More Gains to Moving for College Educated Workers
More Gains to Moving for College Educated Workers
By Austin Herrick
Internal migration of working-age people in the United States has fallen by more than a third – from 18.86 percent in 1994 to 11.44 percent in 2018 (see Figure 1). This phenomenon has received significant scholarly attention – Cooke (2013)1 for example, attributes this decline to the rise of information and communication technologies. Alternatively, Molloy, Smith, and Wozniak (2011)2 point to broad macroeconomic shifts.
Reduced internal migration has important economic implications, particularly for labor markets. We find that wages grow faster for movers, especially for those with a college degree.
Source: Current Population Survey (1994 – 2018)
Figure 2 below shows observed wage growth for movers and non-movers over the last twenty years. As seen by the difference between the red and blue solid lines, the wage premium3 for moving over the last two decades has remained relatively constant, despite recessions (shaded gray) during the intervening years. The wage premium for moving suggests that migration improves matching between employers and employees.
Note: The shaded blue and red ares represent 95 percent confidence intervals.
Source: Current Population Survey (1994 – 2018)
We might expect declines in migration to be correlated with changes in the labor market. As labor markets weaken, the wage premium for movers may decrease. Firms may be less willing to pay a premium to attract new workers in an economy where many current workers are being laid off.
The available data has some limitations. We only observe movement behavior of individuals who make the decision to move, which fewer people did in the wake of the financial crises in 2001 and 2008. It is possible that the wage premium for moving for all workers has decreased over time, even as the marginal wage premium for moving has remained steady. That is, given fixed costs of moving, workers offered lower wage premiums for moving decided against movement altogether, biasing upward the average wage premium for moving. The premium’s increase in recent years indicates labor markets seeking new talent in a population of households still wary about movement.
This suggested pattern of firms searching for new movers becomes more plausible when examining movement patterns of adults based on educational attainment. In Figure 3, we see that the movement rate for college-educated adults fell significantly faster than their non-college educated peers beginning in the early 2000s. Interestingly, this decline stabilizes around 2011.
Source: Current Population Survey (1994 – 2018)
Further evidence of education’s effect on movement is shown in Figures 4 and 5, which display the wage premium of movers, conditioned on college degree attainment. From Figure 4, we see that the movement premium for non-college educated workers is often not statistically significantly above 0. Companies seem relatively unwilling to offer these less-educated workers a significant premium to induce movement, largely independent of economic conditions.
Note: The shaded blue and red ares represent 95 percent confidence intervals.
Source: Current Population Survey (1994 – 2018)
By contrast, in Figure 5, we see a highly significant wage premium for moving offered to college-educated workers. The data seems to reflect that college-educated workers are the primary recipients of the observed wage premium for movers.
This conclusion matches the results found by .4 In that study, Gould finds that higher urban wages are primarily driven by white-collar workers (a rough proxy for college education). Critically, the leveling-off of the college educated movement rate observed around 2011 maps onto the significant increase in the wage premium for moving observed for these workers in Figure 5. This observation supports the conclusion that wage premiums are successful in inducing movement.
Note: The shaded blue and red ares represent 95 percent confidence intervals.
Source: Current Population Survey (1994 – 2018)
We’ve identified that the wage premium for moving influences the decision to move. However, declining premiums only explain some of the puzzle - increasing premiums only stabilize, rather than reverse, internal migration rates, and movement has fallen dramatically over the past two decades. This suggests significant non-labor market factors affecting movement for American workers. We know the decision to move is influenced by the cost of moving and housing, among other things. A forthcoming post will explore the reasons why people report that they move.
-
Cooke, T. J. (2013). Internal Migration in Decline, The Professional Geographer, 65(4), 664-675. doi:10.1080/00330124.2012.724343 ↩
-
Molloy, R., Smith, C. L., & Wozniak, A. (2011). Internal Migration in the United States, Journal of Economic Perspectives, 25(3), 173-96. ↩
-
The wage premium measures the ratio of wage increases for individuals who moved compared to non-movers. For example, if the average non-mover received a 2 percent raise, compared to a 3 percent raise for movers, the wage premium is 50 percent. ↩
-
Gould, E. D. (2007). Cities, Workers, and Wages: A Structural Analysis of the Urban Wage Premium, The Review of Economic Studies, 74(2), 477–506. doi:10.1111/j.1467-937X.2007.00428.x ↩