The Collateral Rule: Evidence from the Credit Default Swap Market (with Agostino Capponi, Allen Cheng, Richard Haynes and Stefano Giglio), Journal of Monetary Economics, forthcoming [pdf]

We explore a novel dataset of daily cleared credit default swap (CDS) positions along with the posted margins to study how collateral varies with portfolio risks and market conditions. Contrary to many theoretical models, which assume that collateral constraints follow Value-at-Risk rules, we find strong evidence that collateral requirements are set an order of magnitude larger than what standard Value-at-Risk rules imply. The panel variation in collateralization rates of CDS portfolios (over time and across participants) is well explained by measures of extreme tail risks, related to the maximal potential loss of the portfolio. We develop a model of endogenous collateral in CDS markets to interpret these empirical findings. The model predicts that the conservativeness of collateral levels can be explained through disagreement of market participants about the extreme states of the world, in which CDSs pay off and counterparties default.

Working Papers

Collateral requirements in central bank lending, Job Market Paper, [pdf]

Central banks around the world are intervening aggressively to cushion the impact of COVID-19. I develop a model of central bank lending in collateralized credit markets. In this model, borrowing constraints amplify adverse productivity shocks during a downturn because firms face a reduction in both their liquid wealth and the borrowing capacity of their collateral. When the downturn is severe, the central bank optimally responds by lending at more favorable interest rates while simultaneously reducing the haircuts imposed on eligible collateral. In doing so, the central bank takes on greater credit risk, but achieves an outcome that is more productively efficient than simply reducing the risk-free interest rate. Lastly, I show that anticipation of such intervention do not always generate greater inefficiencies ex ante.

Capital requirements, the safe real interest rate and the fundamental problem of bank risk taking (with David Miles), [pdf]

We show that bank capital requirements, and variations in the safe real interest rate, operate as imperfect substitutes in countering a tendency for banks to take excessive risks. A tightening of capital requirements, or of the safe real interest rate, can improve ‘prudence’ by disincentivising banks against investing in risky assets with sub-optimally low returns; but only at the cost of decreased ‘participation’ whereby more banks will forego the opportunity to invest. Numerical simulations show that a substantial capital requirement is in general the appropriate policy response. The optimal capital requirement rises as the safe real interest rate falls.

Multinomial max-min theorem: a generalisation of the binomial no-default theorem (with John Geanakoplos)

The Binomial No-Default Theorem in Fostel and Geanakoplos (2015) states that “in binomial economies with financial assets serving as collateral, any equilibrium is equivalent in real allocations and prices to another equilibrium in which there is no default”. I extend this theorem to economies with more than two states of nature. For instance, with three states of nature, borrowers can issue both senior secured debt and junior subordinated debt. The terms and credit risks associated with each creditor tier are endogenously determined. The Multinomial Max-Min theorem states that any equilibrium is equivalent to another equilibrium where the senior tranche never defaults, and the junior tranche only defaults in the worst state of the world. The expanded theorem allows for the application of the endogenous leverage framework to a richer set of models.  

Research in Progress

Supply network fragility (with Agostino Capponi and Joseph Stiglitz)

Capital buffer usability (with Walter Jansson and Rosie Dickinson)

Financial stability implications of the DAI stablecoin (with Kunal Khairnar)