Are we seeing skill mismatch after all?

Peter Orszag writes:

An odd puzzle is taking shape in the labor market: Over the past three years, the number of job openings has risen almost 50 percent, but actual hiring has gone up by less than 5 percent. Companies are advertising a lot more jobs, in other words, but not filling them.

To get some sense of how significant this is, consider that if, since June 2010, hiring had risen a third as much as advertised jobs have (rather than only a 10th), and nothing else were different, job creation would be roughly 500,000 higher each month, and the unemployment rate would already be back to normal levels.

One possibility is that companies are offering jobs, but at wages too low to attract takers:

Some support for this perspective comes from experimental data in the jobs survey, which show that the job-offer rate has risen most sharply (relative to hiring) for establishments with 10 to 250 workers.

Alternatively, companies may be filling positions internally, or they may be advertising without much intention of hiring.  Or we are back to the skills mismatch hypothesis.  Are companies simply expecting too much perfection in their new hires?  If so, why has the demand for (near) perfection gone up?

Here is commentary from Reihan.  And here is commentary by Ashok Rao (did you know he is 18 years old and from Chennai?).


Every job application I've ever looked at asked for someone who doesn't exist. It described the guy who just retired or left for more money plus some ridiculous wish list items: Wanted: mega-nerd with outstanding interpersonal skills, 10 years experience for entry-level position, etc. On top of that, now it costs nothing to proliferate job ads.

They want someone with 50 years of experience, a PhD, speaks five languages, can build a reactor with bubble gum and a swiss army knife in 2.3 seconds, who is interested in an unpaid internship filing papers.

That this question is even a question is a jaw-dropping display of ivory tower ignorance.

Well, a PhD, but only if you have demonstrated a broad array of interests as well as a consistent publication record top journal publications in the newest hot area.

Oh, I almost forgot, also an uncompromising dedication to teaching.

Employers confuse "credentialed" with "smart." What they really want are smarter, more conscientious employees. The trouble is, there are only so many of those, and all we can collectively do is bid up their price. Higher wages can induce more people to earn degrees and such, but can't raise the number of smart people.

There's only one way to get more smart people (*), and as a society, we aren't doing that at an appropriate pace. Partially because the feedback from labor demand to potential parents is so indirect.

(*) You could conceivably also do it through immigration (at least at a national level, but not at a world level), but our immigration policy is not designed to be helpful to the nation or its citizens. Frankly, it looks like it was designed by vandals.

Designed by Vandals is more like it.

University of Idaho ruins everything

'Finch' wrote, "Employers confuse “credentialed” with 'smart.'"

Then again, it could be that employers are intelligent enough to realize that smart employees are a better value than credentialed ones- yet current employment law makes the cost of testing for smarts prohibitive.

"Smart" might be defined as, "able to pick up whatever skills are needed rapidly and well- adaptable, versatile." And there probably are ways to test for that, but, if someone can show that a test has a disparate impact on minorities, and if the employer can't prove that the test is a business necessity, that employer may be taught a very costly lesson.

So, employers fall back on credentialism- if they can't test for "smart," they can at least select for employees who already have the exact skill set that's required to be productive on Day One.

I don't think this is really saying something different. Employers want "smart." Being able to test for "smart" would be a good thing, I completely agree and laws that prevent this are a travesty. But being able to test is not going to make there be more smart people, nor is it going to reduce the differential performance of kinda-smart versus very-smart.

I just went through a job hunt, and the number of employers who wanted "completely trained and ready to go on day 1" versus "could pick up all needed skills within 1 week" was astonishing.

I could understand the former mindset for contractors, but for permanent hires if you hire to fill a specific need you are in trouble as soon as your needs change which is likely to be very soon.

Most employers totally suck at this. It's not because they "can't test for smart." It's because they are a) terrified of making a wrong hire and b) think only hiring 1 out of 10 people means they are getting the best 10%.


Well, NPW, if you can pick your jaw up for a second, could you take some time and explain why this all changed recently?

