Intertemporal CAPM with Conditioning Variables

• 成果类型: Article
• 署名作者: Maio, Paulo
• 署名单位: Hanken School of Economics
• 刊物名称: MANAGEMENT SCIENCE
• ISSN/ISSBN: 0025-1909
• DOI: 10.1287/mnsc.1120.1557
• 发表日期: 2013
• 页码: 122-141
• 关键词: Asset pricing models; conditional capm; ICAPM; linear multifactor models; predictability of returns; Cross-section of stock returns; time-varying risk aversion; momentum; VALUE PREMIUM

摘要

This paper derives and tests an intertemporal capital asset pricing model (ICAPM) based on a conditional version of the Campbell-Vuolteenaho two-beta ICAPM (bad beta, good beta (BBGB)). The novel factor is a scaled cash-flow factor that results from the interaction between cash-flow news and a lagged state variable (market dividend yield or consumer price index inflation). The cross-sectional tests over 10 portfolios sorted on size, 10 portfolios sorted on book-to-market, and 10 portfolios sorted on momentum show that the scaled ICAPM explains relatively well the dispersion in excess returns on the 30 portfolios. The results for an alternative set of equity portfolios (25 portfolios sorted on size and momentum) show that the scaled ICAPM prices particularly well the momentum portfolios. Moreover, the scaled ICAPM compares favorably with alternative asset pricing models in pricing both sets of equity portfolios. The scaled factor is decisive to account for the dispersion in average excess returns between past winner and past loser stocks. More specifically, past winners are riskier than past losers in times of high price of risk. Therefore, a time-varying cash-flow beta/price of risk provides a rational explanation for momentum.