We propose a novel approach to measure the costs of aggregate economic fluctuations, that does not require specification of preferences or the data generating process. Using data on consumption and asset prices, we use an information-theoretic approach to recover an information kernel (I-SDF). The I-SDF accurately prices broad cross-sections of assets and has a strong business cycle component. Using the I-SDF, we find that the welfare benefits of eliminating all consumption fluctuations are large on average, and strongly countercyclical. Moreover, the cost of business cycle fluctuations is substantial, accounting for a quarter to a third of the cost of all fluctuations.
Financial Markets Group Discussion Papers DP 898