Home » Advisor » Michelle Connolly

Category Archives: Michelle Connolly

Heterogeneity in Mortgage Refinancing

By Julia Wu

Abstract
Many households who would benefit from and are eligible to refinance their mortgages fail to do so. A recent literature has demonstrated a significant degree of heterogeneity in the propensity to refinance across various dimensions, yet much heterogeneity is left unexplained. In this paper, I use a clustering regression to characterize heterogeneity in mortgage refinancing by estimating the distribution of propensities to refinance. A key novelty to my approach is that I do so without relying on borrower characteristics, allowing me to recover the full degree of heterogeneity, rather than simply the extent to which the propensity to refinance varies with a given observable. I then explore the role of both observed and unobserved heterogeneity in group placement by regressing group estimates on a set of demographic characteristics. As a complement to my analysis, I provide evidence from a novel dataset of detailed information on borrower perspectives on mortgage refinancing to paint a more nuanced picture of how household characteristics and behavioral mechanisms play into the decision to refinance. I find a significant degree of heterogeneity in both the average and marginal propensity to refinance across households. While observables such as education, race and income do significantly correlate with group heterogeneity, it is clear that much heterogeneity may still be attributed to the presence of unobservable characteristics.

David Berger, Faculty Advisor
Michelle Connolly, Faculty Advisor

JEL codes: D9, E52, G21

View thesis

The Effect of Workforce Participation and Household Income Contribution on Women’s Healthcare Empowerment in Rural Bangladesh

By Hannah Wang

Abstract
Women in Bangladesh have gained increased access to paid work in the past decade yet
still experience limited choices and access to resources, which threatens their ability to exercise
control over healthcare for themselves and their children. Several collective household
bargaining theories hypothesize a link between women’s workforce participation and
empowerment. This paper uses a cross-sectional approach and survey data collected at the end of
a randomized trial field experiment in rural Bangladesh from 2007 to 2017 to examine health
empowerment outcomes for 7,151 young women ages 14 to 32. The results show that women
who work for income are expected to be more health empowered, specifically due to an
increased ability to make their own health decisions. As a woman contributes more income to her
household, her health empowerment is expected to increase, through increased abilities to make
her own health decisions, purchase medicine for herself, and seek medical treatment
independently. Greater mobility and stronger female-positive attitudes towards gender norms are
potential mechanisms through which paid work and household income contribution can translate
into health empowerment. Furthermore, higher total household income, having children, and
being more educated than her husband are expected to increase a woman’s health empowerment.
These results are significant while controlling for the effects of various individual and household
characteristics.

Professor Erica M. Field, Faculty Advisor
Professor Michelle P. Connolly, Faculty Advisor

JEL classification: J1; J16; I15

View Thesis

Short and Long-Term Impacts of a Large-Scale Natural Disaster on Individual Labor Outcomes: Evidence from the 2004 Indian Ocean Tsunami

By Tony sun

Abstract
Natural disasters are often highly disruptive to the livelihoods of impacted populations. This paper investigates the effects of the 2004 Indian Ocean tsunami on male wages and labor supply from its immediate aftermath into the long run. Using fixed effects models that account for individual-specific heterogeneity, I find evidence of significant real wage declines for workers from heavily damaged areas that persist beyond the short-term. This long-term wage effect contrasts with previous literature, particularly in the context of relatively less severe disasters. Male workers also increased their hours-of-work following the tsunami, which suggests reliance on labor markets to smooth income losses and shifted their labor towards less disrupted industries. Additionally, I document the heterogeneity of tsunami impact on wages and hours-of-work by birth cohort and education, as well as by industry and sector of employment.

Professor Duncan Thomas, Faculty Advisor
Professor Michelle P. Connolly, Honors Seminar Instructor

JEL classification: J2; J21; J30; O10; Q54

View Thesis

Peer Effects & Differential Attrition: Evidence from Tennessee’s Project STAR

By Sanjay Satish

Abstract
This paper explores the effects of attrition on student development in early education.
It aims to provide evidence that student departure in elementary schools has educational
impacts on the students they leave behind. Utilizing data from Tennessee’s Project STAR
experiment, this paper aims to expand upon the literature of peer effects, as well as attrition,
in public elementary schools. It departs from previous papers by utilizing survival analysis to
determine which characteristics of students prolonged participation in the experiment. Clustering
analysis is subsequently employed to group departed students to better understand
the various channels of attrition present in STAR. It finds that students who left Project
STAR were more likely to be of lower income and lower ability than their peers. This paper
then uses these findings to estimate the peer effects of attrition on students who remained
in the experiment and undertakes a discussion of potential sources of bias in this estimation
and their effects on the explanatory power of peer effects estimates.

