This dissertation consists of three chapters. The first two chapters investigate how individuals’ perceptions and beliefs shape their economic behavior and their attitudes towards public policies, using large-scale surveys and experiments. In the first chapter, I study how beliefs about search costs, returns to search effort, and outside options relate to the job mobility decisions of employed workers. In the second chapter—coauthored with Alberto Alesina and Stefanie Stantcheva—we investigate how people perceive immigrants and how these perceptions influence their support for redistribution. In the third chapter—coauthored with Alberto Alesina, Gualtiero Azzalini, Carlo Favero, and Francesco Giavazzi—we study the effects on output of fiscal adjustments as a function of the composition of the adjustment—whether the adjustment is mostly based on spending cuts or on tax hikes—and of the state of the business cycle when the adjustment is implemented.

The one person–one vote principle for political redistricting requires balancing populations across districts. We address the matter of simultaneously balancing a second attribute across districts, proving that this is always possible to within reasonable tolerances. Feasibility is demonstrated by formulating the problem as a constrained partitioning problem on graphs. The resulting computational results demonstrate the practicality of obtaining dual-balanced districts whose balance for both attributes is well within reasonable deviations from the ideal values. Applications include attempts to avoid differential population growth leading to malapportionment between decennial census counts or simultaneously balancing total and voting-age populations.

We study the macroeconomic effects of unemployment insurance (UI) benefit extensions in the United States at short and long durations. To do this, we develop a new state level dataset on trigger variables for UI extensions and a “UI benefit calculator” based on detailed legislative and administrative sources spanning five decades. Our identification approach exploits variation across states in the options governing the Extended Benefits program. We find that UI extensions during time periods when UI benefit durations are already long—such as in the Great Recession—have minimal effects. However, UI extensions when initial durations are shorter have substantial effects on the unemployment rate and the number of people receiving UI. Through the lens of a search-and-matching model, we show that our estimates are consistent with microeconomic estimates of the duration elasticity to UI, implying small general equilibrium effects of UI extensions.

Since the Taliban returned to power in Afghanistan in August of 2021, nearly 90,000 Afghans who had fled their country have been resettled in the United States, constituting one of the largest groups of refugee arrivals in the U.S. in recent history. Working from a database I created from the administrative records of a non-profit refugee aid group, I use data and spatial analysis to examine the demographics of Afghans resettled in the DC metro area, the characteristics of the census tracts and counties in which they have been resettled, and their access to public transportation. I find that the resettled Afghan population skews significantly towards youths under the age of 18; that Afghans are typically resettled in the areas within the region where residents are mostly likely to be in poverty, non-citizens, and/or racial minorities; and that while many Afghans are resettled near bus routes, few have convenient access to metro trains. These findings suggest that, despite the sacrifices many Afghans made for U.S. policy objectives abroad, their resettlement patterns mirror those of other refugees and asylees in the United States: because of a widespread reticence to invest in refugee integration programs, these people are likely to encounter structural barriers to finding socio-economic stability and advancement.

Ongoing reliance on state-level, race-neutral job market measures provide an incomplete picture of the health of Georgia’s workforce. On the surface, Georgia has maintained a robust job market with a low overall unemployment rate of 3.3%. This low unemployment rate is partially attributed to a hiring pace that has stayed ahead of layoffs despite federal actions to slow the economy as part of efforts to fight rapid inflation. Beneath that surface, however, disaggregated data shows that an outsized share of Black, Brown and other Georgians with low incomes were harmed by inflation-fighting efforts. After reaching landmark employment levels in mid-2022, Black workers experienced 60% more unemployment spells than white workers. Since summer 2023, among workers in their prime working ages of 25 to 54, Black and Hispanic workers have experienced eroding employment levels, while white workers have continued to see an uptick in employment, surpassing their pre-pandemic peak. Furthermore, Georgia’s economic growth has not meaningfully lifted those earning low incomes, who saw some of the largest pay growth (21%) from 2019 to 2022 but whose buying power only rose by 4% over the same period due to inflation. To address these gaps, Georgia lawmakers must not only view the state’s job market through a racial equity lens but utilize this data as part of a worker-centered change of course to address the many barriers that continue to leave workers behind.

