This dissertation consists of three essays studying the relationship between corporate finance and monetary policy and macroeconomics. In the first essay, I provide novel estimations of the monetary policy’s working capital channel size by estimating a dynamic stochastic macro-finance model using firm-level data. In aggregate, I find a partial channel —about three-fourths of firms’ labor bill is borrowed. But the strength of this channel varies across industries, reaching as low as one-half for retail firms and as high as one for agriculture and construction. These results provide evidence that monetary policy could have varying effects across industries through the working capital channel. In the second essay, I study the effects of the Unconventional Monetary Policy (UMP) of purchasing corporate bonds on firms’ decisions in the COVID-19 crisis. Specifically, I develop a theoretical model which predicts that the firm’s default probability plays a crucial role in transmitting the effects of COVID-19 shock and the UMP. Using the model to evaluate two kinds of heterogeneities (size and initial credit risk), I show that large firms and high-risk firms are more affected by COVID-19 shock and are more responsive to the UMP. I then run cross-sectional regressions, whose results support the theoretical predictions suggesting that the firm’s characteristics, such as assets and operating income, are relevant to understanding the UMP effects. In the third essay, I document that capital utilization and short-term debt are procyclical. I show that a strong positive relationship exists at the aggregate and firm levels. It persists even when I control the regressions for firm size, profits, growth, and business cycle effects. In addition, the Dynamic Stochastic General Equilibrium (DSGE) model shows that in the presence of capital utilization, positive real and financial shocks cause the firm to change its financing of the equity payout policy from earnings to debt, increasing short-term debt.

This dissertation consists of two essays related to dynamic debt contracting and financial economics. The first chapter studies key determinants of inclusion of a financial covenant in corporate loans from theoretical and empirical angles. Using a novel manually collected loan dataset of small to medium-sized publicly-listed U.S. firms, I find that firms that issue loans without financial covenants tend to have (i) lower accounting quality, (ii) lower assets, and (iii) are experiencing faster growth in profitability relative to firms that issue loans with financial covenants. I build a theoretical model of project financing in which there is noisy public information about the project’s profitability, and the lender can privately monitor to improve the information quality. I show that if the signal precision without monitoring is sufficiently low (high), the equilibrium contract does not include (includes) a covenant. Covenant inclusion plays a key role in providing incentives to the lender to monitor. I show that the lender monitors less often relative to the first best. Insufficient monitoring leads to “excessive risk-taking,” namely, bad quality firms continuing with the project too often. Relatedly, I also show that covenants are used less often in equilibrium relative to the first best. The second chapter examines equilibrium consequences of litigation by holdout creditors in sovereign debt renegotiation. I show that given a sufficiently high probability of winning the litigation case against the borrowing country and/or a high enough defaulted sovereign debt, the presence of the holdout creditors increases the expected debt recovery rate, which makes the default option less attractive, and decreases the country’s default probability and the interest rate on the country’s debt. The country responds by borrowing more but defaults less often along the equilibrium path as it wants to avoid default and facing holdout creditors. Having a non-zero probability of successful litigation is welfare improving for the country as it sustains higher debt and defaults less frequently.

This dissertation consists of three essays studying topics in financial economicsthrough the lens of quantitative models. In particular, I provide three examples of the effective use of data in the disciplining of financial economics models. In the first essay, I provide evidence of a significant transitory component of aggregate equity payout. Leading asset pricing models assume exogenous dividend growth processes which are inconsistent with this fact. I find that imposing market clearing for consumption and income in these models induces the relevant behaviors in dividend growth, even when dividend growth is obtained indirectly. In the second essay, I provide a novel decomposition of the unconditional equity risk premium. In the data, the majority of the equity premium is attributable to moderate left tail risks, not those associated with disaster states. In stark contrast to the data, leading asset pricing models do not predict that this intermediate left tail region meaningfully contributes to the equity premium. The shortcomings of the models can be pinned on unreasonably low prices of risk for tail events relative to the data. In the third essay, I document a large dispersion in household allocations to risky assets conditional on age. I show that while standard household portfolio choice models can be made to match the average risky share over the lifecycle, the models fall short of generating sufficient heterogeneity in the cross-section of household portfolios.

This dissertation is a collection of three essays relating household financial obligations to asset prices. Financial obligations include both debt payments and other financial commitments.

In the first essay, I investigate how household financial obligations affect the equity premium. I modify the standard Mehra-Prescott (1985) consumption-based asset pricing model to resolve the equity risk premium puzzle. I focus on two channels: the preference channel and the borrowing constraints channel. Under reasonable parameterizations, my model generates equity risk premiums similar in magnitudes to those observed in U.S. data. Furthermore, I show that relaxing the borrowing constraint shrinks the equity risk premium.

In the Second essay, I test the predictability of excess market returns using the household financial obligations ratio. I show that deviations in the household financial obligations ratio from its long-run mean is a better forecaster of future market returns than alternative prediction variables. The results remain significant using either quarterly or annual data and are robust to out-of-sample tests.