Tyler's asking why a change over time occurred and you're not answering that question. WHY do "they" want all of those things when they didn't before?

Are "they" suddenly more... greedy? Incompetent? Demanding? Optimistic about hiring?

We are recovering from a recession which does make labor a bit of a buyer's market, but this isn't the first recession that we've recovered from...

For the reasons that have already been mentioned here. Keep reading.

The returns to "smart" have been going up for some time, but the number of smart people has been, if anything, going down. Eventually this was going to cause a problem. It was noticeable in the dotcom era and subsequent bust, though it wasn't as obvious.

There is no easy money doing common things anymore. To do handy man work requires working knowledge of electronics, mechanical, chemistry, industrial hygiene, plus a lawsuit proof knowledge of health and safety regulations. And your client is willing to pay you the same amount as they paid the undocumented Mexican. It takes years of working experience in an endeavor to get there, and only one in ten of any group of workers is able or willing.
So you try to hire, you can't afford to pay someone's education debt because there are thousands in other countries who will do it cheaper.
The west is, to put it bluntly, screwed. The bien pensants, rent seekers other words haven't figured it out yet, and continue to support the imposition of more costs. There is a he'll of a lot of seed corn to eat yet, why worry.

Experienced something similar back in the recession in the 80s that's always stuck in my mind. Had a boss who complained the people responsible for filling positions wouldn't hire anyone who didn't meet 13 of the 10 job requirements.

Expectations rise in a hirer's market, sometimes to absurd levels.

National Review famously posted an ad on their site for a receptionist. The job entailed answering phones and greeting people at the door. They wanted someone with a college degree.

Don't know how famous it was, but I looked that up and found this.

I fail to see what's wrong with this.

I wasn't aware that NR was famous for posting such a job ad, but lots of job postings for receptionists state "degree preferred".

Admin assistants and secretaries even more so.

Note based on discussion below - SF Bay Craigslist job ads cost $75 (per category selected) to place.

"On top of that, now it costs nothing to proliferate job ads."

I think this hits the nail on the head. Almost any company is willing to add one more person to headcount if a really strong candidate happens to come along. But if the cost to list a job costs time or money, there's no point advertising for the really rare super-candidate. But if listing a job costs nothing then putting an ad for a super-candidate is like getting a free option. It might only pay off once in a while, but better to have it out there.

It definitely costs money to deal with job applicants.

Most job applications seem to end up in a bitbucket, though, so you may be right.

Is this just the hiring market's equivalent of High Frequency Trader's floating bids that disappear when people want to take them?

When you have a bunch of HR professionals who are in essence trading for talent (getting the most talent for the least price) without looking at fundamentals at all, I think they converge to the same game theory type strategies. It's phantom liquidity in a computerized market.

I would put the scalpers buying up restaurant reservations with no intent of using them with having the same trading mindset.

Rather than speculate, why not ask the job seekers who've become professional interviewees? And given that they've already answered this, why are you'll confused?

Isn't this predictable from a stick-wage / money illusion perspective? Picture demand / supply with a price floor above the equilibrium point. At the floor price, there is a gap between demand and supply, with supply to the right of demand. But if job offers are at a price below the equilibrium point, there is a gap with demand to the right of supply - that is, more offers than takers.

I think of stabilization policy as a classic public good because it solves a pernicious collective action problem. The collective action problem occurs when nominal income contracts, putting downward pressure on wages. Collectively, the rational behavior would be to each accept a equiproportional nominal wage cut. But individually, you have the incentive to guard your sovereign piece of NGDP, and let others take the bigger cut and/or pink slips.

The above model would show superficial evidence of a skill gap. If you are higher skill your probability of losing your spot in the game of musical chairs decreases, meaning the expect value of defending your nominal wage increases. The result would be an apparently novel premium on skill, when in fact they may just be fighting harder to keep their seat.

How are the job ads being measured?

If ads are being measured by Internet traffic then I suspect the numbers are padded.

Easy access, low cost plus bogus (work at home) ads.