Economists have long recognized the importance of human capital in economic growth. In this series of papers, I study how various institutions of human capital formation affected economic development in a variety of contexts prior to 1900. These include both formal institutions of human capital formation such as universities and informal institutions such as apprenticeships. The first chapter examines the role of European universities during the late Middle Ages and early modern eras, starting with the creation of the University of Bologna in the late 12th century. These “formal” institutions of human capital focused on the teaching of law and supplied the civil service and justice systems of their home cities with trained legists. I find that cities which hosted universities had faster population growth than similar cities without universities, suggesting that universities were associated with higher levels of economic development. ix The second chapter contrasts the formal university with an informal human capital institution—the English apprenticeship. Though England was the cradle of Europe’s industrialization, it had far fewer universities per capita than peer nations on the continent. However, it had a well-developed system of apprenticeship which some scholars have theorized was the main means of human capital incubation and transmission during the Industrial Revolution. English cities with more apprenticeships during the 18th century tended to have a greater degree of labor force specialization in the mid-19th century. The final chapter examines the role of the land-grant college system, which was established in the mid-19th century in the United States. In addition to their primary role of providing postsecondary education to students, these institutions added a second channel for the creation of human capital: research to boost the productivity of the nation’s farms. American counties more firmly embedded within the land-grant college network tended to have a better-educated populace and a more specialized labor force; however, counties more exposed to land-grant colleges saw no improvements in agricultural productivity

Early research hypothesized impacts of COVID-19 on agricultural workers, food supply, and rural health systems based on population characteristics from data collected preceding the pandemic. Trends confirmed a vulnerable workforce and limits to field sanitation, housing quality, and healthcare. Less is known about eventual, realized impacts. This article uses the Current Population Survey’s COVID-19 monthly core variables from May 2020 through September 2022 to document actual impacts. Summary statistics and statistical models for the probability of being unable to work reveal that 6 to 8% of agricultural workers were unable to work early in the pandemic and that impacts were disproportionately negative for Hispanics and those with children. An implication is that targeted policies based on vulnerabilities may minimize disparate impacts of a public health shock. Understanding the full impacts of COVID-19 on essential labor remains important for economics, public policy, and food systems in addition to public health.

