Three Essays on the Analysis of Firms' Behaviors Under Staggered Treatment Adoption

This dissertation consists of three essays on firms' behaviors under staggered treatment adoption. The first essay draws information from a micro-lender and a credit bureau to identify the causal effects of small loans on the financial health of a group of small U.S. business owners. To achieve this, we exploit temporal variations in the loan disbursements and use an estimation strategy that controls for potential biases due to treatment effect heterogeneity. The results suggest that even small loans are effective in generating lasting positive impacts on widely accepted financial health indicators, such as Vantage Score (Credit Score), Debt-to-Income Ratio, and Credit Utilization Ratio. We obtain similar robust results for subprime and startup borrowers, who are known to face difficulties in securing credit.

The second essay combines unionization data from the National Labor Relations Board and financial data from Compustat to examine the causal effects of unionization on the financing decisions of publicly traded firms in the United States. In this essay, I exploit temporal variations in the election date of unionization across firms and use a dynamic difference-in- difference estimation strategy to identify the effects of unionization on a range of financial indicators, including the Debt-to-Equity ratio, market leverage, book leverage, long-term book leverage, net leverage and cash to asset ratio. I find that unionization negatively affect firms' financing decisions. For example, after unionization, firms rely less on leverage to raise capital. At the same time, unionization offers incentive to firms to hold more cash in hand. My analysis also suggests that the effects of unionization vary according to the political and institutional structure of the states in which firms operate. For instance, the impacts on the outcome variables are more pronounced for the firms in democrat-led states and for firms which operate in states without right-to-work laws. The effects of unionization are also more noticeable for multi-establishment firms versus one-establishment firms. In addition, we find that the effects vary according to the margin of support for unionization within a firm.

The third essay examines the causal effects of unionization on innovation activities of publicly traded firms in the United States. As in the case of chapters 1 and 2, the analysis uses a dynamic difference-in-difference estimation strategy on a dataset that is compiled using information on unionization data from the National Labor Relations Board, financial data from Compustat and KPSS patent data. My analysis encompasses a wide range of innovation indicators, including the number of patents, number of forward citations, market value of patents, average citations, number of patents to RandD expenditures ratio, number of citations to RandD expenditures ratio, number of patents per 1000 employees, capital expenditures to sales ratio and RandD expenditures to sales ratio. The findings suggest a small positive impact of unionization on most of these innovation indicators, with the exception of market value of patents and number of patents to RandD expenditures ratio. I also find that the effects of unionization vary according to political orientations of states, industry type, firm size and firm age. The results demonstrate that the effects on innovation are more pronounced for smaller and younger firms and for firms operating in democrat-led states as well as manufacturing firms.