This question has been much debated on the Fourth Turning forum, and the consensus is - employers demand perfection in new hires (one poster called it "the purple unicorn employee") because they think they can get perfect employees, with unemployment so high.

And that they say they want people who can think outside of the box, but, like any bureaucracies of long-standing and other private fiefs, actually want people who will go along to get along.

I believe the discussion is on one of the Gen-X threads.

Does that make this just a bad equilibrium, if unemployment causes employers to raise employment requirements?

The paradox of choice?

The same reason I'm never happy with my peanut butter options at the grocery.

Another possibility is that people are declining to take jobs due to the generosity of unemployment benefits. I guess that would be a subset of the too low wage hypothesis.


Anyone who's known some of the long-term unemployed knows there's an element of truth in this. It's a nudge not to settle. But it seems too small an effect to be driving such terrible labor market performance.

One question here: are we talking general job adverts or only full time jobs? Perhaps the 500.000 jobs are part time offers, thus the lower pay, and a result of ACA. The 50 to 250 employees business might be an indication for this.

Unemployment benefits are not remotely "high". Anyone who thinks that is a modern version of Marie Antoinette. If one even qualifies for benefits (most people do not) they pay, on the average, about 1/2 o a person's pre-layoff income-- and no benefits at all.

I was unemployed for 10 months last year in Michigan. My salary was $86,000 a year which qualified me for the states max unemployment payment, $356 ever two weeks or $732 a month. Having to often go into their offices and wait for hours with other people for various reasons, most people were getting $350-500 a month.

My personal opinion having seen and recently received unemployment payments, they are not generous the only reason you might pass up a fulltime job is if you have the ability to make money under the table.

A few possibilities:

1. How much of a divergence is this from the historic norm? Perhaps this is normal?

2. Perhaps this is due to the "latent" labor market of discouraged workers who previously stopped working and stopped counting now returning to the pool.

3. Has there been an increase in job turnover? Perhaps most of the hiring is away from rival firms and not from unemployed - I would call this the ZMP explanation.

No, I did not know that Ashok Rao is 18 years old and from Chenna ... but I love his blog

It does seem that a lot of workers at the lower end, retail, fast food, etc just switch from one employer to the next. Job ads are just a function of high turnover.

I do know one firm that has been advertising for a senior management position for four years, they have yet to hire someone. Why? They just continue to interview while spreading the work load to current staff. They are having a hard time getting candidates they would like to relocate to a depressed community. Their expectations are a bit rigid.

We need an HR flunkie to come in and settle this question once and for all since this keeps coming up as a burning issue.

Anecdotally, companies are asking for the moon: a) because they can, and b) because every once in a while someone stumbles by who actually does meet ridiculous qualifications and is either willing to accept equally ridiculously low wages for whatever reason or just doesn't know his/her market worth. I know this because I know many IT recruiters.

If a company isn't in dire need and can cast a wide net for well-qualified suckers at virtually no cost, why not do it?

I think, increasingly, IT recruiters are paying these Sisyphusian searches no never mind.

I believe there is a lot of this. The HR folds need to keep themselves busy by trolling fro resumes.If a unicorn happens to show up and willing to work on the cheap they will hire them, otherwise they'll just keep looking.

You would expect this sort of problem to be systematic: it's the change we're interested in as far as openings are concerned. Is there any reason to believe job trolling becomes particularly more prevalent in a recession, implying a mirage of openings to begin with?

Here's an image of recruiting intensity per vacancy:

Notice two points: while it is cyclical we've seen a pretty strong recovery since the trough, and the y-axis not inconsiderably overstates the variance in this context. Intensity today is 90% what it is during normal times.

So even if you adjust openings by intensity wouldn't you see a similarly "healthy" recovery?

Mayhaps HR trolling has become cheaper & easier because of software and it has nothing to do with the recesson itself.

It's cheaper and easier because of technology but also because of the abundance of staffing firms who are happy to spin their wheels trying to find the proverbial needle in a haystack. The total cost of those busy beavers doing the heavy lifting to the company: zero unless they hire (or contract) someone.

