Layer 5 · Retirement Planning Theory

Sequence of Returns Risk

The danger that poor market returns early in retirement permanently damage a portfolio, even when the long-run average return is acceptable.

retirementrisksequencewithdrawalSWP

What it is

Sequence of returns risk is the danger that the order in which investment returns occur — not just their average — determines whether your retirement plan survives.

Two retirees can start with the same corpus, earn the same average annual return over 30 years, and withdraw the same amount every year. The one who retires into a bull market is fine. The one who retires into a crash is not.

This is not intuitive. Most people think: “If my average return is 8%, my plan works.” But during the withdrawal phase, averages lie.


The number that explains everything

Consider two retirees, both starting with ₹2 crore, both withdrawing ₹8 lakh/year:

YearRetiree A (lucky sequence)Retiree B (unlucky sequence)
1+28%−38%
2+22%−12%
3+15%+5%
4–30Moderate returnsSame moderate returns
Average return8%8%
Corpus at year 30₹4.1 crore₹0 (ran out at year 14)

Same average. Completely different outcome. The unlucky retiree sold units at the bottom to fund withdrawals, locking in permanent losses before the recovery could help them.


Why it only matters during withdrawals

This risk does not apply during accumulation. If you are in the SIP phase and the market crashes in year 3, you simply buy more units cheaply. A recovery restores your wealth.

During withdrawals, the dynamic reverses. Every time the market falls, you sell units at depressed prices to fund living expenses. You have fewer units to participate in the eventual recovery. The damage is permanent.

The accumulation phase and the withdrawal phase are mirror images of each other. What helps you in one phase (buying low) hurts you in the other (selling low).


The India-specific severity

Sequence risk is more dangerous for Indian retirees than the US research suggests, for three reasons:

1. Higher inflation eats the buffer faster. Indian CPI has averaged 6–7% over the last 30 years. A ₹6 lakh annual withdrawal today becomes ₹34 lakh in 30 years at 6% inflation. There is far less room to absorb a bad sequence.

2. No equivalent of Social Security. US retirees have a guaranteed income floor from Social Security that covers basic expenses regardless of market returns. Indian retirees relying only on corpus SWP have no such floor unless they have a pension, NPS annuity, or rental income.

3. Fewer hedging instruments. Long-duration government bonds that offset equity drawdowns in the US are not easily accessible in India via simple retail products. The SCSS and RBI Bonds provide safety but with limited duration.


The five-year danger window

Research (and India data confirms this) shows that the most dangerous period is the 5 years before and 5 years after retirement. A crash in this window causes the most damage because:

  • The corpus is at its peak — the absolute rupee loss is largest
  • The recovery horizon is shortest — less time for compounding to repair the damage

A crash at age 45 when you have 20 years of accumulation left is survivable. The same crash at age 60 with withdrawals starting is potentially catastrophic.


Strategies to navigate it (India-specific)

1. Bond tent / Rising equity glidepath
Reduce equity to 30–40% in the 5 years around retirement, then gradually increase back to 50–60% over the following 10 years. This limits the damage from a crash in the danger window, then lets growth resume.

2. Cash buffer / Bucket 1
Keep 2–3 years of expenses in liquid funds or FDs. When markets crash, draw from the buffer instead of selling equity. This gives equity time to recover before you touch it.

3. SCSS + RBI Bonds as income floor
Senior Citizens’ Savings Scheme (8.2%) and RBI Floating Rate Bonds provide guaranteed income. This reduces dependence on SWP withdrawals from equity, directly reducing sequence risk exposure.

4. Dynamic withdrawal (Guardrails method)
Instead of withdrawing a fixed amount, set rules: if portfolio value drops >20% below plan, cut withdrawal by 10%. If portfolio is 20% above plan, take a “raise.” This makes the plan adaptive instead of rigid.

5. NPS annuity as the floor
The 40% mandatory annuity from NPS, while modest, provides a guaranteed monthly income regardless of market conditions. It functions as an income floor that reduces sequence risk.


Key takeaways

  • Sequence risk only bites during the withdrawal phase — not during accumulation
  • Two portfolios with identical average returns can have wildly different outcomes based on when the good and bad years happen
  • The 5-year window around retirement is the highest-risk period
  • India’s higher inflation and lack of a guaranteed income floor makes sequence risk worse than in US
  • Mitigation: bond tent, cash buffer, guaranteed income products, and dynamic withdrawals

Further reading on this site

Concept illustrated

The Retirement Timing Trap

Same ₹2 crore corpus. Same 8% average return. Same ₹8 lakh/year withdrawal.
Two different return sequences — two entirely different retirements.

Retiree A — retires into a bull market Retiree B — retires into a crash
5 Cr 4 Cr 3 Cr 2 Cr 1 Cr Yr 0 Yr 3 Yr 7 Yr 11 Yr 15 Years into retirement yr 15 → ₹0 Both start: ₹2 Cr ₹5+ Cr Year 3: gap already ₹2.4 Cr
🏦 Starting corpus: ₹2 crore
💸 Withdrawal: ₹8 lakh/yr (6% inflation-adjusted)
📊 Average return: 8% p.a.
Retiree A
Retires into a bull market
Year
Retiree B
Retires into a crash
+28%
1
−38%
+22%
2
−12%
+15%
3
+5%
Years 4–15: Same moderate 8% average returns for both
Average return: 8%
=
Average return: 8%
₹4.1 crore
After 30 years ✓
vs
₹0
Runs out by year 15 ✗
The 5-year danger window: A market crash in the 5 years before or after retirement causes the most damage — the corpus is at its peak (largest absolute loss) and the recovery horizon is shortest. This window is when sequence risk is most lethal.

Illustrative scenario based on Monte Carlo methodology. Withdrawal grows at 6%/yr (India CPI). Source: freefincal research. Asymmetrica

First published Last updated

Last updated · site changelog