Professor Robert Garlick, Faculty Advisor
Professor Michelle Connolly, Faculty Advisor

JEL Classification: I, I21, I26, H4, J13

View Thesis

The Case for Clemency: Differential Impacts of Pretrial Detention on Case and Crime Outcomes

By George Rateb

Abstract
About half-million of individuals in US jails are detained pretrial while legally presumed
innocent. Using data on quasi-randomly assigned bail judges in the third-largest court system in
the U.S., we study the impact of pretrial detention on defendants’ court and crime outcomes
between 2008 and 2012. We supplement our primary analysis to document patterns on bail
amounts and how they differentially impact Black defendants relative to their white and Hispanic
counterparts. Instrumental variable estimates suggest that pretrial detention increases the
likelihood of being found guilty, mainly driven by the uptake of guilty pleas, especially for
minorities. By linking court and jail data, we provide mechanistic evidence that jail time is
positively correlated with the uptake of these guilty pleas. To the best of our knowledge, these
findings have not been empirically documented due to a lack of previous data availability.

Professor Bocar Ba, Faculty Advisor
Professor Michelle Connolly, Faculty Advisor

JEL classification: C26; J15; K14

View Thesis

A perfect storm: The effect of natural disasters on child health

By Cheyenne Danielle Quijano

Abstract
Typhoons and their accompanying flooding have destructive effects, including an
increase in the risk of waterborne disease in children. Using a spatial regression discontinuity
design, I explore the immediate to short-term effects of flooding as a result
of Typhoon Labuyo on the incidence of diarrhea and acute respiratory infection in the
Philippines by comparing children living in a flooded barangay (town) to children living
just outside of the flooded area. I build on the existing literature by accounting for
both incidence and intensity of the typhoon’s flooding in my model. I construct this
new flooding measure using programming techniques and ArcGIS by manipulating data
collected by the University of Maryland’s Global Flood Monitoring System. This data
as well as health data from the 2013 Philippines National Demographic Health Surveys
were collected the day after Typhoon Labuyo left the Philippines, providing a unique
opportunity to explore the immediate impact of the typhoon on child health. Most of
my results are insignificant, but subgroup analyses show that the effect of flooding on
waterborne disease incidence is less impactful in the immediate term following a flood
and more impactful in the medium-term. This is important, because understanding
the detrimental health effects of flooding is of utmost importance, especially because
climate change will only increase the frequency and intensity of natural disasters.

Professor Erica M. Field, Faculty Advisor
Professor Michelle P. Connolly, Faculty Advisor

JEL classification: I150, O120, O130, Q540

View thesis

Tale of Two Cities An Econometric Analysis of East & West Coast Fine Art Galleries

By Daniella Victoria Paretti

Abstract
In a 2021 report published alongside Art Basel and UBS, renowned cultural economist Dr.
Clare McAndrew posited that the value of art sales in 2020 amounted to an impressive $50 billion
(although this actually marks an over 10-year low). It is no secret that the global art markets are
extremely lucrative, attracting the interest of industry magnates and business tycoons alike.
Though it is important to note that art markets are historically quite distinct from their normal good
counterparts — the sector is laden with issues regarding transparency, high barriers to entry, and
hiding of wealth. Amidst the COVID-19 pandemic, however, the tides began to turn; online
platforms for museums, auction houses, and galleries were employed more than ever before,
effectively modernizing the antiquated industry and expanding its reach to new consumers. How
has this trend of digitalization changed and improved art markets? More specifically, how can data
analytics and other technological resources serve the interests of private galleries? Using sales data
from a parent gallery with multiple locations across the United States (each displaying similar
works/artists), I have conducted a number of qualitative and statistical analyses to identify key
differences between the West and East coast locations. In short, the gallery on the West coast sold
more works and at a lower average cost than its counterpart, providing key insights into this local
market’s consumer base. Beyond this, factors like size, medium, and artist gender were found to
have statistically significant effects on the ultimate sale price and turnover rate of works. My
findings suggest that means of data analytics should be utilized by all actors in the art markets to
optimize their approach to business, as well as understand their consumers better than ever before.