In the third essay, I investigate whether the risk associated with household financial obligations is an economy-wide risk with the potential to explain fluctuations in the cross-section of stock returns. The multifactor model I propose, is a modification of the capital asset pricing model that includes the financial obligations ratio as a ``conditioning down" variable. The key finding is that there is an aggregate hedging demand for securities that pay off in periods characterized by higher levels of financial obligations ratios. The consistent pricing of financial obligations risk with a negative risk premium suggests that the financial obligations ratio acts as a state variable.

This dissertation consists of two essays on corporate policy. The first chapter analyzes whether being labeled a “growth” firm or a “value” firm affects the firm’s dividend policy. I focus on the dividend policy because of its discretionary nature and the link to investor demand. To address endogeneity concerns, I use regression discontinuity design around the threshold to assign firms to each category. The results show that “value” firms have a significantly higher dividend payout - about four percentage points - than growth firms. This approach establishes a causal link between firm “growth/value” labels and dividend policy.

The second chapter develops investment policy model which associated with du- ration of cash flow. Firms are doing their business by operating a portfolio of projects that have various duration, and the duration of the project portfolio generates dif- ferent duration of cash flow stream. By assuming the duration of cash flow as a firm specific characteristic, this paper analyzes how the duration of cash flow affects firms’ investment decision. I develop a model of investment, external finance, and savings to characterize how firms’ decision is affected by the duration of cash flow. Firms maximize total value of cash flow, while they have to maintain their solvency by paying a fixed cost for the operation. I empirically confirm the positive correlation between duration of cash flow and investment with theoretical support. Financial constraint suffocates the firm when they face solvency issue, so that model with financial constraint shows that the correlation between duration of cash flow and investment is stronger than low financial constraint case.

Merton (1987) predicts that idiosyncratic risk can be priced. I develop a simple equilibrium model of capital markets with information costs in which the idiosyncratic risk premium depends on the average level of idiosyncratic volatility. This dependence suggests that the idiosyncratic risk premium varies over time. I find that in U.S. markets, the covariance between stock-level idiosyncratic volatility and the idiosyncratic risk premium explains future stock returns. Stocks in the highest quintile of the covariance between the volatility and risk premium earn an average 3-factor alpha of 70 bps per month higher than those in the lowest quintile.

This dissertation applies the Bayesian approach as a method to improve the estimation efficiency of existing econometric tools. The first chapter suggests the Continuous Choice Bayesian (CCB) estimator which combines the Bayesian approach with the Continuous Choice (CC) estimator suggested by Imai and Keane (2004). Using simulation study, I provide two important findings. First, the CC estimator clearly has better finite sample properties compared to a frequently used Discrete Choice (DC) estimator. Second, the CCB estimator has better estimation efficiency when data size is relatively small and it still retains the advantage of the CC estimator over the DC estimator. The second chapter estimates baseball's managerial efficiency using a stochastic frontier function with the Bayesian approach. When I apply a stochastic frontier model to baseball panel data, the difficult part is that dataset often has a small number of periods, which result in large estimation variance. To overcome this problem, I apply the Bayesian approach to a stochastic frontier analysis. I compare the confidence interval of efficiencies from the Bayesian estimator with the classical frequentist confidence interval. Simulation results show that when I use the Bayesian approach, I achieve smaller estimation variance while I do not lose any reliability in a point estimation. Then, I apply the Bayesian stochastic frontier analysis to answer some interesting questions in baseball.

Schennach (2007) has shown that the Empirical Likelihood (EL) estimator may not be asymptotically normal when a misspecified model is estimated. This problem occurs because the empirical probabilities of individual observations are restricted to be positive. I find that even the EL estimator computed without the restriction can fail to be asymptotically normal for misspecified models if the sample moments weighted by unrestricted empirical probabilities do not have finite population moments. As a remedy for this problem, I propose a group of alternative estimators which I refer to as modified EL (MEL) estimators. For correctly specified models, these estimators have the same higher order asymptotic properties as the EL estimator. The MEL estimators are obtained by the Generalized Method of Moments (GMM) applied to an exactly identified model. The simulation results provide promising evidence for these estimators. In the second chapter, I introduce an alternative group of estimators to the Generalized Empirical Likelihood (GEL) family. The new group is constructed by employing demeaned moment functions in the objective function while using the original moment functions in the constraints. This designation modifies the higher-order properties of estimators. I refer to these new estimators as Demeaned Generalized Empirical Likelihood (DGEL) estimators. Although Newey and Smith (2004) show that the EL estimator in the GEL family has fewer sources of bias and is higher-order efficient after bias-correction, the demeaned exponential tilting (DET) estimator in the DGEL group has those superior properties. In addition, if data are symmetrically distributed, every estimator in the DGEL family shares the same higher-order properties as the best member.  

This paper examines dealers' inventory holding periods and the associated price markups on corporate bonds from 2003 to 2010. Changes in these measures explain a large part of the time series variation in aggregate corporate bond prices. In the cross-section, holding periods and markups overshadow extant liquidity measures and have significant explanatory power for individual bond prices. Both measures shed light on the credit spread puzzle: changes in credit spread are positively correlated with changes in holding periods and markups, and a large portion of credit spread changes is explained by them. The economic effects of holding periods and markups are particularly sharp during crisis periods.