Right-to-work (RTW) laws are one of the most prominent and politically contentious economic policy issues in the U.S. today. RTW laws are enacted by state governments to prohibit what are called “union security clauses” in collective bargaining agreements. These union-supported clauses can require workers at unionized firms to join the union or pay agency fees to the union as a condition of employment. In practice, when a state has RTW protections, workers at unionized firms cannot be required to pay fees to the union in order to retain their job. Both sides of the political divide make emphatic claims about the impacts of RTW laws, and about unions more broadly. Despite many historical studies, there is little consensus on the impact of RTW laws on local economies. RTW proponents, such as business lobbies, argue that RTW laws make states more attractive for investment, and they typically point to faster growth in income and total employment in RTW states over recent decades as supporting evidence. For example, states that were RTW in 1977 experienced aggregate employment and population growth during 1978–2017 of 105% and 90%, respectively, compared with 49% and 35% among non-RTW states.1 Conversely, RTW opponents, including unions and union-supported think tanks, argue that stronger unions benefit both unionized and nonunion workers and that RTW laws weaken unions and thus undermine their ability to produce these benefits. They highlight that wages and nonwage compensation levels in RTW states are lower than in non-RTW states.2 Who is right? Alas, interpreting the statistics behind these competing claims is a fraught task. We cannot merely compare average economic outcomes between RTW and non-RTW states because such comparisons gloss over a crucial fact: RTW and non-RTW states are fundamentally different in numerous ways, some observable and measurable, others not. These differences include variations in climate; population demographics, culture, and cost of living; historical differences in economic development and wage levels; and different exposure to technological change and changes in transportation costs. Because of these fundamental differences, there is no good reason to expect that, except for right-to-work laws, RTW and non-RTW states would have equal economic outcomes or experience economic growth at the same rate. My academic research with Benjamin Austin addresses this issue and serves as the basis of this Manhattan Institute brief.3 We use modern econometric techniques to study a narrow geographic sample where RTW and non-RTW areas are plausibly otherwise similar. In this case, we compare counties that are in different states (one state is RTW, the other is not) but border each other. Then, we estimate the long-term impact of RTW laws, and whether those laws cause the observable economic differences between RTW and non-RTW areas. Our method allows us to measure the impacts of RTW laws on a wide range of economic outcomes. In addition to revisiting debates about the impact of RTW laws on manufacturing employment, overall employment, and wage outcomes for workers, we study how RTW affects migration and where people choose to live. We leverage more recently available data to study the impact of RTW laws on downstream socioeconomic outcomes, including poverty rates and intergenerational mobility. We present evidence about which workers are affected. In this issue brief, I provide a nontechnical overview of this academic work and discuss its policy implications. Our results suggest that RTW laws produce substantial economic benefits, which accrue to workers through stronger local labor markets, and more generally through improved socioeconomic outcomes, especially for children and families that are more economically at risk. Specifically (and consistent with previous research), we find that RTW laws sharply raise a state’s manufacturing share of employment by approximately 28%, or 3.23 percentage points. This increase in manufacturing is not from crowding out other industries; rather, it results in stronger local labor markets in general. Residents of the RTW border counties have a 1.58-percentage-point higher employment rate and a 0.39-percentage-point lower unemployment rate than non-RTW border counties, and more people commute into (than out of) RTW counties for work. Contrary to claims by unions and their advocates, we fail to find any evidence that RTW laws reduce wages—in fact, our results suggest that RTW laws slightly increase per-hour wages, especially for low-income workers. Consistent with this pattern of stronger labor-market outcomes for individual workers, we observe that the RTW sample experienced 19.1 percentage points extra population growth between 1940 and 2010, indicative of stronger labor markets driving greater net in-migration. We find that RTW laws lead to improved social outcomes: childhood poverty rates are lower by 2.29 percentage points in RTW border counties, and there is substantially higher upward income mobility for people who grew up in the bottom half of the socioeconomic distribution.

This work is in three chapters. The first chapter investigates the extent to which county-level property tax changes are capitalized into house prices. The literature mainly focuses on specific tax events, but this chapter generalizes the context to large U.S. cities. The main findings suggest that county-level property tax capitalization occurs and varies along the distribution of house prices. The second chapter measures differences in county-aggregated self-assessed valuations of house prices and countyaggregated sale prices to determine whether there are trends in the level of misperceived house value. Factors like age, time of tenure in a house, and house price all matter when analyzing the difference between self-assessed values and sale prices. Further, the empirical work aims to measure differences in salience for tax policies at different governance levels. The third chapter analyzes whether increases in financial-decision making capabilities has an impact on health insurance purchase decisions and other health-related financial decisions. Financial literacy is shown to reduce undesirable outcomes in these dimensions. These chapters focus on decisions that many households face, and housing and health make up large portions of the typical household budget.

In this paper, we examine the reasons why unemployment insurance (UI) claims have declined so dramatically over the past three decades. The fall in the UI claims rate is concerning because it suggests a reduced countercyclical effectiveness of the UI program. Additionally, weekly initial UI claims are regarded as an important leading indicator of aggregate economic activity, so their meaning has changed. We use a Oaxaca (1973) decomposition approach to identify the main factors for the decline in claims. The procedure suggests what the level of claims would have been later in the period, had values of variables or parameters of the system been at levels observed earlier in the period. Our analysis of stateyear data over the past three decades suggests that the decline in UI claims stems from changes in the industrial and occupational mix of employment interacting with changes in UI program features set by individual states. Employment declines in manufacturing and increases in the health-care and education workforce, along with lower potential UI duration and lower wage replacement rates, contribute to the decline in claims. This decline could be offset by federal rules for states to improve benefit access, replacement rates, and durations. Such changes could improve the relevance of UI to the labor market and help restore UI as meaningful social insurance against job loss and as an automatic stabilizer of the macroeconomy.

