Two unrealistic assumptions underlie standard asset-pricing models. The first assumption is that investors are in complete agreement regarding the probability distributions of future asset payoffs. The second assumption is that investment assets are not used for personal consumption and are selected solely based on their anticipated payoffs. The authors develop a simple framework for assessing how, if both assumptions are dropped, asset prices are affected.

The well-known capital asset pricing model (CAPM), introduced by Sharpe in 1964, and its subsequent iterations (the most notable of which is Merton’s intertemporal CAPM, the ICAPM) are problematic, in that they fail to explain average stock returns, being challenged primarily by the inability to capture the impacts of the value premium and momentum on price. The literature does address the weaknesses of the standard model in terms of payoff disagreement—beginning as far back as 40 years ago with Lintner (Review of Economics and Statistics, 1965)—as well as in terms of asset consumption. The “disagreement” literature tends to be largely mathematical in nature, which the authors seek to overcome by their focus on a simple approach. And the “tastes” literature cites reasons for holding assets other than a pure play on their anticipated payoffs, including holding an employer’s stock, socially responsible investing, and home bias.

The authors’ discussion of how disagreement affects price stands on a market-equilibrium platform. In a state of equilibrium, price operates to induce informed investors in aggregate to overweight (relative to the market) assets that are underweighted (relative to the market) by poorly informed investors. When investors are risk averse, however, equilibrium is only partial and prices retain vestiges of misinformation. Regardless of the persistent actions of informed investors, equilibrium cannot be achieved until misinformed investors become informed. In other words, informed investors have no incentive to bring prices back into equilibrium. The authors characterize their argument as a market-equilibrium version of the “limits of arbitrage” argument of Shleifer and Vishny (Journal of Finance, 1997).

Market equilibrium also regulates price in the realm of investor preferences, or tastes, for assets viewed as consumption goods. As in the case of misinformed investors, asset prices must induce overweighting by investors who do not attribute utility to assets for consumption purposes, offsetting the underweighting by those who do. Equilibrium is only partial (i.e., prices do not conform to the CAPM) when the amount of specific assets held by the two groups is not perfectly offsetting. In an intertemporal setting, equilibrium pricing does exist when investor decision making is driven by the covariances of asset returns with common return factors or state variables.

The authors undertake a series of calibrations of their model as they seek to determine to what degree expected returns and misinformed beliefs impact pricing. The calibrations show that distortions in expected returns can be large when misinformed investors or investors who value assets as consumption goods account for substantial invested wealth, invest in a wide range of assets, take positions far different from the market portfolio, or underweight assets with returns not highly correlated with the returns on the assets they overweight. Although the impact on expected returns can be large, the impact on price was not determinable.

Useful insights gained by the authors through their calibration studies include the following observations. Informed investors earn positive alphas whereas uninformed investors earn negative alphas. And although the price impact of decisions made by investors who have a taste for assets as consumption goods is similar to the price impact of decisions made by misinformed investors, the price effects induced by misinformed investors are temporary whereas those arising from investors with differing tastes are not. Additionally, the authors conclude that because costs are equal-opportunity distortions (i.e., they equally hinder both misinformed and informed investors), the net effect of costs on market efficiency is unclear.