Urban Economics Association

September 25, 2026 9:00 to 11:00 (Location: Iowa)
1A. Public Transit

Chair: Roman Zarate (UC San Diego)

• Akhila Kovvuri
  Moving Opportunity Closer: How Public Transit Transforms Firm Composition and Employment
  Presented by Akhila Kovvuri (Stanford University)
  Discussed by Roman Zarate (UC San Diego)
  Abstract

  Transportation infrastructure can improve workers’ access to existing economic opportunities, but it can also reshape economic opportunity itself by influencing where and what kinds of firms locate. This paper studies how public transit infrastructure influences firm location, composition, and employment at the neighborhood level. We construct novel data tracking over one million establishment entries and employ both difference-in-differences and market access specifications, exploiting the phased expansion of the Delhi Metro Rail in India. Transit access increases firm entry near stations, with larger, established retail and service firms locating first and inducing subsequent entry of other firms. These patterns create new economic hubs in peripheral areas, increasing employment per capita, especially for women in a context of low baseline female labor force participation. Counterfactual decompositions using a quantitative spatial model with estimated gender-specific commute elasticities reveal that compositional shifts toward larger establishments and consumer-facing industries that ex-ante employ more women account for the majority of this differential employment effect. Understanding how infrastructure reshapes the demand side of the labor market is thus critical for predicting and enhancing its distributional impacts.

• Oluchi Mbonu, Christopher Eaglin
  Market Segmentation and Coordination Costs: Evidence from Johannesburg’s Minibus Networks
  Presented by Oluchi Mbonu (The World Bank)
  Discussed by Akhila Kovvuri (Stanford University)
  Abstract

  How does spatial market segmentation affect firms' ability to meet demand across space? We study the market for public transportation in Johannesburg, South Africa, where private associations of minibus owners segment the city into distinct territories. In contrast, the demand for urban mobility is inherently interconnected, with a quarter of commuter trips originating in one association’s territory and ending in another’s. We study the frictions that associations face on these ``between-territory'' routes. Using GPS traces for over 40 million minibus trips and 9 million commuter trips, we present two complementary empirical results that quantify these frictions. First, we use an expected, cyclic mobility demand shock -- the sharp increase in recreational mobility following monthly pay dates -- to trace out the supply curve of minibus services by route type. The supply elasticity is close to 1 on routes contained within an association’s territory but is significantly lower on between-territory routes (0.4). We estimate that if between-territory supply were as elastic as ``within-territory'' supply, aggregate wait time for commuters would decrease by one million minutes per day, or approximately 4 minutes per trip. In our second exercise, we use exogenous fleet reductions due to bus breakdowns and repossessions to show that associations prioritize maintaining service on between-territory routes over within-territory routes, indicating that between-territory routes are more profitable at the margin. In a model of minibus allocation, our observed empirical patterns correspond to more convex costs on between-territory routes, reflecting the need for associations to coordinate with each other on these routes.

• Yizhen Gu, Xiangjie Zhu, Ben Zou
  Congestion Externalities of Public Transit
  Presented by Xiangjie Zhu (University of Toronto)
  Discussed by Oluchi Mbonu (The World Bank)
  Abstract

  We estimate three congestion externalities of public transit using the universe of smart card records from Beijing’s subway and bus systems. We find the delay an additional bus passenger imposes on the bus's other riders is comparable to the delay an additional car imposes on other road users, and an additional subway rider imposes the smallest of the three but still on the same order of magnitude. Both within-transit delays arise outside the vehicle, at platforms and at stops, rather than from slower in-vehicle speed; on the road, adding one bus slows traffic as much as adding $27$ cars. We embed these estimates in a short-run multimodal equilibrium model with heterogeneous travelers and asymmetric road congestion technology. Jointly adjusting the road toll and the two transit fares delivers a welfare gain close to the first-best. The gain comes from a relative-price correction across modes that redirects displaced automobile demand toward the subway rather than the bus, while expanding bus supply adds little. The instrument that relieves the most congestion is not the one that raises welfare the most. Reforms with a road toll are regressive in gross terms; the revenue-neutral regressive in gross terms; the revenue-neutral pricing policy is progressive even without any recycling. Optimal multimodal congestion pricing therefore rests on a balance between efficiency in congestion relief and distributional equity.

• Leticia Juarez, Jordan Mosqueda Juarez, Roman Zarate
  Optimal Transit Networks in Developing Countries
  Presented by Roman Zarate (UC San Diego)
  Discussed by Xiangjie Zhu (University of Toronto)
  Abstract

  A key allocation problem for budget-constrained governments in developing cities is where and how much public mass transit infrastructure to build, relative to allowing privatized and often informal transit operators to operate. Mass transit technologies are fixed-cost intensive but have low marginal operating costs, while minibuses have higher marginal costs and entail externalities, but require little upfront investment. In this paper, we study the optimal transit network problem in general spatial equilibrium, where a planner can choose between technologies to connect locations within a city. We show that the optimal public-private transit mix on any given edge of the network depends crucially on the trade-off between relative marginal and fixed costs across technologies. Leveraging a unique dataset of public and private transit networks in Mexico City and microdata on wages, land use, and commuting flows, we quantitatively study the gains from budget-feasible expansions of the transit system. We find that increasing the infrastructure budget by 50% raises welfare by 1.33% under public-only expansion and by 1.59% when both public and private transit can expand optimally, with the additional 20% welfare gain driven by private transit. In both cases, optimal investments primarily improve connections from peripheral areas to productive outer nodes and toward the existing network. These results suggest that developing cities could benefit substantially by jointly designing public and private networks, rather than public mass transit in isolation.

September 25, 2026 9:00 to 11:00 (Location: Michigan)
1B. Trade and Environment

Chair: Farid Farrokhi (Boston college)

• Tomas Domínguez-Iino
  Critical Minerals, Geopolitics, and the Green Transition
  Presented by Tomas Domínguez-Iino (University of Chicago)
  Discussed by Elliot Kang (Minnesota and PwC)
  Abstract

  The green energy transition will be powered by the mining and processing of lithium, nickel, and cobalt, which are critical for the production of advanced batteries. These minerals are concentrated geographically but traded globally. We study the geopolitical implications of active policy intervention in key mining countries, and we discuss consequences for green technology adoption worldwide. We show that the joint use of critical minerals in advanced batteries generates complementarity, which shapes both producer welfare and battery adoption. We quantify supply chain vulnerability, policy spillovers across mineral markets, and the potential for mineral cartels.

• Elliot Kang
  Global Fisheries: Quantifying the Externalities from Open Access
  Presented by Elliot Kang (Minnesota and PwC)
  Discussed by Farid Farrokhi (Boston college)
  Abstract

  This paper uses a novel geospatial dataset of global fishery catch and develops a quantitative dynamic spatial model to quantify the externalities from open access in global fishing. I first show that (i) the average global fishery stock decreased by 35% between 1980 and 2018, (ii) assigning property rights lead to larger fishery stock, and (iii) fuel subsidies to vessels are positively correlated with high sea fishing. Then, I build a dynamic spatial model of global fisheries and compare two polar cases of open access: the decentralized equilibrium, where atomistic firms have open access to the fishery, and the socially optimal allocation, where the social planner has exclusive property rights. By taking the model to the data in 2018, I find that in the socially optimal allocation, the average global fishery stock at the steady state increases by 88%, and the net present value of global welfare increases by 0.11%, compared to the decentralized equilibrium. The counterfactual analysis shows that fuel subsidies are globally welfare-reducing and decrease the average global fishery stock at the steady state by 3.2%.

• Mark Curtis, Weiting Miao, Felix Samy Soliman, Juan Carlos Suarez Serrato, Daniel Xu
  The Geographical Leakage of Environmental Regulation: Evidence from the Clean Air Act
  Presented by Felix Samy Soliman (University of Zurich)
  Discussed by Tomas Domínguez-Iino (University of Chicago)
  Abstract

  How large is geographic leakage resulting from place-based environmental policy? We study this question in the context of the landmark U.S. Clean Air Act Amendments. We make three contributions. First, using modern event-study techniques and confidential U.S. Census data, we revisit seminal results characterizing the effects of this regulation on directly regulated plants and industries, documenting large and persistent declines in employment and sales driven in part by plant exit. Second, we extend prior research by quantifying leakage to unregulated regions and identifying multi-unit firm networks as a central conduit for this reallocation. We show that firms expand activity at unregulated plants and that counties more exposed to nearby regulation experience positive employment spillovers. Third, we integrate these findings into an industry spatial equilibrium model with intra-national trade and multi-plant production. The model implies that the regulation raised prices by roughly 2% while reducing pollution-related mortality by 1.7%, saving approximately 10,500 lives annually. It further shows that a substantial share of the observed relative employment decline reflects leakage to unregulated domains, underscoring the importance of multiunit firm networks for evaluating the aggregate costs and environmental benefits of environmental policy.

• Farid Farrokhi
  A Framework for Integrating Climate Goals into Trade Agreements
  Presented by Farid Farrokhi (Boston college)
  Discussed by Felix Samy Soliman (University of Zurich)
  Abstract

  A critical tension in global governance is that trade agreements have evolved largely in isolation from climate policy issues. This paper shows that the two domains generate systematic cross-externalities: Larger gains from trade are associated with greater negative climate externalities imposed on partners, while linking carbon taxes to trade agreements produces distributive externalities that undermine the balance of trade concessions. To address this tension, we present a framework to integrate harmonized carbon pricing into the WTO subject to institutional, political-feasibility, and informational constraints. Our proposed implementation centers on a Climate Fund that collects border-related carbon tax revenues and redistributes them to level the carbon tax incidence across countries. Quantitative analysis shows that even a simple fund allocation rule can sustain a harmonized carbon price of $146 per ton of CO2, reducing global emissions by 55%. The main binding constraint is informational: if the reform could identify ex ante the precise carbon tax incidence across countries, it could feasibly support a price as high as $282 per ton.

September 25, 2026 9:00 to 11:00 (Location: LaSalle)
1C. Pollution I

Chair: Edson Severnini (Boston College)

• Simon Greenhill
  Noise Pollution and Infant Health
  Presented by Simon Greenhill (University of California, Berkeley)
  Discussed by Fengqi Liu (University of Illinois)
  Abstract

  Noise, or unwanted sound, is ubiquitous in urban environments. I introduce a new approach to measuring noise at scale using seismometers and compile the largest database of publicly available ambient noise measurements in the United States. Using a research design leveraging idiosyncratic variation in electric passenger rail noise exposure, I estimate the impact of noise pollution on infant health. I find that in utero noise pollution exposure harms infant health. A roughly 2 decibel (5%) increase in average noise levels during pregnancy lowers an overall index measure of infant health by 4% of a standard deviation, equivalent to one-third of the Black-White gap in the index. This effect is driven by nighttime noise, suggesting disruptions to maternal sleep as a possible mechanism. Overnight rail services, which account for under 10% of overall ridership, generate an average externality of $17 per overnight passenger-trip. In per passenger-mile terms, peak marginal overnight rail noise externalities are comparable to marginal rush hour traffic congestion externalities from private vehicles. Using seismic data and machine learning, I produce a novel map of noise for the contiguous United States that substantially outperforms existing national noise datasets, and use this map to assess noise exposure and costs nationally. Nearly all urban residents are exposed to nighttime noise exceeding WHO guidelines. I estimate that the partial social cost of noise pollution due to harms to infant health is $4.9 billion per year. Urban, Black, and Hispanic Americans disproportionately bear these costs.

• Fengqi Liu, Yilan Xu
  Greener on the Track, Dirtier by the Plant: The Air Pollution Cost of High-Speed Rail
  Presented by Fengqi Liu (University of Illinois)
  Discussed by Simon Greenhill (University of California, Berkeley)
  Abstract

  This study highlights an overlooked dimension of high-speed rail (HSR)’s environmental externalities: the spatial mismatch between the accessibility benefits generated by HSR and the environmental costs imposed by the electricity production needed to support its operation. By integrating a GIS dataset of national transportation infrastructure, daily HSR timetables, and high-frequency air pollution data, we quantify the downstream environmental impact of electricity consumption by HSR. The results indicate that, during the 2019–2020 sample period, powering a 1000-km HSR trip increases PM2.5 concentrations by 5.48% within the 30-kilometer downwind buffer of the responsible power plant. The PM2.5 effect of HSR becomes statistically insignificant in the 2025–2026 sample, while the SO2 and NO2 effects are insignificant in both periods. These patterns are consistent with China’s cleaner power generation mix in recent years and its long-standing stringent controls on SO2 and NOx emissions. A nationwide map of HSR-induced health costs in 2019 further reveals that several densely populated regions located in the gaps of the HSR network experience substantial health losses from pollution exposure while receiving limited accessibility gains from HSR. This spatial mismatch appears to have eased by 2025, likely due to the rising share of renewable power contribution and the continued expansion of the HSR network in China. These findings underscore an overlooked dimension of HSR’s environmental externalities and provide empirical support for improving cross-regional ecological compensation policies.

• Yilan Xu, YI Huang, Xiaoyi Han, Linlin Fan, Yucheng Wang
  As the Wind Blows: Dispersing the Environmental Benefits of Zero-Emission Vehicles Across Neighborhoods
  Presented by Yilan Xu (University of Illinois at Urbana-Champai)
  Discussed by Edson Severnini (Boston College)
  Abstract

  This paper estimates the localized air quality benefits of zero-emission vehicle (ZEV) adoption in California’s Bay Area by combining ZIP-level ZEV registrations and commuting flows with quasi-random shifts in wind direction that rotate traffic emissions across EPA monitoring stations. This natural experiment identifies the causal impact of ZEV traffic on downwind pollution: A 10% increase in upwind ZEV traffic reduces NO2 by 0.063%. Projecting these effects spatially reveals that disadvantaged neighborhoods receive disproportionately large pollution reductions, even with lower adoption rates. In 2022, the bottom decile of White-share tracts received 2.26 times the NO2 reduction of the top decile. Removing federal support (e.g., the IRA) reduces total gains and imposes larger absolute and percentage losses on disadvantaged areas. For example, low-income tracts lose 22% of projected NO2 reductions, versus 18% for high-income areas. These results highlight how policies shape the spatial distribution of ZEV adoption and its resulting environmental benefits, with important implications for advancing environmental equity.

• Gabriel Monteiro, Edson Severnini
  Rust in Motion: Impacts of Brazil’s Anti-Scrappage Policy Through Vehicle Ownership Tax
  Presented by Edson Severnini (Boston College)
  Discussed by Yilan Xu (University of Illinois at Urbana-Champai)
  Abstract

  Transportation is a major source of greenhouse gas and air pollution emissions worldwide, with most automobile pollution coming from vehicles older than 10 years. However, many policies discourage the removal of older vehicles from the fleet, with potential consequences for pollution and public health. This paper examines an anti-scrappage policy in Brazil using a border-pair design that exploits state variation in the age thresholds for vehicle ownership tax exemptions. Using municipality-level vehicle registration data from 2013-2020, we find that lower exemption age thresholds increase the average fleet age by up to one year, as they encourage owners to retain vehicles until they qualify for tax exemptions. Combining fleet-based estimates with satellite data, we show that these policies increase on-road CO2 emissions per capita by up to 23% and PM2.5 concentrations by 4%. We also find that the incidence of low birth weight rises by 10% under a 10-year exemption policy and by 2.8% in areas with thresholds between 15 and 20 years. Despite these social costs, we present suggestive evidence that electoral incentives make such policies difficult to reverse.

September 25, 2026 9:00 to 11:00 (Location: Illinois 3)
1D. Urban Upgrading

Chair: Matthew Turner (Brown University)

• Tal Shoshani, Kaihang Zhao, Davide Proserpio
  The Impact of Short-Term Rental Regulations on Urban Spending: Evidence from Transactional Data
  Presented by Tal Shoshani (University of Southern California)
  Discussed by Matthew Turner (Brown University)
  Abstract

  We examine the effects of short-term rental (STR) regulations on urban service markets, focusing on the restaurant industry. Our setting is New York City's 2023 STR regulation (Local Law 18), which triggered an immediate 80% contraction in active Airbnb listings. Using restaurant-month–level credit and debit card transaction data across 19 major US metropolitan areas and a matched difference-in-differences design, we find that the policy led to a 7.5% decline in restaurant spending, a 5.0% reduction in transaction volume, and a 2.3% decrease in average spending per transaction. These effects reflect multiple reinforcing mechanisms: reduced tourist inflows, higher accommodation costs, and depressed local income from lost host revenues and reduced tourism employment. Three complementary analyses support this interpretation: spending declines concentrate among restaurants catering to non-locals and higher-priced establishments, while average spending per transaction falls similarly across restaurant types; NYC airports saw a 3.5%-4.3% drop in passenger arrivals; and neighboring New Jersey experienced a 10.1% increase in restaurant spending, suggesting spatial displacement rather than outright demand destruction. Our findings highlight that STR regulations can have economically meaningful consequences well beyond the housing market, underscoring the importance of accounting for downstream impacts on local service industries in urban policy design.

• Ting Chen, Siying Lin, Changwei Zhan, Fan Zhang
  Street Vendor Evictions, Crime Reduction, and Gentrification: Evidence from Beijing’s Urban Upgrading Campaign
  Presented by Changwei Zhan (The University of Hong Kong)
  Discussed by Tal Shoshani (University of Southern California)
  Abstract

  We study the economic and social consequences of Beijing’s 2017 Urban Upgrading Campaign, which sought to improve urban environments by removing unauthorized vendors and commercial activities from backstreets and alleys. Exploiting the staggered rollout of the policy across subdistricts between 2017 and 2019, we find that the campaign resulted in a 50.00% drop in the number of restaurants and a 46.58% drop in the number of retail establishments, particularly among low-price and non-chain businesses. We also show that the program reduced crime by around 42.42%, particularly property crime, along with evidence of crime spillovers to adjacent cities. Results on neighborhood composition are consistent with gentrification: housing prices rose modestly while transaction volumes declined, and treated subdistricts experienced population sorting toward higher-income, more educated, and more local residents. Overall, our results highlight a central trade-off of urban enforcement—improvements in public order accompanied by the displacement of informal economic activity and selective migration.

• Raj Chetty, Rebecca Diamond, Thomas Foster, Lawrence Katz, Sonya Porter, Matthew Staiger
  Creating High-Opportunity Neighborhoods: Evidence from the HOPE VI Program
  Presented by Rebecca Diamond (Harvard University)
  Discussed by Changwei Zhan (The University of Hong Kong)
  Abstract

  We study whether low-economic-mobility neighborhoods can be transformed into high-mobility areas by analyzing the HOPE VI program, which invested $17 billion to revitalize 262 distressed public housing developments. We estimate the program’s impacts using a matched difference-in-differences design, comparing outcomes in revitalized developments to observably similar control developments using anonymized tax records. HOPE VI reduced neighborhood poverty rates by attracting higher-income families to revitalized neighborhoods, but had no causal impact on the earnings of adults living in public housing units. Children raised in revitalized public housing units earned more, were more likely to attend college, and were less likely to be incarcerated. Using a movers exposure design and sibling comparisons, we show that these improvements were driven by changes in neighborhoods’ causal effects on children’s outcomes. The improvements in neighborhood causal effects were driven in large part by changes in social interaction: HOPE VI increased interaction between public housing residents and peers in surrounding neighborhoods and increased earnings more for subgroups with higher-income peers. Many low-income families in the U.S. currently live in neighborhoods that are as socially isolated as the HOPE VI developments were prior to revitalization. We conclude that it is feasible to create high-opportunity neighborhoods and that connecting socially isolated areas to surrounding communities is a cost-effective approach to doing so.

• Fernando Ferreira, Gillian Meyer, Matthew Turner
  The Effect of LIHTC on Housing Markets: Quantities, Rents, and Welfare
  Presented by Matthew Turner (Brown University)
  Discussed by Rebecca Diamond (Harvard University)
  Abstract

  We estimate the effect of the Low-Income Housing Tax Credit (LIHTC) on local housing quantities, rents, and welfare. We develop a simple equilibrium framework that links LIHTC-induced changes in housing supply to crowd-out, affordability, and welfare, allowing us to measure the program’s effects on renters, landlords, developers, and taxpayers within a common structure. Empirically, we compare neighborhoods with approved LIHTC proposals to otherwise similar neighborhoods with rejected proposals, using a newly assembled dataset of nearly 25,000 LIHTC applications matched to HUD, Census, and ACS data. We find that LIHTC awards increase the local multifamily housing stock by nearly one-for-one with the number of subsidized units placed in service, implying little if any crowd-out of unsubsidized development. Roughly ten years after project approval, average rents in treated neighborhoods are about 6 percent lower than in comparable control areas, with the largest effects concentrated among units renting near the LIHTC price threshold. Our preferred calculations imply that an average LIHTC award raises consumer surplus by about \$47,000 per new unit, transfers about $43,000 per unit from incumbent landlords to renters, and increases developer surplus by about $60,000 per unit. Overall, LIHTC improves affordability largely through market-wide equilibrium effects: most of the gains arise through rent reductions for surrounding renters, not only through the direct allocation of subsidized units.

September 25, 2026 9:00 to 11:00 (Location: Illinois 4)
1E. Building Cities, Villages, and Data Centers

Chair: Sunham Kim (Korea Development Institute)

• Sungwan Hong
  Powering Intelligence: The Geography of AI Benefits and Infrastructure Costs
  Presented by Sungwan Hong (University of Pittsburgh)
  Discussed by Sunham Kim (Korea Development Institute)
  Abstract

  AI services are delivered through broad digital markets, but the data centers that produce them connect to particular local electricity grids. This paper studies the resulting mismatch between the geography of AI-use benefits and the geography of infrastructure costs, using facility-level data on U.S. data centers and a spatial general-equilibrium model in which data centers produce nationally pooled Compute using locally priced electricity. Event-study estimates identify medium-run retail pass-through: industrial prices rise by about 1.4$ per 10 MW in low-headroom grids, with little response elsewhere. Treating the expected U.S. data-center pipeline through 2027 as built and operating, delivered Compute prices fall by about 14% and average real income rises by about 0.8%, while electricity prices in the most exposed markets rise by up to 37%. An AI-use extension shifts gains toward high-skill, business-service places, while the infrastructure burden stays tied to data-center grids. The aggregate mean therefore hides the incidence that matters: Same-BA non-hosts share congestion without a direct income offset, and host compensation depends on local value capture, not grid exposure alone.

• Siyuan Hu
  The Spatial Incidence of Hyperscale Data Centers
  Presented by Siyuan Hu (University of Wisconsin-Madison)
  Discussed by Sungwan Hong (University of Pittsburgh)
  Abstract

  We estimate the local incidence of U.S. data center openings using a stacked difference-in-differences design and nationwide housing transaction data. Hyperscale data centers reduce nearby house prices by 6.8 percent, with effects fading beyond 14 km. Price impacts increase with power capacity and are larger when the facility is the first data center in a county. Large non-hyperscale facilities show little capitalization on average, though negative effects emerge in more residentially exposed locations. Additional evidence is consistent with roles for local utility costs, environmental burdens, and salience. These findings highlight the importance of siting and exposure in evaluating place-based incentives.

• Harrison Mitchell
  The Curse of Connectivity: Evidence from Indonesia's Village Resettlement Program
  Presented by Harrison Mitchell (UC San Diego)
  Discussed by Siyuan Hu (University of Wisconsin-Madison)
  Abstract

  How does spatial connectivity affect structural transformation and economic growth? In this paper, I explore how spatial connectivity ('market access') affects long-run development by exploiting the Indonesian Transmigration program, which created new villages with exogenous variation in initial levels of market access. Contrary to previous work that measures positive growth effects, I find that, even today, villages with initially higher levels of market access have lower GDP per capita. These findings are not explained by price effects, compensating differentials, or differential migration patterns. Rather than limiting structural transformation, higher connectivity accelerates the reallocation of labor out of agriculture, but this happens alongside negative GDP per capita effects. When there are scale externalities in the agricultural sector, rural sectoral reallocation out of agriculture can harm long-run productivity. I leverage exogenous variation induced by the program to provide direct evidence of this scale externality. I then develop a quantitative spatial model with an agricultural scale externality that replicates the key empirical patterns and demonstrates how the standard gains from trade can be reversed at low levels of economic development.

• Seungyub Han, Sunham Kim
  Where to Build New Cities: Optimal Urban Land Supply and Korea’s New Town Projects
  Presented by Sunham Kim (Korea Development Institute)
  Discussed by Harrison Mitchell (UC San Diego)
  Abstract

  Where should governments supply new urban land? We study this question in a multi-region spatial equilibrium model with a government that allocates land across regions subject to a budget constraint. In the short run and at the margin, the welfare return per dollar of procurement cost is equalized across regions: when costs scale with local land prices, location is irrelevant at the margin. However, permanent urban land supply, such as building new cities, changes the steady state itself: labor migrates, capital accumulates and reallocates, and prices adjust to a new long-run equilibrium. The long-run welfare return per dollar is not equalized across regions and depends on regional fundamentals---total factor productivity, amenities, and initial land endowments. We characterize these long-run gains, decompose them into reallocation and price adjustment channels, and use our framework to analyze South Korea's 2nd New Town Project, which supplied 49.6 km^2 of new urban land near Seoul, provided 610,433 housing units at a cost exceeding 10% of 2004 GDP. This project generated positive long-run gains in aggregate welfare and output. The allocation, however, was not optimal: the welfare-maximizing allocation under the same budget would have generated larger aggregate gains while mitigating regional decline.

September 25, 2026 9:00 to 11:00 (Location: Illinois 5)
1F. Economic History

Chair: Suqin Ge (Virginia Tech)

• Alexander Weinberg
  Urban Scars
  Presented by Alexander Weinberg (University of Chicago)
  Discussed by Suqin Ge (Virginia Tech)
  Abstract

  This paper uses data from telephone directories to study the long-run, local impacts of the civil unrest of the 1960s. Commercial blocks damaged by looting and arson remain areas of low commercial activity today, even compared to nearby blocks with similar pre-period observables. Patterns in the data suggest that the mechanism behind the persistent effects is disrupted agglomeration economies. These patterns inform a quantitative spatial model with dynamic firm location choice. I plan to use the model to evaluate alternative place-based revitalization policies.

• Luca Looser
  Family-led Structural Change
  Presented by Luca Looser (Universitat Pompeu Fabra)
  Discussed by Alexander Weinberg (University of Chicago)
  Abstract

  Structural transformation does not unfold in a social vacuum. As workers leave agriculture for new sectors and places, access to those opportunities is mediated by families: some individuals inherit relatives who can open doors to jobs and destinations, while others do not. This paper studies how family ties shaped structural change in the United States during the late nineteenth and early twentieth centuries. I combine full-count census data with population-scale genealogical linkages to build extended kin networks over time and space, and I exploit deaths of sector-specific relatives as shocks to family-based access. These shocks alter occupational choices and propagate within families. Guided by this evidence, I develop and estimate a dynamic spatial overlapping-generations model in which dynastic family capital lowers migration costs to connected places and entry costs into connected sectors. Quantitatively, family ties matter significantly in sectoral and migration switching costs, slow early reallocation out of agriculture and dampen adjustment to a major agricultural shock.

• Jiong Zhu, Per Fredriksson, Gunes Gokmen
  Traditional Religion and Entrepreneurship: Evidence from China
  Presented by Per Fredriksson (University of Louisville)
  Discussed by Luca Looser (Universitat Pompeu Fabra)
  Abstract

  We study the role of traditional religion in the origins of private entrepreneurship in China. Our focus is on the historical and contemporary influence of a prominent Chinese wealth god, Wutong. We leverage the facts that (i) in traditional Chinese folklore, the God of Wealth governs the distribution of wealth, and (ii) a policy reform in 1895 eased restrictions on private enterprise establishment in China. We construct a unique dataset on temples devoted to the Wealth God to capture a more ingrained culture of wealth worship. Combining an instrumental variable approach with a difference-in-differences framework, we find that the 1895 liberalization reform led to a greater number of new private enterprises in prefectures with a Wealth God culture. Heterogeneity analysis reveals that the effects of the 1895 reform was conditional on the urbanization rate, population density, and the transportation network. Moreover, the influence of the Wealth God shows strong persistence. The Wealth God culture still affects the contemporary incidence of registered new businesses, occupational choice favoring entrepreneurship, attitudes towards income, wealth, free markets, and private investment decisions, as well as the naming of children. This paper helps improve our understanding of the enduring impact of previously overlooked cultural and religious factors on entrepreneurship and regional disparities within China. Moreover, it provides evidence on the role played by the cultural context for the result of policy reforms.

• Suqin Ge, Naijia Guo, Zibin Huang, Junsen Zhang, Li Zhang
  Growing without Divergence: The Impact of Innovation on Low- and High-skilled Migration in China
  Presented by Li Zhang (University of Macau)
  Discussed by Per Fredriksson (University of Louisville)
  Abstract

  This paper examines how innovation shapes migration across skill groups. Using Chinese microdata from 2005–2015, we find that cities with faster patent growth attract more low-skilled than high-skilled migrants, opposite to patterns in developed countries. These cities see similar wage growth for both groups but limited amenity gains. We develop and estimate a spatial equilibrium model showing that low-skilled workers prioritize wages, while high-skilled workers value amenities, which rise with the share of skilled workers. Patent shocks draw in more low-skilled workers, reducing amenities and deterring high-skilled migration. Overall, technological growth raised wages and welfare without increasing spatial inequality.

