In discussions about politics and economics, we see a lot of one stage thinking. Virtuous arguments are made without considering factors below the surface or beyond stage one. If we consider these factors, we can fully articulate the issue at hand and make an informed opinion.
Google “college degree compensation” and most of the results will tell you that college degree holders unquestionably earn more than high school. Although true, there are other factors that may change your opinion on the value of a college degree.
Economist Enrico Moretti published award-winning research in the American Economic Journal about the wage inequality between skilled (college degree) and unskilled (high school or less) labor.
According to Moretti, “real wage differences between college and high school graduates have grown significantly less than nominal differences”. Why? The geographic location of different skill groups is driven by relative demand within specific cities, towns, and states. Let’s do a thought experiment to explain.
Bob has a degree in finance. He lives and works in New York City where there is a demand for his labor. As a result, he has a higher cost of living relative to, for example, Lincoln, NE. Bob’s income is indexed accordingly and he earns $100K, which is a lot different in Lincoln to say the least.
Joe barely finished high school and works in agriculture in Lincoln. His nominal income is $50K per year, but his cost of living is also less. So who has more utility and who is “better off”?
Statistical categories, especially at the national level, might lead you to the conclusion that Bob earns twice as much as Joe because he has a college degree. In nominal terms, you would be correct. If we evaluate compensation in real terms, we get a different outcome.
Moretti’s research does two things to find the real difference. First, he develops two Consumer Price Index models that more completely evaluate the cost of living at the local level, as opposed to a national standard. Second, he tracks the college share of labor in municipalities to determine their geographic trends relative to the college premium.
From 1980 to 2000, the college premium increased 20% in nominal terms. In real terms, however, the wage premium increase is significantly less.
In 1980, the cost of living difference between high school and college graduates was only 4%. By 2000, this difference grew to 14%. Moretti found that, using his local CPI, cost of living accounted for 25% of the nominal wage premium for college graduates. The real wage difference is smaller than face value, but the analysis is still not complete.
Utility premium is the next important factor. Going back to our example, if Bob moves to NYC simply because he is following labor demand and doesn’t actually want to live there, then he does not gain any utility from his college degree premium. In other words, the real wage difference is even smaller because living in NYC costs Bob something that does not show up in statistical categories. If Bob moves to NYC under Moretti’s supply push hypothesis (amenities, friends, etc.), then he gains utility and the real wage difference is larger.
So which is it? Does demand pull or does supply push?
Under relative demand, there would be a positive relationship between the college wage premium and college labor share – as one goes up, so does the other. Under relative supply, there would be no such relationship since college grads would be moving to metropolitan areas regardless of demand.
Moretti found a strong positive relationship between college premium and college share. These findings indicate that his demand pull hypothesis of skilled workers to metropolitan areas is valid. From his study, “the empirical evidence suggests that relative demand shocks played an important role in driving changes in the number of skilled workers across metropolitan areas”.
In other words, college graduates concentrate in higher cost of living areas not because they want to, but because that is where they can find work. As previously stated, this indicates an overall loss in utility, thus reducing the real difference in college vs high school income.
To be clear, this does not completely rule out his theory of relative supply. College graduates surely move to big cities because of their amenities, but, empirically, relative demand is more influential. Finally, demand for college grads has disproportionately increased in metropolitan areas relative to lower cost of living areas.
CONCLUSION: While a college degree might have a nominal wage premium, the real premium is significantly smaller than we are commonly led to believe. It is important to always think beyond stage one and consider factors that might not be obvious at first glance. We can apply this same methodology to a multitude of issues.
SOURCE: Moretti, Enrico. 2013. “Real Wage Inequality.” American Economic Journal: Applied Economics, 5(1): 65-103.