- Dartmouth Economic Research Scholars (DERS)
- Fed Challenge
- Jobs, Fellowships and Interships
- Political Economy Project
- Study Abroad
- News & Events
Back to Top Nav
Many undergraduates in the department of economics collaborate with faculty on independent research projects and as research assistants. Here are abstracts of research papers completed by students as honors theses.
This study evaluates how increased access to affordable Long-Acting Reversible Contraceptives (LARCs) impacts high school graduation rates. I utilize the 2009 Colorado Family Planning Initiative (CFPI) as a setting through which to study the relationship. Building on previous studies of CFPI that found a significant decline in teen pregnancy rates in zip codes within seven miles of a Title X clinic, this research measures the increase in graduation rates at high schools near Title X clinics. My results show that the female graduation rates increased by 2.16 percentage points more at schools near Title X clinics than elsewhere in the state following the introduction of CFPI, which were driven by gains in Hispanic and Black female graduation rates. Findings suggest that access to affordable reliable birth control improves a girl's ability to invest in her education.
This paper applies the gravity model framework to nearly seventy years of data to assess whether gender dynamics between de facto world leaders can affect what is commonly considered to be a non-gendered policy issue: trade costs and flows. We find that in the presence of two heads of state of the same gender, bilateral trade flows increase in a significant way on average; however, the magnitude of these effects are sensitive to a bevy of political, geographic, and economic factors. Based on these factors, we find that same gender plays a greater role in bilateral trade dynamics when the destination country is politically violent or has a history of previous conflict. Meanwhile, it has no effect with respect to trade between two democratic countries with high economic mobility and strong regional alliances. These trends indicate that one leader's willingness to cooperate with another purely on the basis of their gender may be deeply rooted in the sociopolitical environments of both their country and their trade partners.
We use 15 years of monthly data from the S&P 1500 trading index to examine the relationship between board composition and firm performance. Our results indicate that having a dual CEO/Chair decreases firm performance during normal market conditions but can protect shareholder value during periods of market turmoil. We document that the practice of CEO/Chair duality is decreasing within the S&P 1500 index over our sample period. These results are consistent with competing theories on CEO/Chair duality and suggest that the impacts of CEO/Chair duality on firm performance may be dependent on market conditions.
I gather two decades of panel data from the S&P 1500 to analyze board diversity and hiring practices. I note that the proportion of female and nonwhite board members has increased substantially over the sample time period. I use a logit model to evaluate the impact of board composition on diverse hiring tendencies within a company. I estimate effects for women, Blacks, Hispanics, and Asians. Overall, the results indicate that when a company's board becomes more diverse, they are less likely to hire an additional diverse director. Additionally, diverse board members are hired with greater likelihood following the separation of a diverse director. The results are consistent with the view that firms use diversity as a signal to make themselves more attractive to investors and new talent and stop hiring once this demand is met. I find limited heterogeneous effects across sectors and inconclusive evidence that California's gender quotas mitigate tokenization.
I estimate the impact of time spent in virtual and hybrid learning modes during the 2020-2021 school year on performance outcomes among third through eighth grade students in Massachusetts public schools. Specifically, I combine district-level learning mode data during the 2020-2021 school year with spring 2017-2019 and 2021 performance outcomes on the Massachusetts Comprehensive Assessment System (MCAS) and Assessing Comprehension and Communication in English State-to-State for English Language Learners (ACCESS for ELLs) assessments. I find that larger schools and those serving higher shares of economically disadvantaged and Hispanic students operated in fully virtual modes for a larger share of the 2020-2021 school year. Additionally, I find that English Language Learners (ELLs) faced more pronounced performance declines on the 2021 MCAS exam compared to non-ELLs. Among both ELL and non-ELL students, these declines were larger among younger students (in third to fifth grade) and more severe in math relative to ELA. Learning losses were concentrated among students exposed to more virtual schooling throughout the 2020-2021 school year, however ELLs were not differentially impacted by virtual learning itself. I estimate that virtual schooling is responsible for a decline in average MCAS scores of 3.457 points (0.15 standard deviations) and 2.398 points (0.10 standard deviations) on the math and ELA sections, respectively, for the average student that was exposed to virtual schooling for 42 percent of the school year. Among ELLs, my findings suggest that virtual schooling is responsible for a 15.54 percentage point decline in the share of students in third to fifth ·grade making progress in English language proficiency in the average district. My findings indicate that teachers, administrators, and policy makers should allocate significant time and resources to those students most exposed to virtual learning during the pandemic to prevent these learning losses from persisting and compounding over time.
This paper examines corporate scandals of both a financial and non-financial nature between 2005 and 2020. I propose a novel date estimation technique to isolate the date the scandal actually occurred to regressed Cumulative Abnormal Returns (CARs) on that scandal's characteristics. My findings suggest that certain scandal characteristics, such as company industry, scandal type and prevalence of media focus have specific and varying effects on CARs. Previous literature has shown that scandal announcements generally lead to significantly negative CARs, but no research has been done analyzing the correlation between CARs prior to announcement date and scandal characteristics. Generally, I've shown it is quite difficult to predict future returns, though there are some that glean important information for investors.