Borrowing money is a core instrument of governments to fund goods with high front costs andlong-term benefits. Scholars have, however, primarily associated public debt with shortsighted policies by office-seeking politicians. The three essays in this dissertation investigate the determinants and outcomes of popular preferences for investment-oriented public debt using novel voting, survey, and budgetary data. The first essay asks: Is a community more amenable to borrowing when its time horizon shortens?Existing theories argue that individuals with shorter time horizons, like seniors, have a higher inclination towards borrowing because they overvalue current consumption and discount future costs. I verify this assumption by studying how population aging affects support for debt-funded investments. Using novel data sets on U.S. state and local bond referendums over six decades, I show that, conversely, aging decreases support for debt-funded investments. Contrary to mainstream predictions, an original conjoint survey experiment further demonstrates that seniors do not have a greater preference for policies with longer repayment maturities and shorter benefit periods. Rather, aging lowers support for investments by increasing fiscal conservatism and shifting consumption away from capital-intensive goods. The effect of aging varies depending on which age groups cohabit with seniors. In particular, aging communities experiencing an influx of nonrelative children show greater opposition to new investments, while increased contact with relative children has no detectable effect on their support for investments. These findings suggest that population aging can complicate the construction of political coalitions over investments, particularly in communities with diverse age distributions. The second essay inquires: Do popular preferences affect how governments fund policies? Policy funding is often presented as technical and hardly influenced by voters. I study this assumption by investigating the effect of population aging on U.S. municipal budgets between 1970 and 2017 with the use of data on municipal finances and mayors’ characteristics. In contrast, I find that aging increases appetite for consumption-oriented policies, leading to more short-term budgeting. When a municipality’s population ages, it substitutes current expenditures for capital spending, shortens its debt maturity, and favors liquid revenues over long-term borrowing. Ultimately, this translates into lower indebtedness and higher property tax revenues. In contrast, its expenditure levels and distribution between policies remain stable as seniors’ fiscal conservatism constrains surges in spending, and seniors’ interest in property values limits cuts in municipal amenities. The effects of aging are not uniform, as municipalities ruled by elderly mayors implement debt policies more aligned with seniors’ preferences. These results contradict the dire budgetary predictions associated with aging and show that seniors’ ideological and economic motives can counterbalance their distributional demands. They also illustrate that the preferences of minorities are better represented when they elect politicians who resemble them. The last essay questions: Do voters care about how a policy is funded? Even if citizens can grasp the technicalities of public finances, policy funding may still not matter to them. Indeed, the Ricardian equivalence argues that people are indifferent about whether a policy is funded by debt or taxation because they internalize the future costs of debt repayment in their bequests. Using a novel dataset of 22,000 local referendums and two original conjoint survey experiments, I demonstrate that, conversely, voters prefer financing policies by small tax increases rather than borrowing. My surveys also reveal that respondents discriminate against policies with longer repayment periods. This result contradicts both the Ricardian equivalence’s assumption that people are indifferent to how policies allocate costs over time and the premise that people overlook future borrowing costs. Time preferences are important to explain opinions regarding debt and taxation, as each funding method distributes costs and benefits differently over time. Specifically, people’s resistance to long repayment periods lowers support for debt-funded projects. Variations in preferences between debt and tax remain after accounting for their temporal differences. My analyses indicate that preferences do not vary by policy content or relative to personal financial investments, although conservative individuals display greater support for borrowing than liberals.

