Abstract
My dissertation consists of three independent yet thematically related chapters on development economics. The first two chapters focus on the digital economy. The first chapter examines gender gaps in the gig economy, while the second investigates how firms respond to digitalization opportunities in developing countries. The third chapter continues to study firm behaviors; specifically, it analyzes how firms leverage political connections to increase resilience during economic crises.
In the first chapter, ``Does the Gig Economy Discriminate against Women? Evidence from Physicians in China,'' I examine gender disparities in the burgeoning gig economy. Using novel data from a major Chinese online healthcare platform, I show that female physicians charge 2.3% lower prices and provide 11.0% fewer consultations than males. Patients appear to discriminate against female physicians despite them having identical observable productive characteristics to those of male physicians. The differential responses of patients to quality signals from female physicians suggest that a portion of this discrimination is statistical in nature. I further find that the platform's design, particularly its ranking algorithm, plays an important role in enlarging gender gaps. The ranking algorithm amplifies and perpetuates the gaps by using past patient behavior (and thus pre-existing discrimination) as a key predictor of future patient behavior, thereby placing fewer females at the top of search results. Additionally, I cast doubt on several other alternative explanations and conducted a series of robustness checks.
In the second chapter, ``Digitalization as a Double-Edged Sword: Winning Services and Losing Manufacturing in India,'' I explore the impacts of digitalization on firms. While digitalization can increase firm productivity, in developing countries with labor market frictions, not all firms are able to capitalize on digitalization opportunities. I use data from India--where a demonetization policy led to a large increase in digital payments--to examine the impacts of digitalization on firms across sectors in a developing country, identifying winners and losers in the short run. I find that service firms experienced growth in income and productivity while manufacturing firms witnessed a decline. I then explore the mechanisms driving this divergence. The results show that service firms invested more in information and communications technology (ICT) capital and hired more complementary skilled ICT labor, whereas manufacturing firms did not. Notably, this influx of skilled ICT workers into the service sector was drawn from the manufacturing sector due to limited spatial labor mobility. During this short-run transitional phase, wages for ICT labor were driven up while remaining stagnant for other workers. These findings underscore how digitalization, in the presence of labor market constraints, can exacerbate short-term sectoral divergence in productivity growth and shed light on its impacts on the growth trajectories of developing countries.
In the third chapter, ``How Do Political Connections of Firms Matter During An Economic Crisis?'', we use a new machine learning-enabled, \textit{social network based} measurement technique to assemble a novel dataset of firms' political connections in India. Combining it with a long panel of detailed financial transactions of firms, we study \textit{how} firms leverage these connections during an economic downturn. Using a synthetic difference-in-differences framework, we find that connected firms had 8-10% higher income, sales, and TFPR gains that were persistent for over a three-year period following the crisis. We unpack various novel mechanisms and show that connected firms were able to delay their short-term payments to suppliers and creditors, delay debt and interest payments, decrease expensive long-term borrowings from banks in favor of short-term non-collateral ones, and increase investments in productive assets such as computers and software. Our method to determine political connections is portable to other applications and contexts.
