Paying Skilled Workers More Would Create More Skilled Workers
Harvard Business Review, 19 May 2016
Thijs van Rens
When computers made their way into workplaces in the 1980s, typists had a problem. As computers replaced traditional typewriters, the skills of typists who did not know how to work with a word processor grew obsolete. Nevertheless, few would argue that information technology permanently increased unemployment. While the unemployment rate spiked in the ‘80s—as it typically fluctuates sharply between recessions and booms—it went back down, so that the average unemployment rate in the 1990s was similar to that in the 1970s. The labor force adjusted to a superior new technology replacing an older one.
In the wake of the 2008 financial crisis, there was a lively debate among policy makers and academics about whether a similar gap “between the skills workers have and the skills businesses say they need” contributed to the increase in unemployment. Research has since shown that the skills gap has a cyclical but large effect on unemployment, explaining as much as one-third of the increase in unemployment following the Great Recession.
It is usually taken for granted that the skills gap is a problem of skills supply, and public concerns often focus on a lack of STEM skills and soft skills. So proposed solutions tend to involve reforming education and worker training programs. The most popular approach has been to reduce tuition fees for selective fields of study, usually STEM majors.
However, I argue that this view is not correct. Research suggests that the main reason that the skills gap persists is that employers are unwilling or unable to pay market price for the skills they require.
There are three possible reasons for why a skills gap exists: 1) workers do not adjust to changes in the demand by acquiring new skills, 2) employers do not take the supply of skills into account when they take hiring decisions, or 3) employers do not take into account the relative shortage or abundance of particular skills when they set wages. Using U.S. data on job finding and filling rates, wages and profits across states and industries since 1979, we measure the contribution of each of these three reasons on mismatch unemployment. We find that wage setting is the main culprit.
The workforce can adjust to changes in the demand for skills by acquiring new skills, through training or by replacing older workers with younger ones with up-to-date skills. For example, an unemployed typist looking for work in the 1980s could learn how to use a computer or fill a vacant position left by another typist who moved on to another job or retired.
Firms can also respond to changes in the supply of skills. In the ‘80s, for instance, organizations could train their typists in word processing, or keep some typist positions open. While hiring less skilled workers hurts a firm’s productivity, the data show that companies still did this, in order to take advantage of the fact that hiring these workers is so much cheaper.
Our data show that these kinds of adjustments do indeed happen, and that they happen fast enough to prevent unemployment from going up. There aren’t many occupations that are both easy to find and at the same time high paying, which is what we would expect if the workforce were not adjusting and companies were struggling to find talent. Similarly, there are few jobs that are easy to fill and at the same time generate high profits for the company.
Yet the skills gap still remains. This is because the adjustments that workers and firms make will only eliminate it if wages reflect the relative supply and demand for various skills across occupations. But our data show that this is not happening: many jobs in industries that generate high profits (retail trade, educational services, mining and forestry) tend to pay low wages and are therefore unattractive to workers, whereas jobs in industries that pay higher wages (finance, computer and electronics manufacturing, paper and printing) are not very profitable.
Imagine that a particular set of skills, say STEM skills, enables workers to be particularly productive, but their pay does not go up to reflect this higher productivity. Then it is not surprising that workers do not acquire more of these skills, since they do not reap any of the benefits of their increased productivity. In the UK for instance, less than half of STEM graduates work in scientific occupations, and there is no wage premium for having a STEM degree in other occupations.
Firms, on the other hand, are more interested in hiring workers with these STEM skills, as they are not only very productive but also cheap. Thus, companies open lots of vacancies for STEM positions, but then find it very difficult to fill them.
Companies often advocate for better education to fix the skill gap, but our results indicate that this is unlikely to work. The reason is simple: students have a choice about what skills they acquire in school, and how they use these skills in the labor market. Encouraging universities to educate more physicists and engineers will not make a difference if these additional STEM graduates then choose to work for investment banks that offer higher salaries.
Unfortunately, our research does not provide an explanation for why wages do not reflect relative labor market conditions across occupations or skills. However, the data clearly indicate that wages for workers with scarce skills are too low compared to wages for workers with a more abundant skill set. It would seem that this provides a profitable opportunity for companies that are able to be flexible in their compensation policy. By paying more for certain skills, an employer would have no trouble attracting workers with those skills in sufficient quantity and of the highest quality, giving it an undeniable edge over its competitors.