Asset Pricing.

Members and guests of the NBER's Program on Asset Pricing met in Chicago on November 5. Program Director John H. Cochrane of the University of Chicago chose the following papers for discussion:

George Chacko, Harvard University, and Luis M. Viceira, NBER and Harvard University, "Dynamic Consumption and Portfolio Choice with Stochastic Volatility in Incomplete Markets" (NBER Working Paper No. 7377)

Discussant: Michael W. Brandt, NBER and University of Pennsylvania 

Anthony W. Lynch, New York University, "Portfolio Choice and Equity Characteristics: Characterizing the Hedging Demands Induced by Return Predictability"

Discussant: Jessica Wachter, Harvard University

Gregory R. Duffee, University of California, Berkeley, "Forecasting Future Interest Rates: Are Affine Models Failures?"

Discussant: Michael Johannes, University of Chicago

Owen Lamont, NBER and University of Chicago, and Christopher Polk, Northwestern University, "The Diversification Discount: Cash Flows versus Returns" (NBER Working Paper No. 7396)

Discussant: Narasimhan Jegadeesh, University of Illinois 

Martin Lettau and Sydney C. Ludvigson, Federal Reserve Bank of New York, "A Cross-Sectional Test of Linear Factor Models with Time-Varying Risk Premiums"

Discussant: Ravi Jagannathan, NBER and Northwestern University

Nicholas Barberis, NBER and University of Chicago; Ming Huang, Stanford University; and Tano Santos, University of Chicago, "Prospect Theory and Asset Prices" (NBER Working Paper No. 7220)

Discussant: Kent D. Daniel, NBER and Northwestern University

Chacko and Viceira ask what the optimal savings and portfolio allocations are for long-horizon investors when the variation in stock return volatility is predictable. They find that the optimal portfolio demand for stocks varies with investors' risk aversion, but only slightly with their willingness to substitute consumption for investment. By contrast, the optimal level of savings relative to wealth depends on both risk aversion and willingness to substitute consumption for saving. For long-horizon, risk-averse investors it is optimal to reduce portfolio holdings of stocks when changes in return volatility are negatively correlated with excess stock returns. This helps investors to hedge their exposure to volatility risk. The magnitude of the reduction in portfolio demand increases with the size of the correlation and, more importantly, with the persistence of changes in volatility. For the U.S. stock market, the correlation is negative and large, and low-frequency shocks to volatility are highly persistent, the authors find,

Lynch examines portfolio allocation across equity portfolios formed according to size, book-to-market value, and the like. He assesses the impact of return predictability on portfolio choice for an investor whose relative risk aversion coefficient is 4. Return predictability with dividend yield causes this investor early in life to tilt away from both high book-to-market stocks and small stocks. The investor wants to short-sell a low book-to-market portfolio in order to hold a high book-to-market portfolio. Lynch also calculates the utility cost of using a value-weighted equity index or of ignoring predictability. Investors who use a value-weighted equity index would give up a much larger fraction of their wealth to have access to book-to-market portfolios instead of stock portfolios formed on the basis of size. Finally, Lynch finds that investors would give up a much larger fraction of their wealth to have access to dividend yield information instead of term spread information, but that term spread does hav e incremental benefits over and above using dividend yield alone.

Duffee finds that the class of affine models typically studied does a poor job of forecasting future changes in Treasury yields. He contends that better forecasts are produced by assuming that yields are Martingales. The failure of these models is driven by one of their key features: the compensation that investors receive for facing risk is a multiple of the variance of the risk. This means that risk compensation cannot vary independently of interest rate volatility. Duffee also estimates a class of models that is broader than the affine class. These "essentially affine" models retain the tractability of the standard affine class but allow the compensation for interest rate risk to vary independently from interest rate volatility. This additional flexibility proves useful in forming accurate forecasts of future yields.

Lamont and Polk study diversified firms and find that for discount firms, the whole is less than the sum of the parts; for premium firms, the whole is greater. These differences in value must be attributable to differences in either future cash flows or future returns. The authors find that the expected returns on securities of diversified firms vary systematically with relative value. Discount firms have significantly higher subsequent returns than do premium firms. Slightly more than half of the cross-sectional variation in excess values is attributable to variation in expected future cash flows, with the remainder attributable to variation in expected future returns and to co-variation between cash flows and returns.

Lettau and Ludvigson ask if theoretically based asset pricing models such as the CAPNI and the consumption CAPM -- referred to jointly as the (C)CAPM -- can explain the cross section of average stock returns. Central to their approach is a variable that proxies for fluctuations in the log consumption-aggregate wealth ratio and is probably important for summarizing conditional expectations of excess returns. They demonstrate that such conditional factor models can explain a substantial fraction of the cross-sectional variation in portfolio returns. These models perform much better on portfolios sorted by size and book-to-market ratios than unconditional (C)CAPM specifications and about as well as the three-factor Fama-French model. This CAPM specification can account for the difference in returns between low and high book-to-market firms and shows no sign of residual size or book-to-market effects.

Barberis, Huang, and Santos study asset prices in an economy where investors care about fluctuations in the value of their financial wealth more than is justified by a concern about consumption alone. Their framework can be helpful in explaining the mean, volatility, and predictability of stock returns. In contrast to consumption-based models, the authors also predict that stock returns will be only weakly correlated with consumption. Their model's design is influenced by prospect theory and by experimental evidence on the effect of prior outcomes on risky choice.