The calculation
Four steps, and none of them is difficult.
- Annual expenses in retirement, in today's money. Not current expenses — some end (commuting, children, the home loan), and healthcare rises. The total is usually not much lower than today.
- Inflate to your retirement date. This is where most estimates go wrong, and it is a compounding error rather than a small one.
- Divide by a withdrawal rate you can defend, or equivalently multiply by its inverse.
- Subtract other income — rent, a pension, continuing work — capitalised at the same rate.
What comes out is a number, and the number is far less interesting than how much it moves when you change the inputs.
Which assumption actually decides it
Take a household spending ₹60,000 a month today, retiring in 25 years, planning for a 30-year retirement. At 6% inflation and a 3% real return, the corpus required is around ₹6.1 crore.
Now change one assumption at a time:
| Change | Corpus needed | Difference |
|---|---|---|
| Inflation 7% instead of 6% | ₹7.7 crore | +27% |
| Inflation 5% instead of 6% | ₹4.8 crore | −21% |
| Real return 2% instead of 3% | ₹6.9 crore | +14% |
| Real return 4% instead of 3% | ₹5.3 crore | −12% |
| Retirement of 35 years, not 30 | ₹6.6 crore | +10% |
| Retirement of 25 years, not 30 | ₹5.4 crore | −11% |
A single percentage point of inflation moves the answer by roughly a quarter — more than a point of return, and more than five extra years of life.
That inverts where attention usually goes. People agonise over expected returns and choose an inflation number in seconds. The arithmetic says the opposite ordering is correct, and the reason is that inflation compounds twice: once over the 25 years before retirement to set the starting expense, and again through the 30 years after it.
So choose the inflation number carefully
Two refinements make a large difference.
Use your own basket, not the headline. Retirement spending is weighted towards healthcare, which has historically risen faster than the general index in India. A retiree's personal inflation rate is plausibly above the published one, and the table above shows what a point costs.
Model healthcare separately. Applying one blended rate to a basket that is 20% healthcare at the start and considerably more at the end understates the later years — which are exactly the years the corpus is thinnest.
The honest conclusion is not a specific rate. It is that the inflation assumption deserves more thought than the return assumption, and that erring low on it is the most expensive optimism available in this calculation.
Precision in the wrong place
Given those sensitivities, a corpus stated as ₹6,14,32,500 is false precision. The inputs do not support a figure to five significant digits, and presenting one suggests a confidence that does not exist.
What is genuinely useful is a range and a direction: a plausible band, which assumptions would push you to the top of it, and how far off you currently are. Planning against a band changes behaviour in a way that a single false-precision number does not, because the band is honest about what is unknown.
The corollary is that reviewing the number matters more than getting it right the first time. Recomputed every few years with observed inflation and actual returns, it converges. Computed once at thirty-five and never revisited, it is a guess with a decade of drift.
The risk you cannot diversify
The table treats retirement length as an input. In reality it is unknown, and the distribution is asymmetric in an uncomfortable way.
Planning to average life expectancy means roughly half the time you outlive the plan. And the failure is not symmetric: dying with money unspent is a modest inefficiency, while running out at eighty-five is a different order of problem, at an age when returning to work is not an option.
Life expectancy at 60 is also meaningfully higher than life expectancy at birth, which is the figure people usually have in mind — having reached sixty, you have already survived everything that would have lowered the average.
Two responses. Plan for longer than the average, accepting the cost. And consider covering essential expenses with a guaranteed lifetime income so that the part of the plan that must not fail does not depend on how long you live — the argument for an annuity covering the floor, with the rest funded by withdrawals.
When the number is too large
It usually is, at first. That is the calculation working — it has converted an unexamined worry into a specific gap, and a gap has levers.
| Lever | Effect | Note |
|---|---|---|
| Save more each month | Direct | The obvious one, and often the most constrained |
| Work a few years longer | Unusually large | Adds contributions, adds compounding, and shortens the drawdown — three effects at once |
| Reduce retirement spending | Direct and proportional | A 10% lower requirement is a 10% smaller corpus |
| Part-time income in early retirement | Large | Reduces withdrawals in the years sequence risk bites hardest |
| Higher return assumption | Moderate | The weakest lever — and the only one that is a hope rather than a decision |
Note the last row. Closing a gap by assuming better returns does not change the plan; it changes the spreadsheet. The four levers above it are decisions you control.
Testing the plan rather than the number
Everything above assumes a constant real return. Nothing delivers one, and the difference matters more in the withdrawal phase than the accumulation phase, because sequence risk only exists when you are drawing down.
FNOTrader's Mutual Funds app computes rolling returns across every start date in the full AMFI NAV history — around 34 million NAV rows — alongside maximum drawdown. Two tests are worth running: whether the accumulation plan reaches the corpus using the worst historical window rather than the average, and what a drawdown of that size would do arriving in the first three years of retirement.
A plan that survives both is a plan. FNOTrader is not a SEBI-registered investment adviser and this is not retirement advice.
Common questions
How do I calculate how much I need to retire?
Estimate annual expenses in retirement in today's money, inflate them to your retirement date, divide by a withdrawal rate you can defend, and subtract other income sources capitalised at the same rate. The arithmetic takes minutes; the assumptions are the hard part.
Which assumption matters most in a retirement calculation?
Inflation, by a clear margin. On a typical case, a single percentage point changes the required corpus by roughly a quarter — more than a point of return and more than five extra years of retirement, because inflation compounds both before and after you stop working.
What inflation rate should I assume?
One reflecting your own basket rather than the headline figure, since retirement spending is weighted towards healthcare, which has historically risen faster in India. Erring low is the most expensive optimism available in this calculation.
Should I state my retirement number precisely?
No — the inputs do not support five significant digits, and false precision implies a confidence that does not exist. A plausible range, plus which assumptions push you to the top of it, changes behaviour in a way a single number does not.
How long should I plan my retirement to last?
Longer than average life expectancy, because planning to the average means roughly half the time you outlive the plan — and the failure is asymmetric. Life expectancy at 60 is also meaningfully higher than at birth, which is the figure most people have in mind.
What is the most effective way to close a retirement shortfall?
Working a few years longer is unusually powerful because it does three things at once: adds contributions, adds compounding, and shortens the drawdown. Assuming higher returns is the weakest lever, since it changes the spreadsheet rather than the plan.
How often should I recalculate?
Every few years, using observed inflation and actual returns. Recomputed periodically it converges; computed once at thirty-five and never revisited it is a guess with a decade of drift.
Continue reading
More in Retirement Planning · App: Mutual Funds · Definitions: glossary · Free tools: calculators · All: every article