Expatriates	Year (July)	American Expatriates	Year (July)	American Expatriates	Year (July)	American Expatriates
1994	2,871,428	1999	4,000,000	2004	4,500,000	2009	5,000,000
1995	3,014,286	2000	4,100,000	2005	4,600,000	2010	5,100,000
1996	3,157,143	2001	4,200,000	2006	4,700,000	2011	5,710,000
1997	3,300,000	2002	4,300,000	2007	4,800,000	2012	6,320,000
1998	3,650,000	2003	4,400,000	2008	4,900,000

Sources:  Association of Americans Resident Overseas, U.S. Department of State.

AJSN Revised and Updated - We're Currently 22 Million Jobs Short

Thank you for your comments and feedback!  I have corrected one data error, and have obtained Bureau of Labor Statistics (BLS) and U.S. Census Bureau data released this past Friday.  We now have an American Job Shortage Number (AJSN) of 22.0 million, that on an official unemployment rate of 8.1% and 12,696,000 officially unemployed.  

As for historical data, I am working with the BLS to obtain monthly detail for years past.  Accordingly, I hope to post more later this week.

AJSN - AUGUST 2012	Total	Latent Demand %	Latent Demand Total
Unemployed	12,696,000	90	11,426,400
Discouraged	844,000	90	759,600
Family Responsibilities	302,000	30	90,600
In School or Training	273,000	50	136,500
Ill Health or Disability	181,000	10	18,100
Other (Adjusted for BLS labor statistics rounding error)	961,000	30	288,300
Did Not Search for Work In  Previous Year	3,812,000	80	3,049,600
Not Available to Work Now	658,000	30	197,400
Do Not Want a Job	81,280,000	5	4,064,000
Non-Civilian and Institutionalized, 15+	7,060,840	10	706,084
American Expatriates	6,320,000	20	1,264,000
TOTAL			22,000,584

Introducing the American Job Shortage Number

The Bureau of Labor Statistics releases the official unemployment percentage monthly.  It also offers a variety of other numbers related to employment every year, including the numbers of people with other labor statuses.  Among them are the total count of people working and officially unemployed; people working part-time for economic reasons; those discouraged from looking for work; those not working due to family responsibilities, being in school or training, ill health or disability, or other reasons; people who want to work but did not look for a job in the previous year; those who want to work but cannot now; and those who do not want a job at all.    

In Work’s New Age, I added a number of categories to clarify the roles these people had in relation to the labor force.  I took 40% of those employed and designated them “underemployed due to money, skills, or values,” as a conservative estimate of those working lower than their established or clearly able levels, leaving those working and not underemployed as “solidly employed.”  I determined from census statistics the number of people under age 15, and the number of Americans who were not civilians (e.g. in the military) or institutionalized.  I estimated from other sources the number of American citizens living in other countries, making a total of 15 employment statuses.

After a discussion of latent demand and how it affected the number of people who actually wanted to work, I allocated a percentage of people to each of these groupings, writing as follows: 

So how can we assign demand for jobs to the 15 categories? Let’s start with two assumptions. First, I will name percentages based on a very strong labor market along a wide range of fields, one where almost anyone willing to work could find a job within a matter of weeks. The closest comparison in recent decades I know of was Houston in the early 1980s, when so many people moved there from manufacturing-declining Detroit that a newspaper’s lead sentence said the hottest import there weighed only seven pounds and couldn’t be driven: a Houston Sunday newspaper filled with job ads. Second, I will strive for conservatism.

Although those in the unemployed category are officially looking for work, some are there only briefly between jobs and some frankly prefer not to work, so I will estimate 90% would obtain employment. The three categories of people with jobs would be unchanged. Discouraged workers in a time of ample jobs would function much as those officially unemployed, so I have also assigned them 90%.

Those in the next set of statuses are smaller in number, but are harder to estimate with confidence. Those on the record as wanting to work but have stopped now due to family obligations would mostly continue them, but many would end up with jobs, so I will say 30% would take employment. While continuing school or seeking training is ostensibly a primary goal, many are there for lack of a paying alternative, so I estimate 50% of those would work. People with ill health or disabilities are mostly not able to take a job, but as we can understand from the rising number of disability applications, some would reach the labor force if they knew they could get employment, so I have included 10% as working. In the “other” grouping, I have estimated 30%.

In a job-seeker’s market, those who did not search for work in the previous year but say they want jobs and are ready for them now would be similar to those officially unemployed. Since I think a significant number, though, would choose to continue not looking, I will assign 80% of them to finding work. Many of those “not available to work now,” but still officially wanting jobs, would get them—I say about 30%.

How many people who are out of the labor force and claim not to want work would end up with it in a robust market?  That is hard to tell. The great majority are fully retired, like or accept their lives as they are, and would not seek employment. However, people make plans, even long-term ones, based on what they consider to be realities. In the current market, many assume they could not get a reasonable job even if they wanted one, so we cannot dispose of the category entirely. With more than a quarter of the population, this is the largest single employment-status grouping, so small changes to the estimate would have a large effect on the number of potential employees. I have estimated that 5% of these people would get jobs if they thought they could, which, though possibly way too low, is unlikely to overstate.

We are left with the non-civilian and institutionalized category, of whom some in the armed services but few shut-ins might take jobs elsewhere, so I have assigned them 10%, and American expatriates, many of whom have left because of economic conditions at home, 20%.

Putting together all categories of people without civilian jobs, here are the number in each        group, along with the share who would work if given a clear chance:

Status	Number as of July 2011	Percent of Total	Latent Demand Percent (Est.)	Latent Demand Number (Est.)
Unemployed	13,908,000	4.39%	90%	12,517,200
Discouraged	1,119,000	0.35%	90%	1,007,100
Family Responsibilities	169,000	0.05%	30%	50,700
In School or Training	323,000	0.10%	50%	161,500
Ill Health or Disability	215,000	0.07%	10%	21,500
Other	877,000	0.28%	30%	263,100
Did Not Search For Work in Previous Year	6,437,000	2.03%	80%	5,149,600
Not Available to Work Now	597,000	0.19%	30%	179,100
Do Not Want a Job	79,564,000	25.13%	5%	3,978,200
Under Age 15	61,201,106	19.33%	0%	0
Non-Civilian and Institutionalized, 15+	6,731,811	2.13%	10%	673,181
American Expatriates (Est.)	6,000,000	1.90%	20%	1,200,000
Total	316,591,917	100.00%		25,201,181

Accordingly, rounding off to the nearest 100,000, the American Job Shortage Number (AJSN) for July 2011 is 25.2 million.  This number is a more accurate estimate of how many additional jobs would be absorbed in the United States than anything directly from the official figures.  It represents an improvement from the July 2010 number of 25.4 million, as documented in Work’s New Age, though not nearly as large a drop as the tally of officially unemployed, which moved from 14,829,000 to the above 13,908,000.   The difference is that more and more people who do not qualify as unemployed want jobs, and the AJSN shows how to include them as well.

See this blog for updates to the AJSN as newer data becomes available.