We’re using cookies, but you can turn them off in Privacy Settings. Otherwise, you are agreeing to our use of cookies. Learn more in our Privacy Policy.


Examining a sizable sample of investors at a large Italian bank, the authors explore the effectiveness of two measurements of risk tolerance—namely, an investor’s own assessment of his ability to bear risk and an examination of the composition of the investor’s portfolio. The authors look for inconsistencies between the two metrics and find that inconsistencies are more evident in individuals with certain demographic traits.

How Is This Research Useful to Practitioners?

Financial risk tolerance inversely correlates with risk aversion. It is hard to measure because it is not directly observable. Numerous attempts to gauge it have produced mixed results.

The authors consider the effectiveness of self-assessment, a subjective metric that asks the investor to rate her level of risk tolerance. They then observe the investor’s behavior objectively by evaluating the level of risk in her portfolio. There is often a gap between what investors think their risk tolerance is and how they invest.

The authors’ findings will be useful to wealth managers, financial planners, and students of behavioral finance who wish to guide their clients toward a better understanding of their risk tolerance and investment suitability.

How Did the Authors Conduct This Research?

The authors analyze a sample of 2,374 clients of a primary Italian bank over 2012–2013. Respondents complete a three-part questionnaire. The first part is a single question to obtain a financial risk tolerance self-assessment, the second part is the psychometrically validated questionnaire, and the third part requests a client’s economic and financial situation along with certain sociodemographic items. Answers provided in the Markets in Financial Instruments Directive (MiFID) questionnaire allow the authors to obtain additional portfolio composition information and other socioeconomic variables.

Its flaws notwithstanding (including cognitive biases and strategic motivation), the first measure (subjective) continues to be widely used by practitioners and academics because of its simplicity. The objective risk measure for risk tolerance uses value at risk (VaR), which provides a direct expression of the client’s assumed risk without having to engage in what could be a subjective classification of assets as risky versus nonrisky.

A disparity often exists between how investors think they can tolerate risk and what level of risk their portfolio evidences. For this reason, the authors introduce a third measure of risk tolerance assessment: a psychometrically validated questionnaire to cross-validate the other two metrics by distinguishing self-assessment gaps from those in the composition of the investor’s portfolio. The 13-item questionnaire tests risk comfort and experience as well as investment risk and speculative risk and reduces cognitive bias because its questions are intuitive and easy to answer.

To facilitate comparison, the authors sort all three risk tolerance measures—the self-assessment metric, the VaR metric, and the psychometric questionnaire—into quartiles. They first use the questionnaire quartile to validate the other two measures. If the questionnaire quartile equals the VaR quartile, this result cross-validates the portfolio composition risk and creates a possible self-assessment gap. If the self-assessment quartile exceeds (is less than) the VaR quartile, the result is an overvaluation (undervaluation) gap.

If the questionnaire quartile equals the self-assessment quartile, this outcome cross-validates the self-assessment and indicates a potential portfolio composition gap. If the VaR quartile exceeds the self-assessment quartile, there is an overexposure-to-risk gap. If the self-assessment quartile exceeds the VaR quartile, the opposite is true.

The authors collect additional socioeconomic and demographic data on each respondent to aid their investigation of the gaps. They then use a least-squares estimation to reveal both the determinants of the self-assessment and portfolio composition gaps and the nature of the gaps (e.g., valuation and exposure discrepancies).

The authors draw several conclusions from their analysis. Inconsistencies between financial risk tolerance self-assessments and portfolio composition are more likely with those less literate in finance, those with no children, and those with higher levels of income. Men, homeowners, and those who save extensively exhibit lower self-assessment discrepancies. Individuals with larger amounts of wealth invested over shorter time frames and those with low debt show lower portfolio composition gaps. The financially cautious are more consistent in their manifestation of financial risk tolerance.

Abstractor’s Viewpoint

Accurate assessments of individuals’ financial risk tolerance are a critical input into the wealth management and financial planning process. Because financial risk tolerance is not directly observable, efforts to measure it effectively have been inconsistent. The authors take a more granular approach in evaluating the disparity between objective and subjective assessment measures to better understand why such gaps exist. Looking at the socioeconomic component sheds important light on the problem. It would be interesting to apply the authors’ methodology to samples of investors in different countries to see whether their conclusions remain valid or vary with different socioeconomics and demographics.

About the Author(s)

Marc L. Ross CFA

Marc L. Ross, CFA, is a senior compliance consultant at John Hancock.