Where Might Biases Creep Into Investment Advice?

If a friend of yours were to approach you and ask for financial advice, what information would you want in order to give a reasonable recommendation? Goals, current financial situation, investment and risk preferences, life expectancy, and so on?

What if you were an advisor (which a fair number of our readers here on Morningstar.com are)? What would you take into account if a prospective client asked you for advice? You'd probably include similar factors, right?

However, would you give different advice based solely on the person's gender? That is, after you've already taken into account the person's preferences, life expectancy, and so on? My guess is that for most people, the answer is no. The gender of the person shouldn't matter for the advice we give after we've already taken into account the specific details of the person's finances.

Surprisingly, a number of studies have shown that the advice advisors give does differ by gender after other factors are taken into account. For example, in a 2012 audit study (1), hired actors were instructed to visit financial advisors and pose as prospective clients (yeah, I know …). The researchers then recorded the advice given to each participant and found that people with identical portfolios were nevertheless treated differently. Female investors were asked about their personal and financial situations less often than men, and women were also advised to have more liquidity, less international exposure, and fewer actively managed funds.

Depending on the cause, those results could be very troubling. At Morningstar, we wanted to better understand why that happens--and whether it is an unintentional bias creeping into the advice process, or whether it was something reasonable and explicable.

So, we ran a study, in which we randomly assigned both advisors and everyday people to one of three scenarios. We'll focus on the advisors for now, and return to the study with nonadvisors shortly. The default version for advisors was as follows:

Imagine a prospective client is meeting with you for the first time. The client is 51 years old and has an annual income of $105,000 and investable wealth of $560,000, all of which is currently invested in short-term certificates of deposit. The client also has a moderate risk preference. All of these resources will be dedicated to maintaining the client's standard of living during retirement. The client's goal is to retire at age 65 and is planning for a retirement that lasts 30 years.

In this default version, no gender was specified for the client. The second version used female names and pronouns for the prospective client; for robustness, we tried two different names, Sarah and Amanda. The third version used male names (David and Daniel) and pronouns for the client.

We then asked the advisors a series of questions about recommendations they would give to this prospective client: high-level asset allocation, active/passive mix, international versus domestic, and so on.

Steps in a Process If you think about speaking with an advisor and receiving advice, one can break up the process in terms of a series of steps. For example:

Initial Conversation => Information Gathering => Generate Recommendation => Giving Advice

This study focuses squarely on the "generate recommendation" stage: When an advisor already has the core information about a prospective client, what recommendations are developed?

Our Results Our results were quite encouraging. We found no statistically significant difference between the recommendations given to men or to women (or to the hypothetical prospect for whom gender was not known). In other words, when provided with the financial situation and preferences of the individual, advisors focused on that situation and those preferences, and not gender.

We found the same thing with our analysis of noninvestment advice provided by everyday people. For them, we posed a similar question, asking about their advice for a male, female, or non-gender-specified friend on how to handle their noninvestment finances. We found that given a clear scenario and details about the friend's life and preferences, the respondents did not take gender into account in their recommendations. Which again, is good news.

While this research study cannot conclude the issue, it does point us to other steps in the advice-giving process. If a gender bias exists, it likely does not occur in the "generating recommendation" stage. It may be in the initial conversation or information-gathering stage: Some female clients report being ignored or underserved by their advisor in these interactions (2, 3). We intend to do further research in this area to better understand the advisor-investor interaction and see if and where biases creep in. We believe that a deeper understanding of this interaction can help both parties avoid pitfalls and be better served by the process.

Morningstar's study was part of The Investor Success Project.

References (1) Mullainathan, S., Noeth, M., & Schoar, A. 2012. "The Market for Financial Advice: An Audit Study." The National Bureau of Economic Research. Working Paper 17929. https://www.nber.org/papers/w17929.

(2) Hewlett, S., & Moffitt, A. "Harnessing the Power of the Purse: Female Investors and Global Opportunities for Growth." Center for Talent Innovation. May 1, 2014. https://www.talentinnovation.org/assets/HarnessingThePowerOfThePurse_ExecSumm-CTI-CONFIDENTIAL.pdf.

(3) EY Global Wealth & Asset Management. 2017. “Women and wealth: The case for a customized approach.” https://www.ey.com/Publication/vwLUAssets/EY-women-investors/$FILE/EY-women-and-wealth.pdf.