Since 2019, the United States has had an increase in violence against Asian American Pacific Islander (AAPI) communities along with an increase of mainstream anti-Asian racist rhetoric. Between 2021 and 2022, The Center for the Study of Hate & Extremism reported an overall 164% increase in anti-Asian hate crimes (Report to the Nation, 2021). While racism against black, Indigenous, people of color (BIPOC) communities has been the topic of an ever-growing body of critical discourse, prior to 2019 few publications had addressed racism and injustice regarding language choices and discourse in the context of anti-Asian rhetoric in the US, specifically regarding the intersection of language and anti- Asian sentiment (Chun & Zalokar, 1992; Palumbo-Liu, 2001). The recent increase in violence against Asian Americans has coincided with the discourse of the COVID-19 pandemic and is grounded in the longstanding history of associating Asian immigrants with “disease and filth” (Li & Nicholson, 2020, p. 4), a discourse historically perpetuated in the US press (Li & Nicholson, 2020). Given the context of increased anti-Asian hate crimes, historically rooted anti-Asian racism and the lack of previous research I found the need to conduct this three-part study that used news article data from the Corona Virus Corpus to analyze metaphor, Theme and Rheme, and transitivity in modern anti-Asian discourse. The first part of this study analyzed metaphors at the intersection of COVID-19 and racism and the role of metaphor in anti-Asian rhetoric. In the second part of this study, I created a sub corpus of 100 news articles written about the March 16, 2021 murder of six Asian women in Atlanta, GA (Davies). I ii examined articles for Theme and Rheme to determine how corporate news media emphasized specific information regarding the shooting. Then I used a transitivity analysis to identify how corporate news media framed the racialized victims vs. the alleged killer through the use of marked lexical and thematic choices. This analysis helped uncover the intersection between metaphor and anti-Asian hate as well as how marked grammatical choices such as nominalization and passivation helped obfuscate the agency of the perpetrator and dehumanize the victims of the March 16 shooting. The study explores 1) how historically, news media has been the main medium of disseminating racism against Asians in the American context, and 2) the importance for linguists to continually investigate historicized racism in corporate media.

This work analyzes two fundamental issue at the intersection of higher education and the labor market - the returns to completing a credential and the effects of working while enrolled on subsequent student outcomes, with a focus on students of community colleges. There is a growing consensus that obtaining a sub-baccalaureate credential leads to increased earnings. However, it remains unclear to what degree these returns are driven by increases in productivity or labor supply. Using detailed UI data from Minnesota, I estimate the labor market returns associated with completing sub-baccalaureate credentials from community colleges. First I show that these credentials increase quarterly earnings, consistent with prior findings. I then decompose the proportion of earnings returns attributed to productivity (wages) and hours worked. More than 60% of the earnings returns to completing an associate degree are due to an increase in hours. This effect is largely driven by part-time workers being pulled into full-time work. Additionally, I find that those working few hours pre-college experience little-to-no wage gains from completing a credential, with nearly all of their earnings return stemming from working more hours post-graduation. I then proceed to analyze the relationship between attending college and gaining work experience and the empirical challenges therein. I find, after employing new methods to correct for endogeneity in the choice to work while enrolled, that term-time employment can have substantial detrimental effects on the probability of graduating, but that conditional on graduating, those who work more while enrolled do so faster. Taken together, the findings of these analyses highlight the tradeoffs students face between earning and learning.

Community-wide wildfire mitigation can effectively protect homes from structure ignition. The Firewise USA program provides a framework for grassroots wildfire preparedness. Here, we examine the 500 Firewise USA sites in California to understand participation and demographic trends. We find important regional differences regarding the influence of underlying fire hazard, fire history, and other Firewise sites on new site formation. Sites in the Bay Area and Sierras respond strongly to fire history and proximity to other Firewise sites, while Northern and Southern California have few Firewise sites despite underlying hazardous conditions and large fire history. Firewise sites are often whiter, older, and more well-educated than California’s median population, potentially leaving out many communities that do not meet this demographic profile but face severe risks from wildfires. These findings offer important insights into the factors motivating communities to pursue wildfire protection, particularly important given recent severe and destructive wildfire seasons.

The relationship between the United States and the refugees it welcomes is full of contradictions: the U.S. has resettled more refugees than any other country, yet the refugee admissions process is byzantine and decentralized and the government has little insight into refugee outcomes. Academic understanding of the refugee experience is similarly fragmented, with scholarship consisting largely of ethnographic research into distinct communities and their practices. Furthermore, refugees are an under-theorized population as the migration processes that affect voluntary economic migrants are distinct from the processes of arrival and incorporation in the refugee community. This dissertation aims to bring refugees into the literature on immigrant incorporation, investigating the socioeconomic outcomes of refugees using contemporary assimilation theories to understand the role that race and place play in refugee integration. It contributes to the body of knowledge by using American Community Survey data on three different-race refugee groups – Bosnians, Laotians, and Somalis – and comparing differences in economic attainment to native-born Whites and same-race natives. This dissertation asks three questions: 1. How does refugee labor force participation differ from native-born Whites and same-race natives? 2. How do the wages of refugees differ from native-born Whites and same- race natives? 3. How does refugee socioeconomic differ from native-born Whites and same-race natives? By controlling for individual and state-level characteristics, this dissertation will probe the extent to which refugees differ from other immigrants and native-born Americans who share similar characteristics.

