UC Berkeley

The impact of a subsidy or transfer depends greatly on the market forces surrounding

the economic actors who receive them. This dissertation contributes to the field of labor eco-

nomics by shedding light on two under-studied, but widely dispersed, government subsidies.

These subsidies are the tax exemptions given to nonprofit firms and the Earned Income Tax

Credit (EITC) issued to low-income households. Although one of these subsidies is given to

a firm and the other to a consumer, both aim to alleviate the low capital burdens for under-

served populations. The first chapter examines the effect of a firm’s nonprofit tax status

on workers’ wages using the universe of quarterly wage records from the state of Florida.

The second chapter investigates market effects for a large durable good during the issuance

months of the EITC. The third chapter presents a theoretical model that improves partial

effects estimation for ordered outcome variables, which are often variables of interest in em-

pirical questions of this sphere. This is particularly the case when using publicly available,

binned data.

In the first chapter I explore the nonprofit earnings penalty. To separate the influ-

ence of demand and supply, I leverage workers who change employers in administrative tax

data. The average nonprofit worker earns 5.5 percent less than the average for-profit worker.

Supply-side factors (worker selection) contribute 80 percent of the nonprofit differential.

The remaining 20 percent is from demand (a nonprofit penalty). Within-worker nonprofit

variation generates several insights about the influence of nonprofits on the labor market.

Nonprofits compress the wage distribution and reduce inequality among earners. Nonprofit

penalties are much more pronounced in classic charities than in “commercial” nonprofits,

which sometimes exhibit nonprofit premia. This study is the first to harness administrative

wage records, rather than survey data, to estimate the nonprofit earnings penalty. This

study is also the first to show serious misreporting issues in survey data which could bias

results.

The second chapter of my dissertation uses Texas DMV vehicle registration records to

precisely estimate quantity and price effects for used car sales in the weeks following the

issuance of the EITC. I use a difference in difference framework taking advantage of the act that most EITC returns are issued in February and March, and that the share of the

population receiving the EITC varies greatly across zipcodes. I find that for a zipcode with

the average EITC population share (20 percent), used car sales increase by 33.3 percent. I

find little movement in prices. I conclude that the timing of the EITC return does not result

in a large loss of incidence for the consumer, despite the large increase in demand for used

cars. This is the first study to examine price effects, in addition to consumption effects, of

the EITC.

My first two chapters address wage and price effects of subsidies using empirical appli-

cations. My third chapter presents a theoretical advancement in methods used to recover

partial effects that are of interest in this field of economics. Often outcome variables of

interest such as income, are binned, resulting in “ordered” outcomes. I present an ordered

response model that relaxes rigid distributional assumptions, allowing for less biased and

more consistent estimators. This model allows for flexibility gains while still being tractable

and easy to implement for the applied researcher.