Risk-Tolerance Estimation Bias: Do Married Women and Men Differ?

Abstract

A person’s estimation of their risk tolerance, defined as the willingness to engage in financial activity whose outcome is uncertain, plays an important role in nearly all household financial decisions (Duda, Bruhin, Epper, & Schubert, 2010). Fewer errors would be observed in financial decisions if consumers were accurate in the evaluation of their risk tolerance. However, this is not the case (Shefrin, 2000). The economics and household finance literature is replete with tests of consumer biases. Three generalizations emerge from the literature. First, when people are overconfident they establish guesses about an outcome that are too high. Conversely, consumers who exhibit underconfidence tend to guess too low. Second, consumers often estimate with overconfidence when faced with general knowledge assessments (Lichtenstein, Fischhoff, & Phillips, 1999). Third, men tend to exhibit overconfidence in most household consumer decisions that involve risk.

While a great deal has been written about consumer confidence biases, very little is known about a related cognitive predisposition known as estimation bias. Grable and Roszkowski (2007) defined risk tolerance estimation bias as the systematic over- or underestimation of a person’s financial risk tolerance compared to an independent criterion. The purpose of this paper is to report on a test that measured how well married men and women were able to estimate their financial risk tolerance. This paper adds to the body of literature on sex-based estimation bias as it relates to household decisions that involve financial risk.