Publications

2018
Bai J, Krishnamurthy A, Weymuller C-H. Measuring Liquidity Mismatch in the Banking Sector. Journal of Finance. 2018;73 (1) :51-93. SSRN copyAbstract

This paper constructs a Liquidity Mismatch Index (LMI) to gauge the mismatch between the market liquidity of assets and the funding liquidity of liabilities, for 2882 bank holding companies over 2002 to 2014. The aggregate LMI decreases from $4 trillion pre-crisis to -$6 trillion in 2008. We conduct an LMI stress test revealing the fragility of the banking system in early 2007. Moreover, LMI predicts a bank's stock market crash probability and borrowing decisions from the government during the financial crisis. The LMI is therefore informative about both individual bank liquidity and the liquidity risk of the entire banking system.

2017
Davila E, Weymuller C-H. Optimal Joint Bond Design. 2017.Abstract

We study the optimal design of a joint borrowing arrangement among countries. In our model, a safe country, which has full commitment and never defaults, and a risky country, which lacks commitment and may default, participate in a joint borrowing scheme through which they allocate a predetermined amount of their bond issuance to a joint bond, which may earn a non-pecuniary premium. The joint borrowing scheme is flexible, and highlights the differences between pooled issuance, in which countries share the funds raised through the joint bond, and joint liability, in which one country guarantees the obligations of another one. We develop a simple but general condition that determines whether issuing a joint bond is welfare improving: if the total marginal increase in the amount raised by the countries – holding constant their borrowing decisions – is greater that the value of the joint liabilities that are originated, it is optimal to issue a positive amount of joint bond. We further decompose the welfare effects of varying the size of the joint bond into several distinct channels. We provide a quantitative analysis of joint borrowing agreements and find that Pareto improvements are possible.

Optimal Joint Bond Design
2013
Weymuller C-H. Banks as Safety Multipliers: A Theory of Safe Assets Creation. 2013.Abstract

Why do banks hold so many safe assets such as government bonds? I argue that the economic role of banks is to multiply safety, and their holdings of public debt help them achieve that goal. In a environment with endogenous collateral constraints and multiple assets, risk-neutral banks issue debt securities to cater to the safety demand of risk-averse investors. Private safety creation requires banks to hold on their balance sheets government bonds, whose returns are negatively correlated with macroeconomic shocks. When heterogeneity in risk aversion is large and public debt supply low, there is a safety multiplier in the sense that lowering even more the supply of public debt induces banks to delever, hurting private debt supply. In this regime, public debt crowds-in private investment, because a shortage of public debt constrains risk-sharing. In a dynamic version with maturity choices, I endogeneize the negative correlation of public debt. The expectation of a flight-to-safety transforms long-term securities into hedging instruments. The private equilibrium is constrained inefficient due to an issuance externality. The economy lacks long-term securities, as private agents do not internalize the benefits of the negative beta of their own liabilities. Public debt is non-neutral but there is an interior optimal level of public securities, as their hedging properties deteriorate with their supply.

The model interprets the ongoing European debt crisis as a shortage of public safe assets. Public debt and private debt positively comove in Europe, contrary to the US. Simultaneously, banks increased their holdings of safe public debt to back private debt. As for asset pricing, the spread between public debt yield and private debt yield reveals bank leverage and can be used as a macroprudential tool. In an open economy environment, sovereign risk hurts aggregate private leverage but domestic banks become the natural holders of domestic public debt.

Banks as Safety Multipliers: A Theory of Safe Assets Creation
Weymuller C-H. Leverage and Reputational Fragility in Credit Markets. 2013.Abstract
This paper analyzes the equilibrium on secured debt markets with heterogeneous borrowers. I develop a bargaining model of endogenous leverage when debt capacity depends not only on the nature of collateral but also on borrower's reputation. Agents with higher reputation secure both more and cheaper credit. As a result, haircuts and rates positively comove in the cross-section, a prediction that cannot be delivered in a model where the heterogeneity among borrowers is about collateral, such as beliefs. I endogeneize reputation by identifying it to continuation value in a repeated borrowing game. Leverage is procyclical because of the endogeneity of reputation. Leverage is high and stable when agents are sufficiently capitalized. However, as their net worth depletes, reputation vanishes in a non-linear way, and leverage then is limited and unstable. Reputation acts as a stabilization mechanism in good times but as an amplification mechanism in bad times. In this dynamic environment, long-term contracts stabilize leverage by avoiding margin calls in the states in which the borrower has low net worth. Reputation also provides an economic rationale to a long intermediation chain, in order to monetize intermediaries franchise. The empirical analysis on repo transactions of money market funds confirms the predictions of the model. First, financial intermediaries with high reputation write repo contracts with lower haircuts and lower rates. Second, haircuts are more sensitive to borrower's reputation when borrower's net worth is low. Third, long-term relationships stabilize funding. It provides evidence that reputation and bilateral relationships matter even in secured funding markets.
Leverage and Reputational Fragility in Credit Markets