Pricing power in real estate markets can reduce housing supply and redevelopment relative to the social optimum. We show how pricing power interacts with popular redevelopment subsidies and zoning regulations. Using building-level rental income data from NYC, we find that increased concentration is correlated with increased rents. Finally, we use the model to estimate the first building-level housing elasticity, finding that markups account for between one-sixth and one-third of rents in the city.

The presence of pricing power in housing markets significantly impacts our understanding of the housing supply. It biases estimates of housing production functions, supply elasticities, the effects of land-use policies, and the results of quantitative spatial models. We test for the existence of pricing power in the New York City rental market. Using tax policy changes, we conduct complementary difference-in-differences and instrumental variable analyses. An idiosyncratic increase in a single building's costs leads to a proportional rent increase, holding market-level rents constant. Our findings support the existence of pricing power and challenge the prevailing perfect competition framework.

International lending flows are often intermediated through banking hubs and complex multi-national routing. We develop a dynamic stochastic general equilibrium model where global banks choose the path of direct or indirect lending through partner institutions in multiple countries. We show how conflating locational loan flows with ultimate lending biases results both by attributing ultimate lending to banking hubs, and by missing ultimate lending that occurs indirectly via third countries. We next study the effects of global banking complexity. Indirect lending allows countries to bypass shocked lending routes via alternative countries; however, it dilutes their ability to diversify sources of funds after shocks. The quantitative analysis reveals that banking complexity can exacerbate credit instability when countries feature heterogeneous banking efficiency.

Why do cities differ so much in productivity? A long literature has sought out systematic sources, such as inherent productivity advantages, market access, agglomeration forces, or sorting. We document that up to three quarters of the measured regional productivity dispersion is spurious, reflecting the “luck of the draw” of finite counts of idiosyncratically heterogeneous plants that happen to operate in a given location. The patterns are even more pronounced for new plants, hold for alternative productivity measures, and broadly extend to European countries. This large role for individual plants suggests a smaller role for places in driving regional differences.

We document that plants belonging to small and mid -sized firms are geographically concentrated, while large firms are much more dispersed. These differences are sizable; firms with 2 plants have a dispersion that is 5 log points lower than predicted by industry location patterns, while the corresponding figure is less than 2 log points for firms with 40 plants and less than a half log point for firms with 100 or more plants. These patterns are qualitatively robust across industries, time periods, and alternative specifications. We also find that plants that are farther from the firm headquarters employ less workers than closer plants within the same firm, and that this relationship is attenuated in large firms. We interpret these findings through the lens of a model of plant location in which more productive firms endogenously choose to lower their cost of geographic expansion.

Entrepots are hubs that facilitate trade between multiple origins and destinations. We study these entrepots, the network they form, and their impact on international trade. We document that the trade network is a hub-and-spoke system, where 80% of trade is shipped indirectly―nearly all via entrepots. We estimate indirect-shipping consistent trade costs using a model where shipments can be sent indirectly through an endogenous transport network and develop a geography-based instrument to estimate economies of scale in shipping. Counterfactual infrastructure improvements at entrepots have on average ten times the global welfare impact of improvements at non-entrepots.

Industry agglomeration can be indicative of agglomeration forces and is correlated with firm outcomes. Because multi-plant firms tend cluster their establishments in space, industry agglomeration could in part be driven by forces internal to the firm rather than across-firm spillovers. We propose and implement a decomposition of the industry agglomeration measures into within and across-firm components using U.S. census microdata. The within-firm component makes a small contribution to observed industry agglomeration for most industries and spatial scales, but accounts for 20% or more of observed agglomeration at short spatial scales for a subset of industries.

We investigate the sources, scope, and implications of landowner market power. We show how zoning regulations generate spillovers through increased markups and derive conditions under which restricting landownership concentration reduces rents. Using newbuilding-level data from New York City, we find that a 10% increase in ownership concentration in a Census tract is correlated with a 1% increase in rent. Market power is substantial: on average, markups account for nearly a third of rents in Manhattan. Furthermore, pecuniary spillovers between zoning constraints and markups at other buildings are appreciable. Up-zoning that results in 417 additional housing units at zoning-constrained buildings reduces markups on policy-unconstrained units and generates between 5 and 19 additional units through increased competition.

Why do cities differ so much in productivity? We document that most of the measured dispersion in productivity across US cities is spurious and reflects granularity bias: idiosyncratic heterogeneity in plant-level productivity and size, combined with finite plant counts. As a result, economies with randomly reallocated plants exhibit nearly as high a variance as the empirical economy. Stripping out this bias using our nonparametric split-sample strategy reduces the raw variance of place effects by about two thirds to three quarters. For new plants, about four fifths of the dispersion reflects granularity bias, and new plants’ place effects are only imperfectly correlated with those of older plants. These US-based patterns broadly extend to the 15 European countries we study in internationally comparable firm-level data.

Geography models have introduced and estimated a set of competing explanations for the persistent relationships between firm and location characteristics, but cannot identify these forces. I introduce a solution method for models in arbitrary geographies that generates reduced-form predictions and tests to identify forces acting through geographic linkages. This theoretical approach creates a new strategy for spatial empirics. Using the correct observables, the model shows that geographic forces can be taken into account without being directly estimated; establishment and employment density emerge as sufficient statistics for all geographic forces. I present two applications. First, the model can be used to evaluate whether geographic linkages matter and when simplified models suffice: the mono-centric model is a good fit for business services firms but cannot capture the geography of manufactures. Second, the model generates reduced-form tests that distinguish between spillovers and firm sorting and finds evidence of sorting.

There are persistent differences in self-reported subjective well-being across US metropolitan areas, and residents of declining cities appear less happy than others. Yet some people continue to move to these areas, and newer residents appear to be as unhappy as longer-term residents. While historical data on happiness are limited, the available facts suggest that cities that are now declining were also unhappy in their more prosperous past. These facts support the view that individuals do not maximize happiness alone but include it in the utility function along with other arguments. People may trade off happiness against other competing objectives.

While economic geography is concerned chiefly with proximity, models in urban economics eliminate proximity as a relative metric in order to preserve tractability. I introduce a new method of solving spatial models that allows for the consideration of proximity in an economic geography setting while retaining much of the tractability of the urban framework. The first chapter in this thesis introduces the solution method for continuous space geography models and shows how it reduces the complexity of the equilibrium conditions and allows such a model to generate more predictions than was previously possible. In this chapter, I build a model of firm location decisions in a spatial setting in order to provide a new explanation for the relationship between productivity and density: sorting of heterogeneous firms for market access. This geographic model of sorting breaks observational equivalence between firm sorting and agglomeration forces: under specific conditions, positive shocks to density can negatively affect average productivity through changes in the local composition of firms, inconsistent with models of agglomeration forces without sorting. Using restricted access establishment-level Census data, I document strong intra-city relationships between location and firm characteristics predicted by the model. I test for evidence of composition effects, instrumenting for the supply of new non-residential real estate construction using the geographic distribution of multi-city real estate developers, and find evidence of firm sorting.