"How big a corpus do I need to retire?" The honest answer is "it depends" — but there's a clear framework to work it out, as long as you adapt it to Indian realities rather than importing a US rule of thumb wholesale. Let's build the number step by step, with an example. (Illustrative; not personalised advice.)
The 4% rule, and where it comes from
The well-known 4% rule comes from US research (the "Trinity study" and its successors): withdraw 4% of your corpus in the first year of retirement, then increase that rupee amount with inflation each year, and historically the money lasted around 30 years across most US market histories. Inverted, it implies a corpus of roughly 25 times your first-year expenses (since 1 ÷ 4% = 25).
It's a useful anchor. It is not a law of nature, and it was calibrated on US inflation, US asset returns, and ~30-year retirements.
Why India needs a more conservative number
Several differences all push the safe withdrawal rate down (and the required corpus up):
- Higher inflation. India's general inflation — and especially healthcare inflation, often in the low-to-mid teens — has historically run well above the US. A fixed-real withdrawal erodes faster when prices rise faster.
- Longer retirements. Earlier retirement ages and rising longevity can mean funding 30–40 years, not 30.
- No robust safety net. There's no broad state pension or social security to catch you if the plan falls short.
- Sequence risk (below) bites harder without that safety net.
For these reasons, many Indian planners use a more cautious starting withdrawal of around 3% to 3.5%, implying a corpus closer to 28–33 times first-year expenses rather than 25.
A worked example
Take someone spending ₹60,000 a month today — ₹7.2 lakh a year — planning to retire in 20 years, with a long retirement ahead.
Step 1 — inflate expenses to retirement. At ~6% general inflation, ₹7.2 lakh today becomes roughly ₹23 lakh a year by the time they retire in 20 years. (This is the step most people skip — planning around today's expenses dramatically under-saves.)
Step 2 — choose a corpus multiple. Using a conservative ~3.3% withdrawal (≈ 30×), the target corpus is about ₹23 lakh × 30 ≈ ₹6.9 crore at retirement.
That number looks alarming, but two things soften it: it's two decades away (so inflation has lifted both the target and, presumably, your income and contributions), and part of it is already being built through EPF, NPS and existing investments.
Step 3 — work out the run-rate. Reaching ~₹6.9 crore in 20 years needs a monthly investment that grows over time. A useful discipline is to estimate the SIP required at a conservative assumed return, then step it up as income rises — even a 5–10% annual increase in the SIP amount dramatically lifts the final corpus and makes a large target achievable.
Don't forget the healthcare corpus
General expenses and healthcare don't inflate at the same rate. It's wise to plan a separate buffer for medical costs, which rise faster and arrive unpredictably late in life — and to hold adequate health insurance with a super top-up so that one illness doesn't force large withdrawals from the corpus at the wrong moment. Treat health cover as part of the retirement plan, not a separate afterthought.
Sequence-of-returns risk — the hidden danger
Two portfolios can earn the same average return over retirement and yet one runs out while the other thrives — purely because of the order in which good and bad years arrive. A market crash early in retirement is far more dangerous than the same crash later, because you're selling units to live on while prices are down, permanently shrinking the base that has to recover.
The same two crashes that are an opportunity for a working saver (cheap units, decades to recover) are a threat to a new retiree. This is why a retiree can't simply hold the same aggressive portfolio they did at 40.
The bucket strategy
The standard defence against sequence risk is to segment the corpus by when you'll spend it:
- Bucket 1 — cash/short debt (2–3 years of expenses). This is what you actually draw your monthly income from. Because it's safe and liquid, you never have to sell equity in a downturn.
- Bucket 2 — debt and hybrid (next ~5–7 years). Medium-term stability that refills Bucket 1.
- Bucket 3 — equity (the long tail). The growth engine that keeps the corpus ahead of inflation across a 30-year retirement.
In good years, you top Bucket 1 back up from gains; in bad years, you live off Bucket 1 and leave equity alone to recover. (A balanced-advantage or conservative-hybrid fund can serve as a simpler one-fund version of this for those who prefer fewer moving parts. See our piece on systematic withdrawal plans for the income mechanics.)
The corpus is not a fixed deposit
A corpus large enough to last decades must keep growing through retirement to outpace inflation. Parking everything in FDs feels safe but, after tax and inflation, the real return can be negative — the surest way to run out slowly. Retirement money stays partly in equity for exactly this reason; the bucket strategy is what lets you hold that equity calmly.
Common mistakes
- Planning around today's expenses (forgetting to inflate).
- Using 25× uncritically when Indian inflation argues for a higher multiple.
- Ignoring healthcare inflation and going underinsured.
- De-risking to all-debt at 60, then watching inflation erode the corpus over a 30-year retirement.
- Forgetting taxes on withdrawals and pension income.
Build in flexibility
The single most powerful lever isn't a perfect forecast — it's flexibility. Retirees who can trim discretionary spending in bad market years, delay a big one-off, or work part-time for a while make a given corpus stretch dramatically further. A plan with a little give in it beats a rigid plan built on optimistic assumptions.
The bottom line
Estimate your annual expenses, inflate them to your retirement age, and multiply by roughly 28–33× (leaning higher if you'll retire early or expect heavy healthcare costs). Hold a 2–3 year cash cushion, keep the rest invested for growth, insure your health properly, and stay flexible. Running out at 80 is a far worse error than over-saving at 60 — so when in doubt, aim higher and keep the corpus working.