Financial planning for women: three constraints, and what each costs

What actually differs, and what does not

Nothing about compounding, diversification or cost changes with the reader's sex. What changes are three inputs to the same calculations: how many years the money must last, how many years it was contributed for, and how many people know where it is.

That is worth saying plainly, because an article that opens by explaining what a mutual fund is — to this reader and not to any other — has already made a claim about who needs the explanation. The arithmetic below assumes the reader can follow it, and gets on with the part that is actually different.

The three constraints are structural rather than behavioural, which is what makes them worth computing. A longer expected retirement is a longer stream to fund, and a stream's cost is a multiplication. An interrupted contribution schedule loses compounding time, and compounding time is an exponent. Where one person holds the whole picture of a household's money, the picture has a single point of failure, and that is a design fact rather than a character flaw.

One demographic claim underlies the first of the three, and it is the only one in this article. India's official abridged life tables, published under the Sample Registration System, report a longer remaining life expectancy at age 60 for women than for men. The size of that gap is a number worth reading from the source rather than from an article, and none is quoted here. Its direction is what changes the arithmetic — and the arithmetic works identically whoever the extra years belong to.

Everything below is what follows mechanically. Where a situation is relevant it is named as one case among several, because the reader of this page may be single or partnered, salaried or self-employed or between the two, supporting dependants or supported, and no line here needs to know which.

A longer retirement is a bigger multiplier, and it is bounded

Funding an income for life is a multiplication. Take the first year's spending and multiply it by the amount of capital needed today to pay one rupee a year, rising with inflation, for as long as the money must last. That multiplier is the annuity factor, and extra years of life enter the calculation through it and nowhere else.

Here it is on illustrative inputs: ₹6 lakh a year of retirement spending, withdrawals at the end of each year, indexed to inflation, discounted at a return of 2% after inflation. That real 2% is an assumption and not a forecast; substitute your own. How the corpus number is built in the first place — and the unit error that halves it — is worked through separately and not repeated here.

Read the marginal steps rather than the levels, because that is where the useful shape is. Moving from 25 years to 30 costs 15%. The next five years cost 12%, the five after that 9%. Each additional year is discounted harder than the one before it, so the factor climbs towards a ceiling of 1 ÷ 0.02, or 50, instead of running away.

Which turns an alarming framing into a priced one. Five extra years of expected retirement is not a doubling of the capital required; on these inputs the step costs about a seventh more at 25 years and under a tenth more at 35. That is a real cost, and it is the size of a few years of contributions rather than the size of a different plan.

The version of this that catches people out is the joint one. Where two people fund a single retirement, the money has to last until the later of two deaths, not until the average of them — the later of two dates is always at or beyond each one separately. A plan built on one person's horizon is short by construction, and a lifetime income bought on one person's life pays nothing to whoever outlives them.

That is also why a joint-life annuity quotes a lower income per rupee than a single-life one. The lower figure is not a worse deal being offered; it is the same arithmetic as the table above, charging for a longer expected stream. What the extra capital buys is an income that does not depend on how long anyone lives, and that trade is worth understanding in full before it is accepted or rejected.

A contribution gap is priced by its date, not its length

Career interruptions happen for reasons that have nothing in common with each other: a parent who needs care, an illness, a relocation for someone else's job, a child, a return to study, a redundancy. The financial consequence is identical in all of them, and it is not the salary forgone. It is the compounding time that the missed contributions never get back.

Which sounds like a general principle until you put numbers on it, at which point it stops being general at all. Take an illustrative ₹20,000 a month invested for 30 years at an illustrative 10% a year — an assumption, not a forecast — which finishes at about ₹4.52 crore. That is a nominal figure, in the rupees of that final year, and not comparable with the inflation-adjusted corpus numbers in the table above. Now remove exactly 36 contributions, three times, at different points.

The same ₹7.2 lakh of missed contributions costs ₹91.2 lakh in one case and ₹11.3 lakh in another — a factor of eight, decided entirely by the date, not the length. The 36 instalments are worth ₹8.36 lakh at the moment the gap ends in every row; what differs is how many years remain for that sum to compound.

Now the part that is almost never stated, and it inverts the obvious response. Repair the early gap by raising the contribution for the remaining 24 years and it takes about ₹7,700 a month on top of ₹20,000. Repair the late gap in the three years that remain and it takes about ₹27,000 a month. The expensive gap is the cheapest to repair, because the same exponent that punished the missing rupees rewards the replacing ones.

So the mistake worth naming is not taking the break. It is measuring the break by its length, concluding that three years out of thirty is a tenth of the plan, and doing nothing on return because the number felt small. What the arithmetic favours after an early interruption is a permanent step-up in the contribution once income resumes — modest per month, and it has two decades to work. After a late one it says something different: the gap barely moved the answer, which is worth knowing before deciding to take more risk near the end of a horizon.

