Local job multipliers revisited

Abstract

There has been a recent surge in papers estimating local multiplier effects. However, existing studies rely on arbitrary periods of observation, limit samples to more populous regions, and commonly use relatively aggregated industrial categories. When we address these and other methodological issues, we find that, in the United States, each new traded sector job adds half a nontraded job to a local economy, and that the addition of each high-tech job adds less than one job to the local nontraded sector. Furthermore, we find that the multiplier effect of the manufacturing sector is no higher than the multiplier effect of the average traded sector. We provide robust evidence that higher-paying traded sectors yield more nontraded jobs than lower paying sectors, and that multiplier effects are higher in larger cities. Furthermore, we generate IV estimates that remedy weak instrument problems in the existing multipliers literature. These findings offer needed clarity on the likely employment impacts of incentive policies aimed at attracting industries in the traded sector of the economy.