Home/Systematic Trading/Behavioral Portfolio Construction

Behavioral Portfolio Construction

portfolio-constructionLevel 3 — Advanced

Prerequisites

Portfolio ConstructionEdge Identification

What It Is

Behavioral portfolio construction applies prospect theory and utility function modeling to build portfolios that match investors' actual psychological preferences — not just their stated risk tolerance — including loss aversion, reflection effects, and target-relative preferences.

Correct Execution

  • Diagnose loss aversion separately from risk aversion — they have different magnitudes and different effects on portfolio construction
  • Use asymmetric utility functions: investors feel roughly twice as much pain from being 20% below target as they feel satisfaction from being 20% above
  • Build around investment targets, not abstract risk tolerance scores — "I want $1.25 in 5 years" is a more actionable constraint than "moderate risk"
  • Risk questionnaires that sum scores without connecting to utility functions are medically invalid — treat them like a doctor who diagnoses by average temperature
    Do not conflate reflection with loss aversion: reflection (risk-seeking when in the loss domain) and loss aversion (asymmetric pain/pleasure) are different phenomena with different portfolio implications

Progression Levels

Diagnostic Tree

Coaching Cues

  • "If you run an optimization with only risk aversion and then add moderate loss aversion, you often see 20-30% difference in equity allocation. That's not marginal — that's a different portfolio." — David Berns, FWM S4E16
  • "Be as medical as possible in diagnosing risk preferences. Researchers have given us 50 years of tools we're mostly ignoring." — David Berns, FWM S4E16
  • "The target is what separates this from mean-variance. Once you have a target and asymmetric preferences, standard finance theory breaks down." — Martin Tarlie, FWM S6E13

Common Errors

  1. Using sum-score risk questionnaires to calibrate loss aversion: Sum scores cannot distinguish risk aversion from loss aversion; the result is systematically wrong portfolios → Build or use questionnaires that separately measure the distinct dimensions of risk preference
  2. Treating all behavioral preferences as time-varying: Some apparent preference changes are genuine (life events); most are just measurement error from the original assessment → Distinguish signal from noise in expressed preferences; don't rebuild the portfolio every time a client expresses fear
  3. Not connecting the behavioral diagnosis to a utility function: "High loss aversion" is meaningless unless you can translate it into a portfolio weight → Connect each behavioral measurement to a mathematical parameter in the utility function that produces a specific portfolio

Edges

⚡ Conventional Wisdom Is Wrong

Loss Aversion Can Shift Optimal Equity Allocation by 20-30 Points — Risk Questionnaires Miss This Entirely

portfolio-constructionbehavioral-portfolio-construction →

Standard risk questionnaires produce a risk tolerance score that maps to a generic equity allocation (e.g., "moderate" → 60% equity). These questionnaires measure risk aversion (trade-off between expected return and variance) but not loss aversion (asymmetric pain from losses relative to gains). For an investor with moderate risk aversion but high loss aversion, the optimal equity allocation incorporating loss aversion can be 30-40% — not 60%. The questionnaire gives 60% and the client fires their advisor after the first major drawdown. The miscalibration is at the diagnostic stage, not the portfolio stage.

What most people do
Administer a sum-score risk questionnaire; map the score to a standard allocation bucket; proceed with portfolio construction.
What the best do
Separately measure risk aversion and loss aversion using distinct questions that isolate the asymmetry in the client's response to gains and losses; build the portfolio from the client's actual utility function parameters.
Why it's an edge: Advisors and portfolio managers who correctly diagnose client behavioral profiles retain clients through drawdowns because the portfolio was built for their actual pain threshold, not their stated one. Client retention IS alpha.
How to exploit: Add one specific question to client onboarding: "If your portfolio fell 20% in a year, how would you feel (1-10) vs if it rose 20% in a year (1-10)?" The ratio of the two scores is a rough loss aversion estimate. If the fall feels worse by more than 2x, the client has above-average loss aversion and standard equity allocations will be too high for them.
Cross-domain parallel
In expectancy frameworks for sports betting, risk of ruin modeling distinguishes bankroll drawdown tolerance from statistical edge — the same concept as risk aversion vs loss aversion applied to different domains.
David Berns, "How Do You Build Portfolios for Human Beings," FWM S4E16, 2021-08-16
🔑 Hidden Causal Lever

When Clients Change Their Mind During a Drawdown, the Diagnostic Was Wrong — Not the Market

portfolio-constructionbehavioral-portfolio-construction →

When a client calls during a 30% drawdown and wants to reduce equity exposure, the standard framing is "the client changed their risk tolerance." The correct framing is "our initial assessment of their risk tolerance was wrong — or we measured the wrong variable." Risk preferences don't actually change meaningfully based on market events; what changes is the salience of risk that was always there but not felt. The behavioral failure was in the onboarding process, not in the market. Responding by modifying the portfolio in the drawdown punishes the client for the advisor's diagnostic error.

