Recessions are a Time to Shine: A theory of attention allocation over the business cycle

A contentious debate in finance revolves around whether investment managers add any value for their clients. Presumably, households delegate the investment decision because these managers are able to process information and use it efficiently to generate additional return. The raises the question: What information should a manager process? We build an equilibrium model where fund managers allocate attention between macroeconomic and stock-specific payoff shocks and then use the information they learn to form mean-variance efficient portfolios. We find that what managers choose to learn depends on the state of the business cycle. In booms, aggregate shocks are not very volatile, so they learn about stock-specific payoff shocks. When two managers get signal realizations about a particular stock that are very different, their holdings of the stock in question differ. But unless their signals on every stock are very different, their portfolios will be largely similar. In recessions, attention shifts to macroeconomic shocks because they become more volatile. When two managers get very different signals about the state of the macroeconomy, they take opposite positions in every asset. This makes portfolios more dispersed in recessions. A number of the model’s predictions are supported by the data. In recessions, funds’ styles change: There is more dispersion in fund strategies and excess returns, less reliance on stock-specific earnings announcements, and more correlation with future returns for the market as a whole.