Let's begin by looking at two key maps created by the MPI's Zara Matheson. The maps are based on data for all U.S. metros, but identify the 46 largest metros (those with more than one million people) by name.
The first map (above) shows average hours worked per week. Among large metros, Washington, D.C. tops the list with an average of 36 hours per week. Las Vegas is next, followed by San Antonio, Virginia Beach, and Seattle. Kansas City, Nashville, Miami, Phoenix, and San Jose round out the top ten. At the opposite end of the spectrum are Buffalo, Rochester, Minneapolis, Columbus, Philadelphia, and Pittsburgh.
The second map charts the highest-earning metros -- measured in terms of average hourly earnings. Among large metros, San Jose -- Silicon Valley -- tops the list, followed by San Francisco, D.C., New York, Minneapolis, and Philadelphia. The lowest-earning metros are Oklahoma City, Las Vegas, Louisville, Buffalo, Memphis, Jacksonville, Cleveland, and Pittsburgh.
What's most striking is how little overlap there is between the two maps. The locations where people put in longer hours are also the ones where workers earn less.
So, I decided to take a closer look at what might be behind these patterns. With the help of Charlotta Mellander, we ran a series of scatter-graphs and performed a simple correlation analysis to probe the effects of various demographic and economic factors on metro-level working hours and earnings. We ran the analysis both for all 340 U.S. metros and for the 46 large metros (again, those with more than one million people). As usual, I point out that our analysis points to association between variables only. It does not imply causation, and other factors may complicate the picture. Still, as with states, a number of patterns are striking.
First and foremost, we find a total lack of correlation between hours worked and earnings across U.S. metros. This holds true both for the largest metro areas and for all U.S. metros: there was no statistical significance at all for the correlation between these two variables in either case.
Second, hard work and long hours do not translate into economic wealth. There was no correlation whatsoever between working hours and economic output, measured as gross metropolitan product per capita, for either large metros or all metros.
Third, as we found in our earlier analysis of states, when it comes to earning power, working smarter trumps working harder across the board.
Source: Human capital data from the U.S. Census Bureau, 2006.
One way to measure smart work is by the level of human capital -- that is the percentage of a metro's workforce with a bachelor's degree and above. Human capital is closely associated with metro earnings. The correlation is .53 for all metros and even higher, .74, for large metros. Silicon Valley -- that is the San Jose-Sunnyvale-Santa Clara metro -- literally jumps off the chart. San Francisco, Seattle, D.C., greater New York, Minneapolis, and San Diego are all above the fitted line in the upper-right hand quadrant of the graph: they combine above-average human capital with above-average wages. Louisville, Oklahoma City, Memphis, Pittsburgh, Milwaukee, and Nashville are all below the line -- combining low levels of human capital with low wages -- wages, in fact, that are even lower than their human capital levels would predict. Interestingly, Detroit, New Orleans, Phoenix, and Las Vegas have significantly higher wage levels than their human capital levels would predict.
Source: Creative class data are from the Bureau of Labor Statistics, 2006. Creative Class definition as in Rise of the Creative Class.
Another way to gauge smart metros is by the share of their workforce in creative, professional, and technical jobs -- that is, metros with high percentages of workers in the creative class. Creative class metros have significantly higher average earnings. In fact, the correlation between earnings and the percentage of workers in creative class jobs is slightly higher than that for human capital -- .58 for all metros and .78 for large metros. The line on the scatter-graph runs quite steeply upward. Silicon Valley is again a huge outlier, way up in the right hand corner of the graph. San Francisco, Seattle, and greater New York are all above the line -- with high levels of creative class work and high average wages. D.C., interestingly enough, is slightly below the fitted line: its average wages are slightly less than its percentage of creative class workers would predict.
Source: Working-class data from Bureau of Labor Statistics, 2006. Working-class definition as in Rise of the Creative Class.
Earnings are far less in more traditional industrial economies. Metro earnings are negatively correlated with blue-collar, working-class jobs -- the correlation is -.31 for all metros and -.5 for large metros.
Source: Data on share foreign-born are from the U.S. Census Bureau, 2006.
