Skip to main content
Behavioural finance

Loss aversion in retirement drawdown

Daniel Kahneman and Amos Tversky showed in 1979 that humans feel losses roughly 2 to 2.5 times more intensely than equivalent gains. For working-age investors, this rarely matters — you keep contributing through bear markets. For retirees in drawdown, it's catastrophic: panic-selling near the bottom of a market crash can permanently lock in losses you'd otherwise have recovered from. Here's the mechanic, the real UK data on how much it costs, and the structural commitments that can save you from yourself.

Educational only. Behavioural finance research isn't financial advice; individual circumstances vary. Consider regulated retirement planning advice. Not financial advice.

What loss aversion actually means

The classic experiment: would you accept a coin flip where heads wins you £150 and tails loses you £100? Statistically the expected value is positive (£25 per flip on average). But most people refuse the bet. The pain of losing £100 outweighs the pleasure of winning £150.

Kahneman and Tversky's Prospect Theory quantified this asymmetry. People typically need a potential gain of roughly 2-2.5x the potential loss to take an even-money bet. This is "loss aversion" — not irrationality, but a deep evolutionary feature of human decision-making.

Why this matters for investment: when your portfolio drops 20%, the emotional weight of that loss feels equivalent to a 40-50% gain that never happened. The pain creates a strong impulse to "stop the bleeding" by selling.

Why drawdown phase makes loss aversion much more dangerous

In accumulation phase (working years), market crashes are actually opportunities: your monthly contributions buy more units at lower prices. You don't depend on the portfolio's current value — you depend on its long-run growth.

In drawdown phase, the maths inverts:

Result: the same 30% market crash that's a buying opportunity for a 35-year-old can be a retirement-ending event for a 65-year-old who panic-sells.

The real UK retirement-drawdown data

The 2020 COVID crash and 2022 bond/equity correction give us recent real-world data on UK retiree behaviour:

March 2020 COVID crash

Global equity markets fell ~34% in 5 weeks. SIPP / pension drawdown statistics (FCA, HL, ii data):

Estimated lifetime cost of panic-selling in March 2020: 1-3% of total retirement wealth for a typical 65-year-old, depending on portfolio mix and re-entry behaviour.

Late 2022 bond/equity correction

The 2022 episode was different: bonds AND equities fell simultaneously (the "60/40" portfolio had its worst year since 1937). Many traditional balanced portfolios fell 20-25%.

The exact behavioural mechanism

Research from Vanguard, Morningstar and the FCA suggests a typical sequence during a market correction:

  1. Week 1-2 of decline: investor anxious but holds. Tells themselves "it's just a correction".
  2. Week 3-4: news intensifies. Investor checks portfolio daily. Anxiety builds.
  3. Week 4-6 (or peak fear point): investor's loss aversion peaks. The pain of further losses feels unbearable. They decide "I can't take any more" and sell.
  4. Week 6-12: market recovers. Investor in cash. They wait for the "right time" to re-enter, but the right time never feels right (markets always seem to have just rallied).
  5. Week 12-26: investor finally re-enters, typically near the previous peak. Total loss vs holding: usually 15-30% of portfolio value.

This sequence repeats across every market crisis. The fundamental error isn't the initial concern (that's rational) — it's the action taken at peak fear.

Why "just stay calm" advice doesn't work

Telling a panicking retiree to "remember the historical data" or "trust the long-run market" rarely helps. The reason: loss aversion is a deep psychological feature, not a knowledge gap. The investor often knows intellectually that they should hold; they just can't override the emotional pull.

Some specific reasons the standard advice fails:

Structural solutions that work

If pure willpower won't override loss aversion, the answer is to build structural commitments that make panic-selling difficult or impossible. Three approaches that genuinely work:

1. The 2-3 year cash buffer

Keep 2-3 years of essential spending in cash, money market funds, or short-dated gilts. When equity markets crash, you draw down the buffer for living expenses rather than selling equities. The buffer protects you from being a forced seller at the worst time.

Practical implementation:

This is the single most important structural commitment for drawdown investors. It directly addresses the sequence-of-returns risk and removes the financial necessity to sell during downturns.

