Correlation neglect in financial decision-making

Abstract

In many arenas, good decision-making requires the consideration of correlation. Notably, optimal portfolio theory depends upon a sophisticated understanding of the correlation between financial assets. In this paper we test people's understanding of correlation in a sequence of portfolio-allocation problems. We test the hypothesis that people neglect correlation, treating correlated random variables as uncorrelated. Our experiment uses pairs of portfolio-choice problems that have the same asset span---identical sets of attainable returns---but differ only in the assets' correlation.While any outcome-based theory of choice makes the same prediction across paired problems, the data exhibit large differences between pairs. We find strong evidence for correlation neglect as well as for the simple "1/n heuristic" of investing half of wealth in each of two assets. Keywords: portfolio choice, correlation neglect, heuristics and biases