CV.pdf

Research:

Is the Banking Sector Too Big?

The paper presents a model of signalling through bank financing, in which the bigger the bank loan a project obtains, the lower the book rate of the deal and the higher the project's quality signalled, capturing well empirical findings. Based on the model, the paper examines the effects of the banking sector size when banks' ability of discerning risks is not given but invested. The paper shows that so long as ex post risk shifting is contained, (a) banks over-invest in the ability; (b) the bigger the banking sector, the lower the investment; therefore the sector is never too big; (c) but the bigger the sector, the riskier its assets, because the lower investment the worse the risk discerning ability. The paper suggests that a universal payment cap upon the financial sector could improve social efficiency.


THE DYNAMICS OF NAMES: A MODEL OF REPUTATION, International Economic Review 52(2011), 1039-58

The paper studies reputation of names and its dynamics. Firstly, the paper shows that pure names, backed by nothing intrinsic, can bear reputation. Although the context is purely of adverse selection, a mechanism in the spirit of Kreps (1990) helps sustain reputation. Secondly, it examines how the dynamics of reputation affects the extent of sorting and the level of social efficiency, and derives the dynamics in the equilibria with the highest efficiency. Lastly, it finds a comparative statics result which empirically predicts that brandnames can be fully established sooner in an industry where high-end products have larger profit margins.


The Allocation of Liability, Delegated Monitoring, and Modes of Financing

Based on the allocation of liability, the paper examines all the modes of organizing finance and delegated monitoring in an economy with costly state verification. Besides the mode of financial intermediation (FI), a mode of direct finance, Conglomeration, accommodates delegated monitoring and obtains the benefit of diversification also, which, therefore, do not necessarily make FI viable, as the extant literature claims. The paper empirically predicts that the prevalence of bank financing could increase with the costs of banks' services or the rates of banks' loans. The paper is the first to examine the market organization of a financial service through the allocation of liability.


Risk Sharing, (Over)Leverage, and the Dimensions of Competition

The paper examines the leverage of financial intermediaries in a general equilibrium framework. The paper's approach to leverage is driven by risk sharing and captures two features: Debt serves to boost the return of equity and equity to "safe net" debt. The paper finds that if entrepreneurs compete only in the dimension of price, the equilibrium leverage rate is above the social best one; and if they compete in both price and investment scale, the two rates coincide, even under the market segmentation which drives financial intermediation. In the latter case, the credit market is cleared not by price, but by contract which specifies both price and scale. The paper argues that overleverage is more likely to occur where firms are small and dispersed while banks are large and concentrated, and shows that it can be rectified by proper capital adequacy regulation.


Ownership, Control, and Incentive

The paper shows that the principal can enhance her control over the agent's human capital by acquiring the physical capital that is critical for him to create value. However, the enhancement in the control necessarily reduces his incentive to make human capital investment ex ante and to exert effort ex post. This trade-off between control and incentive thus decides the boundary of the firm. The paper also presents a rationale for M-form firms: centralized ownership of physical capital to facilitate coordination, and dispersed payoff rights to incentivize divisions.


Collusion, Incentives and Information: The role of Experts in Corporate Governance with Sanjay Banerji

The paper shows that the severer moral hazard problems of the CEO and poorer ability of the outside expert not only diminish the value of the firm on their own but also worsen the problem of collusion which is more likely to arise in times of boom. The shareholders then incur extra costs, attributed to increased incentive and severance payments to the CEO. The expert's incentives to acquire reputation or the public signals from media etc. mitigate such problems but cannot undo them completely. The paper derives empirical predictions on experts' turnovers, fees, and their impacts on corporate restructuring.


Reputation and Scale

The paper offers an explanation for why large scales are necessary to establish reputation. In the model, each producer yields one widget and chooses the quality investment each period. Low quality widgets always fail and high quality succeed with some probability. Due to noisiness of single performance, a producer cannot establish reputation in the market. This noisiness problem is resolved by the Law of Large Numbers, as the performance of many widgets is informative of the average quality. When the scale is large, however, defaulting to invest a small number of widgets seems undetectable, which, if true, will break down the equilibrium expectation of the average quality. This incentive problem is resolved by Central Limit Theorem, which ensures increasing marginal return of making the investment. Thus the operators of large scales are always able to establish reputation. The paper sheds light on the existence of professional firms and middlemen.

Teaching:

EC351 Mathematical Economics

EC908 Topics in Financial Economics