September 25, 2026 9:00 to 11:00 (Location: Directors)
1G. Student Prize I

Chair: Raven Molloy (Federal Reserve Board)

• Colin Williams
  Boundary Issues: Cities and The Geography of Mobile Homes
  Presented by Colin Williams (University of Virginia)
  Abstract

  I document a novel stylized fact: the mobile home share triples from roughly 2% to over 8% just outside municipal boundaries, with no corresponding change in site-built home prices. To interpret this pattern, I develop a sufficient statistics framework to assess distortions to the composition of the housing stock. The framework leverages a CES demand model over differentiated housing varieties in which municipal zoning regulations act as a differential implicit tax on mobile homes. I then maps two reduced-form statistics — the discontinuity in mobile home shares and the discontinuity in relative prices at city boundaries — to two structural parameters: the implicit regulatory tax and the elasticity of substitution between mobile and site-built homes. Using a boundary regression discontinuity design, I estimate that mobile homes face an 11% implicit tax from municipal regulations. The model implies an elasticity of substitution of roughly 11, indicating that households are broadly indifferent between housing types conditional on size, vintage, and lot size. The high substitutability means that even a modest regulatory tax eliminates more than half of the mobile homes that would otherwise exist in cities, though the welfare costs to the average household are small. My results suggest that municipal land use regulations, not household sorting, drive the ruralization of mobile homes.

• Abhisit (AJ) Jiranaphawiboon
  Police Allocation, Crime, and Residential Welfare: Evidence from Chicago
  Presented by Abhisit (AJ) Jiranaphawiboon (Stanford GSB)
  Abstract

  Municipal police deployment trades crime reduction against the protection of high-value neighborhoods that anchor the local tax base. I build a quantitative spatial model in which police allocation maximizes a weighted municipal objective, offenders relocate across neighborhoods in response to enforcement, and residents sort endogenously. Using granular patrol and incident data from Chicago, I exploit a budget-induced district consolidation that reassigned patrol independently of local crime trends, identifying how crime and migration respond to enforcement. The estimated model recovers the objective weights implicit in observed allocation, with crime reduction dominating property-value protection. Shifting to pure crime minimization lowers aggregate crime, only marginally raises welfare for Black residents, but sharply reduces welfare for White residents, revealing a fiscal channel through which property-value-weighted allocation generates racially disparate effects.

• Micah Baum
  When Opportunity Moves to You: Housing Deregulation and the Distribution of Child Outcomes
  Presented by Micah Baum (University of Michigan)
  Abstract

  A large literature documents the economic benefits of moving children to low-poverty neighborhoods. However, it is not clear if children will benefit when their current neighborhood improves around them, especially for households paying market rents. I study how neighborhood revitalization affects incumbent children, focusing on a 1998 Houston housing reform that incentivized building single-family homes downtown. The reform increased median household incomes by ten percent and rents by seven percent as affluent, college-educated households moved into these homes. Using administrative data from the U.S. Census Bureau, I examine impacts on children's long-run human capital attainment, economic self-sufficiency, and criminal justice involvement. Renters moved out of revitalized areas to avoid higher rents, meaning their children grew up in similar neighborhoods as untreated individuals. In contrast, homeowner children lived in better neighborhoods and parent home equity increased. Young renter children had worse labor market outcomes, suggestive of disruption costs, while owner children attained more education and economic self-sufficiency. Owner children were also charged with crimes at higher rates, consistent with increased policing in gentrified areas. The reform improved incumbent well-being on average, but large losses for renters mean that Pareto-improving transfers are practically challenging. These findings highlight the trade-offs faced by policymakers aiming to revitalize urban neighborhoods.

• Ian Sapollnik, Dustin Swonder
  Shared Appreciation Loans for Homebuyers: Evidence from a California Lottery
  Presented by Dustin Swonder (University of California, Berkeley)
  Abstract

  Prospective homebuyers choose from a limited menu of mortgage products. How does mortgage market incompleteness affect homebuying decisions? This paper studies an initiative which expanded the financing menu for certain California homebuyers, offering a down payment assistance loan for up to 20% of a home's purchase price in exchange for a share of future appreciation. Leveraging random lottery variation, we find that shared appreciation loan offers increase homebuying by 42.5 percentage points within six quarters. Among homebuyers induced by offers to buy homes using shared appreciation loans, 70% would not buy homes otherwise. Buying with shared appreciation loan financing does not cause credit distress for these marginal homebuyers, but it does cause them to sacrifice neighborhood quality to attain homeownership. Inframarginal homebuyers use shared appreciation loans to decrease their mortgage borrowing by about $43,000 on average, rather than to buy more expensive homes.

September 25, 2026 9:00 to 11:00 (Location: Moskow)
1H. Urban Dynamics After Big Policy Pushes

Chair: Tianfang Cui (University of North Dakota)

• Vinicios Sant'Anna, Cihang Wang
  International Development Finance Shapes Local Urbanization
  Presented by Cihang Wang (South Dakota State University)
  Discussed by Tianfang Cui (University of North Dakota)
  Abstract

  We examine how international development finance shapes urbanization in Sub-Saharan Africa at the very local level. Using data on 1,592 georeferenced Chinese development projects, we analyze the effect of development projects on the evolution of built-up surface and volume at 100-meter grid cells within a 2-kilometer microregion. Our staggered difference-in-differences approach reveals that international development projects significantly increase local urbanization, with effects that decrease with distance from the projects. Treatment effects are mostly driven by residential development, particularly in previously underdeveloped areas. We contribute to understanding the consequences of international development finance for urban transformations in the developing world.

• Beau Bressler
  Building Segregation: The Long-Run Neighborhood Effects of American Public Housing
  Presented by Beau Bressler (Yale University)
  Discussed by Laura Weiwu (UC Berkeley)
  Abstract

  This paper studies the long-term neighborhood effects of the construction of mid-century American public housing. I construct a new national dataset tracking the locations, completion dates, and characteristics of over 1 million public housing units built between 1935 and 1973, which I link to neighborhood-level data from 1930 to 2010. I document that public housing projects were systematically targeted towards poorer, more populated neighborhoods with higher Black population shares, consistent with the program's slum clearance goals and racialized site selection politics. Using a stacked matched difference-in-differences design, I find that public housing construction caused large, persistent increases in Black population shares and substantial declines in median incomes and rents. Geographic spillovers to nearby neighborhoods were modest: Black population shares increased slightly, driven primarily by white population decline rather than Black inflows. I find evidence consistent with neighborhood tipping: neighborhoods with initially moderate Black shares experienced substantial white population outflows. Finally, linking to intergenerational mobility data, I show that children from low-income families who grew up in tracts containing public housing experienced significantly lower rates of upward mobility than those in comparable control areas. These findings demonstrate that mid-century public housing, despite intentions of neighborhood revitalization, reinforced existing patterns of racial and economic segregation with lasting consequences for economic opportunity.

• Laura Weiwu
  Opportunity in Motion: Infrastructure, Job Access, and Intergenerational Mobility
  Presented by Laura Weiwu (UC Berkeley)
  Discussed by Beau Bressler (Yale University)
  Abstract

  This paper studies how transportation infrastructure improves job access and, in turn, shapes neighborhood economic development and the intergenerational outcomes of children. I use a newly linked administrative dataset covering the near-universe of families with children born between 1964 and 1979. Increased commuting access from Interstate highway construction raises average neighborhood income through both direct income gains for existing residents and compositional changes driven by selective migration. Using a movers design, I show that increases in neighborhood income benefit children and reduce intergenerational inequality, with larger gains for children from lower-income families. However, migration responses generate negative spillovers: areas with smaller access improvements experience declines in peer quality, adversely affecting children in those locations. I develop an accounting framework to quantify the aggregate effects on intergenerational mobility and find that the direct economic benefits of improved access outweigh these negative spillovers. Nevertheless, changes in neighborhood composition increase the spatial inequality of opportunity.

• Beau Bressler, Tianfang Cui
  Taming the Growth Machine: The Long-Run Consequences of Federal Urban Planning Assistance
  Presented by Tianfang Cui (University of North Dakota)
  Discussed by Cihang Wang (South Dakota State University)
  Abstract

  We study how the federal Urban Planning Assistance Program, which subsidized communities in the 1960s to hire urban planners to draft land-use plans, affected housing supply. We digitize program records and link them to a municipality-level panel on housing and zoning outcomes. We identify causal effects using variation in program eligibility and differences in state agencies' capacity to approve funding. Planning assistance caused municipalities to build 20\% fewer housing units per decade over the 50 years that followed. Regulatory innovation steered development in assisted areas away from apartments and towards larger single-family homes. Drawing on text data from newspaper articles that cover local regulation and development politics, we show that recipients since the 1980s placed greater burden on developers to fund local amenities. These findings demonstrate that subsidizing planning expertise under decentralized land-use authority generated durable market frictions, with lasting consequences for national housing affordability.

September 25, 2026 11:30 to 13:00 (Location: Iowa)
2A. Urban Theory

Chair: Pierre-Philippe Combes (Sciences Po, Paris)

• Mohamed Bahlali, Raouf Boucekkine, Quentin Petit
  When Lions meets Krugman: a mean-field game theory of spatial dynamics
  Presented by Mohamed Bahlali (AMSE)
  Discussed by Pierre-Philippe Combes (Sciences Po, Paris)
  Abstract

  We propose a mean-field game (MFG) approach to study the dynamics of spatial agglomeration in a continuous space-time framework where trade across locations may follow a broad class of static gravity models. Forward-looking intertemporal utility-maximizing agents work and migrate in a two-dimensional geography and face idiosyncratic shocks. Equilibrium wages and prices depend on their common distribution and adjust statically according to the underlying trade model. We first prove existence and uniqueness of the static trade equilibrium. We then prove existence of dynamic equilibria, and discuss conditions for uniqueness. In the case of a circular economy, we obtain closed-form solutions for small perturbations around the steady state, and we identify the sets of parameters that lead to agglomeration or dispersion. We exploit the MFG structure of the model to explicitly quantify how uncertainty and forward-looking expectations contribute to aggglomeration and dispersion. In particular, we show that, regardless of the static trade model, forward-looking expectations always promote agglomeration, but cannot reverse the dominant pattern that would arise under myopic behavior.

• Amy Zhang
  The Spatial Dynamics of Environmental Regulation
  Presented by Amy Zhang (The Pennsylvania State University)
  Discussed by Mohamed Bahlali (AMSE)
  Abstract

  I develop a novel spatial environmental growth model in which comparative advantage endogenously shifts across regions. Technology investment can be directed toward clean and dirty sectors, and toward regions with high concentrations of either. Environmental regulation affects polluting production across regions and redirects technology investment, shifting the comparative advantage of dirty sectors toward regions with low regulatory exposure. Cross-regional and cross-sectoral labor reallocation interacts with shifts in comparative advantage to determine local employment and pollution patterns. I apply the model to evaluate local labor market consequences of the Clean Air Act 1990 Amendment, a place-based regulation with differential compliance costs across regions. Quantitative results show that conventional estimates that ignore general equilibrium effects substantially overstate the Act's negative impacts on polluting jobs, especially over the longer run.

• Frederic Robert-Nicoud, Pierre-Philippe Combes, Gilles Duranton, Laurent Gobillon
  A Unified Urban Model With Non-Homothetic Housing Demand
  Presented by Pierre-Philippe Combes (Sciences Po, Paris)
  Discussed by Amy Zhang (The Pennsylvania State University)
  Abstract

  Canonical urban models fail to jointly account for flexible housing demand, nondegenerate city sizes, and observed urban systems. We introduce a unifying urban framework based on price-independent generalized linear (PIGL) preferences in which housing is a necessity. Non-homothetic housing demand generates income effects that cause urban costs to scale more strongly with population than wages, restoring a unique interior efficient city size under standard assumptions. The framework nests existing canonical models, remains tractable, and is consistent with key empirical regularities, including Zipf’s law and observed housing price and income elasticities. We also show that a standard monocentric city model maps exactly into our framework.

September 25, 2026 11:30 to 13:00 (Location: Michigan)
2B. Wages I

Chair: Daniele Coen-Pirani (University of Pittsburgh)

• Ying Feng, Tatjana Kleineberg, Tommaso Porzio, Radine Rafols, Jordan Rosenthal-Kay
  Are workers moving to high-wage regions?
  Presented by Jordan Rosenthal-Kay (Federal Reserve Bank of San Francisco)
  Discussed by Daniele Coen-Pirani (University of Pittsburgh)
  Abstract

  We construct a new regional dataset integrating wage information with migration flow data, covering 1,257 regions across 76 countries over a span of six decades. Using this dataset, we document patterns of labor reallocation within countries over the life cycle of cohorts. By age 40, approximately 25% of individuals no longer reside in their birth location—a pattern that holds across both high- and low-income countries. While migration generally flows toward higher-wage regions, nearly one-third of movers relocate to areas with lower wages. Assuming wage differences reflect productivity gaps, we estimate the contribution of labor mobility to aggregate growth. Over the life cycle of a typical cohort, observed migration to high-productivity regions explains only a 2-3% increase in countries' average wage. We then develop a simple spatial model to quantify the gains from lowering moving costs while allowing the direction and magnitude of migration flows to adjust. This exercise yields larger aggregate gains in low-income countries when the extensive margin of migration is allowed to respond.

• Aakash Bhalothia, Gavin Engelstad, Gaurav Khanna, Harrison Mitchell
  The Global Value of Cities
  Presented by Gavin Engelstad (Northwestern University)
  Discussed by Jordan Rosenthal-Kay (Federal Reserve Bank of San Francisco)
  Abstract

  What is the economic value of each city in the world, and how much could incomes rise if workers were reallocated towards more productive locations? We answer these questions using a new global dataset of 513 million workers across 220,000 cities in 191 countries. Leveraging detailed job histories, we implement an event-study movers design to estimate the causal effect of cities on earnings, separating place-based productivity from worker sorting. We find that place matters substantially. Across international moves, 93\% of wage differences between cities translate into gains for movers, while within countries, 45–73\% do. These estimates imply large global disparities in city-level productivity. More productive cities are larger, more industrially diverse, and allocate more workers to high-productivity firms. We use these estimates to quantify the gains from reallocation. The dispersion of city effects is substantially larger in low-income countries, implying significant unrealized gains from migration. Reallocating workers across cities and firms to match the US distribution yields meaningful income gains in developing economies. Together, our results highlight the central role of location in shaping global income differences and point to spatial and within-city allocation as key barriers to development.

• Daniele Coen-Pirani, Mikhail Zavarzin
  Spatial Dispersion in Factor Shares: Evidence and Implications
  Presented by Daniele Coen-Pirani (University of Pittsburgh)
  Discussed by Gavin Engelstad (Northwestern University)
  Abstract

  We document a substantial and persistent disparity in the labor shares of value added across U.S. states and regions. Most notably, the Southern labor share has remained consistently below that of the Rust Belt and the rest of the country for over a century - a divergence that is especially pronounced in the manufacturing sector. To better understand this differential, we also construct regional capital and profit shares of value added using newly digitized historical capital expenditure data from the Census of Manufactures. Our analysis reveals that both the capital and profit shares are larger in the South than in the Rust Belt and the rest of the U.S. We interpret these patterns through the lens of a quantitative spatial model. We use the latter to measure the contribution of Rust Belt manufacturing unions' greater bargaining power to the labor share gap between the Rust Belt and the South.

September 25, 2026 11:30 to 13:00 (Location: LaSalle)
2C. Pollution II

Chair: Andrew Waxman (University of Texas at Austin)

• Luis Lopez, Dermot Murphy, Nitzan Tzur-Ilan, Sean Wilkoff
  Up in Smoke: The Impact of Wildfire Pollution on Healthcare Municipal Finance
  Presented by Luis Lopez (University of Illinois Chicago)
  Discussed by Manpreet Singh (Georgia Institute of Technology)
  Abstract

  Industries face unique risks that can affect their cost of capital and investment. In healthcare, wildfire smoke pollution (Smoke) increases service demand, but the credit risk effect depends on the relative influx of underinsured patients. We show that Smoke increases borrowing costs in the healthcare municipal bond market, especially in high-underinsurance counties. A natural experiment using the staggered expansion of Medicaid supports a causal interpretation. Smoke increases relative demand from underinsured patients and uncompensated care costs, further supporting the underinsurance channel. Hospitals respond by reducing investment. Out-of-state Smoke similarly increases borrowing costs, indicating unchecked wildfires impose negative externalities on other states.

• Jaeyong (Danny) Han
  In My Backyard? Really? The Impact of Information Provision on Pollution Avoidance
  Presented by Jaeyong (Danny) Han (University of Massachusetts Amherst)
  Discussed by Luis Lopez (University of Illinois Chicago)
  Abstract

  This paper examines how information provision through local news influences household behavior in a hedonic housing market. I construct a novel dataset of local television, radio, and newspaper coverage of the Environmental Protection Agency’s Superfund sites across the United States. Leveraging variation in media exposure, I document substantial heterogeneity in how nearby housing markets respond to environmental risk. To address potential endogeneity in coverage, I instrument using the distance between Superfund sites and local television broadcast stations. I find that house prices fall by more than 3% near sites that receive news coverage, while sites without coverage show muted responses; the estimated effect is even larger when instrumented. These results show that salient, high-reach media—especially television—sharpen household awareness of environmental hazards. Ignoring this channel leads to biased estimates of willingness to pay for environmental quality, underscoring the importance of media access in shaping economic responses to environmental risk.

• Gunjan Seth, Manpreet Singh
  Pollution, Pooling, and Premiums: Spillovers from Toxic Plant Entry on Corporate Healthcare Costs
  Presented by Manpreet Singh (Georgia Institute of Technology)
  Discussed by Jaeyong (Danny) Han (University of Massachusetts Amherst)
  Abstract

  Employer-sponsored health insurance is a major operating cost, yet little is known about how local industrial activity affects premiums. We show that toxic plant entry reduces premiums for nearby incumbent firms by about 10%, even though premiums are higher cross-sectionally in more polluted areas. The decline is larger when entrants add more workers, smaller when entrants are more toxic, and absent for self-insured incumbent firms and single-insurer markets, consistent with geographic risk pooling. Non-toxic plant entry leads to larger premium declines, while carcinogen reclassification reports increase premiums by 14%, confirming that insurers also price health-risk information. Overall, employer health insurance costs depend not only on firm characteristics, but also on the composition of the local insurance pool and insurers’ perceptions of environmental health risk.

September 25, 2026 11:30 to 13:00 (Location: Illinois 3)
2D. Housing Market Dynamics

Chair: Lu Han (University of Wisconsin - Madison )

• Neil Bhutta, Lisa Dettling
  Rising Rents and Consumer Debt
  Presented by Neil Bhutta (Federal Reserve Bank of Philadelphia )
  Discussed by Lu Han (University of Wisconsin - Madison )
  Abstract

  In the aftermath of the COVID-19 pandemic, house prices and rents surged. While research has shown clear links between house prices, consumption, and the financial conditions of home owners, there is little evidence on how rising rents affect renter households. To help address this gap, we construct a new dataset which links administrative credit card data to asking rents at the apartment-building level in order to study the borrowing, spending, and mobility response to rising rents. Exploiting within-county variation in rent growth across apartment buildings, we find that renters buffer rent shocks with increased borrowing on credit cards, especially for those who were already rent burdened. Among those already rent burdened, rising rents lead to increases in revolving balances, finances charges, and requests for credit line increases. Renters are also more likely to become delinquent and incur late fees on their credit cards after a rise in rents. Finally, we find evidence that some renters move out of their building as rents go up. All told, our results provide new insight into the distributional effects of shelter inflation and how renters cope financially with rising housing costs.

• Thomas Bezy, Antoine Levy, Timothy McQuade
  Insuring Landlords
  Presented by Timothy McQuade (Haas School of Business, UC Berkeley)
  Discussed by Neil Bhutta (Federal Reserve Bank of Philadelphia )
  Abstract

  This paper shows that non-payment risk deters landlords from renting to fragile tenants, and that insuring owners against it improves the upward geographic mobility of constrained renters to expensive, high-wage neighborhoods. Theoretically, we show that subsidizing rental insurance below actuarially fair pricing can reduce the spatial misallocation of tenants under heterogeneous financial constraints. Empirically, we test this prediction by studying Visale, a large-scale publicly funded rent guarantee insurance policy in France, free for eligible tenants and landlords. We combine tax and administrative sources on all French households, data on the universe of Visale beneficiaries and claim payouts, and sharp quasi-experimental eligibility variation by exact age. The guarantee increased access to private-sector rental housing for eligible tenants, with stronger effects for immigrants, newly formed households, and low-income renters who typically fail landlords’ screening criteria. By providing ex ante insurance, the scheme eased long-distance mobility of renters toward higher-quality neighborhoods at a moderate fiscal cost per marginal move, though potentially displacing comparable ineligible tenants through higher rents.

• Lu Han, Rachel Ngai, Kevin Sheedy
  Housing Cost Channel: Search Friction and Rents
  Presented by Lu Han (University of Wisconsin - Madison )
  Discussed by Timothy McQuade (Haas School of Business, UC Berkeley)
  Abstract

  This paper identifies a housing cost channel through which contractionary monetary policy raises rents even as it lowers house prices. We trace high-frequency monetary policy shocks through short-term rates into mortgage borrowing costs and then estimate how these shocks propagate jointly across U.S. rental and ownership markets. Higher mortgage rates weaken ownership demand and lengthen time to sell, while rents rise and rental units lease faster. New rental listings and buy-to-rent purchases also increase, yet active rental inventory initially falls as new supply is rapidly absorbed. We rationalize these joint responses in a general equilibrium model with search frictions, endogenous tenure choice, and investor entry. Higher borrowing costs shift marginal households into rental search, raising rental-market tightness and rents. Higher rents, in turn, induce investors to reallocate housing toward rental use, expanding rental supply while cushioning the decline in house prices. With search frictions, however, the added rental units are absorbed before market tightness dissipates. We calibrate the model to quantify the role of search frictions in generating the magnitude and dynamics of the observed responses across rental and ownership markets.

September 25, 2026 11:30 to 13:00 (Location: Illinois 4)
2E. Place-Based Policies

Chair: Stephen L Ross (University of Connecticut and NBER)

• Dimitria Freitas
  The Effect of Public Sector Relocations on Regional Development in Germany
  Presented by Dimitria Freitas (Humboldt University & Thuenen Institute)
  Discussed by Joseph Aguilar-Bohorquez (University of Minnesota)
  Abstract

  As regional economic disparities within countries grow, governments are increasingly experimenting with public employment reallocation as a place-based policy. In this paper, I estimate the causal effect on local labor markets of a German policy that relocated about 3,000 public sector jobs to lagging regions. Using novel data on 60 agency relocations from 2015–2025, I estimate employment and population effects for receiving and sending municipalities and use the results to discipline a quantitative spatial model that simulates spatial reallocation and welfare effects. I find that relocations increased private employment shares by up to 2.3% (1.3 percentage points), reduced unemployment by up to 11.9% (0.33 percentage points), and raised population by 1.6%, implying a public-to-private jobs multiplier of 1.08 in receiving locations. Sending locations also see an increase in private sector employment that is rationalized in the model by within-private-sector spillovers exceeding public-to-private sector spillovers.

• Chiara Motta
  The Labor Market Incidence of Place-Based Policies
  Presented by Chiara Motta (University of California, Berkeley)
  Discussed by Adam Scavette (Federal Reserve Bank of Philadelphia)
  Abstract

  Place-based subsidies are a central instrument of regional policy, yet where their gains ultimately land-at the participating firm, among its workers, or at non-participating competitors-depends on how firms compete for workers in local labor markets. We provide the first unified quasi-experimental analysis of this full incidence. Using Italy's EU-funded Incentives for Disadvantaged Areas (IDA, 1996-2007) linked to the universe of Italian employer-employee administrative records, we exploit quasi-random assignment near within-ranking funding cutoffs to document three sets of effects. Marginally funded firms expand employment by 7-13% and raise pay by 1-2%, with these gains not driven by changes in workforce composition. Within these firms, gains accrue to incumbent stayers-consistent with improved retention-while entry wages for new hires remain unchanged. Combining the program's budget-allocation rules with simulation-based exposure measures, we trace spillovers through local labor markets: a 10% expansion in local market employment raises wages at non-participating competitor firms by approximately 1.3% over a multi-year horizon. Together, these findings show that labor market imperfections-operating through retention frictions within firms and scale effects across them-shape the full incidence of place-based policy in ways not accounted for by standard welfare frameworks.

• Joseph Aguilar-Bohorquez
  American Indian Casino Operations and Firm Long-Run Employment and Sales
  Presented by Joseph Aguilar-Bohorquez (University of Minnesota)
  Discussed by Chiara Motta (University of California, Berkeley)
  Abstract

  American Indian reservations face long-standing barriers to private-sector development. The Indian Gaming Regulatory Act of 1988, which enabled the establishment of casino operations on American Indian tribal lands is one of the most consequential sovereignty-driven, place-based interventions in Native communities. We study its impact on local business revenues and job creation by linking geocoded establishment-level microdata to federally recognized reservation boundaries and a panel of tribal casino openings from 1990 to 2019, using a staggered-adoption event study with never-treated reservation controls. Casino openings lead to economically meaningful and persistent growth in reservation business activity: average establishment sales increase by 47% and employment increases by 50% after opening. Importantly, incumbent establishments also expand, with sales increasing by 17% and employment increasing by 26%, indicating that development benefits extend beyond new entry and beyond the casino itself. When excluding gaming-adjacent industries, employment effects remain positive while sales effects attenuate and are less precisely estimated. Overall, the results provide evidence that tribal casino development leads to sustained private-sector expansion inside reservation boundaries, consistent with local demand, business-to-business spending, and visitor-spending spillovers, and highlight a mechanism through which Indigenous self-determined enterprise can contribute to long-run economic development.

September 25, 2026 11:30 to 13:00 (Location: Illinois 5)
2F. US History

Chair: Allison Shertzer (Federal Reserve Bank of Philadelphia)

• Anais Galdin, Quan Le
  Urban Migration, Public Health Amenities, and Local Newspapers in 1870-1940 U.S.
  Presented by Quan Le (Harvard)
  Discussed by Allison Shertzer (Federal Reserve Bank of Philadelphia)
  Abstract

  We study how local newspapers shaped rural-to-urban migration in the United States, 1870–1940. We combine newly digitized full-count U.S. censuses with data on daily newspaper presence and the flow of news articles across places to measure how migrants respond to information about destination amenities. We focus on municipal water filtration, which sharply reduced waterborne disease mortality. We show that filtration reduces waterborne disease coverage in the news and increases in-migration to cities, with substantially larger effects from origins served by a daily newspaper. Using data on news article flows, we further show that migration is higher toward destinations with greater news exposure, but not when that coverage concerns waterborne disease. We interpret these findings as evidence of information frictions in migration decisions and of local newspapers’ role in alleviating them.

• Michael Coury, Christopher Severen, Allison Shertzer
  Waterfronts as Amenities: Evidence from the Chicago River Reversal
  Presented by Michael Coury (University at Buffalo)
  Discussed by Quan Le (Harvard)
  Abstract

  Although waterfront property now commands a premium in modern cities, rivers and lakes near dense settlements were befouled with sewage for much of human history. This paper studies how the removal of human waste from surrounding waters transformed the urban environment using the engineering feat of the Chicago River reversal in 1900, which produced rapid improvements in surface water quality in a dense urban area. We find that rental prices were highly responsive to sewage exposure. The permanent reversal of the Chicago River increased listed rental prices by over 100\% for properties nearest to the lake shore. This effect decreases with distance from the lake and is persistent over time. We also provide descriptive evidence that improved water quality led to an influx of professional workers to residences near the lakeshore.

• Prottoy Akbar, Rene Livas, Allison Shertzer
  Electrified Transit and the Transformation of the American Downtown
  Presented by Allison Shertzer (Federal Reserve Bank of Philadelphia)
  Discussed by Michael Coury (University at Buffalo)
  Abstract

  Did the transportation innovations of the late 19th century transform the American downtown? We study the impact of electrified transit in Philadelphia on the specialization of economic activity within the city beginning with the first subway and elevated railway in 1907. To measure worker place of residence, workplace locations, and commuting patterns, we construct a panel dataset of roughly one million entries from Philadelphia's city directories over the 1887--1930 period. Using a market access approach, we find that electrified transit caused blocks near the central business district to specialize as workplaces while blocks further away specialized as residences. This effect was especially pronounced for workers in professional services, whose jobs remained in the downtown while their places of residence shifted to the urban periphery. Our preliminary results suggest that electrified transit led to increasing specialization in land use.

September 25, 2026 11:30 to 13:00 (Location: Moskow)
2H. Housing Policy

Chair: Patrick Kennedy (UCLA, NBER)

• Hector Blanco, Luca Perdoni
  New Deal Public Housing and Racial Segregation in U.S. Cities
  Presented by Hector Blanco (Rutgers University)
  Discussed by Evan Soltas (Princeton University)
  Abstract

  Racial residential segregation in U.S. cities rose sharply during the first half of the twentieth century. We study whether early federal public housing contributed to this rise by examining the first projects built by the Public Works Administration in the mid-1930s, most of which were racially designated. Using newly assembled data on project locations linked to full-count Census records from 1910 to 1950 and comparing built to planned-but-not-built projects, we find that public housing reinforced the racial composition of project sites but had little effect on surrounding neighborhoods. The results suggest that early public housing played a limited role in shaping neighborhood-level segregation.

• Evan Soltas, Allan Lee
  Not So Affordable? Low-Income Housing and the Price of Public Capital
  Presented by Evan Soltas (Princeton University)
  Discussed by Patrick Kennedy (UCLA, NBER)
  Abstract

  Capital projects in the U.S. public sector have become costlier but also possibly more complex. To what extent do rising costs reflect quality or compositional changes versus true changes in the price of public capital? Using new data from 3,400 competitive proposals for low-income housing projects, we find real construction costs rose 80 percent from 2013 to 2024, holding fixed project size and location, an increase not explained by input prices. In ongoing analysis, we take a utility-theoretic approach to quality, estimating government preferences from funding decisions and using reapplicants to link competitions over time. Preliminary results suggest quality is stable over time, implying a substantial decline in the purchasing power of tax credits.