How do the staffing firms make money, if all they do is look for purple unicorns but fail to find any?


a) For the contractors or FTEs they do manage to place, the markups are enormous.
b) Not every position that recruiters are trying to fill has the purple unicorn description. Those descriptions are for positions that don't exist in reality and don't have a dire need. If the right person does come by, maybe the company actually hires that person, but HR isn't actively looking, or even holding out hope that said purple unicorn will grace the company's doorstep. You won't see crazy requirements for 3rd shift Data Center warm bodies, for example.

Several components to the challenges of judging this by job ads:
- Companies are posting positions and then not hiring, and often leaving these positions open so that they can claim them when they get budget cuts next quarter. For example: My friend just got a reject e-mail for a job that he applied for in 2011. The company had left it open and either finally purged the system or it expired.
- Many jobs have moved to contract positions, especially in IT, and especially in the Kansas City area. Companies no longer hire off the street but instead hire from the contractor pool. This gives them "try before you buy" advantages. This impacts the job ad postings because many contracting companies post for the same position. If they get the right person they will submit them to the company. But it is not surprising to see three-five-plus contractor postings for the same job - same job description, everything.
- I agree with the previous posters that employers want ridiculous combinations of skills - often skills that are not complimentary. For example, in IT a person is not going to be a super experienced Windows Dotnet software developer and a Linux/LAMP developer. Just doesn't usually happen.
- A lot of job ads are for jobs that are or will soon be filled internally. The employer already knows who they will hire and either has made that decision and is not interviewing, or is interviewing just to "go through the motions".

For example, in IT a person is not going to be a super experienced Windows Dotnet software developer and a Linux/LAMP developer. Just doesn’t usually happen.

As an integration guy, that's exactly what I am (assuming the "P" is "Perl" and not that devil spawn "PHP"), plus Java. I've worked with a few other guys that were similarly experienced. I would posit that as an organization grows larger, the probability that they will have a heterogeneous computing infrastructure approaches 1 (which I not-so-modestly call "Cole's Law"). Therefore, if you work in a larger organization, you're more likely to have had exposure to multiple competing platforms.

This impacts the job ad postings because many contracting companies post for the same position

In the past few weeks I had two different placement firms trying to sell me the same position at another company. Those probably count as two open jobs.

I hate to always come back to this, but these data are consistent with a story about the decline of reallocation rates making labor markets less efficient at matching firms with workers. Permanent obsolescence of skills could cause the Beveridge shift, but so could inability of markets to reallocate. Probably a lot of both is going on, among other things.

Given secular trends in rates of worker flows, job flows, establishment and firm flows, and migration, we should not be surprised to see labor markets struggling. Hopefully we'll figure out what's causing all of this, but in the meantime I think it should be a source of caution among those in the blogosphere who are so quick to confidently dismiss structural stories about labor markets.

We've made it really expensive to hire someone, and even more expensive to fire someone. And it's really, really cheap to list a job.

I was going to say "How has this changed over the last couple of years?" But then I thought, well, yeah, we have made it a lot more expensive to hire in the past couple of years, at least at the lower tier of the labor market.

Who could have ever seen this coming?

Macro rule #1: You get the behavior you subsidize, whether you mean to or not.

But if it is so expensive to hire, then why even bother advertising the positions?

Over the past three years, the number of job openings has risen almost 50 percent, but actual hiring has gone up by less than 5 percent.

Absent some historical perspective, it's hard to know if this is even an important issue or not. What's the historical ratio of job postings to job hirings? And how do the current numbers compare with the numbers prior to 2007? And what are the quality metrics of the current job postings? Is there a higher than normal ratio of low wage to high wage postings? Etc.

Most jobs don't require that high of a level of skills. Many business owners I speak with complain about not being able to find applicants who can perform the basics, like getting to work on time.

Most business owners you speak with don't offer enough money.

Maybe not, but this wasn't an issue for them in the past.

