This page had been redirected to a new URL, please update any bookmarks.

Font Size: A A A


« Will labor force participation continue to rise? | Main | Fed purchases of government debt: Flow-share versus stock-share »

June 07, 2012

The skills gap: Still trying to separate myth from fact

Peter Capelli has looked at the skills gap explanation for labor market weakness and sees more myth than fact:

"Indeed, some of the most puzzling stories to come out of the Great Recession are the many claims by employers that they cannot find qualified applicants to fill their jobs, despite the millions of unemployed who are seeking work. Beyond the anecdotes themselves is survey evidence, most recently from Manpower, which finds roughly half of employers reporting trouble filling their vacancies.

"The first thing that makes me wonder about the supposed 'skill gap' is that, when pressed for more evidence, roughly 10% of employers admit that the problem is really that the candidates they want won't accept the positions at the wage level being offered. That's not a skill shortage, it's simply being unwilling to pay the going price."

To some extent, the issue is semantic:

"But the heart of the real story about employer difficulties in hiring can be seen in the Manpower data showing that only 15% of employers who say they see a skill shortage say that the issue is a lack of candidate knowledge, which is what we'd normally think of as skill. Instead, by far the most important shortfall they see in candidates is a lack of experience doing similar jobs. Employers are not looking to hire entry-level applicants right out of school. They want experienced candidates who can contribute immediately with no training or start-up time..."

In the language of economists, Capelli is defining skill as the possession of generalized human capital, while businesses are defining skill as the possession of firm- or job-specific human capital. In more familiar language, Capelli appears to be focused on innate skill levels and education, while businesses are looking for the types of skills that would be attained through past on-the-job training. In even more colloquial language, Capelli wants businesses to appreciate book-learning, and businesses prefer those who have already survived the school of hard knocks.

We have recently completed our own version of the Manpower survey Capelli references. Our results are based on the responses of about 100 businesses in the Sixth Federal Reserve District represented by the Atlanta Fed, and we do not claim that they are conclusive. But we do think they are instructive.

Of those firms that said they experienced an increase in hiring difficulty over the last year, our poll respondents confirm the notion that businesses are looking for candidates with specific skills:

The lack of technical skills is the only factor that really jumps out as an issue that businesses have with the pool of job applicants. We often hear anecdotal complaints about job seekers' lack of "soft skills," or the difficulty in finding applicants who can pass required background checks. But only 14 percent of all selections indicated too few applicants with required interpersonal skills, and only 7 percent indicated a problem with applicants passing screening requirements like drug-use or credit checks.

On the other hand, our poll found scant support for Capelli's claim that businesses are "unwilling to pay the going price." Only 9 percent of respondents reported that too few applicants would accept the offered compensation package.

Despite the fact that we see some evidence consistent with skill mismatch, it is far from clear that this issue is the smoking gun that explains the current anemic state of job growth. When asked if a dearth of skilled applicants is a persistent problem, our survey respondents overwhelmingly answer "yes." But when asked if they have had more difficulty hiring over the past 12 months, the overwhelming majority answered "no":

Even among the minority of businesses that report recent hiring difficulties, only half indicate that this difficulty is restraining growth:

We infer a couple of lessons from all of this information. First, it does appear that there is a long-term skill level problem in the U.S. economy. Adopting Capelli's definition of skill does not mean the existence of skill mismatch is a myth.

But turning to the short run, we've been pretty sympathetic to structural explanations for the slow pace of the recovery. Nonetheless, we have yet to find much evidence that problems with skill-mismatch are more important postrecession than they were prerecession. We'll keep looking, but—as our colleagues at the Chicago Fed conclude in their most recent Chicago Fed Letter—so far the facts just don't support skill gaps as the major source of our current labor market woes.

David AltigBy Dave Altig, executive vice president and research director at the Atlanta Fed, and

John RobertsonJohn Robertson, vice president and senior economist in the Atlanta Fed's research department

June 7, 2012 in Employment, Labor Markets | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference The skills gap: Still trying to separate myth from fact:


You need to spend more time in corporate America. Having been on both sides of the process recently, i can tell you, he is 100% correct. It would be deeply eye opening for you. HR policies are very rigid and in some cases i know the hiring manager only gets "pre-screened" applicants and is not allowed to review resumes. pick a name from a hat! And these are deeply skilled people with masters and doctoral degrees in economics, finance...

Second, we actually have somewhat of a time series for the manpower talent survey.

How does difficulty filling positions track through time based on the manpower talent survey?
2006 44%
2007 41%
2008 22%
2009 19%
2010 14%
2011 52%
2012 49%

In other words, the data is cyclical: when the economy is growing and employers actually have positions, employers report some difficulty filling them.

You can also see which are the top ten jobs. The 2006 talent survey said the top ten were: Sales Representatives, Engineers, Nurses, Technicians, Accountants, Administrative Assistants, Drivers, Call Center Operators, Machinists, Management/Executives.