• Patrick Kennedy, Harrison Wheeler
  Neighborhood-Level Investment from the U.S. Opportunity Zone Program
  Presented by Patrick Kennedy (UCLA, NBER)
  Discussed by Hector Blanco (Rutgers University)
  Abstract

  We are in the process of updating this draft with up-to-date evidence, which will be ready before the conference in September. We use de-identified federal tax records to document evidence on housing and labor market responses to the Opportunity Zone (OZ) program, a federal place-based policy that provides tax incentives for capital investments in more than 8,000 low-income neighborhoods across the United States.

September 25, 2026 14:00 to 15:30 (Location: Iowa)
3A. Economic Geography

Chair: Simon Fuchs (FRB Atlanta)

• Aleksandr Gevorkian
  Universities in Space
  Presented by Aleksandr Gevorkian (UCLA)
  Discussed by Simon Fuchs (FRB Atlanta)
  Abstract

  Higher education raises human capital, and both its acquisition and use are spatial. This paper studies how universities shape the spatial allocation of talent, and how a central educational planner would design the system. I build a dynamic spatial general equilibrium model with a system of colleges differentiated by geography, quality and size. Returns to attending a particular university come from (1) learning, which depends on college characteristics and peer composition, and (2) adjusted college premia in all locations. Mobility frictions at both stages shape how individuals of different ability and parental wealth sort into local labor markets. Using Swedish administrative microdata together with open-source data, I quantify the model and decompose the role of college selection and labor-market sorting in shaping welfare. I then use the model to study aggregate and distributional effects of alternative allocations of university spots, including Sweden's 1977--1984 higher-education expansion reform. I repeat the exercise for commonplace educational policies like moving subsidies. Finally, I characterize the optimal allocation and compare it to the status quo.

• Miquel Lorente Palau
  Quantifying the effects of the 2025 US trade war on Europe
  Presented by Miquel Lorente Palau (Universitat Pompeu Fabra)
  Discussed by Aleksandr Gevorkian (UCLA)
  Abstract

  This paper quantifies the macroeconomic and welfare implications of a permanent decline in globalization triggered by the 2025 US trade war. The analysis focuses on European regions, while also reporting aggregate effects for the United States. I develop a tractable dynamic general equilibrium model featuring bilateral trade, multinational production, capital flows, and migration. The dynamic structure is central, as it captures intertemporal adjustments in investment and international capital allocation. I find that US tariffs reduce the trade deficit only temporarily. While higher tariffs attract some foreign production to the United States, they simultaneously reduce export-oriented activity and lower real value added. Over time, the US capital stock declines as investment goods become more expensive and US production faces higher taxes. Real GDP and welfare fall in both the United States and the European Union, both initially and in the long run. The effects are highly heterogeneous across European regions, even within countries. EU retaliation only partially mitigates welfare losses, while further escalation of the trade war would amplify them on both sides of the Atlantic.

• Simon Fuchs
  Urban Welfare: Tourism in Barcelona
  Presented by Simon Fuchs (FRB Atlanta)
  Discussed by Miquel Lorente Palau (Universitat Pompeu Fabra)
  Abstract

  We develop a tractable framework for measuring the heterogeneous welfare effects of small short-run urban shocks using a sufficient-statistics approach that captures spatial general-equilibrium propaga- tion without requiring a fully parametric city model. We apply the framework to seasonal tourism demand shifts in Barcelona using high-resolution data on residents’ spending, income, and commut- ing, together with shift-share variation generated by differences in when tourists from different origin countries visit the city and where they spend within it. On average, the implied real-income effect between low and high tourist seasons is modestly negative, but the incidence is highly heterogeneous: welfare changes range from−1.46 percent at the 10th percentile of locations to +1.14 percent at the 90th percentile. Inner-city neighborhoods bear the largest losses, while some peripheral neighborhoods benefit from indirect income gains.

September 25, 2026 14:00 to 15:30 (Location: Michigan)
3B. Wages II

Chair: Donald Davis (Columbia University)

• Klaus Desmet, Wookun Kim, Keyoung Lee
  Decomposing Gender and Racial Wage Gap across the U.S.: Workers, Firms, and Cities
  Presented by Wookun Kim (Southern Methodist University)
  Discussed by Donald Davis (Columbia University)
  Abstract

  We study the spatial dispersion of gender and racial wage gaps across U.S. cities. First, we document substantial geographic dispersion in both wage gap measures across U.S. metropolitan areas, and show that cities with larger gender wage gaps also tend to have larger racial wage gaps. Second, we build a model of worker and firm location featuring group-specific Nash bargaining over wages and task bundles in production. The key implication of the model is the log-linear wage structure which motivates a three-way fixed-effects decomposition of wages into worker, firm, and city components, an extension of the standard AKM model, which we estimate using matched employer-employee data. Even after absorbing worker and firm heterogeneity, the same city offers different returns to different groups: group-specific location fixed effects account for 15.8 percent of the spatial variance in the wage gap. Our findings suggest that place is an independent determinant of gender wage inequality, and that understanding spatial variation in local labor market conditions is essential for explaining persistent wage gaps across different groups of workers.

• Pauline Carry, Benny Kleinman, Elio Nimier-David
  Location Effects or Sorting? Evidence from Firm Relocation
  Presented by Elio Nimier-David (Cornell University)
  Discussed by Wookun Kim (Southern Methodist University)
  Abstract

  What role do firms play in geographic wage disparities? This paper exploits establishment mobility as a novel source of identification to separate firm sorting from ``location effects'' (e.g., infrastructure, agglomeration, amenities). Using data from France and the U.S., we first document relocation patterns: 4% of establishments move each year, retaining their activity and structure while adjusting their workforce and wages. Combining establishment and worker mobility in France, we estimate a model that decomposes wage variation due to workers, firms and locations. Spatial wage differences are primarily driven by sorting: worker composition accounts for 30%, firm composition for 17%, and their co-location for 34%. Location effects themselves explain only 2–4%. Revisiting the urban wage premium, we find that establishment sorting accounts for most of the higher level and faster growth of wages in larger cities.

• Donald Davis, Eric Mengus, Tomasz Michalski, Elie Vidal-Naquet
  Wage Growth Within and Across Cities
  Presented by Donald Davis (Columbia University)
  Discussed by Elio Nimier-David (Cornell University)
  Abstract

  A growing literature debates whether urban wage premia stem from the sorting of firms and skilled workers or from true location effects. We argue that resolving this requires examining finer spatial scales than the traditional commuting zone, focusing specifically on job-dense central business districts that promote knowledge spillovers. Using geocoded French administrative data, we analyze spatial heterogeneity both within and between cities by distinguishing high and low job-density locations at a highly granular level. We partition jobs based on analytical and interpersonal skill requirements and estimate localized, occupation-specific returns to experience. We find that a small fraction of highly dense areas within large cities drives both the dynamic and static components of the urban wage premium. Specifically, after ten years of experience, workers in highly analytical and interpersonal occupations located in the 50 densest square kilometers of the Paris metropolitan area experience dynamic wage gains 3.4 times higher than similar workers elsewhere in Paris (+88pp versus +26pp, relative to workers outside large cities). Notably, discernible wage premia for these workers only occur within the 315 densest square kilometers—just 0.3% of France's working area. Occupations with high computer use show particularly strong gains, indicating that task content and local density jointly shape wage trajectories. These results remain robust when controlling for establishment fixed effects or the local evolution of job types and industry wages.

September 25, 2026 14:00 to 15:30 (Location: LaSalle)
3C. Climate and Environment

Chair: Adrien Bilal (Stanford University)

• Rodrigo Adao, Arthur Bragança, Jonathan Dingel, Esteban Rossi-Hansberg
  Whose Tree Is It Anyway? Property Rights and Deforestation in the Brazilian Amazon
  Presented by Esteban Rossi-Hansberg (University of Chicago)
  Abstract

  We study how the de-jure assignment of property rights over public land shapes deforestation in the Brazilian Amazon. All Amazonian land is government-owned, and deforestation is illegal everywhere, yet two legal categories that coexist on the same forest differ in one critical respect: under Brazil's 1988 Constitution, perpetual public land---conservation units, indigenous territories, and military areas---cannot be privatized once deforested, while other public land can ultimately be granted to occupants and resold to agribusiness. Using satellite data on a panel of 4.4 million 1km cells covering 1987--2024 and the official public forestry registry, we document that deforestation, privatization, and land-use change vary discontinuously across the border between perpetual and other public land, while enforcement activity does not. We then build a dynamic spatial model in which farmers choose whether to deforest and occupy frontier plots and agribusinesses buy privatized deforested plots for scale-intensive, export-oriented use. The model formalizes the option value of acquiring formal use rights as the wedge that makes a forested plot of other public land more attractive to deforest than an otherwise identical plot of perpetual public land. The quantified model replicates the cross-sectional discontinuity in deforestation despite targeting only aggregate moments, and predicts that removing the privatization path for other public land would reduce cumulative deforestation by roughly 33\% by 2150.

• Ashwin Rode, Rachel Baker, Tamma Carleton, Michael Greenstone, Solomon Hsiang, Andrew Hultgren, Amir Jina, Robert Kopp, Ishan NathMatthew Pecenco, James Rising
  Is workplace temperature a valuable job amenity? Implications for climate change
  Presented by Tamma Carleton (University of California, Berkeley)
  Abstract

  Using data representing one-third of the world's population, we find that extreme hot and cold days cause substantial labor supply declines for weather-exposed workers, but not for weather-protected workers. With these results and a simple theoretical framework, we calculate that the value of a weather-protected job's thermal comfort varies widely globally but is worth 2.9% of annual income on average. We project that climate change will increase worker thermal discomfort by 1.8% of global GDP in 2099 under a very high emissions scenario and 0.5% under an intermediate scenario, demonstrating the importance of this new category of climate damages.

• Adrien Bilal, Santiago Franco, Esteban Rossi-Hansberg
  Climate Risk in Dynamic Spatial Equilibrium
  Presented by Adrien Bilal (Stanford University)
  Abstract

  We study the economic implications, over time and space, of the aggregate and local risks arising from climate change. We develop a dynamic spatial model of the U.S. economy disaggregated at the county level that features costly forward-looking migration and capital investment decisions, aggregate risk due to uncertain warming trajectories, and local risk due to uncertain exposure to extreme weather. We achieve tractability by leveraging the ‘Master Equation’ representation of the economy. We estimate structural elasticities and damage functions by matching reduced-form estimates linking population, income, and investment to coastal storms and heat waves. Counterfactual exercises show that local risk generates sizable welfare losses for capitalists in exposed coastal areas and in the aggregate, while raising capital through a precautionary savings mechanism. By contrast, the real option embedded in capital investments implies that aggregate risk leads to modest capital depletion in some southern coastal counties and negligible aggregate welfare losses.

September 25, 2026 14:00 to 15:30 (Location: Illinois 3)
3D. International Housing Policy

Chair: Albert Saiz (MIT)

• Alison Lodermeier
  Who Pays Higher Rent and Why? Price Differentials in the Rental Housing Market
  Presented by Alison Lodermeier (University of Chicago)
  Discussed by Albert Saiz (MIT)
  Abstract

  This paper tests whether different groups of renters pay different prices for housing units of the same quality. I construct a monthly panel of nearly 2 million Australian rental units matched to tenants, landlords, rent prices, operating costs, and rent payments. Using a within-unit repeat rent design, I estimate that immigrant renters pay an 0.9% rent premium and renters with a disability receive a 1.2% rent discount. The immigrant premium is driven entirely by non-European immigrants and recent arrivals. The disability discount can be explained by lower costs when units are occupied by renters with a disability, but the immigrant premium cannot—landlords pay lower operating costs and earn higher net operating income when renting to immigrants. In ongoing work, I leverage property ownership changes during ongoing tenancies to test whether the remaining immigrant premium can be explained by differential price elasticities or landlord bias. Taken together, these results have implications for understanding disparities in housing cost burdens and targeting housing assistance.

• SUMIT Agarwal, Ying Deng, Yi Fan, Qi GAO, Jing Li, Lin Ma
  Externalities of Housing Upgrading in High-density Cities
  Presented by Jing Li (Singapore Management University)
  Discussed by Alison Lodermeier (University of Chicago)
  Abstract

  We estimate housing externalities in a high-density urban setting, using Singapore’s nationwide public housing upgrading initiative 1990-2006. Exploiting staggered implementation and controlling non-random neighborhood exposure, we find upgrading raises resale prices within 500 meters by 2.1%, while driving up the price by 11.7% upon completion; effects decay to zero beyond 1,000 meters. A model with distance-decaying externalities shows that in-kind upgrading yields positive externalities large enough to justify distortions it imposes on dense settings with amplified spillovers. Under lower-density counterfactuals, this advantage dissipates. Upgrading disproportionately retains older incumbent residents, pointing to age-specific amenities as a channel through which externalities propagate.

• Albert Saiz
  Building More Than Homes: How Interior Housing Upgrades Catalyze Welfare and Investment
  Presented by Albert Saiz (MIT)
  Discussed by Jing Li (Singapore Management University)
  Abstract

  We evaluate a qualitative housing improvement program---upgrades to kitchens, floors, and bathrooms---using a cluster randomized controlled trial with 1,163 low-income urban households across three Colombian cities. The Hogares Saludables program, implemented by a private-sector firm, combines physical improvements with a 40-hour construction and life skills training course. Using difference-in-differences with propensity score matching to address baseline imbalances, we find that the intervention improves mental health (+0.054, $p < 0.01$), reduces fever among children under six (20.2~pp, $p < 0.05$), lowers unemployment (8.6~pp, $p < 0.01$), decreases debt (15.7~pp, $p < 0.01$), and raises income expectations (16.0~pp, $p < 0.05$). Effects on cleaning and aspirations are larger for female-headed households and those with children. We find no effects on domestic violence or family harmony. Thirty-one percent of treated households undertook additional home improvements after the intervention, suggesting that interior housing upgrades act as a catalyst for continued investment and community engagement.

September 25, 2026 14:00 to 15:30 (Location: Illinois 4)
3E. Risk and Property Insurance

Chair: Nitzan Tzur-Ilan (Federal Reserve Bank of Dallas)

• Augusto Ospital
  Housing Supply, Property Insurance, and Exposure to Wildfire Risk
  Presented by Augusto Ospital (LMU Munich)
  Discussed by Nitzan Tzur-Ilan (Federal Reserve Bank of Dallas)
  Abstract

  In the past two decades, about half of the new homes in the United States were built in areas at risk of natural hazards. Why is residential development exposed to such risk? I argue that regulated property-insurance pricing and land-use regulations contribute to this pattern. I study this mechanism in the metropolitan area of San Diego, California, where insurance rules compress the premium gradient with respect to wildfire risk and safer locations are highly regulated and built out. Using detailed spatial data on zoning, wildfire risk, housing, commuting, and premiums, I estimate a quantitative urban model of household location choice, housing supply, and insurance supply. The estimates imply that wildfire premiums are 10.5% below actuarially fair pricing, that the average amenity cost of current wildfire risk is equivalent to 3.5% of income, and that the total present-value welfare cost of current wildfire risk, including property damages, is $17.5 billion. This aggregate cost masks substantial incidence heterogeneity, as owners of safe land benefit from equilibrium scarcity effects. Counterfactuals show that housing supply and insurance pricing interact in determining incidence. In the benchmark specification, targeted housing reforms leave the aggregate effect of cost-based insurance nearly unchanged while attenuating its burden on workers: relative to baseline, workers' wildfire costs rise by 2.3% under insurance reform alone, but fall by 0.9% under the joint reform. Moving away from compressed wildfire prices toward more risk-based pricing is a central component of California’s recently implemented Sustainable Insurance Strategy.

• Stephen Buschbom, Evan Eastman, Chongyu Wang, Tingyu Zhou
  Climate Risk and Commercial Property Insurance
  Presented by Tingyu Zhou (Florida State University)
  Discussed by Augusto Ospital (LMU Munich)
  Abstract

  This paper provides the first comprehensive analysis of the trends, drivers, and economic incidence of insurance costs across all major commercial real estate sectors in the United States. Using a novel panel of over 137,000 commercial properties from 2010 to 2023, we exploit lender mandates for comprehensive, non-negotiable coverage to identify the true market price of climate risk, unconfounded by the under-insurance biases common in residential studies. We document a structural break in pricing beginning in 2018, with premiums rising sharply across all sectors and doubling in multifamily and lodging. These escalations are strongly linked to expected climate risk. Using a stacked difference-in-differences design, we find that realized climate shocks cause a persistent 3% increase in insurance costs. Finally, we estimate the pass-through of these costs to rents and find that landlords absorb the majority of the cost shock. However, there is a pronounced asymmetry in high-climate-risk markets, where supply constraints allow landlords to pass a significantly larger share of costs to tenants. These results highlight how insurers price both expected and realized climate risk, with implications for asset valuation, lending, and policy in an increasingly uncertain risk environment.

• Shan Ge, Stephanie Johnson, Nitzan Tzur-Ilan
  The Effect of Rising Insurance Premiums on Mortgage Delinquency and Household Relocation
  Presented by Nitzan Tzur-Ilan (Federal Reserve Bank of Dallas)
  Discussed by Tingyu Zhou (Florida State University)
  Abstract

  Homeowners' insurance premiums in the U.S. have risen dramatically. Using household-level microdata and employing a novel instrumental variable, we document three main findings. First, consistent with mitigating the impact of higher premiums, households experiencing larger premium hikes are more likely to relocate and tend to move to areas with lower insurance costs. Homes experiencing larger premium hikes are also more likely to be acquired by corporate buyers, who obtain lower premium rates. Second, for households unable to mitigate effectively, higher premiums increase mortgage delinquencies. Third, the relocation effect is stronger among less financially constrained households, who can better afford the upfront costs of moving, while the delinquency effect is concentrated among more constrained households. Our findings reveal how rising insurance costs—increasingly driven by climate change—can threaten household financial resilience, while highlighting an adaptation mechanism through relocation.

September 25, 2026 14:00 to 15:30 (Location: Illinois 5)
3F. History of Housing

Chair: Sarah Quincy (Vanderbilt University)

• Jeffrey Lin, Dan Moulton, Isaac Rand, Robyn Smith
  Automated Digitization of the Censuses of Housing Block Statistics, 1940–1970
  Presented by Isaac Rand (Federal Reserve Bank of Philadelphia)
  Abstract

  Beginning in 1940, the Censuses of Housing Block Statistics represent the earliest, comprehensive, small-geography US Census Bureau estimates of the quantity and condition of housing units in US cities. We develop automated methods to digitize both tabular data and block maps from page scans of the original census publications, 1940--1970. These methods include both traditional image-processing techniques and machine learning and build on existing NHGIS tabular and geospatial census tract data. We automate map georeferencing, text detection and recognition, and raster to vector conversion. We plan to distribute code, training data, and digitized block statistics for 16 major cities.

• Rowena Gray
  Rental Prices and the Cost of Living in the United States, 1914-2006
  Presented by Rowena Gray (UC Merced)
  Abstract

  The Rent of Primary Residence (RoPR) series constructed by the Bureau of Labor Statistics (BLS) implies that nominal rental prices increased by just 2.6% per year from 1914 to 2006 while overall prices grew by 3.3%. We show that this ”falling real rents” puzzle can be explained by the evolving treatment of shelter in the Consumer Price Index (CPI). In this paper we construct a new, methodologically consistent shelter price series using the Historical Housing Prices (HHP) Project rental index. We also construct a revised set of shelter weights going back to 1914 and combine them with the price series to create an alternate CPI that applies the owners’ equivalent rent (OER) concept of shelter consistently across time. The HHP shelter price series increases by a factor of 28.4 (compared with the 10.7 increase in RoPR) and lifts average CPI growth from 3.3% to 3.6% per year. The revised series eliminates the long-run decline in real rents in the CPI and provides a new benchmark for assessing trends in the cost of living and real income in the U.S. over the twentieth century.

• Asaf Bernstein, Peter Koudijs, Sarah Quincy
  Curing the Mortgage Debt Disease: Evidence from the Great Depression Foreclosure Crisis
  Presented by Sarah Quincy (Vanderbilt University)
  Abstract

  Before the Great Depression, there was substantial variation in mortgage contracts. Apart from now-common fully amortizing mortgages, interest-only contracts were also widespread in a large cross-section of borrowers, often depending on lending institutions' policies that varied across space and time. Further, mortgage maturities, and therefore the speed of amortization, could differ substantially. Using different sources of variation, we show that mortgage amortization led to fewer foreclosures after 1929. Results suggest that this was not driven by reduced roll-over risk or virtuous selection into amortization. Our findings highlight the importance of mortgage design for financial stability and suggest that the Great Depression's housing crisis was exacerbated by the widespread use of interest-only mortgages.

September 25, 2026 14:00 to 15:30 (Location: Directors)
3G. Imperfect Competition in Real Estate

Chair: Elisa Giannone (CREI)

• Dayin Zhang
  Market Power in Mortgage Pricing: the Role of Referral Lending
  Presented by Dayin Zhang (University of Wisconsin-Madison)
  Discussed by Elisa Giannone (CREI)
  Abstract

  Despite intense competition among mortgage lenders, borrowers continue to face high mortgage rate spreads and substantial price dispersion. We argue that realtor–loan officer referral networks are an important source of lender market power: by steering homebuyers toward a narrow set of loan officers, these networks limit effective borrower choice even in otherwise competitive markets. Using a novel dataset linking 81,306 realtors to 102,860 loan officers across 41 states, we document that these networks are pervasive and highly concentrated: 85% of realtors direct more than 40% of their clients to fewer than four loan officers. This concentration persists—and even rises—in markets with more lenders, suggesting that referrals constrain borrower choice independently of market structure. Exploiting variation in referral concentration, we find that borrowers matched with referred loan officers pay 18.6 basis points higher mortgage rates, equivalent to $2,609 in upfront costs for the average $306,000 loan. The referral premium is substantially larger for Hispanic borrowers, and is also elevated for Black borrowers and financially constrained households. On average, referral lending raises rate spreads by 36.5% and explains roughly half of the residual standard deviation in mortgage rate spreads. We identify two mechanisms: referrals reduce borrowers’ lender search, and referred loan officers retain pricing power relative to other officers within the same institution. Efficiency-based explanations, such as faster processing or reduced denial risk, do not fully justify the premium. Our results show that referral networks are a hidden source of market power in mortgage lending, with meaningful financial and distributional consequences.

• Felipe Barbieri, Gregory Dobbels
  Market Power and the Welfare Effects of Institutional Landlords
  Presented by Felipe Barbieri (Dartmouth Tuck)
  Discussed by Dayin Zhang (University of Wisconsin-Madison)
  Abstract

  In the last decade, large financial institutions in the United States have purchased hundreds of thousands of homes and converted them to rentals. This paper studies the welfare consequences of institutional ownership of single-family housing. We build an equilibrium model of the housing market with two sectors: rental and homeownership. The model captures two key forces from institutional purchases of homes: changes in rental concentration and reallocation of housing stock across sectors. To estimate the model, we construct a novel dataset of individual homes in metropolitan Atlanta, identifying institutional owners of each house and collecting house-level daily prices, rents, vacancies, web page views, and customer contacts from Zillow. Overall, we find that institutional acquisitions decrease rents and increase rental transactions, leading to large welfare gains for renters. This net benefit reflects two opposing forces: while higher concentration raises rents, higher rental supply lowers rents enough to more than offset the effect of concentration, pushing rents down overall. These renter gains come at the expense of homebuyers, whose welfare falls. On the supply side, institutional acquisitions benefit house sellers but harm the average landlord.

• Elisa Giannone, Xinle Pang, Clara Santamaria
  Landlords in Space
  Presented by Elisa Giannone (CREI)
  Discussed by Felipe Barbieri (Dartmouth Tuck)
  Abstract

  How does housing ownership concentration affect affordability? This paper provides new evidence and a new mechanism. Using Spanish administrative panel microdata that link household balance sheets to dwelling characteristics, location, rental income, transactions, inheritances, and tax-reported use, we show that non-primary housing in Madrid and Barcelona is highly concentrated among wealthy owners. But the channel is not that rich landlords charge higher rents for comparable homes: rent premia largely disappear within narrow location-quality cells, while wealthy owners earn lower current yields and hold properties with higher realized price growth. Motivated by the evidence, we build a dynamic urban equilibrium model with homeownership and a key new feature: households can accumulate wealth by investing in multiple dwellings in different locations and need not enter the long-term rental market. Owners choose whether to rent, sell, or keep units available for option value and capital gains. Calibrating the model to Barcelona’s districts, we find that concentrated ownership therefore changes effective rental supply, rather than demand alone. The mechanism raises prices more than rents and redistributes welfare from renters and constrained buyers toward incumbent owners and investors, pushing poor renters toward more peripheral areas.

September 25, 2026 14:00 to 15:30 (Location: Moskow)
3H. Institituions and Neighborhood Effects

Chair: Daniel Hartley (Federal Reserve Bank of Chicago)

• Dionissi Aliprantis, Stefanie DeLuca
  Unlocking Opportunity: The Remarkable Success of the Baltimore Regional Housing Partnership
  Presented by Dionissi Aliprantis (ENS de Lyon)
  Abstract

  Existing research shows that the neighborhood where a child grows up has a causal effect on their later economic prospects. Yet in many cities, low-income nonwhite families with children often live in the lowest opportunity neighborhoods. Over the past few decades, a number of housing mobility programs using vouchers to assist families in making moves to higher-resourced communities have shown considerable success toward increasing neighborhood opportunity, garnering significant policy support. As public housing policy is increasingly influenced by Housing Mobility Programs (HMPs), we study the HMP that has generated the largest improvements in neighborhood characteristics. Participants in the Baltimore Regional Housing Partnership (BRHP) reside in neighborhoods with schools performing 40 percentile points higher on state tests than poor Black residents of Baltimore City – ten times the difference between experimental and control groups in Moving to Opportunity. Oaxaca-Blinder decompositions attribute the majority of the BRHP’s success to its regional design, which allows participants to access all opportunity neighborhoods in metropolitan Baltimore. The BRHP breaks strong neighborhood sorting by income and race. BRHP households live in neighborhoods with socioeconomic status comparable to the highest income Black households and live in more racially-integrated neighborhoods than Black households at any income level. BRHP improvements in neighborhood characteristics are durable.

• Bocar Ba, Iris Chang, Tony Cheng, Maxime Menuet, Alexander Whitefield
  When Protests Work: Institutional Change in Public Safety After 2020
  Presented by Bocar Ba (duke)
  Abstract

  Under what conditions do protest movements produce lasting policy change? We study this question in the context of the Black Lives Matter protests following the murder of George Floyd in May 2020. Exploiting cross-county variation in protest exposure in a difference-in-differences framework, we document a sharp disconnect between political demand and realized policy along the most obvious margin for change: police retrenchment. Protest exposure reduced support for maintaining or increasing police funding by 3--4 percentage points in 2020, an effect driven entirely by Democratic respondents, but these effects dissipated within two years. Consistent with this pattern, the protests had no measurable effect on police staffing. We show, however, that the protests induced substantial and persistent change along a novel institutional margin: nonpolice crisis response programs. We construct a new county-level dataset tracking the rollout of such programs between 2016 and 2023 and find that protest exposure increased the probability of their adoption by 5.9 percentage points (pp) on average and by up to 11 pp in 2023. These effects are concentrated in civilian programs operating outside police departments and in Democratic-leaning counties. Rather than reducing policing directly, the protests reallocated policy change toward alternative, more feasible institutional margins. Their long-run impact has operated not through police retrenchment but through a durable reconfiguration of the institutional structure of public safety.

• Nathaniel Baum-Snow
  Gentrification and Neighborhood Effects: Evidence for Children and their Parents
  Presented by Nathaniel Baum-Snow (Rotman School of Management, U Toronto)
  Abstract

  This paper investigates the consequences of neighborhood gentrification for incumbent children and their parents. Using labor demand shocks oriented toward college and less than college workers as sources of exogenous variation, we provide evidence that growing up in a gentrifying neighborhood improves children’s labor market outcomes when they in their 20s and 30s. Positive effects are concentrated among the children growing up in the least educated neighborhoods with the lowest income parents, with null effects for most other groups. We also see larger effects in central cities than in suburbs. While some of these treatment effects may be mediated through associated improved parent outcomes, relative magnitudes of estimates indicate a central role for neighborhood effects in driving these positive long-run outcomes among children.

September 25, 2026 14:00 to 15:30 (Location: Illinois 1)
3I. Chinese Integration

Chair: Andrew Waxman (University of Texas at Austin)

• Victor Couture, Benjamin Faber, Cecile Gaubert, Yizhen Gu, Ming Li, Wei Lin
  Mobile Internet and Market Integration
  Presented by Victor Couture (UNIVERSITY OF TORONTO)
  Discussed by Ben Zou (Purdue)
  Abstract

  We study how the rollout of mobile internet integrated China's domestic market between 2014 and 2018. We use the universe of firm-to-firm VAT transactions from China's State Taxation Administration, combined with a new county-level measure of bilateral internet coverage built from Taobao user data. A 10 pp increase in bilateral coverage raises bilateral trade flows by about 3.4 percent, with origin coverage contributing roughly twice as much as destination coverage. The effect operates almost entirely through the formation of new buyer-seller matches, and is stronger between more distant county pairs. We embed these estimates in a quantitative trade model with matching frictions: the observed 38 pp average rise in bilateral coverage raised China's real GDP per capita by 3.3 percent, with gains concentrated in initially richer and denser regions.

