We propose a latent variables approach within a present-value model to estimate the expected returns and expected dividend growth rates of the aggregate stock market. This approach aggregates information contained in the history of price-dividend ratios and dividend growth rates to predict future returns and dividend growth rates. We find that returns and dividend growth rates are predictable with R-squared values ranging from 8.2% to 8.9% for returns and 13.9% to 31.6% for dividend growth rates. Both expected returns and expected dividend growth rates have a persistent component, but expected returns are more persistent than expected dividend growth rates. We propose a latent variables approach within a present-value model to estimate the time series of expected returns and expected dividend growth rates of the aggregate stock market. Specifically, we treat conditional expected returns and expected dividend growth rates as latent variables that follow an exogenously specified time-series model, and we combine this model with a Campbell and Shiller (1988) present-value model to derive the implied dynamics of the price-dividend ratio. Then, using a Kalman filter to construct the likelihood of our model, we estimate the parameters of the model by means of maximum likelihood. We find that both expected returns and expected dividend growth rates are time-varying and persistent, but expected returns are more persistent than expected dividend growth rates. The filtered series for expected returns and expected dividend growth rates are good predictors of realized returns and realized dividend growth rates, with R 2 values ranging from 8.2% to 8.9% for returns and 13.9% to 31.6% for dividend growth rates.We consider an annual model to ensure that the dividend growth predictability we find is not simply driven by the seasonality in dividend payments. 1 However, using an annual dividend growth series implies that we need to take a stance on how dividends received within a particular year are reinvested. Analogous to the way in which different investment strategies lead to different risk-return properties of portfolio returns, different reinvestment strategies for dividends within a year result in different dynamics of dividend growth rates. 2We study two reinvestment strategies in detail. First, we reinvest dividends in a 30-day T-bill, which we call cash-reinvested dividends. Second, we reinvest dividends in the aggregate stock market, which we refer to as market-reinvested dividends.Market-reinvested dividends have been studied widely in the dividend-growth and returnforecasting literature. 3 We find the reinvestment strategy to matter for the time-series properties of dividend growth. For instance, the volatility of market-reinvested dividend growth is twice as high as the volatility of cash-reinvested dividend growth. Within our model, we derive the link between the time-series models of dividend growth rates for different reinvestment strategies. This analysis demonstrates that if expected cashreinvested div...

No abstract

We study an institutional investment problem in which a centralized decision maker, the Chief Investment Officer (CIO), for example, employs multiple asset managers to implement investment strategies in separate asset classes. The CIO allocates capital to the managers who, in turn, allocate these funds to the assets in their asset class. This two-step investment process causes several misalignments of objectives between the CIO and his managers and can lead to large utility costs for the CIO. We focus on (1) loss of diversification, (2) unobservable managerial appetite for risk, and (3) different investment horizons. We derive an optimal unconditional linear performance benchmark and show that this benchmark can be used to better align incentives within the firm. We find that the CIO's uncertainty about the managers' risk appetites increases both the costs of decentralized investment management and the value of an optimally designed benchmark. Disciplines Finance | Finance and Financial Management CommentsAt the time of publication, author Jules van Binsbergen was affiliated with Stanford University. Currently, he is a faculty member at the Wharton School at the University of Pennsylvania. ABSTRACTWe study a decentralized investment problem in which a CIO employs multiple asset managers to implement and execute investment strategies in separate asset classes. The CIO allocates capital to the managers who, in turn, allocate these funds to the assets in their asset class. This two-step investment process causes several misalignments of objectives between the CIO and his managers and can lead to large utility costs on the part of the CIO. We focus on i) loss of diversification ii) different appetites for risk, iii) different investment horizons, and iv) the presence of liabilities. We derive an optimal unconditional linear performance benchmark and show that this benchmark can be used to better align incentives within the firm. The optimal benchmark substantially mitigates the utility costs of decentralized investment management. These costs can be further reduced when the CIO can screen asset managers on the basis of their risk appetites. Each manager's optimal level of risk aversion depends on the asset class he manages and can differ substantially from the CIO's level of risk aversion.

No abstract

No abstract