Greg Howard

Abstract: We show a new fact: the t-year interstate migration rate is close to proportional to the square root of t. This is a puzzle for simple moving cost models which predict an approximately linear relationship. We propose a model based on persistent and spatially-correlated unobservable match-specific characteristics that matches the square root fact, as well as the frequency and gravity pattern of internal migration. Compared to the standard model, the new model is better at forecasting individuals' locations, has different implications for welfare effects of migration policies, and makes different predictions for long-run population changes in response to shocks. 

(IZA Discussion Paper 15622)

Media Coverage: (Washington Post) (The Chronicle of Higher Education) (Brookings) (VoxEU) (U of I News Bureau)

Abstract: Regional public universities educate approximately 70 percent of college students at four-year public universities and an even larger share of students from disadvantaged backgrounds. They aim to provide opportunity for education and social mobility, in part by locating near potential students. In this paper, we use the historical assignment of normal schools and insane asylums (normal schools grew into regional universities while asylums remain small) and data from Opportunity Insights to identify the effects of regional universities on the social mobility of nearby children. Children in counties given a normal school get more education and have better economic and social outcomes, especially lower-income children. For several key outcomes, we show this effect is a causal effect on children, and not only selection on which children live near universities. 

Draft available via email request

Abstract: The internal migration literature typically estimates average moving costs to be several times larger than annual income. How should economists interpret this estimate? I show that in the steady-state of a standard model, average moving costs are proportional to the Shannon entropy of next period's location minus the Shannon information of staying in the same location. Therefore, average moving costs are a helpful statistic about the model's predictive power regarding future moves but are not an estimate of the literal cost of moving. This alternative interpretation makes sense of the range of moving costs in the literature.

Abstract: The distribution of the population of cities in the United States has a fat tail. I show that this pattern extends to cities in smaller geographic areas such as Census regions, divisions, and states, indicating a fractal nature. To explain this phenomenon, I propose a location-demand-based theory in which people have heterogeneous geographic attachment. I illustrate the mechanism with a toy model and then estimate an empirical location demand model with this feature. The quantitative model matches key features of migration data and has the ability to generate the observed tail distribution. An implication of the theory is that the elasticity of population to shocks or policies is larger in bigger cities. 

(DOI)

Media Coverage: (U of I News Bureau) (Higher Ed Dive)

Abstract: We use a novel identification strategy to investigate whether regional universities make their local economies more resilient. Our strategy is based on state governments using similar site-selection criteria to assign normal schools (to train teachers) and insane asylums between 1830 and 1930. Normal schools became larger regional universities while asylum properties mostly continue as small state-owned psychiatric health facilities. We find that a regional university roughly offsets the negative effects of manufacturing exposure. We show the resilience of regional public university spending is an important mechanism, and we show correlations consistent with bachelor’s degree share also playing a mediating role.

(Replication Data and Code) (DOI)

Abstract: Who is rationally inattentive, and in what situations? Models of rational inattention allow agents to make mistakes in their actions, but assume they optimize their allocation of attention. Using millions of online chess moves, I test this assumption, by comparing the marginal benefits (better moves) and marginal costs (less time for future moves) of attention. I find that high-skilled players equalize marginal benefit and cost, but low-skilled players have higher marginal cost, i.e. they spend too long on moves. I also find that having less time leads to deviations from rational inattention. A simple intervention improves players’ attention allocation. 

(Replication Data and Code) (DOI) (EIG Executive Summary)

Abstract: Remote work has increased the demand for housing and changed the demand for the location of that housing. Because housing supply is heterogeneous across space and more elastic in the long-run, the effects on rents and populations may differ over time. We use the lens of a spatial housing model with heterogeneous housing supply elasticities to identify the housing and location demand changes from 2020-2022, and show that the same shocks will have different effects in the long run. Even though rents and prices increased significantly in the short-run, we estimate that in the long-run, increased housing demand will increase rents by only 1.8 percentage points, and that changing location demand will decrease rents by 0.3 percentage points, with a more negative impact on cities in which CPI is measured and cities that were initially expensive. 