• Junni Pan, Ben Zou
  Migration Restrictions, College Choices, and Spatial Skill Sorting
  Presented by Ben Zou (Purdue)
  Discussed by Andrew Waxman (University of Texas at Austin)
  Abstract

  When choosing where to attend college, students may also consider future job opportunities in the city where the college is located, especially when post-graduation mobility is limited by migration frictions. This paper examines how such frictions shape college choices in China, where mobility is constrained by both formal migration restrictions and informal barriers. Using a national administrative dataset on four-year college admissions from 2005 to 2011, we show that relaxing migration restrictions through hukou reforms enabled colleges in cities that implemented reforms to attract higher-quality students. The largest gains occurred at colleges located in cities that are economically more developed than students’ origins, consistent with improved local labor market prospects as the underlying mechanism. Counterfactual analysis based on a college choice model indicates that expanding hukou access to higher-wage cities strengthens the sorting of stronger students toward those destinations. However, when informal migration frictions are lower, students face lower post-graduation migration costs and become less sensitive to wages in college cities; as a result, sorting toward high-wage destinations weakens. Aggregate welfare increases as migration restrictions are eased, although the gains are unevenly distributed. These results highlight the role of both formal and informal migration frictions in shaping spatial skill sorting and welfare.

• Andrew Waxman
  A Long Way from Home: Migration and Commuting in Urban China
  Presented by Andrew Waxman (University of Texas at Austin)
  Discussed by Victor Couture (UNIVERSITY OF TORONTO)
  Abstract

  A long literature in labor economics has documented differential behavior of migrants, but the mechanisms linking migration to commuting and location choice remain underexplored. We document several features of migrant commuting in modern China: migrants live closer to work and consume less housing than non-migrants, within-province migrants behave differently than long-distance migrants, and migration reduces the compensating differential of commuting time on income. To identify the effect of migration on household location and commuting behavior, we exploit the rollout of high-speed rail (HSR) connections between Chinese cities as a source of exogenous variation in migration costs. We use the implied migration costs and commuting elasticities to parameterize a quantitative spatial model integrating inter-urban migration — between rural areas and tier 1–3 cities — with intra-urban location and commuting choices. Migration and housing decisions are modeled conditional on consumption and amenities at origin and destination. We simulate counterfactual commuting patterns under alternative \textit{hukou} policies, differential commuting technology, and changes in migration costs.

September 25, 2026 15:45 to 17:15 (Location: Iowa)
4A. Empirical Methods

Chair: Sean McCulloch (Brown University)

• Dave Donaldson, Federico Huneeus, Vincent Rollet
  Quantifying Aggregate Impacts in the Presence of Spillovers
  Presented by Vincent Rollet (MIT)
  Discussed by Yi Niu (Capital University of Economics and Business)
  Abstract

  A widespread threat to the validity of standard policy evaluation tools is the presence of spillovers between treated and untreated groups such as spatial regions. Economic interactions across units of analysis---due to the flow of goods, factors, and payments to and from the government, for instance---result in bias in standard estimates of objects of interest such as the average treatment effect or the total effect of a program. In this paper, we develop a suite of approaches that can enable researchers to use economic theory and data about economic flows and distortions in order to overcome this bias. We apply this methodology to estimate the aggregate economic impact of a large earthquake that struck Chile in 2010.

• Tobias Großbölting, Soonwoo Kwon, Michael Pollmann, Aurel Rochell
  Segment Fixed Effects in Spatial Regression Discontinuity Designs
  Presented by Michael Pollmann (Duke University)
  Discussed by Sean McCulloch (Brown University)
  Abstract

  When a policy varies discontinuously across geographic space, researchers often identify its effects by comparing outcomes on either side of the policy border in a spatial regression discontinuity design. For estimation, most regression specifications control for location along the border using border-segment fixed effects, and for location away from the border using a bandwidth and sometimes linear adjustments in distance. We clarify the conditions under which segment fixed effects are needed, and propose a modeling technique to characterize the benefits and costs of fixed effects in finite samples. We provide theoretical and practical guidance: Many prior studies chose border segments that are likely longer than optimal, and a common local linear adjustment for distance to the border can yield distant comparisons. We illustrate the proposed auxiliary analyses and robustness checks by reanalyzing empirical studies.

• Jiawei Chen, Yi Niu, Matthew Shum
  Reassessing the Firm Selection Hypothesis: New Evidence from Chinese Highways
  Presented by Yi Niu (Capital University of Economics and Business)
  Discussed by Michael Pollmann (Duke University)
  Abstract

  This paper re-examines the firm selection hypothesis—that larger markets eliminate more low-productivity firms—by extending the multi-market framework of Combes, Duranton, Gobillon, Puga, and Roux (2012) to allow for asymmetric trade costs. A city with both a larger local market and better accessibility to other cities has a larger overall market size and therefore stronger selection. Exploiting substantial heterogeneity in highway access across cities during China’s massive highway expansion, we test this prediction using firm-level manufacturing data and find stronger selection in large cities with highway access than in small cities without it. The results remain robust to alternative specifications and to controls for the potential endogeneity of highway placement. We further show that less productive firms are more likely to exit after gaining highway access, consistent with the selection mechanism. These findings highlight transportation infrastructure as a key determinant of market integration and competitive selection.

September 25, 2026 15:45 to 17:15 (Location: Michigan)
4B. Transportation and Infrastructure

Chair: Kentaro Nakajima (Hitotsubashi University)

• Daniel Centuriao, Palak Suri
  The Effectiveness of Speed Enforcement Policies in Congested Cities
  Presented by Daniel Centuriao (Penn State University)
  Discussed by Kentaro Nakajima (Hitotsubashi University)
  Abstract

  Large, congested cities face a persistent policy dilemma: how to reduce motor vehicle collisions without generating unintended congestion or spillover effects. Speed reduction policies such as slow zones, speed humps, and automated speed enforcement systems aim to improve traffic safety by lowering driving speeds, yet their spatial impacts remain poorly understood. This study evaluates these interventions using New York City as a case study. I estimate causal effects through a spatial difference in differences framework, leveraging daily street level data on motor vehicle collisions, injuries, and fatalities and exploiting spatial and temporal variation in policy implementation across treated and untreated street segments. The results reveal substantial heterogeneity across policy types. Stricter enforcement through automated speed cameras is associated with increases in collisions and injuries in treated areas, accompanied by spillover effects that raise collision rates on nearby streets. In contrast, moderate interventions such as slow zones and speed humps generate statistically significant reductions in collisions and injuries, with benefits extending beyond treated segments. No statistically significant effects are found for fatalities. Using estimated treatment effects and monetary valuations of collision costs from the literature, I conduct a cost benefit analysis of each intervention. The results suggest that slow neighborhood zones and speed humps are more cost effective strategies for improving traffic safety than automated speed cameras. These findings emphasize the importance of accounting for localized and spillover effects when designing urban traffic safety policies and suggest intervention intensity should be tailored to congestion conditions and urban spatial structure.

• Kentaro Nakajima
  Economies of Density in Public Service Provision: Evidence from the Time Allocation of Waste Collection Trucks
  Presented by Kentaro Nakajima (Hitotsubashi University)
  Discussed by Daniel Centuriao (Penn State University)
  Abstract

  How does urban density generate cost savings in spatially organized public services? This paper studies this question using one-second GPS records from municipal waste collection trucks in Japan. The data reveal the sequence of station visits and allow me to decompose truck operating time into service time at waste stations and non-service time associated primarily with spatial traversal between stations. I document three facts. First, active collection accounts for only about one-third of truck operating time. Second, station-level service exhibits economies of scale: service time increases with the population served by a station, but much less than proportionally. Third, higher population density is associated with substantially lower per-capita non-service time, making the traversal margin quantitatively central. As an interpretable benchmark, a one-percent increase in population density is associated with a cost reduction equivalent to an 11.13-percent reduction in waste generation per resident. Counterfactual exercises show that the fiscal consequences of population decline depend on both where residents are lost and how the station network adjusts. A uniform five-percent population decline raises aggregate per-capita operating cost by 0.83 percent when station density is fixed, but this increase is almost fully offset when station counts adjust with population. Population loss concentrated in high-density districts raises per-capita operating cost, whereas population loss concentrated in low-density districts reduces aggregate operating cost among the remaining population. These results highlight spatial concentration and service-network adjustment as important determinants of the fiscal sustainability of local public services in shrinking cities.

September 25, 2026 15:45 to 17:15 (Location: LaSalle)
4C. Congestion Pricing

Chair: Shoshana Vasserman (Stanford University Graduate School of B)

• Devansh Jalota, Xuan Di, Adam Elmachtoub
  Simple vs. Optimal Congestion Pricing
  Presented by Devansh Jalota (Georgia Institute of Technology)
  Abstract

  Congestion pricing has emerged as an effective tool for mitigating traffic congestion, yet implementing welfare or revenue-optimal dynamic tolls is often impractical. Most real-world congestion pricing deployments, including New York City's recent program, rely on significantly simpler, often static, tolls. This discrepancy motivates the question of how much revenue and welfare loss there is when real-world traffic systems use static rather than optimal dynamic pricing. We address this question by analyzing the performance gap between static (simple) and dynamic (optimal) congestion pricing schemes in two canonical frameworks: Vickrey's bottleneck model with a public transit outside option and its city-scale extension based on the Macroscopic Fundamental Diagram (MFD). In both models, we first characterize the revenue-optimal static and dynamic tolling policies, which have received limited attention in prior work. In the worst-case, revenue-optimal static tolls achieve at least half of the dynamic optimal revenue and at most twice the minimum achievable system cost across a wide range of practically relevant parameter regimes, with stronger and more general guarantees in the bottleneck model than in the MFD model. We further corroborate our theoretical guarantees with numerical results based on real-world datasets from the San Francisco Bay Area and New York City, which demonstrate that static tolls achieve roughly 80-90% of the dynamic optimal revenue while incurring at most a 8-20% higher total system cost than the minimum achievable system cost.

• Cody Cook
  The Short-Run Effects of Congestion Pricing in New York City
  Presented by Cody Cook (Yale)
  Abstract

  We study the impacts of New York City's Central Business District (CBD) Tolling Program, the first cordon-based congestion pricing scheme in the United States. Using a generalized synthetic controls approach that compares outcomes in NYC to contemporaneous outcomes in other cities, we find that the policy increased speeds on CBD roads by 11%, with little-to-no effect on air quality, transactions at shops and restaurants, or overall foot traffic in the CBD. Speeds also increased outside the CBD, especially on roads commonly traversed by drivers traveling to/from the CBD. These spillovers lead to faster trips throughout the metro area, including for many unpriced trips. We develop a simple model to bound the effects on driver welfare. If drivers have a Value of Travel Time (VOTT) of $40/hour, then we estimate that driver welfare increased by at least $14.3 million/week, before any revenue recycling or environmental benefits. Passenger vehicles headed to the CBD are only better off if their VOTT exceeds $153/hour, but the modest speed improvements for the many unpriced trips reduce the overall 'break-even' VOTT of drivers to at most $21/hour. Finally, we show how characteristics of local travel patterns and road networks can inform the potential impacts of introducing cordon-based congestion pricing in other cities.

• Itai Ater, Brad Ross, Adi Shany, Shoshana Vasserman
  Road Pricing Under Heterogeneity
  Presented by Brad Ross (The Wharton School, University of Pennsylvania)
  Abstract

  Congestion pricing policies are widely debated and critics often argue that toll costs outweigh congestion relief. We show theoretically that welfare gains depend on two forces: driver heterogeneity, which allows tolls to improve travel times for drivers with high value of time by pushing drivers with low value of time off of congested roads during in-demand times, and the extent of revenue recycling. We quantify the effect of both forces using data from a large-scale experiment that charged 10,000 Israeli drivers per-kilometer driving fees over two years. Causal estimates show that pricing reduced congested driving by roughly 10%, mainly through fewer cross-city trips and shifted departure times, with the strongest responses among drivers with greater schedule flexibility and higher baseline pricing exposure. We then develop and estimate a model of time-varying traffic equilibrium with heterogeneous drivers and a granular road network, combining experimentally identified demand with nonparametrically estimated congestion technology. We use the model to simulate the equilibrium impact of the experiment’s per-kilometer fee schedule on traffic patterns in the Tel Aviv commuting zone and compare it with optimized per-kilometer fees and cordon prices. Finally, we assess how preference and traffic exposure heterogeneity and the degree of revenue recycling affect the value of congestion pricing policies.

September 25, 2026 15:45 to 17:15 (Location: Illinois 3)
4D. Housing Supply I

Chair: John Mondragon (Federal Reserve Bank of San Francisco )

• Chandler Lutz, Ben Sand
  Highly Disaggregated Land Unavailability
  Presented by Chandler Lutz (University of North Carolina, Charlotte)
  Discussed by John Mondragon (Federal Reserve Bank of San Francisco )
  Abstract

  Standard empirical proxies contradict the canonical prediction that supply constraints amplify house price cycles. We resolve this puzzle by constructing the Land Unavailability--Machine Learning (LU-ML) Indices using high-resolution satellite imagery. Unlike existing measures, our indices capture the nonlinear and heterogeneous impacts of physical geography. We document two main facts: (1) physical constraints are a key determinant of cross-sectional price dynamics, driven largely by the intensive margin (steep slopes); and (2) looser supply constraints significantly mitigate the price effects of demand growth, overturning the "null result" in recent literature.

• Chao Liu, Edmund Lou, Wei Xiang
  Income Inequality, Housing Supply, and House Prices
  Presented by Chao Liu (Chinese University of Hong Kong)
  Discussed by Chandler Lutz (University of North Carolina, Charlotte)
  Abstract

  This paper studies a supply-side channel through which income inequality increases house prices in the United States. We develop a conceptual framework in which greater income inequality leads to more restrictive housing regulation, tighter housing supply, and higher house prices. Using county-level data from 1990 to 2017 and a novel Bartik-style instrument for the Gini coefficient, we find that a one standard deviation increase in the Gini coefficient raises house prices by 26 percent, accounting for roughly 8 percent of the house price increase over this period. Consistent with our framework, more unequal areas exhibit both higher house prices and fewer housing units. A one standard deviation increase in the Gini coefficient is associated with a 58 percent decline in building permits issued over the subsequent decade and a 2 percentage point decrease in the homeownership rate. We provide evidence that more unequal areas exhibit greater opposition to pro-supply housing reform and more restrictive land-use regulation. Our findings highlight how income inequality shapes housing market dynamics through its impact on new housing supply and housing regulation, with important consequences for housing affordability.

• Schuyler Louie, John Mondragon, Rami Najjar, Johannes Wieland
  House Prices and Quantities Around the World
  Presented by John Mondragon (Federal Reserve Bank of San Francisco )
  Discussed by Maxence Valentin (ETH Zurich)
  Abstract

  We establish several first-order facts using house price and quantity data for middle- and high-income countries. First, housing units and rooms per person rise with per capita income and then asymptote so that additional increases in income are uncorrelated with housing quantities. Second, growth in housing units is strongly correlated with population growth at all incomes. Third, real house prices and housing expenditure shares rise in near-constant proportion with per capita income. A model with non-homothetic demand for home space, homothetic demand for housing quality, and upward-sloping supply of the unit-quality housing bundle can explain these relationships. Our findings are not indicative of tighter unit-supply constraints, as a model where the quantity of housing units asymptotes because unit-supply is increasingly inelastic does not match the facts.

September 25, 2026 15:45 to 17:15 (Location: Illinois 4)
4E. Housing Affordability

Chair: Jeffrey Zabel (Tufts University)

• Huan Deng, Hanchen Jiang, Xi Yang
  Housing Assistance and Labor Supply: The Case of Rent Control
  Presented by Hanchen Jiang (University of North Texas)
  Discussed by Saba Akhavansadr (Rutgers University)
  Abstract

  Housing affordability pressures have renewed interest in rent regulation, yet little is known about how rent control affects tenants’ labor supply decisions. This paper studies the labor-supply effects of rent stabilization in New York City, a setting with a long history of rent regulation. Using the New York City Housing and Vacancy Survey (NYCHVS), which provides administratively verified rent-stabilization status and detailed labor market outcomes, we estimate the impact of rent stabilization on labor supply. To address nonrandom selection into rent-stabilized units, we adopt an instrumental variables strategy that exploits quasi-random variation in the relative availability of vacant rent-stabilized units at the time of move-in. Our preferred IV estimates indicate substantial reductions in hours worked: rent stabilization lowers weekly hours by 7.7 hours on average, or 20 percent, when including nonworkers, and by 5.6 hours, or 13 percent, among employed tenants. A comprehensive set of robustness and validity checks supports these findings. We further estimate marginal treatment effects (MTE) to assess whether the labor-supply disincentive varies with unobserved resistance to entering rent stabilization. The MTE curves are approximately flat, indicating no essential heterogeneity: the IV estimates generalize beyond compliers to the average treated and untreated tenants.

• Gonzalo Respighi Grasso
  Housing affordability and economic well-being: evidence from California
  Presented by Gonzalo Respighi Grasso (University of California, Santa Cruz)
  Discussed by Jeffrey Zabel (Tufts University)
  Abstract

  This paper studies the effects of modern rent control on tenants’ credit scores and the mech- anisms through which these effects emerge. I exploit the institutional constraints imposed by the Costa-Hawkins Rental Housing Act, which establishes a construction-year cutoff for rent control eligibility across California jurisdictions. I leverage this cutoff in a regression disconti- nuity design pooling observations across twelve California cities that adopted rent stabilization policies between 2016 and 2022. To implement the analysis, I construct a novel longitudinal dataset linking individuals to housing units by combining individual-level consumer credit panel data from the University of California Consumer Credit Panel with parcel-level property records from LightBox. I document four main findings. First, rent control initial eligibility improves ten- ants’ economic well-being, increasing credit scores among incumbent tenants by 1.1 percentage points. Second, the underlying behavioral responses are consistent with a price-ceiling liquidity channel: rent control reduces the probability that credit card balances exceed 30 percent of avail- able credit by nearly 10 percent and lowers credit card balances by roughly 3 percent. Third, rent control increases tenant turnover by 17.5 percentage points, with larger effects among low- socioeconomic-status tenants, and these effects do not appear to be explained by landlord exit, condominium conversion, or persistent vacancy, consistent with a vacancy decontrol mechanism. More importantly, residential turnover appears to be a central force shaping rent control effect on credit scores. Over time, the gap in reduced-form effects on credit scores between high- and low-socioeconomic-status tenants widens in favor of higher-status households, mirroring an in- creasingly pronounced turnover differential in which lower-status tenants are disproportionately affected. Fourth, four years after policy implementation, high-socioeconomic-status tenants are 17.2 percentage points more likely to reside in rent-controlled units than in comparable non- covered units, providing additional support to the the hypothesis under which vacancy decontrol is the central mechanism governing the policy’s effect on turnover. Taken together, rent control creates meaningful financial benefits for incumbent tenants, but these benefits become pro- gressively less targeted over time as differential turnover and replacement dynamics alter who ultimately remains in covered units.

• Seung Hyeong Lee, Younggeun Yoo
  "Giving Up": The Impact of Decreasing Housing Affordability on Consumption, Work Effort, and Investment
  Presented by Seung Hyeong Lee (Northwestern University)
  Discussed by Erik Hembre (Federal Reserve Bank of Minneapolis)
  Abstract

  Housing affordability has declined sharply in recent decades, leading many younger generations to give up on homeownership. Using a calibrated life-cycle model matched to U.S. data, we project that the cohort born in the 1990s will reach retirement with a homeownership rate roughly 9.6 percentage points lower than that of their parents’ generation. The model also shows that as households’ perceived probability of attaining homeownership falls, they systematically shift their behavior: they consume more relative to their wealth, reduce labor effort, and take on riskier investments. We show empirically that renters with relatively low wealth exhibit the same patterns. These responses compound over the life cycle, producing substantially greater wealth dispersion between those who retain hope of homeownership and those who give up. We propose a targeted subsidy that lifts the largest number of young renters above the "giving-up threshold." This policy yields welfare gains of 3.2 times greater than a uniform transfer and 10.3 times greater than a transfer targeted to the bottom 10% of the wealth distribution, while also improving homeownership rate, labor effort, and reducing social safety net reliance.

September 25, 2026 15:45 to 17:15 (Location: Illinois 5)
4F. History and Structural Change

Chair: Luke Heath Milsom (KU Leuven, IFS)

• Michael Neubauer, Gabrielle Grafton
  The Great Migration and Those Left Behind
  Presented by Michael Neubauer (Reed College)
  Discussed by Luke Heath Milsom (KU Leuven, IFS)
  Abstract

  Despite the extensive literature on the Great Migration, very little is known about its economic and social consequences for Southern non-migrants. We study the impact out-migration imposes on those who remain behind, focusing on Black Americans who remained in the South during the first decade of the Great Migration. We build a shift-share instrument for the net Black out-migration rate using shocks to manufacturing employment in Northern cities. We find that both Black and White non-migrants experienced increases in wages due to out-migration. Black out-migration also led to a decrease in the number of farms as well as less racial hostility in Southern communities. We identify labor supply shocks, occupation switching, and changes in the production process as the mechanisms driving our results. At least in the short term, out-migration can be beneficial for non-migrants.

• Martin Schlesinger
  Industrialization of the Alps
  Presented by Martin Schlesinger (University of Graz)
  Discussed by Michael Neubauer (Reed College)
  Abstract

  We study factors that facilitate the diffusion of technology following the Industrial Revolution. Our novel dataset covers parts of Central Europe with high spatial resolution from 1782 to 1930. Leveraging instrumental variables and the staggered expansion of the railway network, we document that industrialization flourished particularly where transport infrastructure and resources coincided, highlighting strong complementarities between these endowments. We show that coal mining served as the predominant driver of this relationship. We only find find a negative relationship between pre-existing artisanal skills and railways in predicting growth. Finally, we demonstrate that electrification - though positively associated with growth - primarily reinforced existing patterns.

• Luke Heath Milsom
  The Long Shadow of a Sudden River
  Presented by Luke Heath Milsom (KU Leuven, IFS)
  Discussed by Martin Schlesinger (University of Graz)
  Abstract

  In 1134, a storm created the Zwin, a river connecting Bruges to the North Sea. By 1500, it had silted up and become unnavigable. I exploit this natural experiment to show that whilst open, the Zwin more than tripled Bruges’s population. To measure its long-term shadow, I develop a dynamic quantitative spatial model incorporating merchants and Malthusian dynamics, and build a retrieval-augmented generation large language model to extract merchant data from 120 qualitative historical sources. The model shows that the Zwin altered the distribution of economic activity across the Low Countries, both when navigable and for hundreds of years after.

September 25, 2026 15:45 to 17:15 (Location: Directors)
4G. Intergenerational Mobility

Chair: Evan Mast (University of Notre Dame)

• David Card, Jesse Rothstein, Moises Yi
  How Do Neighborhoods and Firms Affect Intergenerational Mobility?
  Presented by Moises Yi (U.S. Census Bureau)
  Discussed by Shruti Agarwal (University of Warwick)
  Abstract

  We use data from the Longitudinal Employer Household Dynamics linked to the 2000 Census to study intergenerational earnings mobility in the United States. We augment the standard intergenerational transmission model relating children’s log earnings to those of their parent with an additional term representing mean log parent earnings in the childhood neighborhood. The between-neighborhood intergenerational relationship is twice as strong as the within-neighborhood relationship, even after adjusting for measurement error in parents’ earnings. Moreover, mean earnings of the parents in a neighborhood capture over 80% of the variation in unrestricted neighborhood effects that reflect differences in “absolute mobility”. Next, we use an AKM framework to decompose parents’, children’s, and neighboring parents’ earnings into person effects and establishment premiums. Children’s person effects are mainly influenced by parents’ and neighbors’ person effects, whereas children’s establishment premiums are mainly influenced by parents’ and neighbors’ establishment premiums. These patterns point to separate channels for human capital and access to jobs in the intergenerational transmission process. Finally, we explore the implications for the Black-white earnings gap. Neighborhoods explain 30% of the Black-white gap in children’s earnings conditional on parents’ earnings, operating largely through gaps in average person effects. Conditional on neighborhood average earnings, children from neighborhoods with higher Black shares achieve higher adult earnings.

• Ruoyu Chen
  Moving Without Mixing: City Size and Residential Mobility
  Presented by Ruoyu Chen (University of Southern California )
  Discussed by Moises Yi (U.S. Census Bureau)
  Abstract

  Large US cities have higher residential mobility rates, but whether this produces socioeconomic mixing remains unclear. This paper examines whether larger cities enable stronger assortative residential mobility—the tendency of residents to move between neighborhoods of similar socioeconomic status. Using population-scale Infutor address histories (2011–2020), I find that assortative residential mobility increases systematically with city size. Individual fixed effects confirm this is a city-size effect, not driven by time-invariant individual characteristics. An average move in a 5-million population metro covers 30% fewer percentile points in the local SES distribution than one in a 100-thousand population metro. The findings challenge assumptions that high mobility promotes mixing, with implications for housing mobility programs and place-based policy.

• Shruti Agarwal
  Public Access, Feasible Choice, and Social Sorting: Evidence from Public Schooling in Rural India
  Presented by Shruti Agarwal (University of Warwick)
  Discussed by Ruoyu Chen (University of Southern California )
  Abstract

  Can expanding access to local public goods increase segregation? I study this question in rural India, where caste remains a central dimension of social stratification and public primary schooling is formally free. I assemble geocoded administrative school records covering more than a decade and overlay them with high-resolution population grids to construct village-level local education markets. These data measure spatial access to public schools, within-market caste segregation across public schools, and students’ exposure to out-group peers. For identification, I exploit India’s Right to Education proximity rule, which required a primary school within one kilometer and was most binding in villages initially farther from a nearby public school. Comparing differentially exposed villages before and after the reform, with village and district-by-year fixed effects, I find that access expansion increased segregation. Villages more exposed to the rule gained nearby public-school coverage, but dissimilarity rose by about 4–5 percent of the baseline mean and Scheduled Caste students became less exposed to non-Scheduled Caste peers. Instrumental-variables estimates imply that a 10 percentage point increase in within-1 km public coverage raises dissimilarity by 0.026, or 7.3 percent of the mean. Mechanism evidence points to within-market sorting rather than mechanical new-school entry. Enrollment shifts away from mixed-composition schools, school caste composition fans out, and the results exceed what nearest-school assignment would predict from residential geography alone. The paper shows that expanding spatial access to public services can reduce distance barriers while reshaping local peer environments.

September 25, 2026 15:45 to 17:15 (Location: Moskow)
4H. Optimal Spatial Policy

Chair: Levi Crews (UCLA)

• Shinnosuke Kikuchi, Daniel O'Connor
  Granular Origins of Agglomeration
  Presented by Daniel O'Connor (Michigan)
  Abstract

  Large firms dominate many local labor markets. How does this granularity shape economic geography and optimal place-based policy? We develop an economic geography model with granular firms facing idiosyncratic shocks and show that average wages rise with labor market size. Using establishment-level data on Japanese manufacturing, we estimate the model and find evidence consistent with our mechanism. Granularity explains 10-20% of estimated agglomeration externalities in the smallest commuting zones, but only 2-4% in Tokyo. Optimal industrial and wage policy increases the population in the smallest cities; firm effects depend on labor market conduct.

• Miguel Godinho de Matos, Joao Guerreiro, Sergio Rebelo, Pedro Teles
  Foreign Residents and the Future of Global Cities
  Presented by Joao Guerreiro (UCLA and the Federal Reserve Bank of Min)
  Abstract

  Global cities are attracting a growing number of tourists and foreign residents. This influx generates capital gains for property owners but adversely affects renters, creating potentially important production, congestion, and amenities externalities. We study the optimal policy regarding local and foreign residents in a model with key features emphasized in policy debates. Using this model, we provide sufficient statistics to calculate the optimal tax and transfer policies that internalize agglomeration, congestion, and other potential externalities. We find that it is not optimal to impose zoning regulations or to restrict, tax, or subsidize home purchases by foreign residents. However, it may be optimal to charge an entry fee to foreign residents.

• Levi Crews
  Skills, Sorting, and Optimal Dynamic Spatial Policy
  Presented by Levi Crews (UCLA)
  Abstract

  I study optimal spatial policies in a quantitative dynamic spatial model with forward-looking migration decisions, human capital investments, and local human capital spillovers. I characterize the constrained efficient allocation and show how to decentralize it with a simple set of policy instruments: a location-contingent "body tax" and a location-by-skill-contingent "brain subsidy." Using data on US cities and existing estimates of the spillover elasticities, preliminary results suggest that the US economy would benefit from further concentrating its high-skilled workers in a small number of currently high-wage cities.

September 25, 2026 15:45 to 17:15 (Location: Illinois 1)
4I. Human Capital

Chair: Greg Howard (University of Illinois)

• Elio Nimier-David
  Local Human Capital and Firm Creation
  Presented by Elio Nimier-David (Cornell University)
  Discussed by Sarah Quincy (Vanderbilt University)
  Abstract

  Do college expansion policies promote local economic development? This paper exploits the massive construction of new colleges in France during the 1990s to assess the local effect of higher education establishments. I first examine their impact on education, business dynamism, employment and wages at the city level. Leveraging the staggered implementation of the policy in an event-study design, I find a persistent rise in educational attainment of the local labor force. Subsequently, firm creation increased by 10% on average across all major industries. The rise in tradable and skill-intensive industries indicates that the supply of educated workers played a major role in increasing firm entry. On the contrary, incumbent firms experienced lower growth and a higher exit rate, suggesting crowding out effects. Overall, the positive effects dominate, leading to increased economic activity in treated cities. While employment remained constant on average, this masks large composition effects between young and older workers. In addition, province-level analysis suggests that new colleges had non-negative spillover effects on surrounding areas. I complement the city-level results with evidence on the long-run effects on individuals. Relying on differences between cohorts induced by the timing of the policy, I find that cohorts directly exposed to new colleges became more educated, more likely to be employed and, more likely to hold skilled positions. As a result, the policy generated new opportunities for the local population.