Earnings are also associated with open metros. One measure of openness is the share of immigrants or foreign-born people. Over the past several years, we've seen growing anti-immigrant sentiment in some quarters of America. Such sentiment is fueled by notions that immigrants take jobs away from American-born workers, drive down wages, and pose economic and fiscal burdens on states and cities. Our analysis suggests such notions are unfounded. Instead, we find that metropolitan earnings are positively associated with the percentage of immigrants. The correlation is .34 for all metros and .57 for large metros.
Another gauge of openness is the percentage of gay and lesbian people living in a metro. Gay marriage may remain a political hot button, but our findings show that locations with larger gay populations are considerably better off economically. Our measure of gay populations is the gay index -- initially developed by Dan Black, Gary Gates, Seth Sanders, and Lowell Taylor, which we updated based on more recent data. The correlation between the gay index and earnings was .4 for all metros and .56 for large metros. It's hard to specify the exact connection between gays -- or immigrants -- and metro-level earnings. It might be that higher-paying metros are more attractive to gays, immigrants, and other Americans. My own view is that more open locations are better able to compete for more talented and skilled people across the board.
Source: Data as in Rentfrow, Gosling, Potter (PDF), 1999-2005.
There's one more data point that reinforces the connection between openness and earnings. We looked at the correlations between earnings and what psychologists have dubbed the "Big Five" personality types -- conscientiousness, agreeableness, extroversion, openness-to-experience, and neurotics. Only one of these five types had a positive effect on earnings: openness-to-experience. The correlation was .23 for all metros and .3 for large metros. Open-to-experience people, as the name implies, are exceedingly "open" to new experiences of all sorts. They are highly clustered in urban, bohemian neighborhoods and are high on creativity and innovativeness. Interestingly, three other big-five personality types -- extroverted, conscientious, and agreeable people -- were all negatively associated with earnings. "One possible explanation for these findings," notes Jason Rentfrow, a personality psychologist at Cambridge University who has studied the relationship between personality and locations, "is that the knowledge economy is driving growth, and creative, resourceful, and imaginative people are crucial to that growth."
Source: Happiness data are from the Gallup Organization, 2009.
That brings us to happiness. A big question is how does the trade-off between hard work and earnings affect the happiness and well-being of metros? The answer is pretty straightforward. Earnings play a bigger role. The correlation between metropolitan well-being and earnings is positive and statistically significant: .41 for all metros and .67 for large metros. The correlation between happiness and hours worked is negative for all metros (-.2) and statistically insignificant for large metros. This pattern is different than that for states, where the negative correlation for hours worked was stronger than the positive one for earnings.
Once again, we find that working smarter, and not working harder, is what brings wealth and well-being to metros. Longer work hours do not translate into more wealth or higher earnings. Higher earnings turn on higher levels of human capital and higher levels of creative class jobs. Smart metros are also doing much much better in surviving and prospering in the Great Reset. Our findings question the notion that protecting or creating more blue-collar factory jobs will lead to better earnings and improved economic circumstance. Rather, we find that metros with high levels of working-class jobs have significantly lower average earnings.
We also find a close connection between openness and tolerance and earning levels. While some continue to cling to the belief that immigrants drag down wage levels, that's not what we find at all. To the contrary, metros with higher levels of immigrants have higher average earnings. So do metros with greater concentrations of gays and lesbians. It's impossible to pinpoint exactly what is causing what here -- if richer locations offer more opportunity to immigrants, gays, and everyone else, or if these more open locations reflect an underlying economic system and set of industries that generate higher earnings. The way I see it, more open and tolerant metros benefit from their enhanced ability to attract ambitious and skilled people of all sorts and from all backgrounds, races, ethnicities, and sexual orientations. And, smarter metros not only generate higher earnings, they provide greater levels of happiness and well-being to their residents.
It's high time we put aside the notion that hard work is the key to economic success. It's working smarter, not working harder or longer, that drives both earnings and economic development. The president, his economic advisors, and city officials need to keep this top-of-mind in their ongoing efforts to create jobs and spur economic growth.
This article available online at: