Financial independence, and the imported number

Independence is a number; retiring is a choice

Financial independence means your assets can fund your living costs without you needing to work. That is all it means.

It is frequently bundled with early retirement, and the two are separable. Reaching independence and continuing to work — because the work is interesting, or part-time, or lower-paid but better — is a common and entirely coherent outcome. Independence removes the necessity, not the option.

Worth separating because they demand different things. Independence is arithmetic: a corpus, a withdrawal rate, a set of expenses. Early retirement adds a much longer drawdown period and a set of questions arithmetic does not answer.

Where 25x and 4% came from

Both are the same statement. If you withdraw 4% of a corpus in year one, the corpus must be 25 times your annual expenses — 1 divided by 0.04.

The figure originates in American research examining historical US market returns and asking what withdrawal rate would have survived a 30-year retirement across the worst historical starting points, with withdrawals rising each year with inflation. Roughly 4% survived nearly all of them.

It is a genuine piece of work and it is routinely overstated. What it says is: given this market's history, this asset mix, this length of retirement and this inflation record, this rate would have survived. Every one of those inputs is a condition, and the conclusion does not transfer to a different set of them.

Why it travels badly

Four reasons the imported figure deserves scepticism here, and none of them is reassuring.

Different inflation history. The rule assumes withdrawals rise with inflation. A higher or more variable inflation path means withdrawals grow faster than the study assumed, and the corpus is drawn down harder. Personal inflation compounds this — healthcare, which dominates late-life spending, has historically risen faster than the general index.

Different market history, and less of it. The research leans on a long US series. Applying a rate derived from one market's record to another is an assumption, not a finding.

Longer retirements. Thirty years was the study's horizon. Somebody retiring at forty is planning for substantially longer, and withdrawal rates that survive thirty years do not automatically survive fifty — the failure probability rises with duration.

No comparable safety net. The original context sits alongside social security and employer health cover. Where healthcare is funded privately, a single medical event can force a large unplanned withdrawal at exactly the wrong time.

The honest conclusion is not a replacement number — it is that 4% is an imported assumption rather than a finding about India, and anybody using it should know they are extrapolating.

The risk the average hides

This is the mechanism that makes withdrawal harder than accumulation, and it is badly under-appreciated.

Two retirees each start with the same corpus, withdraw the same amount, and earn the same average return over twenty years. The one whose poor years came first runs out; the one whose poor years came last does not.

The reason is mechanical. Withdrawing during a drawdown means selling more units to raise the same rupees, so the corpus is permanently smaller when the recovery arrives. Identical averages, opposite outcomes — this is sequence-of-returns risk, and it applies only when you are withdrawing.

It is why a plan cannot be tested against an average return. What matters is how it behaves when the bad years land first, which is a question about the distribution rather than the mean.

What a more careful plan looks like

Not a different magic number. A set of adjustments, each of which reduces reliance on the assumption.

• A lower initial withdrawal rate, which is the same as a larger multiple of expenses. The cost is working longer; the benefit is not depending on an imported figure.
• Flexible withdrawals. Reducing spending in bad years — rather than raising it mechanically with inflation — improves survival dramatically, because it attacks sequence risk directly.
• A cash buffer of a few years' expenses, so a downturn is not funded by selling into it. This is the bucket idea, and it exists to solve sequence risk rather than to raise returns.
• Health cover treated as infrastructure, not an expense line. One uninsured event can undo the plan.
• Any continuing income counted honestly — part-time work, rent, a pension. It reduces the required corpus more than most people expect, because it reduces the withdrawal at the most fragile moment.

Getting your own number

1. Annual expenses in retirement, not today's — some fall, healthcare rises, and the total is usually not much lower.
2. Inflate them to your start date, then remember they keep rising throughout.
3. Choose a withdrawal rate you can defend, given the length of your retirement and the points above.
4. Divide. That is the corpus.
5. Subtract other income sources, capitalised.
6. Work backwards to a monthly contribution, as in deriving a savings rate.

The result is usually larger than expected. That is information, not discouragement — and it is arriving while there is still time to act on it.

Testing it against bad sequences

A plan tested against an average return has not been tested. The question is whether it survives the bad windows, which is exactly what historical data can answer.

FNOTrader's Mutual Funds app computes rolling returns across every start date in the full AMFI history — around 34 million NAV rows — alongside maximum drawdown. Read the worst window, not the median. A plan that survives the worst historical sequence is a plan; one that works on the average is an assumption.

This is arithmetic and history, not advice. FNOTrader is not a SEBI-registered investment adviser, and what withdrawal rate is appropriate for you is not a question an article can answer.

Common questions

Continue reading

More in Personal Finance Basics · App: Mutual Funds · Definitions: glossary · Free tools: calculators · All: every article