• Zachary Bleemer, Sarah Quincy
  Junior Colleges and Local Development
  Presented by Zachary Bleemer (Princeton)
  Discussed by Greg Howard (University of Illinois)
  Abstract

  The world's first two-year community colleges opened as early 1900s local industrialization fueled American growth. We use linked census data and staggered difference-in-difference designs around 300 college openings to show that ``junior'' college access increased male college-going and Bachelor's attainment by 20\%. Occupational upgrading followed, especially into self-employed agriculture. County-level farm productivity and wages grew by 5--6\%. Persistently higher agricultural production in junior college towns then facilitated reallocation towards skilled occupations nearby. Improved human capital and transmission of innovative knowledge, absent innovation or legislative mandates, spurred spatial structural transformation.

• Greg Howard, Russell Weinstein
  Does Training Teachers Locally Affect Teacher Shortages? Evidence from Regional Public Universities
  Presented by Greg Howard (University of Illinois)
  Discussed by Elio Nimier-David (Cornell University)
  Abstract

  We study whether training teachers locally increases nearby teacher supply. We use the historical assignment of normal schools and insane asylums to identify the effect of university proximity. Normal schools, built to train teachers, became regional universities while asylums mostly continue as small psychiatric facilities. Our evidence suggests greater teacher supply in normal school counties: lower teacher wages and more teachers per student. Asylum counties have more teachers with emergency credentials and fewer who majored in education---suggesting they mitigate lower supply by hiring in different pools. Normal school counties have higher high school test scores and graduation rates.

September 25, 2026 17:45 to 18:45 (Location: Illinois rooms)
5. Keynote: Quantitative Urban Economics by Stephen Redding

Chair: Stephan Heblich (University of Toronto)

• Stephen Redding
  Quantitative Urban Economics
  Presented by Stephen Redding (stanford)
  Abstract

  Quantitative Urban Economics

September 26, 2026 8:30 to 10:30 (Location: Iowa)
5A. Skilled Sorting and Macroeconomics

Chair: Yoonsoo Lee (Seoul National University)

• Pablo De Llanos
  Housing Market Segmentation and Monetary Policy
  Presented by Pablo De Llanos (Columbia University)
  Discussed by Emiliano Harris (David Eccles School of Business, University of Utah)
  Abstract

  Contractionary monetary policy can raise housing rents even as house prices fall and goods inflation declines. I argue that this response reflects a relative-price adjustment driven by tenure reallocation in segmented housing markets. A monetary tightening reduces credit access for constrained households, shifting housing demand from ownership toward renting. Whether rents rise or fall depends on the strength of this demand shift, which is shaped by the substitutability between owning and renting, and on the extent to which landlords can absorb it by expanding rental supply, as limited by the degree of housing market segmentation. I first develop this mechanism in a simple model and then embed it in a dynamic stochastic general equilibrium (DSGE) model with heterogeneous households, collateral constraints, nominal rigidities, and costly reallocation of housing between the owner- and renter-occupied sectors. Estimated on quarterly U.S. data, the model reproduces the empirically observed positive rent response to monetary tightening and implies rental-supply elasticities consistent with external evidence. Because the resulting rent inflation reflects a relative-price adjustment rather than generalized excess demand, Taylor rules that target shelter-inclusive inflation overreact and reduce welfare in the estimated model. Rules that exclude shelter deliver a weak Pareto improvement, while rules that respond separately to goods inflation and relative shelter prices generate larger aggregate gains but redistribute welfare across agents.

• Damien Capelle, Adriano Fernandes, JJ Kruger, Peter McAdam
  Barriers to an European Banking Union
  Presented by Adriano Fernandes (International Monetary Fund)
  Discussed by Robert Reed (University of Alabama)
  Abstract

  We quantify barriers to cross-border bank lending to firms within the euro area and their consequences for credit allocation and output. Using loan-level data from the European credit registry (AnaCredit) and group structures (RIAD), we estimate barriers to relationship formation, loan pricing, and banks’ branching decisions at the country-pair level. We find that barriers to cross-border relationships between banks and firms and cross-border bank entry are large while wedges on interest rates and loan quantities are comparatively small. The estimated wedges are strongly associated with differences in national banking regulations, measured using a novel dataset on regulatory distances. We embed our estimates into a quantitative spatial general equilibrium model with heterogeneous banks and firms subject to cross-border frictions in relationship formation, loan pricing, and bank entry. Partially relaxing frictions predicts sizable and heterogeneous output gains across euro area countries. These gains are primarily driven by increases in capital and labor rather than improvements in allocative efficiency.

• Yifan Zhang
  College Graduates in the Labor Market: Geographic Mobility and Sorting into Firms and Occupations
  Presented by Yifan Zhang (The Chinese University of Hong Kong)
  Discussed by Dayin Zhang (University of Wisconsin-Madison)
  Abstract

  This paper integrates data from three online labor market platforms—LinkedIn user profiles, Burning Glass job postings, and Glassdoor wages—to study how college graduates’ job outcomes vary across universities of different rankings and locations, and analyze factors that determine the outcomes. Estimating a model of graduates’ choices over firms, occupations, and locations, we find strong positive sorting: graduates from higher-ranked universities match to more cognitively intensive jobs and high-amenity cities. Distance reduces job-matching probabilities for most universities, whereas its impact is markedly attenuated for elite institutions. We also find a substantial geographic premium for universities in major cities.

• Ivy Feng, Jing Xue, Dayin Zhang
  Unlocking Innovation: Household Mobility, Inventor Careers, and Productivity
  Presented by Dayin Zhang (University of Wisconsin-Madison)
  Discussed by Yifan Zhang (The Chinese University of Hong Kong)
  Abstract

  This paper studies how household mobility frictions shape inventors’ career and innovation productivity. The question is increasingly relevant given the sharp decline in US household mobility over recent decades, which might constrain job switching and geographic reallocation of inventive talent. To study this question, we construct a new longitudinal dataset linking US inventors’ credit profiles, employment histories, and patenting records. We exploit mortgage rate lock-in as a plausibly exogenous source of variation in household mobility. We find that a one–standard–deviation increase in mobility raises the probability of job switching by 4 percentage points, and approximately 70% of the induced moves involve a promotion or pay increase. Mobility also leads to higher innovation productivity, raising both patent counts and citations. Mechanism tests show that the effects are concentrated among inventors who appear to be initially underplaced, consistent with mobility improving the quality of inventor–firm matching. The effects are also stronger for younger inventors and those working in “thin” technology labor markets, where local matching frictions are more severe. Together, the findings document a previously underexplored household-finance channel through which mobility constraints impact the reallocation and productivity of innovative human capital.

September 26, 2026 8:30 to 10:30 (Location: Michigan)
5B. Work Locations

Chair: Weihua Zhao (University of Louisville)

• Mikhail Komov
  The Effect of Remote Work on Employers’ Monopsony Power in UK Labour Markets
  Presented by Mikhail Komov (University of Trento)
  Discussed by Weihua Zhao (University of Louisville)
  Abstract

  The widespread adoption of remote work after the COVID-19 pandemic blurs the geographical boundaries of local labour markets, which may weaken employers' wage-setting power. This paper examines the effect of remote work adoption on employers' monopsony power in UK labour markets. Using job posting data from 2015 to 2024 and a difference-in-differences strategy, we find that labour market concentration has a negative effect on posted wages, and that this effect attenuates significantly after the pandemic for occupations with high levels of remote work adoption. We estimate that wages in high-WFH occupations are around 20% less sensitive to local labour market concentration in the post-pandemic period, and that this attenuation increases with the share of WFH adoption. Finally, the weakening of monopsony power is most pronounced for occupations that were more geographically clustered before the pandemic.

• Yanyan Liu, Kangkai Wang, Qi Wu, Jianhuan Xu
  Transit Access and Labor Market Efficiency: Evidence from Mobile Phone Data
  Presented by Kangkai Wang (The University of Hong Kong)
  Discussed by Mikhail Komov (University of Trento)
  Abstract

  How does urban rail transit reshape labor-market matching inside a large city? This paper studies the opening of a metro system in a northern provincial capital city in China, using high-frequency individual mobility records, block-level vacancy postings, and simulated pre- and post-opening commuting times between urban blocks. The empirical design combines reduced-form difference-in-differences evidence with a structural spatial search-and-matching model. The reduced-form results show that blocks receiving larger post-opening commuting-time savings experience higher employment, more unemployment-to-employment transitions, more job-to-job transitions, and fewer flows into unemployment. Bilateral gravity estimates imply substantial commuting frictions: a one-minute increase in residence-to-job commuting time lowers unemployment-to-employment match flows by about 5.4 percent, while old-job-to-new-job commuting time lowers job-to-job switching flows by about 5.3 percent. In the structural model, metro access affects labor-market efficiency by reallocating directed applications across space and by changing the acceptance margin for employed searchers. Across selected calibrations of on-the-job search participation and application intensity, the metro-induced commuting improvements reduce the city-proper mismatch index by 0.114 to 0.251 index points, corresponding to 15 to 53 percent of pre-metro mismatch. The results suggest that transit investment can improve labor-market efficiency not only by raising access to jobs, but also by reallocating search effort toward better-matched vacancies.

• Matthew Hockert
  Do Local Income Taxes Cause Residential Sorting? Evidence from Pennsylvania’s Piecewise Earned Income Tax
  Presented by Matthew Hockert (University of Minnesota Twin-Cities)
  Discussed by Kangkai Wang (The University of Hong Kong)
  Abstract

  Do local income taxes cause residential sorting? Pennsylvania’s Earned Income Tax follows a piecewise max-rule where a resident pays the higher of the home-side rates and the workplace rate. The rule creates variation that is used to identify a causal effect. I decompose statutory increases by whether the home-side stack already binds, becomes binding, or never binds, and the never-binding cell is a within-design placebo. I estimate a continuous stacked difference-in-differences design on block-level LEHD/LODES data aggregated to municipalities over 2002–2019. The average causal response to a binding increase is a 0.537 log-point decline per percentage point in the count of workers who both live and work in their own municipality, about a 12% decline at the average cumulative increase of 0.24 percentage points, while the placebo cell is null. The results are robust to multiple tests and specifications. While the headline results originate from a linear model, I run a polynomial and spline specification as well. The two are able to trace the full dose-response, the total and marginal effect at each dose: the effect accumulates steeply across the increases and flattens as they grow. The largest decline comes from the first increase, and each subsequent increase moves fewer workers, consistent with a shrinking pool of residents willing to relocate. All three specifications return similar results. The binding cell passes a set of pre-trend tests and shows no effect at three fake-timing placebo dates. The property tax’s effect is about a third of the EIT’s per percentage point.

• David Agrawal, Weihua Zhao
  The Spatial Effects of Taxes with Telecommuting
  Presented by Weihua Zhao (University of Louisville)
  Discussed by Troup Howard (University of Utah)
  Abstract

  This paper studies how the rules governing where income is taxed shape the spatial structure of cities in an era of widespread telework. When the place of work and residence are geographically decoupled, as is common at the local level, the U.S. tax systems often assign taxing rights on earning to both the place of employment and the place of residence. We show that these “sourcing rules” are a central but underappreciated determinant of urban form and local public finances. Using a quantitative spatial general equilibrium model disciplined by data from the Columbus, Ohio metropolitan area, we find that alternative sourcing regimes generate highly uneven effects across municipalities. Shifting to purely residence-based taxation substantially amplifies the migration of teleworkers to lower-density communities and redistributes tax revenue gains toward these areas. In contrast, employment-based taxation dampens these shifts and results in revenue losses outside major job centers. These effects are not confined to income taxes with various sourcing rules inducing spillovers to property taxes, consumption taxes, and local public services, making general equilibrium responses critical for evaluating welfare. Our results highlight a new margin of tax policy that has distributional effects on population, economic activity, and fiscal capacity across space. As telework continues to reshape where people live and work, the choice of where income is taxed will play a key role in determining which communities are the fiscal winners and losers.

September 26, 2026 8:30 to 10:30 (Location: LaSalle)
5C. Innovation and Ideas

Chair: Enrico Berkes (University of Maryland Baltimore County)

• Ina Ganguli, Jeffrey Lin, Vitaly Meursault, Nicholas Reynolds
  Spreading Out Across Expanding Idea Space
  Presented by Jeffrey Lin (Federal Reserve Bank of Philadelphia)
  Abstract

  This paper investigates how the expanding frontier of knowledge reshapes innovation. If the space of possible inventions grows as knowledge accumulates, inventors could work on increasingly different things, facing less direct competition. However, a growing spread between an invention and its applications could force greater investment in every invention — with implications for whether collective research effort translates into growth. We find that US inventions have indeed become increasingly dissimilar over nearly two centuries (1836–2023), corroborated by 150 years of declining patent interference rates. Documenting this spreading-out requires measuring similarity correctly, as standard text approaches yield the opposite conclusion. Our validation framework, the first systematic comparison for patent text, identifies which representations are accurate and which are misleading. We develop a parsimonious model of positioning in idea space that unifies spreading out with several previously disconnected patterns. Rising research investment, invention quality, patent values, and declining research productivity all emerge from a single spatial mechanism. Other evidence — the coupling of spacing and research investment and quasi-experimental estimates of how proximity shapes knowledge spillovers — can also be explained in this framework. A calibrated decomposition attributes roughly 40% of the decline in research productivity to spatial forces, explaining why productivity growth did not accelerate through the 20th century despite an enormous expansion in aggregate R&D. Where inventors stand relative to each other matters as much for growth as how many of them there are.

• Lukas Rosenberger, William Hanlon, Carl Hallmann
  Innovation Networks in the Industrial Revolution
  Presented by William Hanlon (Northwestern University)
  Abstract

  We offer a new approach to understanding technological progress and economic growth during the Industrial Revolution using modern growth theory combined with detailed micro-data. We develop a multi-sector endogenous growth model that we use to study one leading theory of British advantage during the Industrial Revolution: knowledge access costs. We apply our model to patent data from Britain and France in order to estimate key parameters, including vectors of country and technology-specific knowledge access parameters. We validate our estimates and then use the model to study their implications during the transition to modern economic growth. We show that, relative to France, British inventors faced lower knowledge access barriers, a difference that generated meaningful growth rate differences on the transition path, even when accounting for cross-country knowledge and technology flows.

• Enrico Berkes, Ruben Gaetani
  The Decline in the Transmission of Scientific Ideas
  Presented by Ruben Gaetani (University of Toronto)
  Abstract

  We document that the diffusion of new scientific ideas beyond their field of origin has declined substantially over the past four decades. This contraction is closely linked to increasing specialization in scientific language: research that employs more technical terminology tends to be adopted less broadly. We develop a theory of scientific discovery in which the diffusion of new ideas depends on the degree to which potential adopters can understand and process them. When introducing their discoveries, scientists face a tradeoff between technical communication targeted at their immediate peers and more accessible language meant to reach broader audiences. As knowledge accumulates and research at the frontier builds on deeper layers of prior work, this tradeoff increasingly favors specialized language, limiting diffusion. Policy interventions that align scientists’ incentives can broaden adoption and increase the social value of scientific research.

• Caterina Chiopris
  The Diffusion of Ideas
  Presented by Caterina Chiopris (Columbia University)
  Abstract

  The rise of knowledge and the flow of innovation have long been recognized as an important factor for economic development. How do increased spatial connections affect the generation and diffusion of ideas? I study knowledge production in Germany in the 19th century, which witnessed an explosion of innovation. The analysis relies on several novel largescale datasets, including the universe of bibliographic records, covering all published texts in all fields of knowledge, and detailed and comprehensive railway statistics. New ideas – both novel concepts and new combinations of existing ideas – are measured thanks to recent advances in machine learning and topology applied to data analysis. Exogenous delays in railway construction enable the identification of the effect of the railroad network. I show that the railroad contributed to innovation, explaining 11% of the increase in knowledge production. Scholars’ mobility led to the formation of specialized clusters, and thus to cities’ specialization. New ideas are formed by combining ideas coming from cities connected by the railroad. Diffusion of new ideas increases within field. The findings shed light on the causes for specialization in knowledge production, on the organization of modern science, and on the diffusion of information in dense networks.

September 26, 2026 8:30 to 10:30 (Location: Illinois 3)
5D. Housing Dynamics

Chair: David Jinkins (CBS)

• Yurou He
  Land Assembly: Outsider Bargaining with a Network
  Presented by Yurou He (Cornell University)
  Discussed by David Jinkins (CBS)
  Abstract

  Urban redevelopment often requires assembling contiguous parcels from multiple landowners, generating holdout problems that raise costs and impede efficient land use in dense cities. This paper studies how spatial structure and bargaining dynamics shape outcomes in decentralized land assembly. I model landowners as a parcel-adjacency network and analyze a sequential bargaining game in which a developer must acquire a connected set of parcels. I introduce two network-based concepts—pivotality and assembly centrality—to characterize how spatial position and bargaining order determine veto power and surplus division. Using over 300,000 U.S. land assembly transactions from 2007–2022, I show that pivotal parcels command price premia of 16–23\%, and that deviations from a center-to-periphery negotiation order substantially increase total acquisition costs. The results highlight how frictions in decentralized bargaining distort urban redevelopment.

• Jacob Herbstman
  Regulatory Discretion and Housing Supply: Evidence from Chicago's Aldermanic Privilege
  Presented by Jacob Herbstman (University of Chicago)
  Discussed by Yurou He (Cornell University)
  Abstract

  Land use regulation constrains housing supply, but formal zoning rules are only part of the regulatory environment developers face. I study informal regulatory discretion through Chicago’s “aldermanic privilege,” under which city council members can effectively block, delay, or reshape development in their wards. I construct alderman-level stringency scores from permit processing times and compare places near the same ward boundary, where the neighborhood is similar but the alderman changes. At ward borders, a one standard deviation increase in aldermanic stringency lowers multifamily construction density by roughly 12-14%. A complementary event study approach shows that census blocks reassigned to more stringent aldermen experience a 7.4% decline in permits. These effects on housing supply and built density appear to filter through to prices, as I find evidence of higher listed rents and higher home sale prices on the more stringent side of ward boundaries. Together, the results show that informal regulatory discretion can reduce housing supply and harm affordability, calling into question the assumption that zoning reform alone will necessarily lead to more housing construction

• Yuki Kanayama, Taisuke Sadayuki
  Housing Dereliction in a Superstar City: Future Gains, Today’s Neglect, and Disamenity Effects
  Presented by Yuki Kanayama (London School of Economics)
  Discussed by Jacob Herbstman (University of Chicago)
  Abstract

  Why do derelict houses persist in urban areas despite strong demand, and what are their effects on neighbourhoods? This paper sheds light on a simple mechanism — speculation over future real estate value increases — that explains this puzzle, and employs empirical strategies to test this mechanism and quantify the disamenity effects of derelict houses. First, we develop a dynamic discrete choice model showing that expectations of future urban regeneration can incentivise property owners to delay redevelopment, thereby prolonging dereliction. Using variation induced by urban regeneration plans in central Tokyo, we find that properties located in designated regeneration areas are 6-14% more likely to be derelict. Second, we estimate the effect of derelict houses on nearby rents using future regeneration plans as an instrument — affecting dereliction but not current rents directly. Our 2SLS results show that one additional derelict house within 80 metres reduces rents by 1.5% on average. The effect is magnified to up to 4.5% in areas with low accessibility to public safety services, suggesting that renters’ concerns about fire and crime risks amplify the disamenity effects of derelict properties. Our findings suggest that rational forward-looking behaviour can reduce effective housing supply and generate negative neighbourhood externalities, highlighting an unintended consequence of place-based urban regeneration policies.

• Filip Sangild Beck, David Jinkins, Ismir Mulalic
  Public housing and the Dynamics of City Structure
  Presented by David Jinkins (CBS)
  Discussed by Yuki Kanayama (London School of Economics)
  Abstract

  This paper examines how public housing interacts with the dynamics of city struc- ture. Exploiting individual-level panel data from Denmark along with a large-scale quasi- experimental privatization of public housing in Copenhagen, we document the effects of public housing on city structure. We develop and estimate a dynamic quantitative urban model with both publicly allocated and private housing as well as endogenous amenities and find that reductions in public housing substantially amplify gentrification patterns, altering household sorting, housing prices, and local amenities.

September 26, 2026 8:30 to 10:30 (Location: Illinois 4)
5E. Environmental Disasters

Chair: Luca Perdoni (ifo Institut, LMU)

• Haeseong Park
  Agglomeration under Shock - Labor, Networks, and Trade in Urban Disaster Resilience
  Presented by Haeseong Park (University of Colorado Boulder)
  Discussed by Luca Perdoni (ifo Institut, LMU)
  Abstract

  This paper examines how various urban agglomeration effects adjust natural disasters' short-run economic impacts. Using wind strength to proxy disaster intensity and electricity consumption to measure economic activity, I show that natural disasters reduce aggregate economic activity by 3 percent at the metropolitan statistical area (MSA). Lower job diversity amplifies losses, while centralized road networks and greater internal goods sourcing mitigate them. Leveraging granular industry specialization profiles, I find that increases in specialization in non-tradable local industries magnify the adverse economic effects of natural disasters.

• Victoria Wang
  Adaptive Migration in the Face of Wildfires
  Presented by Victoria Wang (UCLA)
  Discussed by Sowon Kim (Ohio State University)
  Abstract

  Natural disasters are expected to impact a large and increasing number of people with climate change. Adaptive migration – movement from risky to safe areas – is theorized to be a key strategy for minimizing the costs of natural disasters. This paper investigates how people migrate in response to wildfires in California, where fires have become more frequent and severe. I provide novel empirical estimates of the extent of adaptive migration, up to seven years after the disaster. Using detailed individual-level geographic data, I estimate the effect of wildfires on migration using a difference-in-differences (DID) event study design by comparing the migration behavior before and after a fire of individuals in blocks that are burned for the first time with that of those in never-burned blocks within a census tract. I find that an individual experiencing a first fire is more likely to move and to stay away, even when their home is not destroyed. They are also more likely to be in safe areas. This effect is driven by wildfires that destroyed buildings. Moreover, people who were previously homeowners are less likely to be homeowners after the disaster, suggesting long-term assessments of a home as an asset.

• Michael Fogarty, Sarah Bass
  Sorting Over Wildfire Risk Information
  Presented by Michael Fogarty (UW Madison)
  Discussed by Nitzan Tzur-Ilan (Federal Reserve Bank of Dallas)
  Abstract

  We study how climate risk information affects housing market outcomes through its impact on household risk perceptions and residential sorting decisions. We develop and estimate a structural model of neighborhood choice in which households have heterogeneous preferences over housing prices and wildfire risk and may hold biased perceptions of their risk exposure. We use the introduction of a wildfire risk disclosure policy in California as an information shock that causes households to update their beliefs about wildfire risk (without changing true underlying risk) that alters location choices and housing prices in equilibrium. We find that, prior to disclosure, households substantially underestimate wildfire risk, discounting true risk levels by about 60\%. The disclosure policy causes households to update their beliefs, which leads to declines in demand and prices in high-risk areas and induces systematic re-sorting. More risk averse, mortgage-financed buyers shift away from areas newly perceived as higher risk. More price sensitive and less risk averse buyers, particularly cash financed buyers and those purchasing homes for secondary uses, sort in. As a result, disclosure reallocates wildfire exposure across different types of household rather than reducing it overall. Our findings highlight that information policies can have large aggregate effects on housing markets when misperceptions are widespread, while also reshaping the distribution of households that experience climate risk through equilibrium channels and heterogeneous sorting responses.

• Richard Bluhm, Luca Perdoni, Julian Wichert
  Rethinking redlining: Environmental inequality within and between U.S. neighborhoods
  Presented by Luca Perdoni (ifo Institut, LMU)
  Discussed by Rhiannon Jerch (University of Wisconsin Madison)
  Abstract

  Environmental inequalities such as unequal exposure to pollution and climate risks persist across racial and socioeconomic groups in the United States. This paper re-examines the role of the Residential Security Maps created by the Home Owners’ Loan Corporation (HOLC) in the 1930s, which graded neighborhoods by perceived mortgage risk and have been widely linked to long-run racial segregation and environmental disadvantage. A common view holds that these maps not only reinforced residential segregation but also directly shaped the spatial distribution of environmental hazards, including air pollution, flood risk, and extreme heat. We evaluate this claim using a causal framework that combines machine-learning predictions of counterfactual HOLC grades in unmapped cities with a spatial difference-in-differences design. Our results confirm that the maps modestly increased racial sorting and segregation, consistent with prior work. However, we find no evidence that HOLC mapping independently affected the siting of environmental or climatic hazards. Differences in air pollution, flood risk, heat exposure, and mortality across historical grades are quantitatively similar in mapped and unmapped cities. These findings suggest that contemporary environmental inequalities primarily reflect residential sorting and discriminatory practices operating broadly across U.S. cities, rather than an additional siting effect uniquely induced by the HOLC maps, which we do not detect.

September 26, 2026 8:30 to 10:30 (Location: Illinois 5)
5F. Risk, Uncertainty, and Growth in Spatial Equilibrium

Chair: Kyle Mangum (Federal Reserve Bank of Philadelphia)

• Maximiliano Dvorkin
  Heterogeneous Agents Dynamic Spatial General Equilibrium
  Presented by Maximiliano Dvorkin (Federal Reserve Bank of St. Louis)
  Discussed by Weiliang Tan (National Bureau of Economic Research)
  Abstract

  I develop a dynamic model of migration and labor market choice with incomplete markets and uninsurable income risk to quantify the effects of international trade on workers employment reallocation, earnings, and wealth. Macroeconomic conditions in different labor markets and idiosyncratic shocks shape agents’ labor market choices, consumption, earnings, and asset accumulation over time. Despite the rich heterogeneity, the model is highly tractable as the optimal consumption, labor supply, capital accumulation, and migration and reallocation decisions of individual workers across different markets have closed-form expressions and can be aggregated. I study the asymmetric impact of international trade on the evolution of employment, earnings, and wealth, and decompose the frictions workers face to reallocate across U.S. sectors and regions into those with a transitory effect and those with long-lasting consequences.

• Marcos Sora
  Labor reallocation during booms: The role of duration uncertainty
  Presented by Marcos Sora (University of Illinois-Urbana Champaign)
  Discussed by Kyle Mangum (Federal Reserve Bank of Philadelphia)
  Abstract

  A salient aspect of sectoral booms—prevalent in commodities, construction, or tech—is that the end of the boom phase is difficult to predict. I study how this uncertainty shapes labor mobility across sectors during the boom phase. Using a model of sector-specific human capital accumulation, I show that workers can exhibit risk-loving attitudes towards duration, leading to ambiguous effects of uncertainty on labor supply. Then, I turn to an empirical investigation of the effects of duration uncertainty during the boom in mineral prices of 2011–2018, driven by a construction boom in China. I estimate the model using financial data and novel administrative micro-data from Australia, an exporter of mineral products to China. I use the quantified model to study a counterfactual perfect-foresight economy in which the mining boom was temporary and duration certain. I find that the mining share of employment in Australia would have increased from 3.7% to 4.4%, and the relative wage in the sector would have been substantially lower, leading to a decrease in labor income inequality. Changes in the age composition of the mining sector indicate heterogeneous attitudes towards risk across age groups.

• Weiliang Tan
  Directed Innovation across Space
  Presented by Weiliang Tan (National Bureau of Economic Research)
  Discussed by Marcos Sora (University of Illinois-Urbana Champaign)
  Abstract

  I introduce directed innovation into a quantitative spatial model. Profit incentives direct innovation across regions through the interaction of research productivity, gradual technology diffusion, trade, and worker mobility – extending the concept of directed technical change (Acemoglu, 2002) to a spatial setting. I show that any steady-state distribution of innovation is consistent with balanced growth and determines the economy’s long-run growth rate. I then analytically decompose the welfare impact of any shock to economic fundamentals into its transitory and long-run growth components. I apply the model to the post-1990 rise in the spatial concentration of American innovation, which was clustered in high-skill regions and driven primarily by the growth of innovation in information and communication technologies (ICT). The model decomposes the contributions of demand-side forces – changing innovation incentives across regions – and supply-side forces – the sorting of skilled workers – and quantifies the aggregate consequences for growth and welfare.

• Kyle Mangum, Daniel O'Connor
  Dynamic Economic Geography With Risk
  Presented by Kyle Mangum (Federal Reserve Bank of Philadelphia)
  Discussed by Maximiliano Dvorkin (Federal Reserve Bank of St. Louis)
  Abstract

  This paper presents a framework for incorporating risks and non-convexities into dynamic spatial equilibrium models. Such models are hard to solve using conventional methods because of the large state spaces introduced by the economic geography. Hence, researchers usually sidestep the difficulties using very specific assumptions about laws of motion for key variables such as factors of production--assumptions that preclude a serious treatment of risk in a model with non-convexities. We show that even mild departures from these assumptions invalidate the solution methods of the current state of the art. We then present an economic geography model with forward-looking capitalists who face frictions to readjustment of capital. Our base case is regions that interact via trade and migration, but have fixed-in-place capital. We present an approach for solving the equilibrium model through a mild relaxation of rational expectations, as advocated by Moll (2026), in which agents track as state variables the prices the face but aggregate information on the determination of these prices. We apply the model to study local regions facing negative demand and/or depreciation shocks, including population loss and natural disasters.