Throughout US history, mothers played an important role in educating their children. However, this role is obscured because limitations in data and methods have prevented researchers from including women in intergenerational studies. By lever-aging data from Social Security applications, we build a large panel that tracks both men and women over time despite marital name changes. We measure intergenera-tional mobility as the share of variation in child outcomes explained by parental background (R 2), which we decompose into mothers' and fathers' separate contributions. We find that a mother's human capital is a key determinant of her child's outcomes, often surpassing the influence of fathers, especially for daughters and Black children. More generally, maternal human capital is more important for children with limited school access. Incorporating mothers' human capital suggests that, contrary to current evidence, intergenerational mobility increased over the 19 th century.

Prior studies have shown that people of color (POC) in the United States are exposed to higher levels of pollution than non-Hispanic White people. We show that the city of Denver, Colorado, displays similar race- and ethnicity-based air pollution disparities by using a combination of high-resolution satellite data, air pollution modeling, historical demographic information, and areal apportionment techniques. TROPOMI NO2 columns and modeled PM2.5 concentrations from 2019 are higher in communities subject to redlining. We calculated and compared Spearman coefficients for pollutants and race at the census tract level for every city that underwent redlining to contextualize the disparities in Denver. We find that the location of polluting infrastructure leads to higher populations of POC living near point sources, including 40% higher Hispanic and Latino populations. This influences pollution distribution, with annual average PM2.5 surface concentrations of 6.5 μg m-3 in census tracts with 0-5% Hispanic and Latino populations and 7.5 μg m-3 in census tracts with 60-65% Hispanic and Latino populations. Traffic analysis and emission inventory data show that POC are more likely to live near busy highways. Unequal spatial distribution of pollution sources and POC have allowed for pollution disparities to persist despite attempts by the city to rectify them. Finally, we identify the core causes of the pollution disparities to provide direction for remediation.

In this paper, we examine the question of how trade drives educational attainment in a cross-country panel setting. If a country experience a positive demand shock for exports or a negative supply shock for imports that is intensive in educated labor, the demand for education is expected to increase. We motivate the analysis with a theoretical framework that underscores the relevance of factor intensities across sectors, including the skill acquisition sector, in determining how the composition of trade a⁄ects the demand for education. We show that, consistently with our expectations, variations in the educational composition of exports and imports that follow from supply and demand shocks in partner countries a⁄ect educational attainment. We also show that an increase in exports of capital intensive products has a positive impact on the demand for educated labor, presumably through capital-skill complementarity.

A number of recent papers have investigated the impact of hurricanes on economic growth. However, there is limited understanding of the investment component of local growth after hurricanes. Using hand collected and web-scraped statutory property tax rate data in the U.S., I find that local governments respond to hurricane impact by raising tax rates. I find the hike in tax rates is persistent for 3-4 years after hurricane impact. The response is four times larger for major hurricanes compared to minor hurricanes. However, the increase in tax rates is not expected to be large enough to cause significant out-migration after the average hurricane. I supplement these findings with a novel data set of firm facility-level hurricane impact. I find that firms initially decrease investment in the quarter following hurricane impact and increase it in the final quarters of the second year after impact. Taken together, my paper presents a novel set of stylized facts on government and firm mitigation investment response that can be interpreted in light of recent general equilibrium models with disaster risk.

Headline numbers have shown that the US labor market has recovered the jobs lost during the pandemic. Nevertheless, there is significant variation in the recovery across states and counties and across occupations and industries. Using the available data from the monthly Current Population Survey and the Bureau of Labor Statistics’ State and Metro Area Employment, Hours, and Earnings for January 2019 to August 2022, we present the changing patterns in the labor market. We also highlight some possible underlying reasons that are correlated with the varying patterns across groups and space. Finally, we look at the spatial distribution of the employment across states and micro and metropolitan areas. Results are in line with an uneven recovery across areas, while at odds with a narrative based on working arrangements making economic activity more even across space.