What Drives Investors’ Portfolio Choices? It’s Not What You May Think

Investors are often seen as choosing portfolios based on their preferences. Each decides how much risk to take, allocates between stocks and safe assets, and builds a portfolio that reflects their beliefs. But in reality, observed portfolios tell only part of the story. Many investors who could invest in stocks simply do not, even when theory suggests they should. This paper introduces a new perspective. Portfolio choices are not driven by preferences alone, they are also shaped by frictions, small costs, inertia, and default options. The result is a subtle but powerful mechanism. What investors hold is not always what they want. Understanding this gap is key to explaining how portfolios are actually formed.

What Drives Investors’ Portfolio Choices? Separating Risk Preferences from Frictions

• Taha Choukhmane, Tim De Silva
• The Journal of Finance, 2026
• A version of this paper can be found here
• Want to read our summaries of academic finance papers? Check out our Academic Research Insight category

Key Academic Insights

Most investors actually prefer stocks
The paper shows that over 90% of investors prefer stock market participation. This is much higher than what we observe in real-world portfolios. The gap suggests that nonparticipation is not primarily driven by risk aversion.

Defaults strongly influence behavior
Investors react very differently depending on how they are initially allocated. Those defaulted into safe assets gradually move into equities. Those defaulted into equities rarely opt out. This asymmetry reveals preferences. Investors tend to keep stocks when given them, but delay acting when they must opt in.

Trading behavior reveals hidden preferences
Active decisions are the key signal. Investors who move away from a zero-equity default reveal a preference for stocks. Investors who move out of equity reveal the opposite. By comparing these behaviors across different defaults, the paper recovers preferences that are otherwise unobservable.

Frictions, not risk aversion, drive nonparticipation
The model estimates a moderate level of risk aversion (≈ 2.5). At the same time, even a relatively small adjustment cost (~$156) is enough to explain why many investors remain out of the market. This overturns the standard interpretation. Investors are not avoiding stocks because they fear them. They are avoiding them because acting is costly.

Observed portfolios misrepresent true preferences
Observed participation rates are significantly lower than preferred participation. Similarly, observed equity allocations differ from preferred ones across the life cycle. This creates a wedge. What we see in the data is not what investors actually want.

Life-cycle patterns align with theory once frictions are removed
Preferences follow a standard pattern of high participation at all ages and equity exposure declining with age. Observed data does not match this. But once frictions are accounted for, behavior aligns with textbook life-cycle models.

Practical Applications for Investment Advisors

Rethink how portfolio choices are formed

Do not assume that portfolios reflect preferences. Many investors are underexposed to equities not by choice, but because of inertia and friction.

Pay close attention to defaults
Initial allocations matter. They have long-lasting effects on portfolio outcomes. Structuring better defaults can significantly improve investor behavior.

Reduce complexity wherever possible
Even small barriers can prevent action. Simplifying decisions, reducing steps, and lowering cognitive load can help investors move closer to their preferred allocations.

Be cautious when interpreting risk tolerance
Low equity exposure does not necessarily imply high risk aversion. It may reflect inaction rather than preference.

How to Explain This to Clients

“Many people think that if someone is not investing in stocks, it means they are too risk-averse. But that is often not the case. In reality, most investors actually want to be invested in the market. They just do not always take action. Sometimes because it feels complex. Sometimes because they never get around to it. When people are automatically invested, they usually stay invested. This shows that the issue is not fear of risk, but the difficulty of taking the first step.”

The Most Important Chart from the Paper

Figure 2 plots the observed portfolio responses for employees hired within 12 months of their employer changing the default asset allocation τ years after they were hired. The left panel shows the stock market participation rate and the right panel shows average unconditional stock shares of current employer retirement wealth. The blue lines are employees automatically enrolled in a money market fund; the red lines are employees automatically enrolled in a TDF.

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged and do not reflect management or trading fees, and one cannot invest directly in an index.

Abstract

This paper studies why many investors do not participate in the stock market despite a positive equity premium. It separates risk preferences from participation frictions using quasi-experimental variation in 401(k) default asset allocations. By comparing investor behavior under different defaults, the authors identify preferences from active choices. They find that most investors prefer stock market participation and exhibit moderate risk aversion, while limited participation is primarily driven by adjustment frictions. The results highlight the importance of defaults, inertia, and small costs in shaping portfolio decisions and reconciling observed behavior with standard life-cycle models.