There is significant overlap in the 2012 survey (what skills does "driver" need? just a commercial drivers license).

Finally, the point you need to recognize is that most training in the US is given on the job (OJT). People with masters, PhDs, MBAs, sorry, even for them its old fashioned OJT. Companies are more willing to train and less willing to be "picky" when they are not getting 6 applicants for every position. 10 applicants, 3 make it to interviews, 1 job.

The purest measure of cyclical unemployment is for young, with a college degree or above - these are the highly mobile highest skilled workers. And unemployment among this segment is still atrocious.

Posted by: dwb | June 07, 2012 at 05:03 PM

I would like to see a survey of the recent wage history of these so-called "skilled workers", as defined by employers. If these workers are in such short supply, shouldn't their wages be rising rapidly?

Posted by: rab | June 09, 2012 at 11:12 AM

Outsourcing!! Look at H1-B salaries and everything will be clear.

Posted by: vv | June 11, 2012 at 02:12 AM

"On the other hand, our poll found scant support for Capelli's claim that businesses are 'unwilling to pay the going price.' Only 9 percent of respondents reported that too few applicants would accept the offered compensation package."

Applicants that don't accept the compensation package, after applying and interviewing, aren't really the problem. While some employers don't advertise wages in a posting - others do. Those offering sub market wages are going to attract the least qualified, most desperate applicants.

Further, our experience working with the unemployed and employers suggests that a significant percentage screen out the long-term unemployed and/or anyone unemployed at all. Screening out the unemployed will bias the sample. Given the number of mass layoffs since 2008, affecting "good" and "bad" workers alike, many highly skilled employees will be automatically screened out.

Some firms use computer programs to pre-screen applications meaning no human ever sees the application before it's "accepted" by the firm. These programs will screen out those with salary expectations higher than that which the firm is willing to offer (as well as the unemployed).

Posted by: Bob | June 11, 2012 at 06:40 AM

I sincerely hope that the researchers have a good idea of how outsourcing works. It is NOT JUST the employees who work in the US at all!

Let us take one of the good job categories which support around 4 to 5 other jobs in the economy (Computer Engineering) as an example and it is the best way to describe this phenomenon. The way the firms reduce costs is by employing a TOKEN H1-B visa candidate in the US (from one of the outsourcing firms) and make this person manage a pool of 30 offshore workers (paid around $20/hour offshore wage as opposed to $50/hour in the US). The outsourcing firms also train all their employees unlike the US where the employers need to train them. So For every H1-B visa issued, there are over 30 high paying jobs lost in the US (which results in over 150 other jobs lost indirectly) and the work is done at a one-third of the cost. So basically employers are stunned when US citizens ask for more than $20/hour wages for any job (or say they haven't worked in that field) which can be offshored and don't want to pay and say it is a skills mismatch. Of course, lawyers, doctors and dentists have got it made
since there is no technology to pull the teeth thru offshore labor so far.

On a side note, I suspect the productivity figures in the US are also showing large surges because of this (a large pool of employees in offshore locations NOT included in the productivity calculations). Profit margins are also skyrocketing because the pool of employees in offshore locations are paid a pittance in weaker currencies. So as long as the offshore labor pool is available at low wages, companies here can have skyrocketing productivity and have high margins until the whole system collapses due to lack of purchasing power.

Posted by: vv | June 11, 2012 at 09:31 AM

If the lack of proper skills is what is holding back hiring, then the most sought-after people in the labor market would be recent college graduates. Sadly, as many graduates, college placement officers, and parents of graduates can tell you, this is not at all the case.

This suggests, then that skill mismatch is not the problem and the refrain of "If we only had employees who were ready for the new economy . . ." is a smokescreen. Instead of looking on the labor side of the hiring equation, perhaps giving attention to the management side of the equation might help.

Managers with whom I speak tell me of memos and conversations with executives that present two very powerful forces at work:

(1) "You have to squeeze more productivity out of existing workers, to keep the bottom line looking good."

(2) "Don't add head count until you are absolutely certain that our sales are on the rise. It's better to lag behind in hiring than to get out in front of the recovery."

Fear of being wrong about the need for new employees is a huge motivator not to hire.

Posted by: Peterr | June 11, 2012 at 11:09 AM

Seconding rab - structural problems should lead to significant pockets of rapidly increasing wages.

Posted by: Barry | June 12, 2012 at 04:16 PM

Employers are cost cutting to the bone when it comes to employees. It started with wage stagnation, increasing workloads while minimizing hiring, elimination or reduction of benefits, and minimizing training costs. There appears to be a number of influential employers that do not want to incur any costs (i.e. unemployment, social security, medicare, or workers compensation) in acquiring an employee other than paying enough to meet their reservation wage which has probably been reduced due limited opportunities to negotiate or change positions. It almost makes you wonder whether the preponderance of scientific management and propensity to go public encourages these practices.

Posted by: LB | June 20, 2012 at 11:12 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in