Most arguments about the 4% rule in India get stuck on the wrong question. People debate whether Indian equity returns support a higher or lower rate, quote Nifty CAGR since 1996, and go round in circles.
The rate is not really the problem. The second half of the rule is, and almost nobody discusses it. Withdraw 4% in year one, then raise that rupee amount by inflation every year after. In a 2-3% inflation economy that adjustment is a slow drift. At 7% it is a compounding claim on your corpus that doubles your withdrawal in ten years and quadruples it in twenty.
Your inflation is not the headline inflation
India's CPI prints somewhere in the 4-6% range in a normal year. Plan around that and you will be wrong, because CPI is a basket built for the average Indian household, where food carries roughly 46% of the weight. Your basket looks nothing like that.
A retired urban professional spends comparatively little on cereals and pulses, and a great deal on the categories that inflate fastest. Medical treatment. Insurance premiums. Private school fees. The cook and the driver. Restaurants, travel, services generally. Services inflation in India runs structurally above goods inflation because it tracks wage growth in a fast-growing economy rather than commodity prices.
| Category | Typical inflation | Share of a FIRE household's spend |
|---|---|---|
| Healthcare and insurance premiums | 10-14% | Climbs steeply with age |
| Education, if you have children | 8-10% | Large, then drops to zero |
| Domestic help, services, repairs | 8-9% | Steady, tied to wages |
| Rent | 5-7% | Often the single biggest line |
| Food and groceries | 5-6% | Moderate |
| Electronics, appliances, telecom | 0-3% | Small |
Weight those by what a FIRE household actually spends and the effective personal inflation rate lands near 7%, not 5%. That is the number your plan has to survive. It is also why we use 6% as a planning assumption and then stress-test above it.
Inflation Calculator
What will your monthly expenses be in 10 years?
Factor in category-wise inflation for healthcare, education, and daily needs to find your real effective inflation rate.
What 7% does to a 4% plan
₹5 crore corpus. Year one withdrawal at 4% is ₹20 lakhs, or ₹1.67 lakhs a month. Portfolio earns a blended 9.5%, which is a 60:40 mix. Withdrawals rise 7% a year to hold purchasing power steady.
| Year | Annual withdrawal | Corpus at year end | Withdrawal as % of corpus |
|---|---|---|---|
| 1 | ₹20.0 L | ₹5.26 cr | 4.0% |
| 10 | ₹36.8 L | ₹6.66 cr | 5.5% |
| 20 | ₹72.4 L | ₹5.42 cr | 13.4% |
| 27 | ₹1.16 cr | ₹1.9 cr | 61% |
| 29 | ₹1.33 cr | ₹0 | Depleted |
Look at what happens between year 1 and year 10. The corpus grows, from ₹5 crore to ₹6.66 crore. Everything looks fine. The retiree at year 10 feels vindicated. They have more money than they started with and they have been living off it for a decade.
Then years 15 to 25 happen. The withdrawal, compounding at 7%, catches the portfolio, which is compounding at 9.5% minus withdrawals. Once the withdrawal crosses about 6-7% of what is left, the decline is fast and it does not reverse. Depletion at year 29 means somebody who retired at 45 is out of money at 74. That is the base case for a 4% plan under Indian inflation, not some unlucky tail.
The dangerous part
A failing plan looks exactly like a working plan for the first decade. Better, often, because the corpus is visibly growing. By the time the trend shows up in your account balance you are in your sixties, your ability to go back to work has thinned, and fifteen years of compounding are gone. This particular failure gives you no early warning, which is the whole reason to stress-test it before you retire rather than after.