What most people do
Adjust the portfolio to reflect the client's "new risk tolerance" during drawdowns; re-optimize toward a more conservative allocation.
What the best do
Use the drawdown moment as new information about the original diagnostic quality — update the behavioral model, not the portfolio in the moment; the correct time to rebalance toward the client's true preference is after the market stabilizes, not at the bottom.
Why it's an edge: Advisors who hold the portfolio through the drawdown — because they built it correctly in the first place — generate better outcomes. The advisor who adjusts at the bottom locks in the client's maximum loss.
How to exploit: Build a pre-commitment document with each client before investing: "We have designed this portfolio for a maximum drawdown of X%. If the portfolio falls to -X%, we have agreed in advance not to reduce equity exposure during the drawdown. The decision to reduce exposure, if any, will be made at the next annual review with full market context." Get the client's signature. This forces the diagnostic to happen pre-crisis.
David Berns, FWM S4E16, 2021-08-16; Martin Tarlie, FWM S6E13, 2023-08-07
🔑 Hidden Causal Lever

Investment Targets Create Asymmetric Utility That Mean-Variance Optimization Cannot Handle

portfolio-constructionbehavioral-portfolio-construction →

Mean-variance optimization treats utility as a continuous, symmetric function of returns — more is always better, losses and gains of the same magnitude matter equally. Real investors have targets ("I need $1.5M in 10 years to retire") that create hard asymmetry: being 20% below the target at the deadline is categorically worse than being 20% above it. Once a target is introduced, the utility function has different risk aversion parameters above and below the target. Single-period MVO with a symmetric utility function produces the wrong answer for any investor with an explicit investment objective.

What most people do
Apply mean-variance optimization; set a "risk tolerance" input; arrive at a single efficient portfolio; apply it regardless of whether the client has a target.
What the best do
Build target-relative optimization with asymmetric utility: higher risk aversion below target (preserving survival) and lower risk aversion above target (capturing upside); accept that this requires simulation rather than closed-form solution.
Why it's an edge: The entire financial planning industry operates on MVO with symmetric utility despite the fact that virtually every retail investor has an explicit target (retirement, education funding, home purchase). Applying the correct mathematical framework to this near-universal use case is an underexploited opportunity.
How to exploit: For any client with an explicit financial goal and timeline, define the target wealth at the deadline. Run a Monte Carlo simulation of portfolio outcomes at that horizon. Optimize for the probability of reaching the target (not for Sharpe). Compare the resulting allocation to the MVO output — for most clients with near-term targets, the target-aware optimization will be more conservative than MVO recommends.
Martin Tarlie, "Bridging the Gap Between Financial Planning and Portfolio Management," FWM S6E13, 2023-08-07
🔑 Hidden Causal Lever

Reflection Effect Creates Risk-Seeking Below Target — Portfolio Construction Must Account For This

portfolio-constructionbehavioral-portfolio-construction →

Prospect theory's reflection effect means investors who are below their investment target will take MORE risk, not less — the opposite of what standard risk-aversion models predict. A portfolio optimized for a target-relative investor must have different risk parameters above and below the target. Most behavioral portfolio construction only models loss aversion, missing this second asymmetry entirely.

What most people do
Model investors as uniformly loss-averse. Assume that investors below target are even more risk-averse than usual.
What the best do
Model the reflection effect explicitly: investors above target are risk-averse (protect gains), investors below target are risk-seeking (gamble to recover). Portfolio construction with target-relative utility should be more aggressive below target and more conservative above — the opposite of standard practice.
Why it's an edge: Standard portfolio construction systematically under-allocates to risk for investors below target (when they actually want more risk) and over-allocates for investors above target (when they actually want less). Getting this asymmetry right improves both returns and client satisfaction.
How to exploit: Identify your or your client's target return. Build portfolio rules that increase risk allocation when below target and decrease when above. Test this regime-dependent sizing against a constant-risk approach. The behavioral alignment should reduce client exits during drawdowns.
From Level 4 and Diagnostic Tree symptom 4 — prospect theory reflection effect application

Sources

  • David Berns, "How Do You Build Portfolios for Human Beings," Flirting with Models S4E16, 2021-08-16 — prospect theory in portfolio construction; loss aversion measurement; risk questionnaire critique
  • Martin Tarlie, "Bridging the Gap Between Financial Planning and Portfolio Management," Flirting with Models S6E13, 2023-08-07 — target-relative optimization; asymmetric utility functions; multi-period optimization with behavioral constraints