The 4% rule is the most repeated number in personal finance. It comes from work by William Bengen in the 1990s and the Trinity Study that followed: withdraw 4% of your portfolio in year one, increase it with inflation each year after, and historically the money lasted 30 years.
It is a good rule. It is also routinely misapplied, in two specific ways that make the resulting number far too small.
First correction: the rate is 3.9%, not 4%
Morningstar’s 2026 State of Retirement Income research puts the baseline safe withdrawal rate at 3.9% for a retiree using fixed spending — up from 3.7% in their 2025 work, and modelled for a 90% probability of lasting a 30-year retirement.
The difference from the original 4% is methodological rather than cosmetic. Bengen used historical returns. Morningstar uses forward-looking return assumptions, which currently imply lower equity and bond returns than the twentieth-century record. Their 3.9% figure was derived for a relatively conservative portfolio, in the range of 20–50% equities.
They also note that retirees willing to flex their spending with market conditions — a "guardrails" approach, cutting in bad years and spending more in good ones — can support a starting rate as high as 5.7%. That is a very large difference, and it comes entirely from flexibility rather than from a better portfolio.
Second correction: the one that actually hurts
Here is the arithmetic almost everyone gets wrong.
If you want ₹50,000 a month in retirement, that is ₹6 lakh a year. At a 3.9% withdrawal rate the corpus required is ₹6,00,000 ÷ 0.039 = ₹1.54 crore. At 4% it is ₹1.5 crore. That is the number most people arrive at, and it is the number most calculators produce.
It is wrong, because ₹50,000 is what that lifestyle costs today. If you retire in 25 years, you need what that lifestyle will cost then.
At 6% inflation, ₹50,000 a month today becomes ₹2,14,594 a month in 25 years. That is ₹25.75 lakh a year, and the corpus required to sustain it at a 3.9% withdrawal rate is ₹6.6 crore — not ₹1.54 crore. The inflation adjustment more than quadruples the target. Any retirement number that has not been inflated to the retirement date is not a plan, it is a rounding error.
Where the 4% rule genuinely does not apply
It is worth knowing the assumptions, because several of them fail for a lot of people.
It assumes a 30-year retirement. Retire at 50 and you may need 45 years of income, at which point 3.9% is too aggressive. Retire at 70 and you can safely draw more.
It was derived on US market history. Indian equity and bond returns, and Indian inflation, are a different distribution. The 3.9% figure is a useful reference point, not a law of nature that travels unchanged across markets.
It assumes constant real spending. Real retirement spending is not flat — it tends to be higher early (travel, activity), lower in the middle, and higher again late (healthcare). The "retirement spending smile" is well documented and it means a flat withdrawal model misstates both ends.
It ignores every other income source. Any pension, annuity, rental income, EPS payout or part-time work reduces what the portfolio must produce. Someone with a ₹20,000 monthly pension needs a dramatically smaller corpus than the headline suggests.
Working out your own number
- Start from actual current spending, not income. What you spend is what you need to replace. Some costs fall in retirement — commuting, the loan you will have cleared, supporting children. Others rise, healthcare most of all.
- Inflate it to your retirement date. Multiply by (1 + inflation)years. Use a realistic long-run rate; India has averaged meaningfully higher than developed markets, and medical inflation runs higher than headline inflation almost everywhere.
- Subtract guaranteed income — pension, annuity, rent — in the same inflated terms.
- Divide the remainder by your withdrawal rate. 0.039 for a fixed-spending plan; higher if you are genuinely willing to cut spending in bad years.
- Then work backwards to the monthly contribution that reaches it, which is the only number you can act on this month.
If the number looks impossible
It often does, and there are four levers rather than one.
Spend less in retirement. Mechanically the most powerful, because it cuts the required corpus by roughly 25 times the annual reduction. Trimming ₹5,000 a month of future spending removes about ₹15 lakh from the target in today’s terms.
Work longer. This does three things at once: more years of contributions, more years of compounding, and fewer years to fund. Two extra years typically moves the picture more than a percentage point of return does.
Accept flexible spending. Morningstar’s own research says a guardrails approach supports up to 5.7% against 3.9% fixed. That is not a small optimisation — it is the difference between needing ₹6.6 crore and needing around ₹4.5 crore for the same lifestyle.
Increase contributions with income. A SIP that steps up 10% a year finishes vastly ahead of a flat one, and it is far easier to sustain because it grows out of raises rather than out of sacrifice.