Research

Multinational enterprises (MNEs) increasingly impose "responsible sourcing" (RS) standards on their suppliers worldwide, including requirements on worker compensation, benefits and working conditions. Are these policies just “hot air” or do they impact suppliers and their workers? What is the welfare incidence of RS in sourcing countries? To answer these questions, we combine the near-universe of RS rollouts by MNEs with subsidiaries in Costa Rica with firm-to-firm transactions and matched employer-employee microdata. We find that RS rollouts lead to reductions in sales and employment at exposed suppliers, salary increases for their low-wage workers and a decrease in their low-wage employment share. To rationalize these effects and study their implications in general equilibrium, we develop a simple open-economy model of RS. We show that the welfare effect of RS is ambiguous, depending on the interplay between an export tax (+) and a consumption tax (−), and that RS has distributional implications within worker types. Combining model and evidence for counterfactual analysis, we find that RS has delivered significant gains for low-wage workers at exposed suppliers, but led to adverse indirect effects on wages and the price index for the broader low-wage workforce.

We study optimal income taxation in a spatial equilibrium model with heterogeneous locational preferences, labor supply decisions, and competitive housing and labor markets. Expressions characterizing the optimal tax schedule in each community are provided that capture the fiscal externalities associated with migration and the effects of redistribution between households and landlords. Correlation between skill and locational preferences yields optimal transfers to poor areas, while sorting based on comparative advantage can motivate transfers in either direction. A calibration to areas targeted by the U.S. Empowerment Zone program yields sizable optimal spatial transfers that are sensitive to assumed levels of migration responsiveness.

We study how political preferences shaped California’s High-Speed Rail (CHSR), a large transportation project approved by referendum in 2008. Voters’ support responded significantly to the projected economic gains in their tract of residence, as measured by a quantitative model of high-speed rail matched to CHSR plans. Given this response, a revealed-preference approach comparing the proposed network with alternative designs identifies strong planner’s preferences for political support. The optimal politically-blind design would have placed the stations nearer to California’s dense metro areas, where it was harder to sway votes, thus increasing the projected economic gains.

We revisit the rationale for place-based policies using a canonical urban framework with agglomeration spillovers. We derive six main lessons. First, the spatial allocation is inefficient even when spillover elasticities are constant across regions. Second, under constant and positive spillover elasticities, the optimal policy is a national wage subsidy funded by a lump sum tax, reallocating activity towards higher wage locations. Third, more generally, a region’s optimal labor subsidy rate equals its spillover elasticity. Fourth, place-based policies that favor low-wage locations on efficiency grounds are justified when density has negative spillover effects, spillover elasticities are higher in low-wage locations, or across-skill spillovers favor more mixing in low-wage locations. Fifth, government spending on infrastructure, investment incentives, or housing policies cannot fully correct externalities from labor density. Sixth, housing supply elasticities do not affect the design of first-best place-based policies targeting agglomeration spillovers.

We propose a spatial equilibrium model with heterogeneous households holding general non-homothetic preferences over tradable goods and housing. In equilibrium, desirable and productive locations command high housing prices. So long as housing is a necessity, these locations are disproportionately inhabited by high-income earners who are relatively less affected by high housing prices. We clarify how this source of sorting complements other potential sorting forces in spatial equilibrium models, namely, comparative advantage in production and heterogeneous preferences for locations. We show how to measure changes in welfare inequality across income groups in a theoretically-consistent way when housing is a necessity, extending the approach popular in models with homothetic preferences. We use our framework to track the evolution of welfare inequality between college and non-college graduates in the United States between 1980 and 2020. We find that, accounting for change in prices, it has risen by more than nominal wage inequality, even as college graduates increasingly sort into cities with expensive housing over this time period.

We summarize recent methods to study optimal spatial policies. We center the discussion on policies that implement the optimal distribution of population in the presence of spatial spillovers, spatial transfers to optimally tackle redistribution between rich and poor regions, and optimal transportation investments.

We explore the impact of rising incomes at the top of the distribution on changes in spatial sorting patterns within large US cities. We develop and quantify a spatial model of a city with heterogeneous agents, heterogeneous neighborhoods of endogenous quality, and non-homothetic preferences for locations with different amenities. As the rich get richer, their increased demand for luxury amenities available downtown drives housing prices up in downtown areas. The poor are made worse off, either being displaced or paying higher rents for amenities that they do not value as much. Endogenous provision of private amenities amplifies the mechanism, while public provision of other amenities in part curbs it. We quantify the corresponding impact on wellbeing inequality. Through the lens of the quantified model, the change in income distribution between 1990 and 2014 led to neighborhood change and spatial resorting within urban areas that increased the welfare of richer households relative to that of poorer households by an additional two percentage points on top of their differential income growth.