September 26, 2026 8:30 to 10:30 (Location: Directors)
5G. Student Prize II

Chair: Raven Molloy (Federal Reserve Board)

• Ritabrata Bose
  Production Networks and the Welfare Cost of Highway Congestion
  Presented by Ritabrata Bose (Arizona State University)
  Abstract

  I estimate the welfare cost of highway congestion in India through the lens of production networks. Removing 147 government-identified bottlenecks on India's national highways would raise formal-manufacturing welfare by 3.4 percent in the full general equilibrium model. The Hulten first-order approximation, which already embeds the full input-output structure through Domar weights, captures 55 percent of this gain; higher-order CES reallocation effects account for the rest, consistent with the prediction of Baqaee and Farhi (2019) that first-order approximations understate welfare for discrete shocks. In a Bigio-La'O style decomposition comparing economies with and without intermediate linkages, production networks approximately double the welfare return to decongestion (amplification factor 2.2x). I estimate sector-specific congestion wedges from 471 single-origin products whose source locations are known, exploiting variation in chokepoint exposure along delivery routes. A Hulten-type formula decomposes the welfare value of each bottleneck into route-by-sector contributions, enabling a knapsack optimization over the government's project portfolio. A budget of about $ 0.45 billion (4 percent of total infrastructure cost) captures 29 percent of the welfare gains in the baseline specification. The per-chokepoint welfare values are determined by the production network structure - which sectors each corridor serves and how deeply cost reductions cascade - information unavailable to a planner without an IO model.

• Idaliya Grigoryeva
  Agricultural Productivity, Structural Transformation, & Urbanization: Evidence from Indonesia’s Transmigration Program
  Presented by Idaliya Grigoryeva (UC San Diego)
  Abstract

  In this paper, I study rural-to-urban spillovers, estimating the effect of rural agricultural productivity on urban population and employment growth, migration, and structural transformation in cities, utilizing the quasi-random allocation of migrants to destinations under the Transmigration Program in Indonesia. By exploiting variation in agricultural productivity across villages, I show that cities surrounded by more productive rural areas experience sectoral shifts from agriculture to manufacturing and services, with limited evidence of overall population and employment growth. The primary channel is likely the goods' trade: productive villages supply food to nearby towns, reducing the cost of living and stimulating demand for urban services, while farmers in successful villages tend to stay rather than migrate. As transmigration villages are settled, the areas around more productive villages develop into agro-processing hubs and local market towns. I develop a three-sector spatial equilibrium model with non-homothetic (Stone-Geary) preferences, agglomeration externalities in services, and endogenous food prices to rationalize these findings and decompose the trade and migration channels. Urban structural transformation appears to result from these regional economic spillovers through higher agricultural output and trade rather than outflows of transmigrants from the villages themselves.

• Lovepreet Singh
  Military Procurement and Spatial Structural Change: The Long-Run Effects of the US Civil War
  Presented by Lovepreet Singh (University of Illinois Urbana Champaign)
  Abstract

  This paper provides causal evidence that a temporary government procurement shock can have large and persistent effects on regional manufacturing development. I exploit variation in the placement of Quartermaster facilities during the U.S. Civil War to trace the long-run development of affected counties from 1820 to 2000. Wartime procurement induced a wave of firm entry across manufacturing industries: by the early twentieth century, treated counties exhibited manufacturing employment roughly 50 log points above comparison counties, population 25 log points higher, and county revenue 45 log points higher; these gaps persisted for over a century after procurement ended. The expansion operated on the extensive margin, with establishment counts rising by 48 log points and the number of industries by 21 log points, while average plant size was unchanged. Treated counties also experienced a large relative decline in agricultural employment share and greater intergenerational mobility for men from disadvantaged households, with mobility gains tied entirely to remaining in treated counties. The breadth of the procurement shock shaped its persistence: only Quartermaster sites, where procurement was diverse and induced entry across many industries, exhibit long-run effects, while narrowly specialized Ordnance sites do not.

• Lucas Kaplan
  Should We Flee? The Mortality and Economic Impacts of Hurricane Evacuation Orders
  Presented by Lucas Kaplan (Syracuse University)
  Abstract

  I estimate the mortality and county-level economic impacts of hurricane evacuation orders using a novel instrumental variables framework. Relative to similarly affected but non-evacuated areas, evacuated counties experience fewer deaths but lower personal income, employment, and business establishments for at least one year following the storm. I show that these economic losses stem from a stalled recovery driven by marginal business closures and delayed insurance claims rather than permanent out-migration. I then develop a threshold decision rule that balances the mortality benefits against these persistent economic costs and apply this framework to orders issued from 2014–2022. Incorporating these persistent economic costs drastically alters policy evaluation: only one-third of recent evacuation orders remain justified under this framework, compared to 60 percent when evaluating mortality benefits alone.

September 26, 2026 8:30 to 10:30 (Location: Moskow)
5H. Crime and Law Enforcement

Chair: Matthew Freedman (University of California, Irvine)

• Patrick Bayer, Marcus Casey, Joe Fish, Jonathan Moreno-Medina
  A Framework for Interpreting High-Frequency Hedonic Price Dynamics with an Application to Valuing Homicide Risk
  Presented by Jonathan Moreno-Medina (University of Texas at San Antonio)
  Discussed by Joaquin Urrego (Wake Forest University)
  Abstract

  Do changes in housing prices in response to nearby crime reliably reveal how much households value safety? We develop a Bayesian learning framework showing that observed price responses reflect both preferences and changes in beliefs about neighborhood safety—and that separating these channels is essential for recovering willingness to pay. In our model, neighborhoods oc- cupy latent crime states that evolve as a Markov chain; households observe crime signals and update beliefs via Bayes’ rule. We estimate the model in two steps: recovering beliefs from crime data alone, then combining estimated belief effects with reduced-form price estimates to recover WTP. Using listing-to-sale price changes—which absorb unobservable location quality by conditioning on the asking price—we combine the universe of gun-related homicides from the Gun Violence Archive with MLS records from CoreLogic for California and Texas (2014–2023). A nearby homicide reduces the sale price by approximately 0.3 percent. The income gradient in these effects—larger in richer areas—does not necessarily imply wealthier households value safety more. Our decomposition shows the gradient reflects, in substantial part, the differential informativeness of crime signals: an additional homicide is less surprising where crime is already frequent, producing a smaller belief revision and a smaller price response, even when willingness to pay is comparable.

• Carlos Diaz Guimaraens, Franco Fernandez, Sebastian Fossati, Nico Trajtenberg
  The Limits of ‘Eyes on the Street’: A Threshold in the Surveillance-Opportunity Trade-Off
  Presented by Carlos Diaz Guimaraens (Universidad Alberto Hurtado)
  Discussed by Kristof Dascher (Regensburg University)
  Abstract

  Whether pedestrian density reduces crime by increasing informal surveillance or instead increases it by multiplying targets, offenders, and anonymity remains a central unresolved question in urban economics. A key empirical challenge is that the places that attract people are usually inseparable from the flows they generate, making it difficult to isolate the independent effect of pedestrian concentration. We address this challenge using Santiago de Chile's 246 itinerant street markets, observed over 3,560 days between 2013 and 2022 in a ring-by-day spatial panel. Because these markets operate on fixed weekly schedules, the same residential streets are observed both with and without retail-driven activity, allowing us to compare the same micro-places under different crowding conditions while holding neighborhood characteristics constant. Within this setting, we introduce two sources of identifying variation that move beyond estimating an average market-day effect. First, we exploit day-to-day fluctuations in crowd size using Google Popular Times data, which allow us to distinguish between the presence of a market and the intensity of pedestrian concentration it generates. Second, we leverage high-resolution spatial variation around market boundaries to trace how effects decay across nearby areas. Our results show that theft is unresponsive to pedestrian density below a critical level, and rises sharply with crowding once that threshold is crossed. This pattern suggests that informal guardianship dominates at moderate densities, while opportunity generation takes over above a critical level. More broadly, the findings help reconcile competing views on density and urban safety by showing that the relationship is fundamentally nonlinear.

• Xuan Li, Hantao Wu
  The Social Costs of ICE Enforcement
  Presented by Hantao Wu (Princeton University)
  Discussed by Steven Craig (University of Houston)
  Abstract

  Using geocoded ICE records and foot traffic data across six states from 2025--2026, we provide the first causal estimates of immigration enforcement's community-wide costs. Visits within 0.5 miles of 10,018 enforcement events decline 1.69 percent with no rebound; comparable declines under both contact and non-contact operations indicate ambient deterrence rather than direct confrontation as the primary mechanism. Disruption varies by destination type: education facilities see the sharpest decline, workplaces and commercial venues also contract, and healthcare is unaffected. Effects are largest where enforcement recently reached previously off-limits education and healthcare settings. Residential mobility falls broadly: unique visitors from Hispanic CBGs decline 1.57 percent, and a neighborhood-level analysis finds total outside trips fall 7.45 percent---non-Hispanic neighborhoods recover within four weeks while Hispanic areas show persistent suppression, and geographic substitution is absent. Translating these declines into dollar terms, per-establishment costs total $196,527---over 11 times the DHS benchmark. A structural model finds restoring sensitive-location protections recovers 50.4 percent of welfare losses, versus 18.3 percent for non-contact enforcement.

• Samantha Cervantes, Amy Schwartz
  The Unintended Consequences of High-Intensity Streetlights on Student Outcomes: Evidence from a Crime Deterrence Experiment
  Presented by Samantha Cervantes (Syracuse University)
  Discussed by Matthew Freedman (University of California, Irvine)
  Abstract

  In 2016, the New York Police Department (NYPD) randomly assigned high-intensity streetlights to high-crime public housing developments to deter crime. While prior evaluations document sizable reductions in nighttime crime (Chalfin et al., 2022; Mitre-Becerril et al., 2022), the extraordinary brightness of the lights drew public backlash, with residents reporting glare and sleep disruption, especially on lower floors. Using linked administrative records on New York City’s (NYC) public school students, we provide the first student-level evidence of the intervention’s effects on academic outcomes. High-dosage lighting substantially reduced students’ exposure to nighttime crime, youth-involved incidents, and police contact, yet achievement declined in the same dosage range where crime reductions emerged. To probe the role of direct residential light exposure, we compare effects for students in low-rise versus high-rise buildings within high-dose areas. Negative achievement effects concentrate among students in low-rise buildings and are significantly larger than among their high-rise peers, yet nighttime crime exposure does not differ by building height. Several additional tests bolster confidence in the light intrusion interpretation, including a spillover analysis showing that students living near treated developments experience crime reductions without academic harm, and evidence that achievement losses increase with greater proximity to the lights along both vertical and horizontal dimensions. Additional analyses point to sleep disruption as the mechanism behind the achievement losses. More broadly, these findings highlight the importance of design parameters—including the built environment—to evaluate and design safety interventions.

September 26, 2026 11:00 to 13:00 (Location: Iowa)
6A. Transportation Infrastructure

Chair: Janet Kohlhase (University of Houston)

• Jiming Wang, Ruohao Zhang, Xibo Wan, Zhen Lei, Junxian Tang
  Growth or Redistribution? The Business Effects of Public Infrastructure as a Local Amenity: Evidence from EV Charging Stations in Kansas City
  Presented by Ruohao Zhang (The Pennsylvania State University)
  Discussed by Theodore Papageorgiou (Boston College)
  Abstract

  Public infrastructure can raise nearby business activity by improving local amenities. These observed gains, however, need not represent net economic growth: they may reflect the redistribution of consumers from competing establishments. We study this distinction in the context of public electric vehicle (EV) charging stations, a rapidly expanding form of subsidized infrastructure. Using granular mobile-device restaurant visits from Kansas City in 2019, matched to the universe of public charging stations operating during the same period, we estimate a random-coefficients discrete-choice model with instrumental variables to address endogenous station placement and residential sorting. We then decompose the business effects of charging access into net market growth and redistribution across nearby restaurants. The estimated model delivers three main findings. First, charging access has a large effect on restaurant demand: gaining a public charger within 500 meters raises visits by roughly 21%. Second, consumers value nearby charging access, with an implied willingness to pay of about $2.27. Third, most of the establishment-level gain reflects redistribution rather than net growth: about two-thirds is drawn from competitors. Counterfactual simulations show that targeting chargers to gas stations in historically disadvantaged tracts generates real local growth in areas where such growth is otherwise limited, but the redistribution share remains high and gains become more uneven within targeted communities. These results imply that infrastructure policies that ignore business-stealing effects can substantially overstate social returns.

• Jiming Wang, Jianwei Xing
  Spatial Heterogeneity in Externalities and Private Investment in EV Charging Infrastructure
  Presented by Jiming Wang (The Pennsylvania State University)
  Discussed by Pearl Li (University of Illinois Urbana-Champaign)
  Abstract

  Investment in EV charging infrastructure generates externalities that private charging service providers fail to internalize. However, these externalities vary sharply between high-density urban cores and low-density peripheries, leading to spatial misallocation of investment. We build a dynamic model to study firms' behavior and policy impacts in this complex environment. Two externalities are documented as driving these undesirable outcomes. First, through positive two-sided market spillovers, infrastructure in the urban core generates roughly twice as much EV adoption as infrastructure in the periphery, with part of the induced charging demand leaking to rival providers. We estimate this using a discrete choice model with data from 50 Chinese cities. Second, using hourly port occupancy data to measure the negative business-stealing externality from competition, we find that competition is more pronounced in the periphery: a 1% increase in charging capacity within 1 km reduces the visit rate by 0.26%, compared with 0.07% in the core. Weaker positive spillovers and stronger negative externality from competition lead to excessive investment in the periphery, while the opposite pattern leads to insufficient investment in the core. Embedding these estimates in a dynamic investment game, we further recover that the marginal cost of expanding capacity in the core is roughly 45% higher than in the periphery. Comparing providers' behavior with a counterfactual in which the externalities are internalized, we find underinvestment in the core by 6.6% and overinvestment in the periphery by 1.7%. Spatial heterogeneity in externalities accounts for about half of these distortions.

• Yuyang Jiang
  Strategic Investment in Transportation Networks: Evidence from the U.S. Highway System
  Presented by Yuyang Jiang (Princeton University)
  Discussed by Steven Craig (University of Houston)
  Abstract

  Transportation networks are often built by multiple jurisdictions whose invest- ment decisions may not fully account for cross-border benefits. This paper builds a quantitative spatial framework to evaluate the welfare implications of decentralized investment in the U.S. highway network. The framework highlights an empirical characterization of cooperation between state planners, and features three sources of inefficiency in a decentralized equilibrium: a classical network externality from transshipment traffic, a terms-of-trade externality from trade in differentiated goods, and a fiscal externality from taxpayer migration. I estimate that state planners place a weight of 0.56 on other states’ constituents’ welfare relative to their own. At this estimated level of cooperation, decentralized investment leads to 15% underinvestment relative to the national optimum and generates a welfare loss equivalent to 30% of current highway spending. Raising the federal subsidy rate increases total spending but does not achieve a cost-effective allocation across the network.

• Giulia Brancaccio, Myrto Kalouptsidi, Theodore Papageorgiou, Yixin Zhou
  Transportation Bottlenecks and Aggregate Shocks
  Presented by Theodore Papageorgiou (Boston College)
  Discussed by Yuyang Jiang (Princeton University)
  Abstract

  Do transportation disruptions have significant impact on macro outcomes? If so, in what ways? And, what can we do about it? This paper aims to address these questions by focusing on maritime transportation (ports, and ships), which moves more than 80% of international trade. We first document a number of facts that suggest that the transportation sector has a limited capacity for adjustment in periods of increased demand for shipping services. We formalize this intuition through a model of the transportation sector, viewed as a sequence of queues. Using granular data and insights from queueing theory, our setup identifies transportation bottlenecks, allows us to measure the capacity of the transportation sector and finally quantify the impact of these bottlenecks on prices. We use the setup to shed light on what unfolded during the Pandemic surge in commodity prices, as well other disruptions such as the Red Sea crisis.

September 26, 2026 11:00 to 13:00 (Location: Michigan)
6B. Agglomeration and Knowledge

Chair: Yichen Su (Southern Methodist University)

• Nuri Ersahin, Hyunseob Kim, Seung Hyeong Lee
  Agglomeration at the Top: Headquarters Clusters, Productivity, and Resource Allocation
  Presented by Hyunseob Kim (Federal Reserve Bank of Chicago)
  Discussed by Yichen Su (Southern Methodist University)
  Abstract

  This paper studies how the agglomeration of corporate headquarters affects productivity, resource allocation, and the internal organization of multi-unit firms. Using restricted-use US Census Bureau microdata, we show a robust positive relationship between headquarters agglomeration and productivity at both the plant and firm levels, over and above the well-documented association between productivity and plant agglomeration. Our evidence suggests that agglomeration economies at a headquarters location propagate to operating plants. Plants managed by firms with more agglomerated headquarters exhibit higher investment and capital deepening, specialize their labor and product mix, and engage in more productivity-sensitive restructuring. Notably, these productivity gains do not translate into higher plant worker wages; instead, the headquarters-plant wage gap widens and the plant-level labor share falls. Using openings of large headquarters across competing locations as a quasi-experimental source of variation, we provide evidence that greater headquarters agglomeration raises plant productivity and induces capital deepening. These results identify headquarters agglomeration as an important and previously underappreciated channel through which agglomeration economies shape the performance of multi-unit firms. Given the large role of multi-unit firms in US economic activity and the spatial concentration of headquarters in large metropolitan areas, the findings have implications for aggregate productivity and the allocation of rents.

• Palak Suri, Yao Wang, Yixuan Wang
  The Geography of Agglomeration Forces
  Presented by Yao Wang (Ohio State University)
  Discussed by Hyunseob Kim (Federal Reserve Bank of Chicago)
  Abstract

  We quantify agglomeration economies across 109 U.S. metropolitan areas using a unified spatial equilibrium framework. We estimate an average elasticity of productivity with respect to employment density of 0.03, with substantial heterogeneity across cities. Agglomeration is stronger in larger, more diversified, and service-oriented economies, and in areas with greater human capital and infrastructure. Leveraging this heterogeneity, we show that the employment effects of Opportunity Zones are significantly larger in high-agglomeration areas, despite limited average impacts. These findings highlight the importance of spatial heterogeneity in agglomeration forces for understanding urban productivity and designing effective place-based policies.

• Furkan Kilic
  Spatial Allocation of Inventors, Knowledge Diffusion and Growth
  Presented by Furkan Kilic (University of Chicago)
  Discussed by Yao Wang (Ohio State University)
  Abstract

  Where does innovation truly thrive? Inventive activity in the US is strikingly concentrated in a handful of hubs. This raises compelling questions: Does further agglomeration drive innovation, or could a more dispersed approach better leverage regional spillovers? To investigate, I exploit variation in patent citation lags across US states and develop a novel endogenous growth model with mobile inventors and workers. The model integrates an exogenous knowledge network that facilitates the dynamic exchange of ideas—laying the foundation for future inventions—between locations, revealing that inventors do not internalize how their location choice influences broader knowledge diffusion. These knowledge spillovers call for a targeted, place-based R&D subsidy to unlock latent innovation potential. Calibrating the model to data on inventor and worker allocations—and estimating the knowledge diffusion network from patent citations—I find that optimal policy would further concentrate inventors in established hubs, enhancing welfare by 1.8% in consumption-equivalent terms and boosting the economy’s long-run growth rate by 0.14 percentage points.

• Sitian Liu, Yichen Su
  The Geography of Research: The Trade-off between Knowledge Production and Access
  Presented by Yichen Su (Southern Methodist University)
  Discussed by Furkan Kilic (University of Chicago)
  Abstract

  Universities and research activities generate significant but often highly localized positive spillovers for the economy through improved access to frontier knowledge and knowhow. The localized nature of these spillovers provides a rationale for geographically dispersing researchers to align more closely with the spatial distribution of economic activity. Yet if knowledge production itself benefits from agglomeration economies, dispersing researchers to improve knowledge access may come at the cost of reducing knowledge production. Using a panel of academic researchers constructed from bibliographic data, we find evidence of strong place effects and agglomeration economies in knowledge production: when a researcher moves to a larger research cluster in the same field, her research output and impact rise substantially. We then use a spatial equilibrium model of researchers and non-researchers to examine how reallocating researchers affects aggregate economic output, explicitly accounting for the trade-off between knowledge access and knowledge production. Preliminary results suggest that reallocating researchers toward the geographic center of gravity of the real economy would yield a net positive effect on aggregate output.

September 26, 2026 11:00 to 13:00 (Location: LaSalle)
6C. Disasters and Homeowners

Chair: Edward Coulson (UC Irvine)

• Katharine Harwood, Rachel Meltzer, Ruitian Yan
  Disaster Vultures or Recovery Machines? The Role of Investors in the Wake of Hurricanes Sandy and Harvey
  Presented by Rachel Meltzer (Harvard University)
  Discussed by Philip Mulder (University of Wisconsin Madison)
  Abstract

  Extreme climate events can lead to both physical and economic transformations in neighborhoods. In the case of hurricanes and extreme flooding, the focus of our research, buildings and homes can be destroyed, but reinvestment and rebuilding can soon follow. Who is engaged in this rebuilding? And how does it impact the housing options for existing and new residents in areas hit hard by these storms and extreme flooding? We are interested in identifying the degree to which investor owners play a role in post-disaster neighborhood recovery and change. Investor owners (sometimes referred to as corporate owners) are defined in various ways but usually include landlords or investment entities that own multiple properties and buy-to-sell them or keep the units to rent. Their proliferation has been well-documented, but less is known about how their activity intersects with post-disaster conditions, which seem ripe for investor-owner involvement. Specifically, we will document the behavior of investor owners in the wake of extreme hurricane and flooding events, including (i) changes in their acquisition of properties, (ii) what they do with the properties once they own them, and (iii) the implications for residents in the neighborhood including the concentration of investor ownership and their impacts on housing quality and affordability. We will use two of the largest climate shocks in recent history, Superstorm Sandy and Hurricane Harvey, to test how investor ownership mediates the resilience of housing markets and what it means for housing options moving forward.

• Naqun Huang, Weizeng Sun, Yanmin Yang, Siqi Zheng
  Flash Flood Risk in Dense Cities: Housing Markets, Infrastructure and Adaptation
  Presented by Siqi Zheng (Massachusetts Institute of Technology: MIT)
  Discussed by Edward Coulson (UC Irvine)
  Abstract

  This paper examines how housing markets in dense Chinese cities respond to flash flood risk, emphasizing the vertical dimension of exposure in high-rise buildings, and the multi-layered nature of urban and neighborhood resilience. Using millions of housing transactions with floor-level identifiers, we find that extreme precipitation significantly reduces both rents and prices, with ground-floor units in low-lying areas suffering the largest losses. These impacts are moderated by neighborhood and citywide drainage infrastructure, which acts as a critical public good for flood resilience. In addition, higher-income, better-educated, and urban hukou households are more likely to adapt by relocating vertically to higher floors, while other groups remain disproportionately exposed.

• J. Anthony Cookson, Emily Gallagher, Philip Mulder
  Coverage Neglect in Homeowners Insurance
  Presented by Philip Mulder (University of Wisconsin Madison)
  Discussed by Siqi Zheng (Massachusetts Institute of Technology: MIT)
  Abstract

  Most homeowners lack sufficient insurance coverage to fully rebuild their homes after a total loss. Using contract-level data from a major wildfire in Colorado, we document substantial variation in underinsurance across insurers with more reputable insurers writing more complete coverage. This cross-insurer variation is not explained by differences in policyholder characteristics. Households with total losses insured by low-coverage firms were more likely to sell without rebuilding. We find that homeowners tend to shop on premiums without adjusting for differences in coverage limits across insurers, a phenomenon we call coverage neglect. Coverage neglect reduces consumer surplus by 10% of annual premiums.

• Edward Coulson, Tracy Turner
  Did Dorothy Become a Homeowner?: Childhood Exposure to Disaster and Later Homeownership
  Presented by Edward Coulson (UC Irvine)
  Discussed by Rachel Meltzer (Harvard University)
  Abstract

  We use the Panel Survey of Income Dynamics, combined with the SHELDUS database of US natural disasters to estimate the hazard rate of transitioning by young people into homeownership as a function of childhood exposure to natural disasters. We find that such exposure does markedly reduce that hazard, especially for those with higher levels of exposure. We conclude that the salience of disasters increases the perceived risk of ownership and delays its acquisition. We find additional evidence that migration behavior is potentially done with an eye towards reducing that risk.

September 26, 2026 11:00 to 13:00 (Location: Illinois 3)
6D. Low-Income Housing and Discrimination

Chair: Konhee Chang (UC Berkeley)

• Molly Clark-Barol, Christopher Timmins, Xiaoou Zhan
  Screening by Name: How Tenant Screening Algorithms Sort Innocent Renters into Disadvantaged Neighborhoods
  Presented by Xiaoou Zhan (University of Wisconsin-Madison)
  Discussed by Konhee Chang (UC Berkeley)
  Abstract

  Landlords and tenant screening companies routinely search court records using name-based algorithms that return results for both exact and approximate matches, yet no prior study has estimated whether the errors this imprecise matching produces actually change where people live. We provide the first such estimates. Linking consumer records to the universe of Milwaukee County circuit court cases, we measure each renter's exposure to false association with a stranger's court record and decompose this exposure into three categories: exact-name mistaken attribution, Levenshtein-distance-1 fuzzy matches (names one character edit away), and distance-2 fuzzy matches (two edits away). The exact-match and distance-1 channels independently predict that renters live in neighborhoods with lower rent, lower income, lower educational attainment, fewer parks, and higher area deprivation; distance-2 effects collapse to zero, confirming decay with string distance. Effects concentrate among the 63% of renters with no criminal record of their own. The distance-1 channel---which cannot mechanically contain any individual's own court activity---is statistically null for homeowners who bypass landlord screening, confirming that the mechanism operates through tenant screening. The algorithm's per-case effect is roughly uniform across races, but Black, Hispanic, and Asian/Pacific Islander renters face systematically more matched cases than White renters, producing a disparate-impact pattern: a race-neutral algorithm generates racially disparate outcomes through disparate exposure rather than disparate treatment. Black-male renters bear the largest harm on every outcome. These findings demonstrate that imprecise name matching systematically sorts innocent renters into worse neighborhoods and motivate identity-verification requirements for screening providers.

• Ingrid Ellen, Katharine Harwood, Katherine O'Regan, Sarah Strochak
  The Impact of Source of Income Discrimination Laws on Housing Voucher Search Outcomes
  Presented by Katharine Harwood (New York University)
  Discussed by Xiaoou Zhan (University of Wisconsin-Madison)
  Abstract

  Despite the significant benefits the Housing Choice Voucher program provides, recent estimates show that only 60% of households that receive vouchers use them successfully within one year (Ellen et al., 2025). Landlord discrimination against voucher holders is a key barrier. This paper provides the first direct estimates of the impact of laws that prohibit such discrimination (Source of Income Discrimination laws) on the ability of new voucher recipients to use their vouchers to lease a home. Using a staggered difference-in-differences and matching approaches, we find that the enactment of an SOI law leads to a three percentage point increase in the likelihood of new voucher holders using their vouchers in the four years after enactment. Effects grow over time and, as expected, more stringent laws that allow fewer exemptions have larger impacts, as do those enacted at the state rather than the local level. In addition, estimated impacts are larger for households with children.

• Konhee Chang
  Price Discrimination Against Voucher Tenants
  Presented by Konhee Chang (UC Berkeley)
  Discussed by Katharine Harwood (New York University)
  Abstract

  The Housing Choice Voucher (HCV) program is the largest U.S. rental assistance program, costing more than $30 billion annually. I document systematic pricing differences between voucher and market-rate tenants using administrative unit-level panel data on rents, tenants, and landlords for both market-rate and voucher units. Landlords charge voucher tenants 6% higher rents and 7% higher security deposits for the same unit. At tenant turnover, rents increase by roughly $96 more when the incoming tenant is a voucher holder, nearly doubling the typical rent increase. Leveraging variation in voucher generosity arising from program design, I show that these price differentials reflect a combination of tenant-side price insensitivity and strategic pricing by landlords. Voucher premia are largest where subsidy standards are more generous relative to market rents such that tenants’ out-of-pocket payments are less sensitive to contract rent, and they increase with landlords’ experience participating in the HCV program. These patterns are difficult to reconcile with cost-based explanations and instead point to systematic price discrimination against voucher tenants.

September 26, 2026 11:00 to 13:00 (Location: Illinois 4)
6E. Taxes and Services in Developing Cities

Chair: Tanner Regan (George Washington University)

• Michael Best, Luigi Caloi, Francois Gerard, Evan Kresch, Joana Naritomi, Laura Zoratto
  Greener on the Other Side: Inequity and Tax Compliance
  Presented by Evan Kresch (Oberlin College)
  Discussed by Tanner Regan (George Washington University)
  Abstract

  Governments frequently use observable tags as proxies to measure the tax base or household means. These tags are often imperfect, leading to misclassification and inequities among equally deserving individuals. This paper studies the efficiency effects of such misclassification in the context of the property tax system in Manaus, Brazil. We leverage quasi-experimental variation in inequity generated by the boundaries of geographic sectors used to compute tax liabilities, and a tax reform in a series of augmented boundary discontinuity designs. We find that inequity significantly reduce tax compliance, accounting for 40% of the overall change in compliance at the boundaries. A simple model of presumptive property taxation shows how mistagging affects the optimal tax schedule. Interpreting our findings through this lens implies that optimal progressivity is around 30% lower than without inequity responses. These results underscore the importance of inequity for public policy design, especially in contexts with limited state capacity.