My company has had a listing out for an assistant for me and a new person for my work-partner's team for about three months. The positions were approved and are desperately needed, but for budget reasons they were advertised at a salary that doesn't attract anyone useful. We have interviewed a handful of people, read dozens of resumes but deliberately haven't yet hired anyone. Our thinking is that if we fill these spots and then complain that we don't have the staff or skills to get everything done the company will blame us- after all, we chose our assistants. If we hold out for HR to admit that the applicants we're getting are not skilled enough to be useful then they may approve a higher salary so that we can get the staff we really need.

This process is very very frustrating and I think it is influenced by the fact that we added two people in 2009 who are more skilled and better workers than average for the price points we hired them at. The job market was scary horrible back then and good people accepted offers that they wouldn't today. Those hires have given my president an unrealistic view of what skilled people in my field cost.

From what I'm seeing in the piles of resumes, there are still a lot of unemployed people, but not many of the gems that were available three years ago.

Suggest to someone higher up that you could pay a reasonable salary for the people you need if they got rid of an HR flunky or two.

High labor mobility increases the specificity of hiring requirements.

If you hire somebody now, you have to assume they will move on within a few years. Thus on-the-job training makes less business sense. You can't choose somebody with not-quite-matching skills and train them up, because once you're done training they'll be ready to quit.

Hence you leave out an ad for your perfect employee. And until that unicorn turns up, you pay through the nose for contractors.

[I've seen this pattern very strongly in the tech industry. I'd imagine something similar to apply to other skilled and fast-changing fields, though it wouldn't be such a factor lower down the pay scale]

Experienced something similar back in the recession in the 80s that's always stuck in my mind. Had a boss who complained the people responsible for filling positions wouldn't hire anyone who didn't meet 13 of the 10 job requirements.

Expectations rise in a hirer's market, sometimes to absurd levels.

HR departments often require that jobs be posted publicly for a period of time in order to avoid discrimination lawsuits, even if they don't have any intention of hiring from outside. It makes filling out job applications annoying and depressing.

My company is going through a merger. Apparently there is someone who is designed to eliminate X jobs as a result. So each department created new positions that they haven't filled and aren't planning to fill, but when that person comes around they can "eliminate" the unfilled position without having to fire anyone.

BTW, I get an incredible amount of garbage job offer spam and its annoying. Does anyone actually use something like careerbuilder? The jobs they send me are just hilarious.

The words "dot net" and "C#" do not appear at all on my resume and they still try to make me apply for those jobs.

I thought the standard move here was to claim these are jobs Americans won't do at market wages and call for increased immigration.

Optimism and reality may have something to do with it. My firm (which is quite small) posted a job listing this spring. We fully intended to fill it, held interviews and everything. When it came down to it, though, the principle decided that business wasn't growing enough fund the position.

I have some anecdotal perspective on this since my company has posted a large number of positions this year that have been difficult to fill. Many of these positions are technical positions of various types. I can tell you what explanations frequently trotted out do *not* apply:

- Every single position pays well above the industry average as a matter of policy. I don't think there is a single person at the company who is not making at least six figures. Money is rarely an object for a quality candidate.

- We purposefully have few requirements. For example, the only requirements for software engineers tend to be functional familiarity (not deep expertise) with Linux and C++. We don't care about degrees, certifications, tools, and to some extent, even work experience. Mostly, we look for high-skill engineers that can learn new things, relevant experience doesn't really matter because we'll train that part. Our non-technical positions are the same way.

So why is it so difficult to find people? The key is "high-skill engineers". We literally have no use for lower skill engineers and it takes *years* to turn a low-skill engineer into a high-skill engineer assuming they have the aptitude. Some people have a high-level of skill straight out of school, others can work an entire career and never develop that level of skill. The issue of training is irrelevant -- we train people -- because many of the candidates we see are not meaningfully trainable in a timeframe that matters.

What is happening is that more and more of the hiring is for high-skill candidates across a broad swath of industry. Consequently, there is incredible competition for high-skill employees and a willingness to pay to get them but many companies have rapidly shrinking pools of positions for low-skill employees because those employees can provide no net value.