The spatial segregation of college and non-college educated workers between commuting zones in the U.S. has steadily grown since 1980. We summarize prior work on sorting and location and document new descriptive patterns on how sorting and locations have changed over the past four decades. We find that there has been a shift in the sorting of college-educated workers from cities primarily centered around production in 1980 to cities centered around consumption by 2017. We develop a spatial equilibrium model to understand these patterns, and highlight key places where further research is needed. Our framework helps understand the causes and consequences of changes in spatial sorting, their impact on inequality, and how they respond to, and feed into, the changing nature of cities.

We study trends in income inequality across U.S. states and counties 1960–2019 using a mix of administrative and survey data sources. Both states and counties have diverged in terms of per-capita pre-tax incomes since the late 1990s, with transfers serving to dampen this divergence. County incomes have been diverging since the late 1970s. These trends in mean income mask opposing patterns among top and bottom income quantiles. Top incomes have diverged markedly across states since the late 1970s. In contrast, bottom income quantiles and poverty rates have converged across areas in recent decades.

We use the granular model of international trade developed in Gaubert and Itskhoki (2020) to study the rationale and implications of three types of government interventions typically targeted at large individual firms — antitrust, trade and industrial policies. We find that in antitrust regulation, governments face an incentive to be overly lenient in accepting mergers of large domestic firms, which acts akin to beggar-thy-neighbor trade policy in sectors with strong comparative advantage. In trade policy, targeting large individual foreign exporters rather than entire sectors is desirable from the point of view of a national government. Doing so minimizes the pass-through of import tariffs into domestic consumer prices, placing a greater portion of the burden on foreign producers. Finally, we show that subsidizing ‘national champions’ is generally suboptimal in closed economies as it leads to an excessive build-up of market power, but it may become unilaterally welfare improving in open economies. We contrast unilaterally optimal policies with the coordinated global optimal policy and emphasize the need for international policy cooperation in these domains.

Large firms play a pivotal role in international trade, shaping the export patterns of countries. We propose and quantify a granular multi-sector model of trade, which combines fundamental comparative advantage across sectors with granular comparative advantage embodied in outstanding individual firms. We develop an SMM-based estimation procedure, which takes full account of the general equilibrium of the model, to jointly estimate these fundamental and granular forces using French micro-data with information on firm domestic and export sales across manufacturing industries. We find that granularity accounts for about 20% of the variation in realized export intensity across sectors, and is more pronounced in the most export-intensive sectors. We then extend the model to a dynamic environment featuring both granular and fundamental shocks that jointly shape the time-series evolution of comparative advantage. We find a central role of granular forces in shaping comparative advantage reversals observed in the data.

We study optimal spatial policies in a quantitative trade and geography framework with spillovers and spatial sorting of heterogeneous workers. We characterize the spatial transfers that must hold in efficient allocations, as well as labor subsidies that can implement them. There exists scope for welfare-enhancing spatial policies even when spillovers are common across locations. Using data on U.S. cities and existing estimates of the spillover elasticities, we find that the U.S. economy would benefit from a reallocation of workers to currently low-wage cities. The optimal allocation features a greater share of high skill workers in smaller cities relative to the observed allocation. Inefficient sorting may lead to substantial welfare costs.

Tourism is a fast-growing services sector in developing countries. This paper combines a rich collection of Mexican microdata with a quantitative spatial equilibrium model and a new empirical strategy to study the long-term economic consequences of tourism both locally and in the aggregate. We find that tourism causes large and significant local economic gains relative to less touristic regions that are in part driven by significant positive spillovers on manufacturing. In the aggregate, however, these local spillovers are largely offset by reductions in agglomeration economies among less touristic regions, so that the national gains from tourism are mainly driven by a classical market integration effect.

The distribution of firms in space is far from uniform. Some locations host the most productive large firms, while others barely attract any. In this paper, I study the sorting of heterogeneous firms across locations and analyze policies designed to attract firms to particular regions (place-based policies). I first propose a theory of the distribution of heterogeneous firms in a variety of sectors across cities. Aggregate TFP and welfare depend on the extent of agglomeration externalities produced in cities and on how heterogeneous firms sort across them. The distribution of city sizes and the sorting patterns of firms are uniquely determined in equilibrium. This allows me to structurally estimate the model, using French firm-level data. I find that nearly half of the observed productivity advantage of large cities is due to firm sorting. I use the estimated model to quantify the general equilibrium effects of place-based policies. I find that policies that decrease local congestion lead to a new spatial equilibrium with higher aggregate TFP and welfare. In contrast, policies that subsidize under-developed areas have negative aggregate effects.