Zachary Finn, PhD Candidate in Applied Economics
Northeastern University
​
Research Papers
The Affordable Care Act’s (ACA) Medicaid Expansion has provided millions of low-income Americans with access to affordable health insurance. However, many Americans both with and without health insurance struggle financially when faced with an adverse health event. These individuals often turn to their local communities through online crowdfunding, as a means to help reduce the financial burden of hospital and other health care related debt. This paper examines the impact of North Carolina’s Medicaid Expansion on online crowdfunding activity, testing whether any individuals no longer needed to ask for financial assistance online due to having Medicaid Insurance, and whether any crowding out of private donations to online medical fundraisers due to the public spending associated with Medicaid Expansion occurred. To answer these questions, I use data from GoFundMe on medical fundraisers to compare North Carolina and states that, as of 2024, have not expanded their state's Medicaid programs; I use data on just under 60,000 medical campaigns over the period July 2023 to April 2024, with North Carolina's Medicaid Expansion occurring at the midpoint of that time period, December 1st, 2023. As part of the project, I implement text classification using a Large Language Model, which allows me to understand each medical campaign at a more granular level than other GoFundMe studies, adding specific medical condition categories and the reasons why individuals ask for specific funds to my analysis.
Impact of Medicaid Expansion on Incidence of Medical Tax Deductions
This paper assesses the impact of Medicaid Expansion on the incidence of medical tax deductions. Providing individuals with health insurance should alleviate some financial burden imposed by medical costs associated with a health related event, but it is unclear how many individuals would forego claiming a medical tax deduction because of financial relief provided from insurance, given the relatively high threshold of debt needed to claim a medical deduction. Utilizing differences in state expansion timing, I conduct an event study that uses IRS zip-code level data on medical tax deductions to see whether these types of deductions went down in states after Medicaid Expansion occurred.
A systematic literature review of machine learning to predict
opioid use disorder: An alignment problem in AI for healthcare
There is a growing body of work on developing machine learning models intended to identify patients at risk of opioid misuse, disorder, and/or overdose. Such models are ostensibly intended to inform decision-making around opioid prescription (and some have already been deployed to this end in clinical settings), in which case they should be optimized to predict the casual effect of prescribed opioid therapy with respect to risk of developing opioid use disorders. We conduct a systematic review of the literature on machine learning models for opioid risk prediction, with a focus on how such models are trained and evaluated, and how well this aligns with their envisioned use.
Impact of COVID-19 on Childcare and Gender Equity
This paper examines the impact of the COVID-19 pandemic on labor outcomes, specifically whether there were differential impacts on Females due to childcare responsibilities. Utilizing a survey of over 1500 working parents conducted during the onset of COVID in 2020, we explore how working parents coped with a childcare shock due to the onset of the pandemic, whether their work status changed, and why, as well as what individual and employer resources may have helped alleviate the impact of childcare on work status. We then ask if being more susceptible to a childcare shock has a differential impact on indicators of mental health and job satisfaction
Population Aging, Labor Demand, and the Structure of Wages
One consequence of demographic change is substantial shifts in the age distribution of the working age population. As the baby boom generation ages, the usual historical pattern of there being a high ratio of younger workers relative to older workers has been replaced by a pattern of there being roughly equal percentages of workers of different ages. One might expect that the increasing relative supply of older workers would lower the wage premium paid for older, more experienced workers. This paper provides strong empirical support for this hypothesis. Using U.S. Census data, econometric estimates imply that the size of one’s birth cohort affects wages throughout one’s working life, with members of relatively large cohorts (at all stages of their careers) earning a significantly lower wage than members of smaller cohorts.