THE AVENUE AUGUST 29, 2012
For the first time since World War II, there are fewer jobs three years after the end of a recession than before it began. Our new Brookings report suggests that most of this flat recovery can be attributed to severe losses in housing wealth and jobs in industries such as manufacturing and construction.
Yet education--especially the balance between the demand and supply of educated workers--is the most important factor explaining long-run unemployment in metropolitan and national labor markets.
First, consider the short-run picture. As of the first quarter of 2012, the economy was down 5.1 million jobs from the first quarter of 2008. 71 percent of that jobs deficit--3.7 million jobs--is attributable to just two sectors: construction and manufacturing, which made up only 15 percent of all jobs in 2008. The massive losses in construction jobs devastated metro areas like Las Vegas, while manufacturing losses crippled Detroit and Wichita. Since manufacturing is so export-oriented, it has not helped that growth in Europe--a large trading partner--has been dismal in recent years.
Likewise, housing prices are still down 17 percent from their peak in 2007, and by even larger margins in California and Arizona metro areas. Households have lost roughly $4 trillion in wealth and associated spending power, further holding back job growth in construction and other areas.
Indeed, our analysis finds no evidence to support the popular idea that the recovery is slowed because employers can’t find workers with the necessary skills to fill job vacancies. In the 100 largest metro areas, roughly one-third of all jobs take more than one month to fill, but this ratio has not increased since 2006, is uncorrelated with unemployment rates, and was roughly the same in the 1960s, according to a 1966 study. Moreover, the ratio of job openings to new hires has been no different post- Great Recession than during the housing boom years. There just are not enough openings.
But this doesn’t mean that education has no important short-term effects. Educated workers present short-term advantages to metro economies in terms of higher spending power--which boosts demand for local services provided by less educated workers--and higher rates of business creation. College-educated workers are twice as likely to own a business that employs workers as adults with no more than a high school diploma. As a result, metro economies that are well educated have a significantly higher rate of job openings, higher business growth, and higher job growth.
Over the long term, education has a huge effect on metro and national labor market health. Here, the relationship between the supply of and demand for educated workers is especially important. To measure mismatch between supply and demand in a given metro area, I calculated the average years of education required by vacancies using the Conference Board’s Help Wanted Online series and compared it to the years of education attained by the average worker. No single factor more strongly predicts metro unemployment than the “gap” between those two measures. The unemployment rate difference between those metro areas that are above average compared to those below on this “education gap” index is two percentage points. This is how metro areas like Madison, Washington D.C., Boston, and Charleston S.C. have maintained low relatively low unemployment rates, compared to places like El Paso and Bakersfield.
The mismatch between supply and demand means that there are far too few job openings for less educated workers. In 2007, there 2.9 job openings over the course of the year for each unemployed worker with a high school diploma or less. In 2011, there were just 1.6 openings. Things also got worse for educated workers. In 2007, there were 12 openings for every unemployed worker with a college degree or higher, compared to just 5.6 in 2011.
Yet, these opportunities vary widely across metro areas. In Madison, there were five job openings in 2011 for every unemployed worker with a high school diploma or less. In Riverside, California and McAllen, Texas, there were just 0.5.
The larger context here is that changes in the U.S. economy have created an increasingly specialized workforce, and the infusion of new technologies--such as computer software--has played to the strengths of highly educated workers. Supply used to keep up with this extra demand for education, but there are signs it is faltering. From 1950 to 1980, growth in the number of people with college degrees proceeded rapidly at 5 percent each year, but since 1980, that growth has slowed to 3 percent. Not only has this exacerbated income inequality by increasing the wage premium for college education, but also it has stifled job growth, which slowed from 2 percent to 1 percent over those periods, despite a steady growth rate of people 25 years and older. It’s no surprise, then, that highly educated metros have enjoyed the largest benefits from these trends.
From this perspective, the politically contentious debates dominating Washington about whether the economic stimulus package was too big or too small, whether ObamaCare--or other policies--have increased investor uncertainty, or how quickly we should balance the federal budget seem short-sighted. Over the long run, boosting educational attainment is the only way to lower unemployment, especially in the nation’s hardest-hit regions.