Research

Working Papers

Connections as Jumps: Estimating Financial Interconnectedness from Market Data

Abstract: I develop a new methodology for measuring interconnectedness between financial institutions using readily available asset price data. I address the classic endogeneity problem that arises when using contemporaneous price movements by focusing on connections that trigger substantial spillovers upon default. Using the statistical similarity between spillovers and jumps (sudden, discontinuous reductions in the underlying asset value of the firm), the effects of such connections are found in the jump-like default risk of a firm. Importantly, the remaining default risk, which captures smooth paths to default, is not exposed to such connections. Therefore, under appropriate identification assumptions, regressing jump-like default risk on non-jump-like default risk uncovers causal evidence of direct and indirect exposures. Using the rich distributional information embedded in equity, equity options, and credit default swap data, I apply the methodology to the largest US financial firms during the 2008 financial crisis. I find estimates of connections that are consistent with well-known developments during the crisis: Firms change positions in the network in line with their risk and access to support programs. My estimates further suggest that market participants viewed the collapse of Lehman Brothers as a symptom, rather than the cause, of the crisis. The methodology I develop in this paper provides a new tool for monitoring the financial sector in real time during distress periods using contemporaneous market price changes.

Research work in progress

Regressions with Jumps (draft and R package available soon)

I propose a new method for estimating continuous and jump regression coefficients, such as those used for estimating jump-diffusion betas in empirical asset pricing studies. In contrast to existing methods, which depend on chosen cutoffs and typically require multiple passes of the data, my method based on the MedRV estimate of Andersen, Dobrev, and Schaumburg (2012) requires no additional decisions by the researcher and can be calculated using one pass of the data. In addition, I obtain asymptotic standard errors for the continuous coefficients. Preliminary empirical applications show similar performance to far more complicated existing methods.

Compression as an Alternative to Central Clearing
Joint work with Ben Charoenwong and (soon!) John Kuong.

We study the ability for trade compression, defined as the reduction of gross trades in a market through bilateral or multilateral netting, to reduce default risk in an over-the-counter (OTC) asset market using a tractable model that incorporates nonpayment from the worst-performing financial institution in each state of the world. Although compression reduces capital costs for participants, we show that maximal compression is generally inefficient from a systemic risk perspective. We show that a social planner trying to minimize total defaults will want to compress out any trade cycles but will also want to incorporate some intermediation to help absorb potential losses. Our numerical simulations suggest optimal trade compression can perform similarly well to a centrally cleared market and may outperform central clearing in the presence of highly risky market participants by shielding safer participants from the collapse of riskier participants. Our results suggest that trade compression, when performed to reduce risk rather than simply to reduce gross notional, may be a viable alternative to central clearing for markets where central clearing is infeasible or undesirable.

Archived Research

Flexible Lead-Lag Regressions

I develop a penalized regression technique for estimating flexible lead-lag relationships between time series. Lead-lag relationships are allowed to be time-varying. Simulation shows potential when there is a strong relationship. However, empirical applications so far have mostly shown overfitting.