Working Paper: NBER ID: w21120
Authors: Enrico Moretti; Daniel Wilson
Abstract: Using data on the universe of U.S. patents filed between 1976 and 2010, we quantify how sensitive is migration by star scientist to changes in personal and business tax differentials across states. We uncover large, stable, and precisely estimated effects of personal and corporate taxes on star scientists’ migration patterns. The long run elasticity of mobility relative to taxes is 1.6 for personal income taxes, 2.3 for state corporate income tax and -2.6 for the investment tax credit. The effect on mobility is small in the short run, and tends to grow over time. We find no evidence of pre-trends: Changes in mobility follow changes in taxes and do not to precede them. Consistent with their high income, star scientists migratory flows are sensitive to changes in the 99th percentile marginal tax rate, but are insensitive to changes in taxes for the median income. As expected, the effect of corporate income taxes is concentrated among private sector inventors: no effect is found on academic and government researchers. Moreover, corporate taxes only matter in states where the wage bill enters the state’s formula for apportioning multi-state income. No effect is found in states that apportion income based only on sales (in which case labor’s location has little or no effect on the tax bill). We also find no evidence that changes in state taxes are correlated with changes in the fortunes of local firms in the innovation sector in the years leading up to the tax change. Overall, we conclude that state taxes have significant effect of the geographical location of star scientists and possibly other highly skilled workers. While there are many other factors that drive when innovative individual and innovative companies decide to locate, there are enough firms and workers on the margin that relative taxes matter.
Keywords: state taxes; geographical mobility; high-income earners; star scientists; migration patterns
JEL Codes: H71; J01; J08; J18; J23; R0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| 1% decline in after-tax income in the destination state (H73) | number of star scientists migrating from the origin to the destination state (F22) |
| personal income taxes (H24) | geographical mobility of star scientists (J61) |
| corporate income taxes (G30) | geographical mobility of star scientists (J61) |
| investment tax credits (H25) | geographical mobility of star scientists (J61) |