This paper analyzes the effect of rising U.S. exposure to (1) imports from China on county-level mortality rates from 1995 to 2014 and (2) imports from all countries on health outcomes for individuals over time from 1972 to 2007. Rising import exposure causes higher rates of mortality from all causes and deaths of despair in U.S. counties. Import exposure in a household's state of residence significantly increases the likelihood of an individual developing a work disability but does not significantly affect poor health status or missed work time due to sickness.This effect is of the same magnitude for both sexes but slightly larger for males.Import exposure in a household's main industry has mostly small and insignificant impacts but significantly increases the likelihood of a female developing a work disability.
Recent work finds school integration increased per-pupil spending and educational attainment for black students. We study whether school resegregation has the opposite effect. With new school-level data, we document substantial variation in per-pupil expenditures inside school districts. We exploit quasi-random variation in the timing of the lifting of federal school desegregation orders to identify the effect of order dismissals on resource disparities within school districts. The lifting of desegregation orders is associated with a modest rise in segregation and a small ($137) but insignificant increase in the black-white gap in per-pupil expenditures.
Understanding the casual relationship between national income and population health has significant implications on the allocation of resources for the optimization of overall wellbeing. The past two centuries have seen enormous changes to both national income and population health. Since the 1940s, the poorest countries have made remarkable improvements in population health, measured in life expectancy. At the same time, growth economists have found little evidence of a catch up on the economic front. This paper uses a bivariate Markov transition model to analyze the nonlinear growth patterns of population health and national income as they relate to each other from 1970 to 2019. Key findings of this paper hint at the causal relationship between improvements in population health and national income growth. However, this relationship is not found to move in the opposite direction.
The debate on nurse practitioner scope of practice (NP SOP) expansions lack evidence on labor market implications. Leveraging cross-state differences in degrees of SOP independence in 2000-2019, I analyze time-variant effects of NP SOP changes on NP labor market outcomes and supply. My results indicate temporary increases in NP work hours as well as graduate nursing program completions 2-4 years after a state grants fully independent SOP. Furthermore, expansion of prescriptive authority is associated with an increase in the total number of NPs in states with lower access to the out-of-state health care labor market. Overall, I find that SOP changes only impact the NP labor market in the short-run. I also conclude that the NP labor force has a high wage elasticity as well as high responsiveness to demand shifts. Thus, targeting long-term labor demand should be the critical goal for NP-related policymaking.
Whether or not to complete the acquisition of a company is one of the most important decisions any corporation must make. Using data from CRSP / COMPUSTAT, this paper analyzes whether these transactions generate value. This paper analyzes these M&A transactions from 1998 to 2022 and conducts both an event study and a difference-in- difference analysis with comparable firms that have not engaged in a transaction. Through these analyses, this paper aims to examine whether aspects like revenue synergies or cost synergies are realized from these deals, along with how the market reacts to transactions.
This paper evaluates the impact of fluctuations in the price of gold on enrollment and household expenditure in Tanzania. Using a difference-in-differences strategy that compares households that vary in their proximity to a gold mine, I find that a sixty-two percent increase in the price of gold leads to an 18.6 percentage point decline in enrollment among individuals fourteen to seventeen years old living within thirty kilometers of an artisanal gold mine. I also find that wealth increases for households living within thirty kilometers of a gold mine.
Ski racing is a complex sport to study as it is among the most "elitist" and factors such as gender, age, ability, socioeconomic background, injuries, and home location all play a vital role in an individual's ability to persist in the sport. By working with an original data set which contains every US ski race over the past 23 years I am able to study the exogenous, individual and peer effects with respect to one's success and persistence in the sport. Through my research I was able to make the following findings: 1. Peers are extremely important to one's ability to persist. "Superstar" peers who compete at a high level have a negative impact on non-superstars ability to persist in the sport, while "superstar" peers who are an age category ahead have a positive impact on non-superstars ability to persist. 2. Although top placement in younger ages suggests future high-level performance, it does not guarantee persistence. Marginal placement of 4th and 5th are stronger indicators of persistence compared to podium finishes. 3. Trends are similar across both genders.
Using data from the Gallup U.S. Poll, this paper estimates the effects of aggregate unemployment on the subjective well-being of U.S. residents. First, this paper finds that a rise in the state-level unemployment rate negatively affects subjective well-being, even after controlling for state and time fixed effects and personal characteristics, such as employment status. This suggests that a rise in state-level unemployment has negative spillover effects on all residents of that state and not just those who lose their job. Second, this paper finds that the presence of strong welfare, measured by the base generosity of state unemployment insurance benefits, does not impact how individuals react to changes in the unemployment rate. In other words, statutory generosity does not appear to protect the subjective well-being of individuals during economic downturns. Third, this paper finds that the perception of strong welfare, proxied by the adoption and implementation of Medicaid expansion, has a statistically significant impact on the way in which individuals react to changes in the unemployment rate. Specifically, individuals living in states that are perceived to have strong welfare experience a smaller loss of positive emotions and a smaller gain of negative emotions during recessions.