Governments go to great lengths to attract foreign multinationals because they are thought to raise the wages paid to their employees (direct effects) and to improve outcomes at local domestic firms (indirect effects). We construct the first U.S. employer-employee data set with foreign ownership information from tax records to measure these direct and indirect effects. We find the average direct effect of a foreign multinational firm on its U.S. workers is a 7% increase in wages. This premium is larger for higher-skilled workers and for the employees of firms from high GDP per capita countries. We find evidence that it is membership in a multinational production network—instead of foreignness—that generates the foreign-firm premium. We leverage the past spatial clustering of foreign-owned firms by country of ownership to identify the indirect effects. An expansion in the foreign-multinational share of commuting-zone employment substantially increases the employment, value added, and—for higher-earning workers—wages at local domestic-owned firms. Per job created by a foreign multinational, our estimates suggest annual gains of US$13,400 to the aggregate wages of local incumbents, two-thirds of which are from indirect effects. Our estimates suggest that—via mega-deals for subsidies from local governments—foreign multinationals are able to extract a sizable fraction of the local surplus they generate.

We quantify the importance of imperfect competition in the US labor market by estimating the size of labor market rents earned by American firms and workers. We construct a matched employer-employee panel dataset by combining the universe of US business and worker tax records for the period 2001–2015. Using this panel data, we identify and estimate an equilibrium model of the labor market with two-sided heterogeneity where workers view firms as imperfect substitutes because of heterogeneous preferences over nonwage job characteristics. The model allows us to draw inference about imperfect competition, worker sorting, compensating differentials, and rent sharing.

Many studies use matched employer-employee data to estimate a statistical model of earnings determination with worker and firm fixed effects. Estimates based on this model have produced influential yet controversial conclusions. The objective of this paper is to assess the sensitivity of these conclusions to the biases that arise because of limited mobility of workers across firms. We use employer-employee data from the United States and several European countries while taking advantage of both fixed effects and random effects methods for bias correction. We find that limited mobility bias is severe and that bias correction is important.

We develop, identify, and estimate a model of imperfect competition in both labor and product markets. Our context is the US construction industry, where firms compete for workers, private market projects, and government procurements. Our empirical approach leverages bidding data from procurement auctions linked to employer-employee tax records. We find imperfect competition in both markets generates a total wage markdown of more than 30 percent and a total price markup of around 45 percent. By contrast, if one erroneously assumed a perfectly competitive product (labor) market, then one would conclude wages (prices) are marked down (up) by only 20 percent (16 percent).

There is no evaluation of the consequences of Disability Insurance (DI) receipt that captures the effects on households’ net income and consumption expenditure, family labor supply, or benefits from other programs. Combining detailed register data from Norway with an instrumental variables approach based on random assignment to appellant judges, we comprehensively assess how DI receipt affects these understudied outcomes. To consider the welfare implications of the findings from this instrumental variables approach, we estimate a dynamic model of household behavior that translates employment, reapplication, and savings decisions into revealed preferences for leisure and consumption. The model-based results suggest that on average, the willingness to pay for DI receipt is positive and sizable. Because spousal labor supply strongly buffers the household income and consumption effects of DI allowances, the estimated willingness to pay for DI receipt is smaller for married than single applicants.