So our candidate flow is almost purely low-skill people, for which we have no use. To fill positions, we are essentially forced to poach high-skill people from other companies. There simply are not enough high-skill people being produced every year to satisfy the growth in demand for high-skill positions and away from low-skill positions.

I think this is the best comment on here. It's an articulate and more detailed version of my "smart people shortage" theory. Part of that is the growth in demand for "smart" and part is the decline (or at least failure of commensurate growth) in supply.

Why would you assume you are paying above market if the market isn't clearing? Isn't that the very definition of not paying enough?

The absolute demand significantly exceeds the absolute supply. There is no ability to produce more supply that would come close to meeting the demand and the demand is growing so the price of that labor is set by what the demand side can afford. One of the reasons we pay as much as we do is to attract high-skill people from companies that also need high-skill employees but cannot afford to pay as much. In my city, and I am not in Silicon Valley, we are now regularly seeing companies offer experienced high-skill software engineers $200-250k base pay and outliers can do much better than that.

Every time we hire an engineer, all that we've done is move the job opening to a different company. Consequently, the total number of open positions would be expected to grow even if we were paying $1 million base salary, the openings would just be distributed differently. The rate of pay increases in tech where I live averages 5% annually, and that is being distributed disproportionately to the high-skill end of the spectrum so the growth rate is even higher there.

I can verify the same thing. Our company has a large number of open positions we can't seem to fill. Most of them are specialized technical positions. We don't just need 'programmers', we need people skilled in the use of large datasets, or security experts, or other people fluent in technical areas that are required for our product development.

I get very annoyed when the level of detail in economic discussion of labor uses high level aggregates to classify workers. Even discussions of "STEM" grads vs non-stem are not nearly precise enough to understand where the dislocations are. A computer engineer is not the same as a person with a degree in Biology or environmental science, yet they're all "STEM" grads. And even if you try to narrow it down to the level of 'computer engineer' you miss the wide degree of specialization within sub-specialties of computer engineering.

I think the problems of labor market dislocation right now are happening at a very fine-grained level that you'll simply miss if you're only looking at broad aggregates or high-level labor classifications.

I visit factories as part of my job, and the first thing that jumps out at you is the age of the work force - one factory I was in recently had an average worker age that was over 50!. Why? Partly because of unionization and seniority rules, but also because according to the factory manager it was very hard to find quality young applicants for open positions. They're not looking for people with degrees in general science or arts, or business majors or people with degrees in journalism or communications. They're looking for machinists, technicians, people trained in quality control, electricians, and other skilled factory workers. And they can't find them.

They see this as a crisis in the making as their old workers retire, so this is not 'fake' hiring or over-stated job requirements. This is simply a mismatch between the needs of the factory and the qualifications of the labor pool in the surrounding area. My understanding is that this is becoming a common problem.

FWIW, I attended this event at Brookings -- -- and if memory serves me right one of the speakers, Eric Roegner of Alcoa, claimed the company would create "thousands" of jobs if it could just find people with the right skill set. More intriguingly, I believe he also claimed that they didn't need math geniuses, but just people with basic skills who could pass a drug test were difficult to find.

Maybe they should stop testing people for marijuana use.

But how do they try find these people?

"Apply at our website where your resume will be filtered out before ever having a human see it!"

Sounds a lot like Tyler's ZMP thesis.


My department recently hired three transportation planners. We're in the exurbs of a fairly large city with low unemployment, so there's strong competition for good talent. We got over 70 applications, but when we started scheduling interviews the day after the opening closed half the top ten had already accepted other jobs.

We also got a lot of applications from people who didn't bother to read the job description.

What sort of proof are you asking of your candidates to verify that they're skilled? I know several young, very intelligent and adept programmers (who I think would expect meet your description, and whom I'd like to inform of you), who've nevertheless been unable to land even a junior position because of the experience paradox. How does an employer identify a candidate as trainable if they haven't been trained?

