New Working Paper (2017)

Trading in the CDS market in this paper occurs because irrational investors have optimal beliefs about the default state of the economy, those investors tend to be overly optimistic that default is less likely. Since the imposition of the CDS ban on member states of the European Union in 2012, not only are investors forced to share risk through an incomplete capital market, optimal beliefs about default can also change. I show two sets of asset pricing implications can occur, one in which stock market values increase and countries' borrowing rates decrease after the ban is imposed, and one in the opposite direction. I find empirical support for the former, and also show the CDS ban is welfare improving in this case. While rational investors would benefit from lifting the ban, the gain is not sufficient to compensate irrational investors for their utility loss.

Published Papers

with Michael Gallmeyer, 2013, Journal of International Money and Finance

We study the equilibrium pricing of a debt contract when the borrower's economy is subject to rare event risk. In a setting where the lender and the borrower have heterogeneous beliefs about the likelihood of the rare event, we study the risk sharing role of a debt contract that defaults at the occurrence of this event. A higher belief by the lender compared to the borrower can lead to countercyclical interest rates and credit spreads in non-default times, and to an increase in the borrower's indebtedness in default times, as typically observed in emerging market economies. When calibrating the model to prices in the credit default swap market, we show that heterogeneous beliefs can account for more than 40% of the variation in CDS spreads associated with shocks to the borrower's economy.

with Thomas Plank, 2012, Review of Finance, Citations

What determines the price of insurance against default of advanced economies? Our laboratory to answer this question is the credit default swap (CDS) market on government debt of 18 advanced economies. The price of credit protection on these countries shows a strong degree of co-movement, has severely increased since the beginning of the financial crisis, and remains at elevated levels. We document that the state of a country's domestic financial system, and since the beginning of the crisis also the state of the world financial system have strong explanatory power for the behavior of CDS spreads, and that the magnitude of this impact depends on the relative importance of a country's financial system pre-crisis. Furthermore, it matters whether a country is a member of the Economic and Monetary Union of the European Union (EMU), in that their sensitivities to the health of the financial system are higher compared to non-EMU members. While one would expect the unconditional risk of default to be low in case of advanced economies, our results suggest the presence of an important economic channel in adverse economic times: a private-to-public risk transfer through which market participants incorporate their expectations about financial industry bailouts and the potential burden of government intervention.

We study the risk sharing implications that arise from introducing a disaster relief fund to the cat insurance market. Such a form of intervention can increase efficiency in the private market, and our design of disaster relief suggests a prominent role of catastrophe reinsurance. The model predicts buyers to increase their demand in the private market, and the seller to lower prices to such an extent that her revenues decrease upon introduction of disaster relief. We test two predictions in the context of the Terrorism Risk Insurance Act (TRIA). It is already known the introduction of TRIA led to negative abnormal returns in the insurance industry. In addition, we show the negative TRIA effect is stronger for larger and for low risk-averse firms - two results that are consistent with our model. The seller's risk aversion plays an important role in quantifying such feedback effects, and we point towards possible distortions in which a firm may even be overhedged upon introduction of disaster relief.

with George Aragon, 2011, Journal of Empirical Finance

In this paper we study the relation between daily stock market trading activity and the Dow Jones Industrial Average's (DJIA) movement around millenary milestones. We find aggregate turnover to be 5% lower when the DJIA level is less than 1% away from the nearest milestone. The effect emerges as the DJIA approaches a milestone from below, and it is stronger for first-time milestones compared to subsequent passages. The aggregate price impact of this investor behavior is large, such that daily stock returns show a negative abnormal performance of -10 basispoints. The milestone effect is robust to several specifications that capture changes in an investor's opportunity set, and is not due to lower risk around milestones or differences in stock market news production. Our findings suggest that millenary milestones of the DJIA play a role in investors' decision making.