2. Annuity floor for essential spending

Buy a level or inflation-linked annuity that covers your essential lifetime spending (basic food, utilities, council tax, basic care needs). Your drawdown portfolio then funds discretionary spending (holidays, hobbies, gifts).

Effect: equity-market value swings only affect discretionary spending. You're not at risk of "I won't be able to pay my mortgage" because the annuity covers essentials. Loss aversion still triggers but with much lower stakes.

Current 2026/27 annuity rates: a 65-year-old can get ~6-7% from a level annuity, ~4.5-5% from an inflation-linked one. A £200,000 annuity purchase generates ~£9,000-£14,000 of lifetime income.

3. Automatic, pre-committed rebalancing

Set up a written investment policy statement (IPS) BEFORE you retire. Specify:

The IPS becomes a pre-commitment device. When markets crash and your equity drops to 50%, the IPS instructs you to BUY more equity, not sell. Following a written rule is easier than overriding emotion in real time.

How a good financial adviser helps

The biggest financial value an adviser delivers isn't asset selection or tax optimisation — it's preventing client behavioural mistakes. Vanguard's "Adviser Alpha" research estimates the average value of an adviser at ~3% per year, of which roughly half (~1.5%) comes from preventing panic-selling and bad behavioural decisions.

Specifically:

For drawdown investors, the £1,500-£3,000/year cost of a good adviser is often more than repaid by a single avoided panic-sell. The economics make sense even for relatively modest portfolios in retirement.

Personal anti-panic checklist

If you're managing your own drawdown:

Worked example: the cost of selling in March 2020

Sarah, 67, £500,000 SIPP in drawdown, 60/40 split (60% global equity / 40% global bonds, GBP hedged). Withdrawing £25,000/year (5%).

Scenario A: Held through March 2020

Scenario B: Sold equities to cash on 20 March 2020 (near bottom)

Ongoing impact

The 4% permanent gap continues compounding. Over a 25-year retirement, that £20,000 today is approximately £55,000 of foregone future wealth (assuming 5% real growth). For Sarah, the March 2020 panic cost her potentially £55,000 of inheritance for her children, or 2 years of retirement spending.

Frequently asked questions

Is loss aversion always bad for investors?

No. It usefully prevents reckless behaviour in working-age investors (you don't put all your savings into Bitcoin because the potential loss feels too painful). The problem is when it triggers panic-selling at exactly the wrong moment. The challenge is overriding it when it's harmful while keeping its protective benefits.

What if there's a permanent change — what if the market really won't recover?

The "this time is different" hypothesis has been wrong in every previous major UK/global market crash (1973-74, 1987, 2000-02, 2008-09, 2020, 2022). Could the next one be the exception? Possible but historically unlikely. Markets reflect ongoing economic activity; as long as developed economies generate output and corporate profits, equity returns recover. The risk of permanent loss is real but much smaller than the risk of panic-selling.

How do I know if I'm panicking vs reasonably re-assessing?

Two tests: (1) Have your underlying long-term plans changed (retirement age, spending needs, time horizon)? If no, your circumstances haven't changed and your portfolio plan shouldn't either. (2) Are you making this decision after a long calm period of reflection, or in response to recent market moves? If recent moves are the trigger, that's panic, not re-assessment.

Should I check my portfolio less often?

Yes, especially in drawdown. The investor who checks daily sees 250 trading days a year, of which roughly 30-50% are losses. The cumulative emotional weight of seeing 75-125 "loss days" per year creates anxiety even if the annual return is positive. Checking monthly or quarterly dramatically reduces emotional volatility.

Does the same logic apply to property values?

Property is less liquid so panic-selling is harder, which partially protects homeowners from their own loss aversion. But it shows up in other ways: refusing to sell at a sensible price because "we paid more than that" (sunk cost fallacy combined with loss aversion), refusing to downsize because the loss of the family home feels too painful even when financially sensible. See our sunk cost fallacy in property decisions page.

Editorial accountability
Open Trust Centre →

Every page is reviewed against the editorial standards, written from primary sources, sourced openly, and corrected publicly. No affiliate revenue. No sponsored content. No paid placements.

Editorial standards Editorial process Corrections policy How we make money Editorial team Methodology
Cookie settings