Is it even possible to use a resume to signal the requisite psychological characteristics? Obviously it's necessary to use some kind of filter to reduce the swarm of unqualified applicants who descend, locust-like, on any advertised opening in this economy, but what steps can be taken to avoid throwing out the untrained baby with the untrainable bathwater? And, asking out of purely selfish concern, what signals trainability?

We use resumes, then call people in for interviews. We don't necessarily look for lots of experience - in fact, most of the people we've been hiring lately have been fairly new grads for the simple reason that we can't find the people with experience so we've had to lower our standards.

One of our more successful hiring initiatives involves mentoring 3rd and 4th year Computer Engineering students. We take them on from Co-op programs or as summer students, and then we get to evaluate them for months to discover if they actually have what it takes. In the meantime, they gain work experience and build their real-world skills. Most of them are sent on their way at the end of their time, but the best ones get job offers at good salary. This is a fairly common practice in the computer industry, and it would be a good thing if other industries adopted the same pattern - if the government and the unions would let them.

Which brings me to another problem: A lot of the new grads I'm interviewing are shockingly ignorant. They can't answer very basic questions of computer science. Maybe that's why employers are looking for lots of experience - they don't trust the quality of the education kids are receiving these days.

Also, you mentioned you know 'programmers'. But that's kind of like saying "You're looking for Neurosurgeons? I know a couple of doctors." Not all programmers are the same. People trained in business applications or web development are next to useless when it comes to, say, system level programming or scalable database development or enterprise security.

This is what I'm talking about when I say the problem is fine-grained dislocation. There are real shortages of people with specific skill sets. For example, the "internet of things" or the 'industrial intranet' is a rapidly growing area. The result is a real shortage of security experts, low-level programmers who can write software for PLCs and microcontrollers, 'big data' database developers, scalable system architects, and other specialties. If you're a programmer and all you've built are web pages or Visual Basic business apps, you're simply not qualified for many of these jobs.

In fact, I'm a programmer myself, so I'm aware of the fields in question, but your point stands. I should've said that I have acquaintances with strong fundamentals and functional familiarity with C++ and Linux, and some with other specializations, who've never found anyone willing to incur the risks needed to discover what I know, which is that they are extremely "trainable."

It may take months to evaluate a candidate, but how do you choose which ones to evaluate? Your company apparently isn't one of them, but elsewhere in this thread, it's been mentioned there are companies who only appear to require competent and intelligent generalists (acknowledging that "generalist" is a relative term), and are still disappointed. Obviously, a method for identifying intelligence is an intractable problem, but it seems to me that an entire generation is going to waste because whatever method these companies currently use produces too many false negatives.

If you require experience with a specific technology, you are screwing yourself. I sometimes considered applying for positions but stopped when they said "required 5 years with MUMPS" or something like that.

Maybe it's worth my time to hunt for the diamonds in the rough by trying to jump through a company's broken hiring process (the food is better at places with ugly waiters), but a dysfunctional hiring process can often indicate a dysfunctional company culture.

The more specifics in a job description, the more I think the company is trying to avoid bad outcomes than catch good outcomes. That is not a good place to work.

This is what you'd expect to see if competition has gone up overall. A business might not survive with an extra employee (or it might not be worthwhile, the margins are lower) unless they happen to find the purple unicorn.

"If so, why has the demand for (near) perfection gone up?"

If this is different from 10, 20, 30 years ago, one thesis is that the cycle of trust between employers and employees has been broken. What is the average time of employment with a company today vs. 20 years ago? People don't trust who they work for to keep them and companies don't trust their employees to stick around long. In consequence, companies are averse to hiring someone they aren't going to get a positive return on from month 1. If the expected length of employment were 5+ years, they'd be willing to hire someone it took 18 months to train, but if the expected length is a single year, it doesn't make sense to hire someone who isn't productive from the get go.

I hired 4 people last year. This post inspired me to go look at the free sites where I posted the openings. The openings are still listed there, but way at the back. I still get a very occasional resumé.