The same table at 3.3%
Identical corpus, identical returns, identical 7% adjustment. Only the starting withdrawal changes: ₹16.5 lakhs instead of ₹20 lakhs. About ₹29,000 a month less.
| Year | Annual withdrawal | Corpus at year end | Status |
|---|---|---|---|
| 10 | ₹30.4 L | ₹7.55 cr | Healthy |
| 20 | ₹59.7 L | ₹9.1 cr | Healthy |
| 30 | ₹1.18 cr | ₹6.6 cr | Declining |
| 36 | ₹1.77 cr | ₹1.4 cr | Survives to 81 |
A 17% cut in year-one spending buys seven extra years of solvency. Small change in withdrawal, large change in how long the money lasts. That is the case for the lower rate, and it is why we default to 3.3%.
Test your own numbers against 7% inflation
Set inflation to 7% and the withdrawal to what you actually plan to spend. If the depletion year lands before age 85, you have a decision to make while you still have options.
Open SWP Calculator →The other half: sequence risk
Everything above assumes a smooth 9.5% every year. Markets do not do that. And when the bad years show up matters far more than the average return across the whole period.
Two retirees, same 30-year average, same withdrawals. One gets a 35% drawdown in years 2-3 of retirement. The other gets the identical drawdown in years 22-23. The first may never recover, because they were selling units at depressed prices to pay for groceries at exactly the moment those units were cheap, permanently removing the shares that would have participated in the bounce. The second had twenty good years of compounding first and barely notices.
Which is why the standard advice to hold 3-5 years of expenses in debt at the start of retirement is not conservatism theatre. It exists for one specific purpose: so you never have to sell equity into a crash in the first five years, the window where sequence risk does nearly all of its damage.
What actually fixes this
A fixed withdrawal rate is a strange thing to commit to. It ignores every piece of information that arrives after you retire. Real retirees don't behave that way and the plan shouldn't either.
Guardrails: skip the raise in bad years
The simplest fix and the most effective. In any year the portfolio fell, don't apply the inflation increase. Spend last year's rupee amount. In a year it rose more than 15%, take the full increase and maybe a bit more. Skipping two or three inflation raises across a retirement adds close to a decade of corpus life, because you break the compounding of the withdrawal at the moments it hurts most.
Percentage-of-portfolio withdrawals
Rather than a fixed rupee amount indexed to inflation, take a fixed percentage of the current portfolio each year. Say 4% of whatever it is worth on 1 April. Mathematically this can never deplete the corpus. The trade is income volatility: a 30% market fall means a 30% pay cut that year. Fine if a decent chunk of your spending is discretionary. Painful if it isn't.
Keep equity high enough
The instinct at retirement is to shift heavily into debt. Against 7% inflation over 35 years, a debt-heavy portfolio earning 7% pre-tax loses purchasing power every year after tax. Most India FIRE plans we have modelled hold 50-60% equity well into retirement. The volatility is uncomfortable. The alternative is worse.
Keep some ability to earn
Even ₹25,000 a month of consulting or teaching during a bad stretch takes enormous pressure off, because it substitutes directly for a withdrawal you would otherwise make at the worst possible price. Having the option is worth more than half a percent of assumed return.
The short version
- Headline CPI understates a FIRE household's real inflation. Weight it for healthcare, services, rent and school fees and you land nearer 7%.
- The damage comes from the annual inflation adjustment, not the withdrawal percentage. At 7% the withdrawal doubles in ten years while returns stay flat.
- 4% at 7% inflation and 9.5% returns depletes around year 29, and looks completely healthy for the first decade. That is what makes it dangerous.
- Dropping to 3.3% costs about 17% of year-one spending and buys seven more years.
- Sequence risk stacks on top. A crash in the first five years does damage a late crash does not, so hold 3-5 years of expenses in debt when you start.
- The real fix is a rule that responds to what happened, not a better fixed number. Skip the inflation raise after down years, keep equity above 50%, keep some way to earn.
If you want the fuller argument for why India needs a lower rate than the US in the first place, the 1998 study it came from, what Social Security did for American retirees and what changes without it, that case is made in our post on whether the 4% rule is valid in India. This one is the mechanism underneath it.