Krista Schwarz, University of Pennsylvania
Mind the Gap: Disentangling Credit and Liquidity in Risk Spreads
Widening interest rate spreads observed during the recent financial crisis could represent deteriorating liquidity or greater credit risk. Schwarz constructs new microstructure measures of credit and liquidity and finds that market liquidity effects explain more than two-thirds of the widening of one- and three-month euro LIBOR-OIS spreads and of intra-euro-area sovereign debt spreads over the sample period. Her new credit risk measure is an indicator of credit tiering in the interbank money market; her new liquidity measure uses the spread between bonds of differing liquidity that are all guaranteed by the German government, and that therefore should not be contaminated by any effects of credit. Over the sample period, her two measures are nearly orthogonal, making it possible to econometrically identify the separate effects of credit and liquidity. Previous literature finds that risk spreads are largely attributable to default risk, but Ishe ascribes this to their mismeasurement of liquidity and credit.
Darrell Duffie, Stanford University and NBER; and Bruno Strulovici, Northwestern
Capital Mobility and Asset Pricing
Duffie and Strulovici present a model for the equilibrium movement of capital between asset markets that are distinguished only by the levels of capital invested in each. Investment in that market with the greatest amount of capital earns the lowest risk premium. Intermediaries optimally trade off the costs of intermediation against fees that depend on the gain they can offer to investors for moving their capital to the market with the higher mean return. Those fees also depend on the bargaining power of the investor, in light of potential alternative intermediaries. In equilibrium, the speeds of adjustment of mean returns and of capital between the two markets are increasing in the degree to which capital is imbalanced between the two markets.
Viral V. Acharya and Douglas Gale, New York University; and Tanju Yorulmazer, Federal Reserve Bank of New York
Rollover Risk and Market Freezes
The crisis of 2007-9 has been characterized by a sudden freeze in the market for short-term, secured borrowing. Acharya, Gale, and Yorulmazer present a model that can explain a sudden collapse in the amount that can be borrowed against finitely-lived assets with little credit risk. The borrowing in this model takes the form of a repurchase agreement ("repo") or asset-backed commercial paper that has to be rolled over several times before the underlying assets mature and their true value is revealed. In the event of default, the creditors can seize the collateral. The researchers assume that there is a small cost of liquidating the assets. The debt capacity of the assets (the maximum amount that can be borrowed using the assets as collateral) depends on the information state of the economy. At each date, in general there is either "good news" (the information state improves), "bad news" (the information state gets worse), or "no news" (the information state remains the same). When rollover risk is high, because debt must be rolled over frequently, the debt capacity is lower than the fundamental value of the asset and in extreme cases may be close to zero. This is true even if the fundamental value of the assets is high in all states. Interpreted differently, the model explains why discounts on ABS collateral in overnight repo borrowing (the so-called "haircuts") rose dramatically during the crisis.
Igor Makarov, London School of Business; and Guillaume Plantin, Toulouse School of Economics
Equilibrium Subprime Lending
Makarov and Plantin develop an equilibrium model of a subprime mortgage market. The model is analytically tractable and delivers plausible orders of magnitude for borrowing capacities, loan-to-income ratios, home prices, and default and trading intensities. The authors offer simple explanations for several phenomena in the subprime market, such as the prevalence of "teaser rates" and the clustering of defaults. In their model, the degree of income co-movement among households plays an important role. They find that both systematic and idiosyncratic income risks reduce debt capacities, although through quite distinct channels, and that debt capacities and home prices need not be higher when a larger fraction of income risk is idiosyncratic.
Stefan Nagel and Kenneth J. Singleton, Stanford University and NBER
Estimation and Evaluation of Conditional Asset Pricing Models
Nagel and Singleton find that several recently proposed, consumption-based models of stock returns, when evaluated using an optimal set of managed portfolios and the associated model-implied conditional moment restrictions, fail to capture key features of risk premiums in equity markets. To arrive at these conclusions we address two methodological issues that are central to assessing the goodness-of-fit of asset pricing models in which the stochastic discount factor (SDF ) is a conditionally affine function of a set of priced risk factors. First, we show that there is an optimal GM M estimator for this class of SDF s. That is, there is a choice of instruments that leads to the most efficient estimator within a class that subsumes virtually all of the GM M estimators used to date in assessing the fit of conditionally affine factor models. Second, for the (often relevant) case where a researcher is proposing a generalized SDF relative to some null model, we show that there is an optimal choice of managed portfolios to use in testing the null against the proposed alternative. The form of the optimal choice of test portfolios is derived directly from the (locally) most powerful Wald and Lagrange-multiplier tests of the null against the alternative specification of the SDF .
Ivo Welch, Brown University and NBER; and Gerard Hoberg, University of Maryland
Better Factor Portfolios and Pricing Book-to-Market Characteristics with the Fama-French Factor Model
Hoberg and Welch suggest forming portfolios by optimizing an objective function instead of by sorting. This is more parsimonious and flexible, and makes better use of the data. Empirically, they confirm the Davis, Fama, and French (2000) conjecture that the Daniel and Titman (1997) result was unique to their 1973-93 sample period. The latter's evidence is obsolete: From 1973-2008, the Fama-French model can price their sort-based incongruence portfolio (spreading HML exposures vs. book-to-market characteristics) almost perfectly. However, these authors show that it could never price optimized incongruence portfolios. Moreover, one can also construct optimized benchmark factors in lieu of the original Fama-French benchmark factors. These alternatives have higher Sharpe ratios, and some models based on them can price their own incongruence portfolios.