(Replication Data and Code) (DOI) 

Abstract: We document a new fact: regional divergence, the rate at which rich states grow faster than poor states, explains most U.S. house price movements since 1939, including the post-2000 boom-bust-boom cycle. An industry-share instrument provides evidence the relationship is causal, implying the location of economic growth affects national house prices. We propose a model to learn why regional divergence and house prices are related. In the model, high interstate inequality raises rents on average because relative demand for living in a high-income state increases and housing supply in low-income states is elastic. Regional divergence leads to higher expected future interstate inequality, which implies higher expected future rents, and therefore, higher current house prices.  The model accurately predicts rents since 1929, which are quite different than prices, as well as cross-sectional moments of prices, rents, construction, and migration. 

(Online Appendix) (Replication Data and Code) (DOI)

Abstract: This paper provides evidence of rational inattention by experienced professionals in strategic interactions. We add rational inattention to a game of matching pennies with state-dependent payoffs. Unlike the full-information mixed-strategy Nash equilibrium, payoffs of different actions need not be equated state-by-state. Moreover, players respond partially to payoff differences, this responsiveness is stronger when attention costs are lower, strategies converge to full-information Nash as stakes increase, and average payoffs across all states are approximately equal across actions. We test these predictions using data on millions of pitches from Major League Baseball, where we observe strategies, payoffs, and proxies for attention costs. 

 (Replication Data and Code) (DOI)

Abstract: How do different local policies in a federal system affect local land values, production, and sorting? We study the question exploiting a large historical policy change: U.S. Alcohol Prohibition in the early twentieth century. Comparing same-state early and late adopters of county dry laws in a difference-in-differences design, we find that early Prohibition adoption increased population and farm real estate values. Moreover, we find strong effects on farm productivity consistent with increased investment due to a land price channel. In equilibrium, the policy change disproportionately attracted immigrants and African-Americans. 

(Replication Data and Code) (DOI)

Abstract: Real rents measured in the United States CPI increased 17.4 log-points from 2000-2018. We present a spatial equilibrium framework to decompose the increase into several channels, including demand to live in housing-supply-inelastic cities. We find location demand contributed significantly: using parameterizations from the literature and a new rent index, we find it is responsible for between 17 and 73 percent of the overall rent increase, and an even larger share in cities where CPI is measured. The wide range is primarily due to a lack of consensus over the population elasticity to rents, so we estimate it by comparing the effects of demand shocks across cities of differing housing supply elasticities. We find that demand changes have similar effects across cities, suggesting a high population elasticity. Therefore, our preferred estimate is that location demand accounts for more than half of the increase. We discuss implications for housing supply policy. 

(Online Appendix) (Replication Data and Code) (DOI)

Abstract: What is the role of migration in regional evolutions? I document that within-U.S. migration causes a reduction in the unemployment rate of the receiving city, over several years. To establish this causal effect, I construct an instrument using outmigration of other places and predict its destination from historical patterns. The decline in unemployment is due to housing. Housing is durable, so increased demand causes a surge of new houses and construction jobs. Additionally, migrants’ housing demand raises prices, increasing borrowing and non-tradable employment. This finding implies the endogenous response of migration amplifies local labor demand shocks by about a third. 

(DOI)

Abstract: Cromwell and Marar [CM] present an analysis of semi-regular (generic) surfaces with a single triple-point and connected self-intersection set. Six of their surfaces are the projective plane, including Boy’s surface and Steiner’s surface. We build on their work by incorporating twists similar to that of Apery’s immersion of the projective plane and show that with a few additional surfaces, all such generic maps of the projective plane are now identified. 

(International Finance Discussion Paper #1037)

Abstract: This paper studies the behavior of recoveries from recessions across 59 advanced and emerging market economies over the past 40 years. Focusing specifically on the performance of output after the recession trough, we find little or no difference in the pace of output growth across types of recessions. In particular, banking and financial crisis do not affect the strength of the economic rebound, although these recessions are more severe, implying a sizable output loss. However, recovery does change with some characteristics of recession. Recoveries tend to be faster following deeper recessions, especially in emerging markets, and tend to be slower following long recessions. Most recessions are associated with a slowing, if not outright decline in house prices, but recessions with large declines in house prices also tend to have slower recoveries. Long recessions and those associated with poor housing-market outcomes can lead to sustained output losses relative to pre-crisis trends. Consistent with microeconomic studies showing permanent income loss to job-losing workers during recessions, we find that the sustained deviation in output from trend is associated with a reduction in labor input, especially linked to declines in employment and labor-force participation following recessions. On net, our results imply that the output/employment gap following a severe, long recessions is considerably smaller than is typically assumed by standard macro models, which in turn may have substantial implications for macroeconomic policy during recoveries.