Professor Michelle Connolly, Faculty Advisor
Professor Hans Van Miegroet, Faculty Advisor

JEL classification: Z11, C10, J11, O33

View Thesis

Withdrawal: The Difficulty of Transitioning to a Cashless Economy

By Praneeth Kandula

Abstract
In 2021, modern payment methods such as mobile pay have increased nearly fivefold since their introduction in 2015. This shift to an increasingly cashless, digital economy has been marked by inequitable financial and technological divides. Historically, Black and Latino adults have had less access to financial systems and are less likely to own traditional computers and home broadband. Without rectifying these issues, a cashless, digital economy only serves to widen divides. Using data from the Diary of Consumer Payment, this study descriptively examines the use of cash and alternative payment methods by different racial and ethnic groups from 2015 through 2020. I also extend this effort to address the effects of COVID-19. I find that racial differences not only exist but also the gap between Black and Latino adults and White adults grows between 2015 and 2019. Still, this paper finds that in 2020 the likelihood to employ cash for a transaction falls for Black adults but not for Latino adults. COVID-19 has been a critical driver of change, forcing both consumers and corporations to shift to a more digital-centric economy. While there have been positive shifts for Black adults, policy ensuring that all racial groups have access to the necessary financial and digital networks will be critical in establishing an equitable economy moving forward.

Professor Lisa A. Gennetian, Faculty Advisor
Professor Michelle P. Connolly, Faculty Advisor

JEL Classification: D1 D31 G20 I24 J11

View Thesis

Financial Inclusion and Women’s Economic Empowerment in India

By Nehal Jain

Abstract
On August 14th, 2014 India’s Prime Minister Narendra Modi implemented the largest ever
financial inclusion scheme to date known as Pradhan Mantri Jan Dhan Yojana (PMJDY). The
program aimed to bank all of India’s unbanked population. Prior to the program, India had one of
the highest rates of unbanked citizens. The program also included measures that prioritized women’s
access to these financial institutions given the gender gap in financial inclusivity. This paper aims
both to understand the effectiveness of PMJDY on granting women equal access as men to financial
institutions and whether financial inclusion results in increased economic empowerment, I find that
PMJDY was successful in increasing access to bank accounts and separately, that access to bank
accounts economically empowers women.

Pengpeng Xiao, Faculty Advisor
Michelle Connolly, Faculty Advisor

JEL classification: J1; G28; I31

View thesis

Does Responsiveness to Mortality Risk Vary by Age? Evidence from Pandemic Health Outcomes and Movement Patterns

By Ryan Jones Hastings

Abstract
When choosing whether to visit venues like stores and restaurants during the
COVID-19 pandemic, individuals faced trade-offs between movement and mortality
risk. This paper analyzes age-specific responsiveness to infection-related mortality
risk in the Philadelphia metropolitan area from March through December 2020.
First, we develop a theoretical model that characterizes potential sources of
heterogeneity in the decisions of individuals choosing how much to move. Next,
we use data on the health outcomes of COVID-19 patients to estimate fatality
rates for different demographic groups. Finally, we use a panel of cell phone data
tracking visits to venues before and during the pandemic along with a revealed
preference approach to estimate an empirical model that relates age to movement
decisions. Our results suggest that older people’s movements are less sensitive
to mortality risk. Under weak assumptions, this implies that older people have
a lower willingness to pay for marginal reductions in the probability of death.
This finding has implications for the cost-benefit analysis of policies that mitigate
adverse health outcomes, such as pandemic movement restrictions and pollution
remediation, and for the value of statistical life (VSL) literature more broadly.

Professor Christopher Timmins, Faculty Advisor
Professor Michelle Connolly, Faculty Advisor

JEL classification: D81; I12; J17; R2.

View thesis

Questions?

Undergraduate Program Assistant
Matthew Eggleston
dus_asst@econ.duke.edu

Director of the Honors Program
Michelle P. Connolly
michelle.connolly@duke.edu