Working papers

Securities lending and information transmission: a model of endogenous short-sale constraints

I study short-sale constraints in a market with asymmetric information. I offer a novel approach endogenizing short-sale constraints by including an asset-borrowing market in my model. Short-sellers have to borrow an asset and therefore reveal information to a lender. The lender trades on her own account in addition to charging fees, which motivates the short-seller to hide the information and hinders short sales. I contribute to the literature by modeling the mechanism behind short-selling in the absence of explicit short-selling restrictions that are currently less relevant in practice. The model has new implications for profit distribution, market efficiency, and volatility.

Informed trading, short-sale constraints, and leverage effect in equity returns

I model informed trading subject to short-sale constraints and find that short-sale constraints can cause the asymmetric volatility effect (also known as the “leverage effect”). I offer an infinite-horizon model with overlapping generations of private information and stationary time series of returns. This model builds up on the framework of Kyle (1985) model augmented with a short-sale constraint. I consider a collection of settings and find that the magnitude of the leverage effect is driven by the assumptions of the probability distribution of the asset’s fundamental value. Additionally I find that fat-tailed fundamental values can generate a persistent volatility effect irrespective of the short-sale constraints.

Information leakages, distribution of profits from informed trading, and last mover advantage

This is a model of front-running in which the front-runner may have an advantage over an informed trader due to the fact that she makes her trading decision later. With endogenous information-acquisition choices, the model predicts that the amount of information generated by the front-runner as well as her trading aggressiveness can be discontinuous in the state of technology. Once, information extraction costs fall below a certain threshold, the aggressiveness of the front-runners increases abruptly. I interpret this as an explanation for an explosive emergence of high-frequency trading in the mid-2000s.

Work in progress

Dissimulation of informed trades on OTC fixed income markets

I propose a model à la Easley and O’Hara (1987). In my model, the traders have a larger decision space than in the Easley and O’Hara model. The traders chose their order size in a continuum of values. If the asset fundamental value is binary, my model yields a unique partially-separating mixed strategy equilibrium. In equilibrium, the price is a non-linear function of the order size, and moreover, a non-smooth function. The orders whose size does not exceed a given threshold have no effect on the price. At the threshold, the price graph (as a function of the order size) has kinks. There is a kink on each side: purchases and sales. For the orders whose size exceeds the threshold, the bid-ask spread expands as a function of the order size.

A model for interest rate risk (Joint with F. Severino)

We propose and analyze a new alternative approach to measuring interest rate risk of bonds and bond portfolios. The methodology is particularly relevant for assessing the consequences of huge interest rate changes (stress test) and for bonds and fixed-income portfolios with long maturity. In these cases, our approach allows to approximate the swings of the bond prices much more precisely than commonly used methods relying on bond duration and convexity. We analyze the relationship between a bond price and a bond Yield-to-Maturity and propose a family of approximating functions for the bond present value profile.

“The Chicken and the Egg of WACC”

The trade-off capital structure theory contends that firms choose their financial leverage so as to balance between exploiting the interest tax shield and avoiding the escalation of expected bankruptcy costs and agency conflicts. In practice, buy-side analysts are in a setting with incomplete information. Managers know about the firm’s capital structure target more than the outsides. Moreover, the insiders are likely to have incentives to mislead the market regarding the capital structure target to manipulate the stock price.

To apply the discounted cash flow technique, an analyst needs the weighted average cost of capital (WACC). Two common recipes are known: (1) rely on the firm’s public announcements about targeted capital structure, (2) find the capital structure from the firm value. The first recipe is straightforward but it suffers from lack of credibility of the public announcements. The second method is more economically sound but it requires  solving an evil loop: the firm value depends on the WACC, which depends on the financial leverage, which in turn depends on the value of the firm. Existing literature proposes an iterative method to solve this circular problem. Usually, this methodology works but the convergence is not guaranteed. I scrutinize the circular problem and identify the conditions for existence and uniqueness of solutions. In addition, I propose an alternative concise method to solve the circular problem featuring much faster and more certain convergence. In practice, this means that financial analysts will be equipped with a simple yet powerful tool to find mutually consistent capital structure and the value of equity in two to three iterations.