• Daniel Agness, Pascaline Dupas, Tigabu Getahun, Fiker Negash
  Don’t know, don’t care? Experimental evidence on barriers to responsive public services in Ethiopia
  Presented by Daniel Agness (University of Maryland, College Park)
  Discussed by Evan Kresch (Oberlin College)
  Abstract

  Using rich survey data spanning citizens, firms and local government officials working in Ethiopia’s decentralized bureaucracy, we document low knowledge of public priorities among local bureaucrats, and substantial mismatch between priorities of bureaucrats and those of the public. We rule out information frictions as the primary cause, by experimentally showing that (a) bureaucrats fail to update beliefs or adapt policy when provided with feedback on citizens’ and firms’ priorities, even in sectors over which they have high autonomy; (b) bureaucrat demand for such information is very low. We also show that bureaucrats do not react when they know feedback from citizens and firms is being elicited, suggesting low accountability. Our results challenge standard assumptions in models of fiscal federalism and highlight the limitations of decentralized bureaucracies in providing locally tailored public services without incentives or intrinsic motivation to meet public demand.

• Joshua Deutschmann, Molly Lipscomb
  The Costs of Sewer Blockages in Cape Town, South Africa
  Presented by Molly Lipscomb (University of Virginia)
  Discussed by Daniel Agness (University of Maryland, College Park)
  Abstract

  As cities in many low- and middle-income countries have grown, urban infrastructure has struggled to keep up. In the case of sewage infrastructure, overburdened infrastructure and household behavior around waste disposal can lead to blockages in open sewers and flooding, with negative implications for local businesses and environmental quality. Cape Town received between 125,000 and 225,000 service calls for blockages in sewers per year between 2017 and 2024.  We use instrumental variables techniques based on the predicted flow of litter downhill to show the substantial negative impact of sewer blockages on water quality, business growth, and property values.   Finding cost-effective ways to reduce blockages is of first-order importance to fast growing cities around the world, but particularly in developing economies with over-burdened infrastructure. 

• Tanner Regan
  Property tax assessment, regressivity, and compliance: Evidence from Uganda
  Presented by Tanner Regan (George Washington University)
  Discussed by Molly Lipscomb (University of Virginia)
  Abstract

  To improve the cost-effectiveness of property tax administration, many governments in developing countries are turning to algoorithmic procedures of property valuation. What are the economic implications of these procedures? Valuation is always imperfect, and discrepancies between true income and valuation result in experienced tax rates at the individual property level that differ from the policy rate. I show generally that any assessment procedure that provides an unbiased prediction of property value will be regressive. Drawing on empirical evidence from Kampala, Uganda I study how the method of valuation impacts equity and revenue through experienced tax rates. I document that the regressivity induced by algorithmic valuation is empirically important in Kampala and likely has knock on consequences for compliance and revenue.

September 26, 2026 11:00 to 13:00 (Location: Illinois 5)
6F. Distressed Renters and Regions

Chair: Hans Koster (Vrije Universiteit Amsterdam)

• Antoine Levy, Mathilde Munoz
  Foreign Opportunity and Local Revival: Evidence from the French Rust Belt
  Presented by Antoine Levy (UC Berkeley )
  Discussed by Hans Koster (Vrije Universiteit Amsterdam)
  Abstract

  This paper studies the short- and long-run adjustment of distressed regions to a positive globalization episode: access to an increasingly thriving Luxembourg labor market for residents of the French "Rust Belt". We document a three-phase expansion of local labor markets, driven by a short-run workplace substitution of incumbent workers towards cross-border commuting, a medium-term rise in labor force participation, and a long-run sustained increase in population via net domestic in-migration. Welfare effects for residents of treated areas are unequal: higher foreign incomes are partly offset by lower domestic employment, rising housing costs and congestion, and reduced fiscal transfers. Improved job opportunities abroad lead to a net decrease in far-right and anti-EU vote shares, muting the rise of populism visible elsewhere in former industrial regions of France. Taken together, our findings suggest that former industrial regions are not inherently rigid, but that their capacity to adjust depends critically on the nature, scale, and spatial incidence of economic shocks.

• Brian Asquith, Kate Pennington, Charly Porcher
  Evicted from Opportunity: Labor Market Effects of Housing Loss in High-Productivity Cities
  Presented by Charly Porcher (Georgetown University)
  Discussed by Antoine Levy (UC Berkeley )
  Abstract

  How important is stable housing near high-paying jobs for long-run career growth? We study San Francisco Ellis Act withdrawals, which remove rent-controlled buildings from the rental market and force all tenants to relocate, and compare displaced tenants with tenants in nearby rent-controlled buildings. Six years after eviction, displaced workers earn about $13, 000 less (20 percent of baseline earnings) and live in lower-value housing and neighborhoods. Losses are largest for younger workers and remain large even for movers of 5–25 km, who transition to smaller, lower-paying firms and face longer commutes. We develop an equilibrium model of frictional housing and job search, costly commuting, localized search, and human-capital accumulation, calibrated to the Bay Area. The model quantitatively matches the eviction effects because central locations offer higher wages and faster wage growth; displacement reduces access to both and slows progression up the spatial job ladder. Counterfactuals show that improving access to central housing reduces losses more than temporary rent or commuting subsidies.

• Guangbin Hong, James Macek
  Redevelopment and Gentrification in General Equilibrium
  Presented by James Macek (University of Alberta)
  Discussed by Brian Asquith (W.E. Upjohn Institute for Employment Res)
  Abstract

  Redevelopment is a major source of housing supply in dense cities, but it is spatially concentrated and often associated with gentrification and displacement. We document three facts using Chicago data: redevelopment is the primary margin of housing supply adjustment in dense neighborhoods, it concentrates in lower-income areas and shifts the stock toward fewer but larger units, and it causally raises neighborhood housing values and incomes without expanding net supply. We develop a dynamic general equilibrium model in which forward-looking landlords choose whether to maintain, renovate, or redevelop, and heterogeneous households sort across neighborhoods and vertically differentiated housing units. Endogenous differences in rent between high and low quality units govern landlords’ redevelopment incentives and neighborhood sorting on income. We calibrate the model to Chicago, disciplining it with quasiexperimental variation from two anti-redevelopment policies, including a tax on housing teardowns. A large counterfactual teardown tax targeting all below-median-income neighborhoods benefits low-income households by preserving affordable housing. However it generates substantial spillovers to untreated areas. Welfare losses are non-monotonic in income: low income households benefit substantially and middle-income households are harmed most, as reduced high-quality supply pushes rents up for most units except for the lowest quality.

• Hans Koster
  The Persistence of Urban Decline: Evidence from France's Largest Coal
  Presented by Hans Koster (Vrije Universiteit Amsterdam)
  Discussed by James Macek (University of Alberta)
  Abstract

  We argue that urban decline persists due to the interaction between sorting and the durability of the housing stock. Using the history of coal production in France's largest mining basin as a negative economic shock to local employment, we examine its effect on urban decline. The basin's geological boundaries and the proximity of housing to mines allow us to exploit local variation in sorting and housing quality. We find that housing prices today drop by $8.5\%$ when entering the basin. This decline is attributable to lower housing quality and the rest to sorting effects. A dynamic spatial equilibrium model with forward-looking developers suggests that urban decline can persist for many decades before reaching a new steady state, leading to increased welfare inequality between low-skilled and high-skilled households in the area.

September 26, 2026 11:00 to 13:00 (Location: Directors)
6G. Local Policy Spillovers and Neighborhood Externalities

Chair: Emily Merola (Princeton University)

• Elior Cohen
  The Effects of Homeless Housing Sites on Local Communities: Evidence from Los Angeles County
  Presented by Elior Cohen (Federal Reserve Bank of Kansas City)
  Discussed by Emily Merola (Princeton University)
  Abstract

  Los Angeles County's homeless population has increased by approximately 40 percent in the past five years. While county voters have supported the goal by approving billions of dollars in bonds that would provide tens of thousands of affordable housing units and services for the homeless, there remains a substantial gap in affordable housing for the homeless and low-income individuals who are at risk of homelessness, driven by fears and stigma in local communities. I investigate the effect of such housing sites on street homelessness, crime, and property values. I construct comprehensive data that geocode the locations of all homeless housing sites in Los Angeles County. Using spatial and time variation in homeless housing sites, I estimate the exposure of a community to homeless housing sites over time and use changes in this exposure to recover the causal relationship. I find that communities with an increase in homeless housing exposure, proxied using distance to the closest site, experience a sizable decline in street homelessness and homeless-related crimes and that housing values in these communities were not affected.

• Claire Nelson
  The Pass-Through of Student Aid to Off-Campus Rents: Evidence from Federal Loan Supply Expansions
  Presented by Claire Nelson (Princeton Univeristy)
  Discussed by Kaan Cankat (Princeton University)
  Abstract

  I study the causal effect of student loan supply on the rents charged by off-campus landlords. To isolate the effect of loans on rents, I propose an identification strategy that utilizes variation in students’ individual federal student loan limits. I find a passthrough effect on local rents of 40 cents on the dollar over the academic year. This landlord aid capture effect is concentrated in the lower end of the rent distribution and the rental units closest to campus.

• Kaan Cankat, Allison Green
  Cities without Suburbs? The Effects of Large Municipal Boundary Expansions on Local Public Finance
  Presented by Kaan Cankat (Princeton University)
  Discussed by Claire Nelson (Princeton Univeristy)
  Abstract

  Between 1945 and 1990, suburbanization dramatically reduced population in American urban cores. In response, the 150 largest cities in America collectively added 11,500 square miles of fast-growing, outlying areas to their boundaries. Using new comprehensive data on city boundaries, municipal finance, and public good provision in a stacked difference-in-differences design exploiting sharp annexation events, we evaluate how boundary expansions impacted municipal finance and public good provision. The average expansion increased the municipal population by 42%. Ten years after an annexation event, per capita expenditures and revenues decrease by −23%. The largest declines occur in current expenditures and labor-intensive services such as fire and policing. We find little evidence of declines in public service provision, suggesting that expansions yielded economies of scale in public good provision. Decreased costs allow treated cities to lower property tax rates. Municipal fiscal efficiency translated to increased aggregate economic activity: counties with annexing central cities see 5% higher population, employment, and establishments relative to counties with comparable, non-annexing jurisdictions

• Emily Merola
  The Impact of Homeless Encampment Closures on Neighborhoods: Evidence from Seattle, WA
  Presented by Emily Merola (Princeton University)
  Discussed by Elior Cohen (Federal Reserve Bank of Kansas City)
  Abstract

  This paper examines the impact of homeless encampment closures on neighborhood-level outcomes in Seattle, Washington. Using novel geocoded data from Seattle Human Services Department “site journals” covering 312 encampment closures between March 2017 and November 2019, I employ a stacked difference-in-differences approach with matched comparison areas to estimate causal effects. Between 2017 and 2019, encampment closures followed a standardized process including 72-hour advance notification, offers of alternative shelter, and police presence during removal. Results indicate that closures produce modest, temporary decreases in police activity (10-16% reduction in weeks 4-10 post-closure, with effects fading afterwards) and substantial, persistent reductions in public service requests (17% decrease overall) within original site boundaries. However, I find no significant changes in fire/EMS calls, crime incidents, or foot traffic. Importantly, closures do not prevent future encampments from forming—46% of sites intersect with previously closed locations, with an average 136-day interval between closures. Supplemental analysis focused on the immediate areas surrounding sites reveals that the modest decreases in police activity persist within 1km of sites, but public service requests appear to shift to nearby areas (200-400m away), suggesting spatial displacement. These findings indicate that while encampment closures may produce meaningful reductions in visible homelessness complaints, they represent an imperfect policy response that does not lead to long term public safety gains or prevent site recurrence, ultimately raising questions about how localities should allocate their resources for addressing unsheltered homelessness.

September 26, 2026 11:00 to 13:00 (Location: Moskow)
6H. Work From Home

Chair: Hannah Rubinton (Federal Reserve Bank of St. Louis)

• Mert Akan, Jose Maria Barrero, Nicholas Bloom, Tom Bowen, Shelby Buckman, Steven Davis, Hyoseul Kim
  The New Geography of Labor Markets
  Presented by Shelby Buckman (Stanford University)
  Abstract

  We study where Americans live in relation to their employer’s worksite using matched employer-employee data, and how that relationship changes with the rise of work from home (WFH). Mean distance from home to employer’s worksite rose more than 70% between 2019 and 2024 in our dataset. Twelve percent of employees hired after March 2020 reside fifty or more miles from their employer by 2024, triple the pre-pandemic share. Distance to employer rose most for those in their 30s and 40s, among highly paid employees, and in Finance, Information, and Professional Services. Especially for the affluent, the pandemic-instigated rise in WFH initiated a multi-year pattern of net migration to areas with cheaper housing and states with lower tax rates. Finally, we show that distant employees exhibit more sensitivity to firm-level adjustments on hiring and separation margins. These developments have implications for residential location, state-level tax revenues, labor markets, and household welfare.

• Morris Davis, Andra Ghent, Jesse Gregory
  Winners and Losers from the Work-from-Home Technology Boon
  Presented by Jesse Gregory (University of Wisconsin)
  Abstract

  We model how an increase in Work-from-Home (WFH) productivity differentially affects workers using a framework in which some workers cannot work offsite, some are hybrid, and some are completely remote. The improvement in WFH productivity increases housing demand and thus housing prices since housing is inelastically supplied. Because workers in non-telecommutable occupations must consume housing but their total factor productivity does not increase, the rise in house prices reduces their welfare. The welfare decline is equivalent to 1-9% of consumption, depending on how substitutable WFH is with onsite work, and it arises despite measured income of all workers increasing.

• Brian Greaney, Andrii Parkhomenko, Stijn Van Nieuwerburgh
  Work from Home and the Evolution of Cities
  Presented by Brian Greaney (University of Washington)
  Abstract

  We study how the rapid rise ofwork from home (WFH) since 2020 has transformed cities and affected the welfare of their residents. We develop a dynamic spatial model with forward-looking households who choose residential location,workplace location, housing tenure, and WFH frequency. The model features commuting, costly migration, uninsurable income risk, housing-market frictions, commercial and residential real estate, and local public finance: local taxes fund endogenous amenities. We interpret the post-2020 economy as a transition to a new long-run spatial equilibrium and characterize the resulting paths of residential and commercial real estate prices, the locations of residents and jobs, income sorting, commuting, local tax revenues, amenities, and welfare. The model provides a framework for quantifying whether WFH triggers an urban doom loop, whereby declines in commercial real estate values erode the local tax base, reduce amenities, and further depress demand for central-city locations. We use the framework to evaluate return-to-office mandates, fiscal transfers, and office-to-residential conversions.

• Hannah Rubinton
  Work From Home and Long Distance Migration
  Presented by Hannah Rubinton (Federal Reserve Bank of St. Louis)
  Abstract

  Full-time work from home (WFH) breaks the link between where workers live and where they work, with important implications for migration and the spatial distribution of labor. We show that WFH increases long-distance migration: WFH workers are 40–50\% more likely to move than commuters, and plausibly exogenous changes in firms’ remote-work policies raise migration. Motivated by these facts, we develop and calibrate a quantitative spatial equilibrium model with migration, remote work, and job search. We use the model to quantify how the rise in WFH affects migration, cross-city changes in population, wages, and rents, and the welfare consequences of remote work.

September 26, 2026 11:00 to 13:00 (Location: Illinois 1)
6I. Networks and Information

Chair: Eleonora Patacchini (Cornell University)

• John Bonney
  How Network Hiring by Entrepreneurs Shapes Firm Formation and Performance
  Presented by John Bonney (New York University)
  Discussed by Chiara Castrovillari (Cornell University)
  Abstract

  Many entrepreneurs rely on their personal networks to hire their first employees. How important is this practice for the formation and performance of new firms? I study this question using Norwegian administrative data that allow me to link entrepreneurs to their firms, employees, and former coworkers. To identify causal effects, I develop an instrumental variables framework that jointly models entry and network hiring, allowing for endogenous selection on both margins. The results reveal three main findings. First, each ex-coworker hired in the firm’s first year raises annual revenues in the following four years by over $250K and crowds in other hires, without reducing average productivity. Second, without the ability to hire ex-coworkers, a quarter of network-hiring entrepreneurs would not have started their firms at all. Third, counterfactual simulations show that, compared to entry subsidies, networks enable entry of entrepreneurs who create substantially more jobs, survive longer, and achieve higher value added per worker. Interpreted through the lens of a simple model, the data suggest that private information about coworker quality is a key driver of network hiring. Taken together, the results show that access to human capital through networks is an important determinant of entrepreneurial entry and success.

• Victoria Mooers
  Social Networks and Voter Information
  Presented by Victoria Mooers (University of Oklahoma)
  Discussed by Lucy Page (University of Pittsburgh)
  Abstract

  Informed voters are essential for government accountability, and social networks are an important avenue through which voters acquire political information. However, U.S. House of Representatives districts do not necessarily align with social networks. This misalignment potentially impacts the ease with which voters learn about their representatives, by altering the chance of encountering friends who provide relevant political information. I study whether the alignment between district boundaries and social networks affects voter knowledge, turnout, and campaign contributions in congressional elections. Using Facebook’s Social Connectedness Index and an event study design, I find that an increase in the share of friends living in the same district increases voters’ knowledge about their representative. For example, a 10 percentage point increase in this share raises the probability that a voter knows their representative’s party by 3.3 percentage points; this represents a 5% increase over the mean. Additionally, a higher share of friends in the same district increases voter turnout in House elections and shifts campaign contributions towards own-district House candidates. These findings suggest that aligning political boundaries with social networks can enhance democratic engagement.

• Lucy Page
  Barriers to Civic Action for Clean Air in Delhi
  Presented by Lucy Page (University of Pittsburgh)
  Discussed by Victoria Mooers (University of Oklahoma)
  Abstract

  Why do Delhi residents fail to push for cleaner air despite severe pollution? We study this question through a repeated survey of Delhi residents paired with embedded experiments. Standard demand-side explanations appear insufficient: 78% of respondents want more government action on air pollution, 40% rank it their top policy priority, and 82% believe citizen pressure has at least moderate influence on politicians. Yet few residents have ever contacted a politician, attended a government meeting, or signed a petition about air pollution. We document two candidate constraints. First, residents substantially overestimate others' civic engagement. The median respondent guesses that 73% of survey participants rank air pollution as their top priority, compared to a true share of 40%, and that 40% will message their local politician about air pollution, compared to a true share of 8%. Second, there is a large gap between stated and revealed willingness to act: 63% say they want to send a message to their local politician through a partner NGO, but only 8% actually do. These patterns suggest that low civic engagement reflects neither low demand for clean air nor pessimism about government responsiveness, but rather overoptimism about others' action and behavioral frictions in converting intentions into action. Forthcoming experimental waves will test informational interventions targeting each of these constraints.

• Eleonora Patacchini
  Face-to-face Meetings, Worker Mobily, and Referrals
  Presented by Eleonora Patacchini (Cornell University)
  Discussed by John Bonney (New York University)
  Abstract

  We study how face-to-face interactions shape worker mobility through social networks. Using granular cellphone geolocation and sociodemographic data on 3.3 million workers in a large urban labor market, we exploit settings in which multiple friends of the same worker relocate to jobs within the same destination area. Within the same worker and destination, face-to-face interaction with a friend increases the probability of moving to that friend’s workplace by 36-43 percent relative to remote communica-tion, with effects substantially larger than those of phone calls or digital messaging. Consistent with referral models, the effect emerges only after the friend joins the desti-nation firm and disappears in pre-move placebo periods. A data-driven heterogeneity analysis using regularized machine learning reveals a pronounced targeting gradient: effects in the highest predicted-return settings are more than twice as large as the population average. We interpret these findings through a simple referral model in which interaction technology shapes the effectiveness of information transmission, showing that mobility depends not only on the presence of social ties but on how information flows within them.

September 26, 2026 14:00 to 15:00 (Location: Illinois rooms)
7. Keynote: The Behavioral Economics of Crime by Jens Ludwig

Chair: Bocar Ba (duke)

• Jens Ludwig
  The Behavioral Economics of Crime
  Presented by Jens Ludwig (University of Chicago)
  Abstract

  The Behavioral Economics of Crime

September 26, 2026 15:30 to 17:30 (Location: Iowa)
7A. Measuring Urban Change

Chair: Laurent Gobillon (Paris School of Economics)

• Carolina Arteaga, Victoria Barone, Stephen Claassen
  Demographic Divergence: The Legacy of the Opioid Epidemic
  Presented by Stephen Claassen (University of Toronto)
  Discussed by Laurent Gobillon (Paris School of Economics)
  Abstract

  We study how the opioid epidemic shaped local population growth in the United States. Exploiting variation in exposure to the epidemic—driven by Purdue Pharma's targeted marketing of OxyContin and proxied by 1996 cancer mortality rates—we find that commuting zones with greater exposure experienced lower population growth. By 2020, a one-standard-deviation increase in exposure reduced population growth among individuals aged 18 to 64 by 2.4 percentage points. Direct mortality from drug-induced deaths made only a limited contribution to these changes. Instead, population losses were primarily driven by migratory responses: exposure increased out-migration rates, especially among college-educated individuals. These responses are consistent with the opioid epidemic operating as a disamenity shock, deteriorating local quality of life. We also document a rise in fertility that, by 2020, partially offsets population losses. Our findings show that the opioid epidemic reshaped local demographic composition and contributed to the long-run divergence in population dynamics across U.S. commuting zones.

• Guangbin Hong, Katsuhiro Komatsu, Xinle Pang, Hoyoung Yoo
  Aging Cities
  Presented by Guangbin Hong (Michigan State University)
  Discussed by Stephen Claassen (University of Toronto)
  Abstract

  The U.S. population is aging rapidly, and public housing policy has increasingly pivoted toward the elderly. This paper studies how population aging reshapes neighborhoods and cities, and what it implies for elderly housing policy. Using two instruments for local aging and an event-study design exploiting the staggered placement of elderly-targeted LIHTC (E-LIHTC) projects, we show that a rising old-age share endogenously reorients local amenities toward services favored by older residents and away from those valued by the young. We then develop a quantitative spatial model with two age groups in which the residential choices of the old reshape local amenities and, by displacing the young, propagate to citywide production and commuting. The optimal siting of elderly housing therefore needs to balance local amenity effects and these citywide spillovers. Calibrating the model to Chicago, we find that the welfare cost of elderly housing is largest in young, high-wage, central neighborhoods, and that current E-LIHTC projects are not sited optimally.

• Simon Greenhill, Nicklas Nordfors, Suraj R Nair, Hannah Druckenmiller, Joel Ferguson, Solomon Hsiang, Cornelia Paulik, Andreas Madestam, Hikari Murayama, Eugenio Noda, Anna Tompsett, Sherrie Wang
  Combining aerial photography and machine learning to map 20th century African urban change
  Presented by Nicklas Nordfors (University of Gothenburg)
  Discussed by Guangbin Hong (Michigan State University)
  Abstract

  Developing world cities are projected to grow dramatically throughout the 21st century. Understanding the causes and consequences of 20th century urban growth can improve 21st century urban policy, but requires detailed data on historical urbanization. To address this need, we develop a machine learning method to extract building footprints from historical imagery, producing high-resolution building maps. We produce 592 building maps for 427 African cities between 1946 and 1978. These data measure long-run urban development at an unprecedented spatial granularity and scale. Importantly, our data allow us to study not only large, well-known cities, but also slums, hinterlands, and newly urbanizing areas, which may not be represented in other data sources.

• Laurent Gobillon, Clément Gorin, Stephan Heblich, Amine Ouazad
  The eye of the beholder: Neighbourhood embeddings and housing prices
  Presented by Laurent Gobillon (Paris School of Economics)
  Discussed by Nicklas Nordfors (University of Gothenburg)
  Abstract

  This paper proposes a framework for measuring neighborhood amenities from streetlevel imagery and embedding them into a quantitative spatial model with limited attention. Using 3,825,241 Google Street View panoramas linked to 1,075,587 housing transactions in Paris over 2010-2024, we construct visual embeddings summarizing neighborhood appearance. In hedonic regressions, these embeddings explain a substantial share of housing price variation beyond standard controls, with spatial spillovers decaying beyond 250 meters. We decompose the image embeddings into interpretable semantic objects like vegetation, building type, walkability, public transport, and nightlife using a vision-language model and sparse group LASSO. Amenities are then embedded into a quantitative spatial model where households allocate limited cognitive attention across dimensions, determining which amenities influence equilibrium prices and residential sorting.

September 26, 2026 15:30 to 17:30 (Location: Michigan)
7B. Trade I

Chair: David Krisztian Nagy (CREI)

• Hans Koster, Ernest Liu, Philip Ushchev, Benjamin Wache
  Tourism and Trade
  Presented by Philip Ushchev (Universite Libre de Bruxelles)
  Discussed by David Krisztian Nagy (CREI)
  Abstract

  This paper examines the impact of tourism on international trade using a 30-year panel of bilateral data. We exploit major international sports events as plausibly exogenous shocks to bilateral tourist flows between countries. Our empirical results show that increases in inbound tourism significantly raise subsequent imports from, and exports to the visited country with an elasticity of about $0.2$. Tourism generates transitory, rather than persistent, increases in trade and operates primarily through the extensive product margin. To rationalize these findings, we develop a parsimonious model in which tourists from exporting countries meet potential buyers and distributors abroad, enabling their home-country firms to sell products in foreign markets that were previously out of reach. Counterfactual simulations indicate that eliminating tourism would result in a non-trivial reduction in aggregate international trade.

• Kyle Hood, Dirk van Duym
  Merchandise Trade by State
  Presented by Dirk van Duym (Bureau of Economic Analysis)
  Discussed by Philip Ushchev (Universite Libre de Bruxelles)
  Abstract

  Understanding intranational trade flows is essential for analyzing regional interconnectedness, evaluating place-based and trade-related policies, and supporting the construction of sub-national supply-use and input–output systems. Unlike in the case of international trade, however, interstate trade flows are not observed as a by-product of customs enforcement, since shipments across internal jurisdictional boundaries are not subject to duties or tariffs. As a result, information on sub-national trade has been lacking. It must be inferred from surveys collected for other purposes—most notably infrastructure and freight planning—which are not designed to measure production-to-consumption trade flows and embed substantial pass-through activity from wholesaling and warehousing. In this paper, we develop a framework for estimating interstate trade in goods by re-purposing infrastructure-oriented shipment data. We factor observed flows into production-related shipments and pass-through activity arising from distribution, storage, and wholesale intermediation, allowing for cross-hauling between regions. Using a set of transparent assumptions and a large-scale numerical solver, we reconcile these components into a single matrix (for each commodity) of production-origin to final-destination trade flows that is internally consistent. The framework provides a policy-relevant representation of interstate goods trade that can be used as an input into broader systems of regional economic analysis.

• Banu Demir, Beata Javorcik, Piyush Panigrahi
  Breaking Invisible Barriers: Does Fast Internet Improve Access to Input Markets?
  Presented by Piyush Panigrahi (International Finance Corporation)
  Discussed by Dirk van Duym (Bureau of Economic Analysis)
  Abstract

  Why is aggregate productivity lower in developing countries? This paper argues that a key part of the answer lies in information frictions that distort domestic production networks and hinder efficient supplier matching. It uses theory and empirics to study how such frictions constrain firm-to-firm trade and distinguishes two channels through which they operate: communication frictions, the cost of transmitting known information, and information acquisition frictions, the cost of discovering and evaluating suppliers. Combining firm-to-firm transaction data from Türkiye with rollout of fiber-optic cable, instrumented by proximity to a pre-existing infrastructure, we show that improved connectivity reshapes production networks along two margins: firms reallocate input purchases toward better-connected provinces and diversify their supplier base within those provinces. We develop and estimate a spatial model with endogenous production networks in which communication cost reductions drive reallocation across origins and information cost reductions broaden firms’ supplier sets. Counterfactuals show that easing these frictions through fiber expansion generates sizeable welfare gains, highlighting the quantitative importance of information frictions for aggregate productivity.

• David Krisztian Nagy, Woan Foong Wong
  Growth, Trade, and the Uneven Rise of Port Regions
  Presented by David Krisztian Nagy (CREI)
  Discussed by Piyush Panigrahi (International Finance Corporation)
  Abstract

  Can spatially neutral productivity improvements have spatially non-neutral consequences? Across three major sources of countrywide productivity growth — increased internet penetration, vaccination campaigns, and improved access to schools –, we document a consistent reallocation of economic activity toward port regions within countries. We rationalize these findings in a spatial model in which productivity growth increases export volumes, which in turn triggers a reduction in import costs through the round-trip effect (Wong, 2022), thus allowing port regions to attract economic activity. This prediction contrasts with a standard model with exogenous trade costs in which spatially neutral productivity growth shifts economic activity toward interior regions that are more isolated from global markets. We calibrate our model and simulate it with and without the round-trip effect. Preliminary results suggest that the round-trip effect accounts for 15.6% of relative population growth in port regions and amplifies welfare gains from TFP growth by 24.1% in developing countries.