2011, Journal of Financial and Quantitative Analysis

This paper provides an equilibrium model subject to heterogeneous beliefs about the likelihood of rare events. I explore asset pricing implications in an incomplete capital market and the effects of market completion. Without explicit rare event insurance, investors insure themselves indirectly through the stock and money markets, the risk premium is countercyclical, and flight to quality effects arise. Upon market completion, the risk premium increases as investors increase their exposure to rare event risk. While market completion leads to a more efficient allocation based on investors' anticipatory utilities, its effect on ex-post efficiency is ambiguous.

with Michael Gallmeyer, Journal of Economic Dynamics and Control, 2005

We study a two-agent pure exchange equilibrium subject to both nondiversifiable diffusive and jump risks. Agents can trade in a financial market consisting of a stock market, a money market, and an insurance market for jump risk. Heterogeneity is introduced through different levels of relative risk aversion. In the framework of standard expected utility we find the surprising result that the less risk averse agent purchases insurance contracts against jump risk from the more risk averse agent. This equilibrium allocation is linked to the non-linear wealth sharing rule in such an economy, and preserves the wealth effects studied by Dumas (1989) in the case of pure diffusive risk. Since the benchmark economy with homogeneous agents generates no excess uncertainty in the stock market, we study the effect on excess volatility and excess jump size solely due to different levels of relative risk aversion. We observe 3% excess uncertainty in jump sizes for a reasonable specification of economic fundamentals.

Working Papers

This paper documents the market reaction to the introduction of the European Financial Stability Facility (EFSF) in 2010 and 2011. The effect on borrowing rates is ambiguous - Greece, Ireland, Portugal, Spain, Italy and Slovenia exhibit a decrease in rates on event days, the remaining Eurozone countries exhibit an increase. The sovereign CDS market shows that not all of the effect on rates can be attributed to the assumption of default risk. CDS spreads do not increase in the case of Germany, France, the Netherlands, and Finland, consistent with a moderate increase in riskless rates. However, the net effect on the value of all Eurozone debt is positive and amounts to EUR 108 billion. Market participants also seem to anticipate an impact on real economic conditions given that the increase in non-financial equity value is EUR 126 billion, or 1.4% of Eurozone GDP.

This paper shows that catastrophe bond returns correlate significantly with economic fundamentals such as consumption. Hence, I build a consumption-based equilibrium model trying to reconcile investor preferences with several features of the cat bond market. The driving force behind the model is a habit process, in that catastrophes are rare economic shocks that could bring investors closer to their subsistence level. The calibration requires shocks with an impact between -1% and -3% to explain a reasonable level of cat bond spreads. Such investor preferences are not only able to generate realistic cat bond returns and price comovement among different perils, they may also explain why cat bonds offer higher rewards compared to equally-rated corporate bonds.

This paper re-examines to which extent catastrophe bond prices can 
be explained via investor preferences. I show that cat bond spreads 
equal between two and three times expected losses after controlling 
for bond-specific characteristics. At the occurrence of Katrina, the 
model predicts a 15-20% increase in the cost of capital of 
reinsurance companies and plausible degrees of comovement among 
different perils. The driving force behind the model is a habit 
process, in that catastrophes are rare economic shocks that could 
bring investors closer to their subsistence level. Such preferences 
may also explain why catastrophe bonds offer higher yield spreads 
compared to equally-rated corporate bonds.

with Spencer Martin, David Skeel, and Deon Strickland

As an alternative to liquidation, corporate reorganization is especially controversial in distressed or declining industries. Recoveries on average are significantly higher in reorganization outcomes, but is the modern machinery of Chapter 11 responsible? We use a detailed set of American corporate bond defaults from the pre-Chapter 11 era to explore the ex post efficiency of reorganizations. We report the new finding that reorganization recoveries are diminished in cases where the industry is distressed, just as liquidation recoveries are. Thus, even though the fire sales modeled by Shleifer and Vishny (1992) do not apply in  reorganizations, impaired asset values do appear through the lower reorganization recoveries. Our results suggest that even in the present day, reorganization may not be a viable mechanism for preserving declining industries.

with Min Ahn and Fabricio Perez

This paper provides a new approach to model the common variation in the term structure of credit spreads. The novelty is that common factors are extracted using canonical relations between credit spreads and observable economic variables. We show how these factors can be used to test if a given set of macroeconomic and financial variables is sufficient to capture all the systematic variation in response variables, such as credit spreads. We find that credit spread innovations are subject to three common factors, two strong factors and one weak factor, and they account for 49% of the total variation. The first strong factor is related to the contemporaneous state of the economy, the second represents expectations about future economic conditions, and the weak factor is mainly related to the error correction processes in short-term spreads.