Three essays on the strategic effects of debt on firms' R&D decisions.

摘要

This dissertation represents an attempt to integrate financial and market structure considerations into a unified model, in order to gain a richer understanding of the relationship between firms' decisions in product and capital markets. Specifically, the dissertation explores how firms' debt financing decisions affect R&D investment decisions and equilibrium capital structure.; We first develop a simple theoretical model of external R&D in which manufacturing firms outsource R&D activities to firms specializing in research. The model has a number of testable implications. First, debt financing has a non monotonic effect on research firms' R&D investments: An increase in debt induces research firms to increase (reduce) the level of R&D investment when the debt level is high (low). Second, research firms choose either low or high levels of debt in equilibrium, depending on the magnitude of strategic commitment effects. Third, under plausible conditions, the equilibrium level of R&D exceeds (falls short of) the social optimum when aggregate debt is high (low).; Next, we extend the framework to allow for internal R&D; that is, we develop a model in which R&D and production activities take place within a firm. Similar to the case of external R&D, we show that debt financing has non-monotonic effects on firms' internal R&D decisions. However, these debt effects critically depend on rival firms' debt levels. More generally, debt financing has a direct impact on a firm's own output decision as well as rivals' best responses in the product market. Thus, the model predicts a link between financial structure and product markets.; The dissertation concludes with an empirical examination of the link between debt and R&D activity. In particular, we test the hypothesis that debt and R&D expenditures have a non-monotonic relationship, based on industry-level data from Korean manufacturing firms. Econometric results, based on data from 2000--2002, indicate that debt and R&D have a U-shaped relationship with a positive threshold level in Korean high-tech industries. These findings are broadly consistent with the predictions of the theoretical models.