September 26, 2026 15:30 to 17:30 (Location: LaSalle)
7C. Retail Agglomeration

Chair: Eunjee Kwon (University of Cincinnati)

• August Chen, Franklin Qian, Qianyang Zhang, Xiang Zhang
  Identifying Agglomeration Spillovers: Evidence from Grocery Store Openings
  Presented by August Chen (Rice University)
  Discussed by Ben Klopack (Texas A&M University)
  Abstract

  We provide the first large-scale causal estimates of demand-side agglomeration spillovers in the non-tradable service sector, leveraging data on 851 grocery store openings across the United States between 2019 and 2022. We combine a convolutional neural network with propensity score matching to identify credible counterfactual opening sites and compare business outcomes between actual and counterfactual locations. Grocery store openings increase foot traffic to nearby businesses within 0.1 miles by 23.4 percent within 6--12 months, with spillovers decaying sharply beyond this radius. The effects are strongest for wholesale and retail establishments and hospitality services, and are substantially attenuated after COVID-19---from 42.7 to 14.6 percent---consistent with a trip-chaining mechanism. Grocery anchors also reshape the local business landscape: within 0.1 miles, establishment growth rates rise by 7.0 percentage points and employment growth rates by 8.8 percentage points, averaged over the first three post-entry years. Decomposing establishment growth rates, the effect is accounted for by higher entry and fewer exits.

• Leah Brooks, Rachel Meltzer, Erica Moszkowski
  Agglomeration Writ Small: New Theory and Evidence on Why and How Consumer-Facing Establishments Cluster
  Presented by Erica Moszkowski (UBC)
  Discussed by Lindsay Relihan (Purdue University)
  Abstract

  Although we understand consumption as a central part of urban growth and vitality, we have limited tools to accurately capture whether this sector is distinct from other production oriented sectors. To understand whether and why the externalities at play in the consumer-facing sector differ from those in other industries, we propose a new measure of establishment clustering and motivate it with a model of firm location choice with spillovers. We lay out three new facts about how the spatial clustering of consumer-facing firms differs from that of other industries using our metric. Using data from the universe of consumer-facing establishments in U.S. metros, we first show that consumer-facing establishments are more likely to have at least one neighbor, relative to establishments in other industries. Second, consumer-facing establishments are among the most clustered type of establishment at the median of their distribution for nearly all metro areas. And, third, the most clustered consumer-facing establishments are usually less clustered than the most clustered establishments in other industries. Our model rationalizes these descriptive facts through variation across industries in spillover strength, which delivers agglomeration, and firm heterogeneity, which delivers within industry sorting. Our model shows how to separate these two forces using information on the distribution of clustering across firms within an industry. Furthermore, since the model incorporates multiple channels through which agglomeration can affect firm profits, we use it to compare the strength of agglomeration and sorting forces across industries.

• Eunjee Kwon
  Retail Rents under E-Commerce Pressure: Evidence from South Korean Commercial Real Estate
  Presented by Eunjee Kwon (University of Cincinnati)
  Discussed by Yichen Su (Southern Methodist University)
  Abstract

  E-commerce is changing how commercial space in cities is used, yet we know surprisingly little about how this shift is capitalized into rents. This paper estimates the effect of online shopping expansion on retail commercial rents by linking transaction-level credit card data to building–floor–industry rent data from South Korea (2015–2023). South Korea is well suited to this question: e-commerce penetration is high and its dense mixed-use buildings and granular rent data allow us to study adjustment both horizontal (across locations) and vertical (across floors) rent gradients. We employ a shift-share instrumental variable strategy that combines demographic-predicted product consumption shares with U.S. online penetration trends as shifts. Our preferred IV specifications reveal that a one-percentage-point increase in the online share reduces real retail rents by approximately 0.6 percent. The aggregate effect conceals important heterogeneity. Goods-oriented retail rents experience the largest declines through direct online substitution, while restaurant rents decline most on the second floor—not from direct substitution, but from vertical resorting as restaurants move into streetlevel spaces vacated by contracting ground-floor retail. E-commerce erodes the steep vertical rent gradient from the bottom up: the shock originates at street level and propagates upward, with diminishing intensity on higher floors.

September 26, 2026 15:30 to 17:30 (Location: Illinois 3)
7D. Housing Supply II

Chair: Elisabet Viladecans-Marsal (Universitat de Barcelona, IEB)

• Alexander Bartik, Arpit Gupta, Daniel Milo, Desmond O'Shaughnessy
  The Origins of American Zoning
  Presented by Desmond O'Shaughnessy (New York University)
  Discussed by Elisabet Viladecans-Marsal (Universitat de Barcelona, IEB)
  Abstract

  We construct the first large-scale longitudinal dataset of American zoning ordinances, spanning nearly a thousand municipalities across one hundred years. We combine historical codes retrieved through direct municipal outreach, newspaper archives, and modern ordinances to document three central facts about the evolution of zoning. First, zoning has shifted away from strict Euclidean use segregation toward flexible and mixed-use provisions, with prohibitions on multifamily housing, accessory dwelling units, and mixed use declining substantially. Second, regulatory attention has migrated toward environmental, aesthetic, and growth-management objectives implemented through procedural requirements such as site plan review and design boards. Third, zoning codes have grown dramatically more complex: lengthening from roughly 20 to 200 pages, accompanied by more zoning districts and greater dispersion in bulk regulations across zones. These facts suggest that the impact of housing regulation is less due to persistence of early zoning rules than the accretion of novel regulatory goals layered into new procedural instruments and complex structures.

• Stephanie Kestelman
  Effects of Transit-Oriented Incentive Programs: Evidence from Los Angeles
  Presented by Stephanie Kestelman (Harvard University)
  Discussed by Desmond O'Shaughnessy (New York University)
  Abstract

  This paper examines the effects of transit-oriented development incentives on multifamily housing development, using the implementation of Los Angeles’s Transit- Oriented Communities (TOC) incentive program as a natural experiment. Introduced in 2017, TOC streamlined the entitlement process for and granted density bonuses to qualifying projects near major transit stops. Using novel, project-level data covering applications, approvals, and permits from 2004 to 2023, I document a sharp increase in development proposals following the program’s rollout. However, this increase in applications did not translate into a rise in the rate of new housing supply. The program did shift the composition of new development: TOC projects were more likely to include income-restricted units and be located in lower-income, renter-occupied neighborhoods. Using a structural framework and estimated approval probabilities, I show that TOC enabled projects that would have faced lower approval chances or higher appeal risk under the previous discretionary regime. These findings highlight how procedural streamlining can shift the type and location of urban development, even if aggregate supply effects are limited.

• Suvy Qin
  Cutting Delays, Not Corners: The Role of Bureaucratic Performance in Housing Supply
  Presented by Suvy Qin (UC Berkeley)
  Discussed by Stephanie Kestelman (Harvard University)
  Abstract

  Regulatory frictions are seen as a constraint on housing supply, yet the role of bureaucratic capacity in enforcing these rules remains largely unstudied. This paper examines how bureaucratic performance in the permitting process affects the timely provision of housing. Using novel administrative data from Los Angeles, I find that a one standard deviation faster bureaucrat issues permits in 15.4% less time on average. Faster bureaucrats are more likely to work voluntary overtime and to be hired in years with a more competitive public sector pay. Higher-performing bureaucrats can manage a workload increase during busy periods without taking longer to review permits. I use an instrumental variables design to estimate how being assigned to a faster bureaucrat affects project quality, finding no measurable effect on the likelihood of a future code violation. Back-of-the-envelope estimates imply that increasing bureaucratic capacity through alternative personnel policies is cost-effective.

• Antoine Grimaud, Ghizlen Ouasbaa, Albert Sole Olle, Elisabet Viladecans-Marsal
  Local Housing Supply Spillovers: Evidence from California
  Presented by Elisabet Viladecans-Marsal (Universitat de Barcelona, IEB)
  Discussed by Suvy Qin (UC Berkeley)
  Abstract

  Housing supply is largely determined by local governments, yet housing markets operate across municipal boundaries. When one municipality expands construction, housing supply shocks may spill over into neighboring jurisdictions. This paper provides causal evidence on such interjurisdictional housing supply spillovers. We exploit close elections between real estate developers and non-developers for city council seats in California as a quasi-random shock to pro-development political representation. Electing a developer increases housing permits in the treated municipality by about 0.5 log points and generates positive spillovers in nearby jurisdictions: permitting rises by roughly 0.28 log points within five miles and 0.19 log points within ten miles, with effects fading beyond that point. Guided by a framework combining housing market substitution with local political incentives, we interpret these results as evidence that development decisions across municipalities can be complementary. The presence of such spillovers highlights the potential importance of cross-jurisdictional externalities in decentralized land-use regulation.

September 26, 2026 15:30 to 17:30 (Location: Illinois 4)
7E. Climate and Cities in Developing Economies

Chair: Simon Buechler (Miami University)

• Alishuba Milna Philip
  Slum Redevelopments and Evictions in a Developing Megacity
  Presented by Alishuba Milna Philip (University of Zurich)
  Discussed by Mélanie Gittard (Stanford Universitiy)
  Abstract

  I study the short- and medium-term effects of slum clearance and redevelopment on incumbent slum residents in Mumbai by assembling a novel residential mobility panel that tracks all 12 million residents of the city over 15 years by digitising and matching electoral roll records. Exploiting variation in the timing of redevelopment approvals across slums, I use a staggered difference-in-differences design to estimate these effects. Slum residents who are entitled to in-situ compensation — ownership of an apartment in the redeveloped neighbourhood — are persistently displaced from the neighbourhood. Redevelopment increases the probability of living in formal housing by 6 p.p., but the effect on the probability of living in other slums is four times as large, shifting slum residents across slum neighbourhoods more than out of slum living. Using digitised project records and high-resolution daytime satellite data, I document substantial non-completion after slum clearance and large lags between slum clearance and completion of units for slum residents. A key feature of the policy design is the timing mismatch between displacement and compensation: residents are cleared before receiving replacement housing, which is delivered only when units are built on the cleared site. Transitions out of slum living are 250% larger in completed than incomplete projects and largest where completion occurs soon after clearance. These results suggest that while slum redevelopments succeed in redeveloping the targeted site, the resulting displacement of incumbent residents to slums elsewhere can undermine the broader goal of curbing slum growth in the city.

• Juliana Pinillos
  Long-Term Effects of Slum Upgrading on Climate Resilience
  Presented by Juliana Pinillos (University of Southern California)
  Discussed by Alishuba Milna Philip (University of Zurich)
  Abstract

  This paper studies whether slum upgrading can serve as a climate adaptation policy. I examine the long-run effects of PRIMED, a large slum-upgrading program implemented in Medellin, Colombia, between 1993 and 2000. In addition to standard neighborhood improvements, PRIMED invested in risk-mitigation infrastructure. I exploit the program’s incomplete rollout to estimate its effects on disaster risk, and long-run urban development. I find that slum upgrading substantially reduces vulnerability to extreme rainfall events. Upgraded neighborhoods experience fewer floods and landslides and less housing damage than comparable non-upgraded slums. At the same time, upgrading changes the spatial pattern of urban development: housing and population density increase in treated areas, while non-upgraded slums continue expanding into higher-risk land. I find no evidence of changes in population compositional, suggesting no gentrification effects. I then use the reduced-form estimates to discipline a quantitative spatial model of the city. The model implies that slum upgrading generates positive citywide welfare effects through improvements in local amenities and reductions in environmental risk exposure. The results show that place-based infrastructure policy in informal settlements can produce long-run gains in both urban development and climate resilience.

• Mélanie Gittard
  Climate Migration and Urban Integration in Africa
  Presented by Mélanie Gittard (Stanford Universitiy)
  Discussed by Juliana Pinillos (University of Southern California)
  Abstract

  This paper studies how climate-induced migration shapes urban settlement patterns and employment composition in African cities. Using census data from fifteen African countries spanning 1987-2019 combined with high-resolution climate data (CHIRPS and CHIRTS), we show that dry conditions increase rural-to-urban migration by up to 24% in less arid regions, with effects concentrated in medium-sized secondary cities rather than major metropolitan areas. The migration response is driven by water scarcity during the agricultural growing season and is concentrated in regions where populations have not developed adaptation mechanisms to drought. We implement a shift-share instrumental variables strategy that exploits exogenous variation in historical migration networks and contemporaneous climate shocks to identify the causal impact of climate-induced migration on urban outcomes, including informal settlement expansion and employment composition. We then examine whether climate migrants disproportionately settle in informal settlements and remain in agricultural or low productivity informal employment compared to economic migrants, and whether this pattern reinforces incomplete structural transformation in African cities. To distinguish whether observed patterns reflect binding integration barriers or migrant preferences, we develop a structural model of migrant integration choices that will allow us to quantify the welfare costs of climate-induced displacement and evaluate targeted policy interventions such as housing vouchers and formal employment access programs.

September 26, 2026 15:30 to 17:30 (Location: Illinois 5)
7F. Financial Frictions and Spatial Allocations

Chair: Jane Dokko (Federal Reserve Bank of Chicago)

• Erica Xuewei Jiang, Yeonjoon Lee, Quinn Maingi
  Credit Without Proximity: Informational Frictions and Unequal Gains from Technology
  Presented by Quinn Maingi (USC Marshall School of Business)
  Discussed by Jane Dokko (Federal Reserve Bank of Chicago)
  Abstract

  We study how the organization of information production---and its response to economic and technological forces---affects informational efficiency, credit allocation, and borrower risk. Using U.S. administrative data linking mortgage applications to loan officers and subsequent loan performance, we show that underwriting facilitated by officers located close to the borrower increases approval rates without worsening ex-post performance or processing speed, but is not always deployed where it is most valuable, because lenders allocate loan-officer labor elastically with respect to local wages. These gains are especially large for observably riskier borrowers. We develop and estimate a model that combines a core information-production problem over latent borrower risk, an endogenous choice over local versus remote underwriting, and equilibrium in mortgage and labor markets. We find substantial baseline credit rationing---up to 15 percent in high-risk segments---with local officers eliminating roughly half of it while also reducing excessively risky approvals. A technology shock that raises the processing productivity of remote officers induces lenders to substitute away from local screening, lowering informational efficiency, increasing excessively risky approvals and expected defaults, and tightening rationing for marginal borrowers despite only modest reductions in interest rates.

• Elena Aguilar
  Credit Constraints, Learning, and Spatial Misallocation
  Presented by Elena Aguilar (Princeton University)
  Discussed by Quinn Maingi (USC Marshall School of Business)
  Abstract

  Human capital is a critical determinant of productivity in modern economies, and we have long understood credit market frictions to be a critical barrier for its accumulation. In this paper I study location decisions as a particular form of human capital investment, where individuals trade long-term benefits for high upfront costs, most notably in the form of housing. I show that when locations differ in the learning opportunities they offer and agents are heterogeneous in their learning ability, credit frictions not only weaken positive sorting of learning ability across space but, under empirically relevant conditions, they will induce negative sorting among individuals that are credit constrained. That is, marginally better learners will optimally choose to reside in locations offering worse learning opportunities. Relying on a novel source of administrative data from Spain, I show that this mechanism is not only empirically relevant, but also quantitatively important. I use the data to document the key mechanisms of the theory and quantify the losses associated with the effect of credit frictions on the spatial distribution of labor. Importantly, these losses arise from distortions in the composition of skill in each city: the dominant source of inefficiency is the spatial misallocation of individuals with high learning-ability. In the presence of negative sorting, standard place-based policies strictly aimed to expand the size of productive cities may have limited effects, making it important to design policies that can better target the composition of heterogeneous workers across space.

• Rachel Bryant
  From Promised Jobs to Net Local Employment
  Presented by Rachel Bryant (University of Chicago)
  Discussed by Elena Aguilar (Princeton University)
  Abstract

  State and local governments spend billions on firm-specific subsidies, frequently touting the number of new jobs the entering firm promises to create. This paper asks how gross jobs at large subsidized tradable firms translate into net local employment, service-sector spending, rents, migration, and incidence. Using subsidy records, linked employer-employee data, mobile-device location data, and commercial brokerage data, I document the following facts: gross entrant jobs are not net local jobs, entrant hires come from multiple source margins, hiring follows structured industry-flow networks, and nontradable demand follows residence-shopping geography. I quantify a spatial equilibrium model with commuting, shopping, low-skill nontradable occupational choice, and land-allocation frictions. The baseline counterfactual decomposes net employment into entrant gross jobs, incumbent crowd-out, and nontradable multiplier effects. Policy counterfactuals show that incidence depends on which firm is targeted, where in the city the subsidy is placed, and how elastically land can be reallocated—margins that single average multipliers ignore.

September 26, 2026 15:30 to 17:30 (Location: Directors)
7G. Home Buying

Chair: Christopher Timmins (University of Wisconsin - Madison)

• Kelly Bishop, Nicolai Kuminoff, Alvin Murphy, Liam Perdue
  Household Sorting, Mobility Costs, and Residential Misallocation
  Presented by Alvin Murphy (Arizona State University)
  Discussed by Christopher Timmins (University of Wisconsin - Madison)
  Abstract

  This paper studies how moving costs affect household mobility, welfare, and residential sorting within U.S. metropolitan areas. We develop a dynamic model of neighborhood choice and estimate moving costs using data on nearly three million homeowners from the American Community Survey. Our results show that realtor commissions account for a large share of within-metropolitan moving costs and substantially reduce mobility, especially in high-cost housing markets. Counterfactual simulations indicate that halving commission rates would significantly increase mobility and generate large welfare gains. However, lower moving costs may also intensify income-based sorting, highlighting tradeoffs between efficiency, mobility, and spatial inequality.

• Paul Carrillo, Jeanna Kenney, Matt Martinez
  Agent Effort in Housing Search: Evidence from Open Houses
  Presented by Jeanna Kenney (Villanova University)
  Discussed by Alvin Murphy (Arizona State University)
  Abstract

  This paper studies the determinants and effects of intermediary effort in the residential housing market. We use open houses - scheduled time windows during which a listed property is open for unscheduled walk-in visits by potential buyers - as a directly observable form of effort and assemble a novel dataset covering the universe of open-house events in the Washington, DC metropolitan area from 2018-2024. Listings are linked to listing histories, agent identifiers, property characteristics, and transaction outcomes. We find open-house effort rises with expected commission dollars and local buyer demand, falls with agent workload, and is concentrated during periods of high buyer availability. Using instrumental variables based on prior-agent open-house propensities and prior-agent availability, we find that early open houses increase the probability of selling within 30 days by roughly 17 percentage points, with no robust price gains. Spillover IV estimates suggest that nearby open houses may reduce focal-listing liquidity, consistent with buyer-attention congestion rather than positive shopping externalities.

• Xinyu Cao, Jinfei Sheng, Tianping Wu
  The Political Economy of Home Buying
  Presented by Xinyu Cao (UBC)
  Discussed by Jeanna Kenney (Villanova University)
  Abstract

  We examine whether partisanship shapes home purchase decisions using a novel dataset that links individual housing transactions to voter registration records in the United States from 2010 to 2023. We find that Republicans are twice as likely as Democrats to purchase homes and realize 0.69 percent higher housing returns compared to Democrats. Individuals are also significantly more likely to purchase a home when their affiliated party controls the White House, highlighting the role of partisan alignment in home-buying decisions. This alignment effect is stronger among male and younger voters. Importantly, partisan alignment also affects purchase timing: aligned buyers tend to accelerate their purchases, which is associated with lower subsequent returns. Evidence from survey data suggests that partisan alignment leads to more optimistic housing market expectations and influences home buying decisions through this expectations channel. These partisan-driven shifts in housing demand also have aggregate effects on local housing markets, leading to higher house prices but lower subsequent housing returns---a pattern consistent with belief-driven overvaluation. Overall, our findings highlight the role of partisanship in shaping housing market participation, timing, and prices.

• Zach Babat, Lu Han, Christopher Timmins
  Not All Bids Are Equal: Racial Gaps in the Housing Market
  Presented by Christopher Timmins (University of Wisconsin - Madison)
  Discussed by Xinyu Cao (UBC)
  Abstract

  We study racial disparities in housing transactions using novel data that describe the full set of offers submitted for residential properties, including both accepted and rejected bids. Leveraging within-listing comparisons across competing offers, we show that while bidders of color submit lower bids on average, a significant “acceptance gap” remains. Even after controlling for bid amount, bid rank, buyer agent, and detailed offer terms, bidders of color are significantly less likely to have their offers accepted for the same listing. We quantify this disparity in the form of the bid-equivalent penalty: to achieve the same probability of acceptance as a white buyer, a buy of color must bid approximately 2 – 3 percent more. These disparities intensify in tighter markets (# buyers to # listing ratio), in listings with more competing bidders, and when the seller is white – scenarios where sellers possess greater market power and where the identity of the marginal buyer is perceived as consequential. Taken together, our results are difficult to reconcile with explanations based solely on buyer preferences, financing constraints, or search costs. Rather, they point to differential acceptance thresholds applied by sellers.

September 26, 2026 15:30 to 17:30 (Location: Moskow)
7H. Transportation Safety and Externalities

Chair: Christopher Severen (Federal Reserve Bank of Philadelphia)

• Enrico Moretti, Harrison Wheeler
  The Traffic Noise Externality: Costs, Incidence and Policy Implications
  Presented by Harrison Wheeler (University of Toronto)
  Discussed by Christopher Severen (Federal Reserve Bank of Philadelphia)
  Abstract

  More than 42 million Americans are exposed to medium or high levels of traffic noise. Despite its potentially large economic toll and unequal distribution, the aggregate costs, incidence, and policy implications of traffic noise have received limited attention in economics. We quantify the economic cost of traffic noise by estimating its effect on homebuyers' willingness to pay for quieter environments. Using quasi-experimental variation based on the construction of noise barriers, we find that reduced traffic noise exposure leads to significant increases in house prices indicating that buyers are willing to pay a substantial premium for each decibel of noise reduction. The effects are largest within 100 meters and decline with distance. We use these estimates to calculate the aggregate economic cost of traffic noise at $110 billion nationwide. The economic burden of the externality is disproportionately borne by lower income and minority households, suggesting that the externality is regressive. The cost varies widely across cities, due to differences in noise levels, property values and density. Using our estimates, we calculate that the socially efficient Pigouvian tax amounts to $974 per vehicle. In addition, we estimate that a broad shift to electric vehicles -- which are quieter than traditional vehicles -- could yield noise reduction benefits of $77.3 billion, concentrated among low-income families in dense urban areas.

• Jonathan Hall, Conor Lennon, Joshua Madsen
  How Do Driver Assistance Technologies Affect Transportation Safety?
  Presented by Jonathan Hall (University of Alabama)
  Discussed by Xibo Wan (University of Connecticut)
  Abstract

  Advanced Driver-Assistance Systems (ADAS), such as Automatic Emergency Braking, Blind Spot Monitoring, and Lane Keeping Assist, are now common in new vehicles. In theory, these new safety features should reduce automobile crashes. However, these technologies may be ineffective in real-world driving conditions or could encourage riskier driving behavior, offsetting some or all of the anticipated safety benefits. Using 2016 to 2020 crash data from 38 U.S.\ states and national-level fatal crash data, we employ a difference-in-differences approach that compares crashes among vehicles with and without driver assistance systems when exposed to an exogenous change in crash risk: precipitation. Our approach mitigates the potential for bias due to selection into vehicles with and without ADAS. We find that ADAS is associated with a 19.1\% reduction in crashes. However, we find a 14.6\% increase in fatal crashes, suggesting some moral hazard effects among drivers using ADAS-equipped vehicles. A regression discontinuity analysis centered on the annual transition to Daylight Savings Time provides supporting evidence. Notably, our estimates are sensitive to the composition of the non-ADAS comparison group, as older vehicles appear to experience more crashes at times of heightened risk.

• Florian Allroggen, John Hansman, Christopher Knittel, Jing Li, Xibo Wan
  Planes Overhead: How Airplane Noise Impacts Home Values
  Presented by Xibo Wan (University of Connecticut)
  Discussed by Harrison Wheeler (University of Toronto)
  Abstract

  Air transportation supports economic growth and global connectivity but imposes localized environmental costs, particularly through aircraft noise. We estimate the causal effect of aviation noise on housing prices using quasi-experimental variation from the Federal Aviation Administration's rollout of performance-based navigation (PBN) procedures and runway reconfigurations at three major U.S. airports. Combining high-resolution flight trajectory data with geocoded housing transactions, we apply a difference-in-differences hedonic framework to identify changes in exposure unanticipated by residents. A one-decibel increase in annual day-night average sound level reduces house prices by 0.6 to 1.0 percent. Among alternative noise metrics, average exposure explains property value impacts most strongly. Willingness to pay for quieter conditions varies systematically with income and race, indicating that aircraft noise externalities have meaningful distributional consequences. Our results highlight the need to incorporate localized environmental costs into aviation and urban land-use policy.

• Dustin Frye, Christopher Severen
  Mortality Consequences of the Interstate Highway System
  Presented by Christopher Severen (Federal Reserve Bank of Philadelphia)
  Discussed by Jonathan Hall (University of Alabama)
  Abstract

  Limited-access freeways are a highly engineered technology that aim to efficiently and safely move large numbers of people, yet the United States, a premier adopted of freeways, regularly experiences very high automobile mortality rates. We use new data to analyze the effect of the Interstate Highway System on automobile mortality in the mid-20th century. To overcome concerns about endogenous placement and timing of constructed highways, we adapt a network theoretic approach in combination with historical planned highway maps. Preliminary estimates indicate that the IHS reduced automobile mortality by 3\% between 1950 and 1988. Heterogeneity analysis and examination of the 1970s oil crises point to varied mechanisms at play: In cities, highways improve driving safety, likely due to lower speeds. In rural areas, highways may facilitate more driving with an accompanying increase in mortality risk.

September 26, 2026 15:30 to 17:30 (Location: Illinois 1)
7I. Homelessness

Chair: Matthew Freedman (University of California, Irvine)

• David Phillips
  Using Algorithmic vs. Human Judgment to Prioritize Access to Homelessness Programs
  Presented by David Phillips (University of Notre Dame)
  Discussed by Winnie van Dijk (Yale University)
  Abstract

  This paper evaluates human versus algorithmic prioritization for homelessness programs. Communities ration access to scarce housing subsidy programs for homeless individuals via a prioritization ranking embedded in the homelessness services coordinated entry system. Working with the homelessness services system in Bexar County, TX (San Antonio), we designed two new, alternative prioritization rankings and implemented them side-by-side: (i) a simple data-driven algorithm using existing homelessness service system data and (ii) the ratings of human experts who staff the homelessness services system. Both systems were prompted to predict risk of persisting in homelessness among a group of clients who were homeless at baseline. The data-driven algorithm predicts homelessness persistence more accurately than both the status quo tool (the VI-SPDAT) and raw human assessments. Furthermore, the two new systems prioritize different populations: algorithms emphasize extensive homelessness history and behavioral health issues, whereas human assessors prioritize families and individuals with no income.

• Stephen Billings
  Does Diversion Improve Housing, Health and Criminal Outcomes
  Presented by Stephen Billings (University of Colorado)
  Discussed by David Phillips (University of Notre Dame)
  Abstract

  This study evaluates the effectiveness of homelessness diversion interventions designed to prevent shelter entry among households experiencing an immediate housing crisis. Conducted across three community-based service organizations in geographically distinct locations, the project uses a large-scale, multi-site randomized controlled trial to examine whether flexible financial assistance and proactive case management improve housing stability and reduce emergency shelter utilization. Eligible participants are randomized into a control group, a financial assistance treatment group, or a combined financial assistance and case management treatment group. As of April 2026, 2,288 clients had enrolled across the three participating sites. Preliminary analyses suggest encouraging trends, including increased likelihoods of exits to stable housing situations such as renting or staying with family and friends, alongside decreased likelihoods of exits to locations not fit for habitation among treatment participants relative to controls. Ongoing follow-up efforts will continue assessing longer-term housing and well-being outcomes.

• Jamie McCasland
  The Economics of Shelter Refusal: Heterogeneous Preferences for Interim Housing
  Presented by Jamie McCasland (University Of British Columbia)
  Discussed by Stephen Billings (University of Colorado)
  Abstract

  Unsheltered homelessness is one of the most visible manifestations of poverty in the United States, yet policymakers continue to debate whether persistent unsheltered homelessness primarily reflects shelter supply constraints, preferences to remain outside, or mismatches between available shelter options and the preferences of unhoused individuals. This paper provides new quantitative evidence on how shelter characteristics shape demand for interim housing among unsheltered adults. We study preferences for interim housing using a discrete choice experiment administered to 605 unsheltered adults across the San Francisco Bay Area. Respondents completed repeated choice tasks comparing experimentally varied hypothetical shelter options and then indicated whether they would prefer their chosen shelter to where they slept the previous night. Shelter attributes included privacy, curfews, drug testing policies, pet and partner restrictions, cleanliness, allowable length of stay, proximity, case management, and resident participation in rule-making. We find substantial demand for shelter overall: respondents choose a hypothetical shelter over their prior sleeping arrangement in roughly three-quarters of choice scenarios. At the same time, shelter characteristics meaningfully affect willingness to come inside. Privacy and low-barrier environments strongly increase shelter demand, while policies regarding pets, partners, and curfews also play important roles. Using a latent class framework, we additionally document substantial heterogeneity in preferences: 51% of respondents appear broadly willing to accept shelter and relatively insensitive to shelter characteristics, 29% are highly responsive to shelter design and policies, and 20% exhibit persistently strong preferences for remaining unsheltered.

• Nour Abdul-Razzak, John Eric Humphries, Stephen Stapleton, Winnie van Dijk
  The Impacts of an Unconditional Cash Transfer for Homeless Families: Experimental Evidence from Illinois
  Presented by Winnie van Dijk (Yale University)
  Discussed by Jamie McCasland (University Of British Columbia)
  Abstract

  This paper studies the impact of a large one-time cash transfer to homeless families with children. We evaluate how such a cash transfer impacts future homelessness and housing stability. Additionally, we study how assistance affects recipients' well-being, mental health, and labor market participation. After recruiting over 1,100 adults with children staying at emergency shelters and transitional housing in Illinois, participants were randomly assigned to a treatment group that received a one-time unconditional $9,500 cash transfer, or a control group that received $500. Results will be disclosed by the time of the conference.