Research

Job Market Paper

 

Strategic Interaction and the Co-Determination of Firms' Financial Policies

Abstract:

With a few notable exceptions, corporate finance studies of firms' financial policies typically rely on a single firm setting, thus overlooking the possibility that firms' financial policies are co-determined by those of their rivals. I develop and test a model in which firms interact by buying and selling productive assets. This interaction affects cash policies because a lack of cash may force a firm to sell assets at a discount, while having surplus cash may enable a firm to take advantage of other firms' asset sales. The model generates sharp and novel empirical predictions at the industry (as opposed to the individual firm) level. I test these predictions using a carefully built methodology that tackles the endogeneity and persistence of firm-level determinants. Precisely as predicted by the theory, I find that both the average cash holdings in an industry, as well as the heterogeneity in cash policies within that industry, depend on two variables: the asset specificity of that industry and industry cashflow volatility. These results point to the importance of strategic interaction as a determinant of corporate financial policies.

PDF Full Text

 

In Preparation

 

Distress Losses, Equity Values and Equity Returns

Abstract:

The value of equity depends on the future potential losses due to asset fire-sales triggered by the firm becoming distressed but not bankrupt. Moreover, equity holders today are exposed to potential future losses due to bankruptcy since future bondholders are rational. In contrast with the literature about the effect of distress risk on equity returns, I investigate empirically the implications of a framework in which equity values depend on the potential losses due to asset liquidations. These losses are a function of the joint probability that both the industry and the firm are in a bad state, and on the valuation that outsiders have for the assets. Notably, the potential losses depend both on firm and industry characteristics. Given that expectations about losses may be systematic, expected equity returns depend on these characteristics as well. Moreover, I investigate implications for the industry returns, options on industry ETFs versus options on individual stocks, the betas with downturn market, and the relation between returns and risk factors.

Volatility Co-Movement (in revision)

Abstract:

Index arbitrage trades may generate common “noise” in both the components of an index and the index’s ETF. Using the Multiplicative Error Model of Engle (2002), I find that S&P500’s and individual components’ volatilities co-move more after the stock becomes part of the index. This effect is prevalent at higher frequencies (i.e. hourly) and becomes weak at daily levels. The return co-movement is not responsible for this result, since I observe a significant return co-movement only at low frequencies but not at high frequencies. In fact, my hypothesis about index arbitrage noise implies a negative co-movement in returns between the ETF and the stock, consistent with my results.