Skills mismatch or a decline in saleable skills? My industry (mechanical trade), faces a severe shortage of skilled labor, albeit considered “blue collar”. A highly skilled tradesman in my market can earn well north of $70,000; in other markets $100,000 + not uncommon. I cannot blame cultural or societal acceptance of the blue collar trade entirely but rather lack of low cost vocational training opportunities.

If employers really wanted to fill all of these positions they've advertised, wouldn't we see rising wages and company training/apprenticeship programs?

We will. But for now, when profit margins are still strong, whining is free.

Except for a brief post-dotcom recession, we had over ten years of unprecedented economic success. During this time, recruiting switched from print classifieds (and the internet in the tech sector) to low-cost internet ads. Due to the low unemployment during most of this time period, an extensive recruiting industry developed to find candidates that might not otherwise apply. Others have noted HR trolling for purple unicorn employees or advertising positions externally that will be filled internally. Recruiters will also advertise non-existent positions to develop a stock of pre-vetted employees for openings that may occur later. So we have cheap, persistent job ads, often for nonexistent positions, by a larger recruiting industry. That is the major change in the last few years. There are other stories:

1. Skills mismatch. The manufacturers I visit here (in a large Southern state) bemoan the lack of skilled workers, but, outside the military, the only extensive training programs are up North. On the flipside, white-collar employers have not yet accepted the fact that over-qualified job-seekers will accept low wage, entry level jobs or they refuse to deal with high turnover. Given the transition of so many manufacturing jobs from the North to the South during the last few years, the former skills mismatch can be expected to persist for sometime until people relocate. As for the other skills mismatch, others have discussed it better. I should also note the experience conundrum. Since so many companies automate resume review, experience, even in low-skill positions, often precludes employment. For example, retailers wanting three prior years of retail experience.

2. . Child custody and geographic servitude. With continuing high divorce rates and the growth of non-marital family structures, many people are bound for a term of years to their present location or they can give up involvement in their child's life. I posit that this involves a change in parental preferences exacerbated by an increase in child custody arrangements.

3. The increasing use of credit scores for job applicants. Once you fall off this track, it is very difficult to get back on, especially in light of the above.

"Except for a brief post-dotcom recession. . ."

Employment number for young have ranged from bad to worse to unfathomably horrible ever since the "brief post-dotcom recession." People in their early 30s and younger have never experienced a healthy, vibrant job market.

The measurement question is important. We have job openings posted to our website, marketed through recruiters, and posted to various job boards. If you just do a simple count of jobs on the job boards, you'll get multiple counts for the same jobs.

Sincere question: Why do economists entertain the idea of a skills gap when wages are flat?

I am currently contracting and I'm looking for a permanent position. I am seeing that employers are looking for perfect fits. They want the exact experience in the job description and generally are not willing to pay remotely near what an experienced person would expect. Hilariously, a recruiter recently told me I would not be considered as I did not have Epic (clinical software) experience. She told me they were having a hard time finding people with those skills. I bit my tongue and did not point out that apparently you can't get Epic experience if you don't have Epic experience. Close is not good enough in this market. I have also noticed positions posted for a long time and re-posted. Hiring managers seem to have unreasonable and inflexible expectations.

Recruiters tend to take all the bad things that HR departments did, and then magnify them to an incredible degree.

They have no idea what close substitutes are for Epic or what, besides Epic, would be a good signal that someone could pick it up quickly. They want the exact right thing right now.

The perception is that a new mediocre experienced employee does not contribute much more than a current employee equipped with a search engine.

Thus the HR departments are holding out for the "purple squirrel".

Where I live software engineers are having to take time off between jobs to learn new technologies.

With other factors contributing as well, this can be considered a problem of frictional and structural unemployment. With the latter being the more severe as you refer to above. Some interesting policy changes to awknowledge are things like the staple act, which would staple a green card on foreigners diplomas who specialize in stem fields; or private or public provided child care at business offices, which would attract more young parents and enable them to realize a higher level of effort toward their careers. Nothing is bullet proof but just a couple ideas to chew on.

So are employers training and apprenticing more people?

Comments for this post are closed