Two structural details make the loss larger than the table shows, and both are worth checking against your own arrangements rather than assumed. Where retirement saving runs through an employment-linked scheme, an unpaid break stops the employer's contribution as well as your own, so the gap in the schedule is wider than the gap in your own cashflow; the mechanics of one such scheme are set out in the article on it. And where a re-entry restarts at a lower base than the exit, every contribution set as a proportion of that base is lower too, so the schedule resumes at less than it stopped at.

The break also restarts clocks an employer was holding open

Health cover provided through an employer is a benefit of the employment. It ends when the employment ends, and the gap it leaves is not simply an uninsured period. It is the loss of accrued time on clocks that only run while a policy is in force.

Two of those clocks decide what a health policy will actually pay for. A pre-existing disease is defined by a lookback of 36 months before the policy, and a fresh individual policy may impose a waiting period on any such condition of up to 36 months. Separately, IRDAI's rules set a moratorium of 60 months of continuous cover, after which the grounds on which an insurer may contest a claim narrow sharply — and that clock runs afresh on any enhanced sum insured, so raising cover is a decision with a cost as well as a benefit.

Put those two rules beside a break and the mechanism is uncomfortable but simple. Anything diagnosed while you are uninsured falls inside the lookback of the policy you buy afterwards. Four years of accrued waiting time on a policy you let lapse is worth nothing on the day you buy a new one; you are back at month zero on a clock that can run 36 months.

The defence is continuity rather than optimisation, and it is one of the few places in personal finance where holding a merely adequate arrangement beats switching to a better one. An individual policy held alongside employer cover keeps its own clocks running regardless of what happens to the job. Where a policy is being changed rather than started, portability preserves accrued credit — but only inside a window, which runs at least 30 days before, and not earlier than 60 days from, the renewal due date, so it is a diary entry and not an intention. How portability actually works is covered separately, as is what a health policy is buying in the first place.

The longevity constraint returns here in a different form. More expected years is more years spent in the age band where health cover is most needed and most expensive to start. IRDAI caps the annual premium revision on individual indemnity health cover for those aged 60 and above at 10% per annum — which limits the rate of increase, not the level, and does nothing at all for someone buying their first policy at that age with conditions already on record.

Records are per person, and a household is not a person

Several of the systems a plan depends on keep their records against an individual, not against a household. Where money is pooled and only some of it stands in your name, the pooling is real and the record is not, and the difference only becomes visible at the moment you need the record.

Credit is the clearest case. A credit score is computed from a credit record, on a scale that RBI now requires all four credit information companies to calibrate to 300 to 900. Someone who has never borrowed or held a card in their own name does not have a poor score. They have no score at all, which is a different problem with a different remedy — a thin file is fixed by originating a small, reported obligation and servicing it, not by paying down debts that were never yours. Which entries actually sit against your name is a question your own report answers: you are entitled to one free full credit report a calendar year from each credit information company, and what the score is made of is set out in full elsewhere.

Deposit insurance keeps records the same way, and the detail catches households that think they have doubled their protection. DICGC cover is ₹5 lakh per depositor per bank, aggregated across every branch and account held in the same right and capacity — and each capacity is insured separately. For joint holdings, the order of names decides whether joint deposits pool: the same two people in the same order are one pool under one ceiling, not two. One caution belongs with that, because secondary coverage rarely carries it. The Corporation pays against the claim list the bank or liquidator submits, so the separation is only ever as good as the bank's own records. The full mechanics are here.

The third case is the one with no paperwork at all. Where work is unpaid — a household run full time, an unpaid role in a family business, care given to a parent or a child — there is no employment-linked retirement account quietly receiving a deduction every month, and no independent contribution history accumulating anywhere. Nothing about that is a failure. It is simply that the automatic mechanism most retirement saving relies on is absent, and an automatic mechanism has to be replaced by a decision that gets made every month, which is a materially harder thing to sustain.

Hence the mistake this section exists to name: reading "the household is saving" as "I have retirement savings". Those two statements differ only in whose name the asset stands, which is exactly the fact that matters at the moments a household stops functioning as one unit — a death, an incapacity, a separation, a dispute. Being the nominee on an account is not ownership either; a nominee receives the money and holds it for whoever inherits it, a distinction worth getting right once.

The silent portfolio, and the test that finds it

Where two people share money — partners, siblings, a parent and an adult child — the running of it may sit entirely with one of them. Where it does, the risk that creates attaches to the role and not to the person, and it runs in both directions: the one who does not manage cannot act if the manager is gone, and the manager's own arrangements are equally unusable by anyone else if they are incapacitated rather than absent.

What makes it a genuine risk rather than an inconvenience is that a forgotten asset does not fail loudly. It goes quiet. An account with no customer-induced transaction for two years is classified as inoperative, and an unclaimed balance transfers to the Depositor Education and Awareness Fund after ten years — from maturity, for a term deposit, rather than from opening. The money remains claimable throughout, and always from the bank rather than from RBI. Call the failure mode the silent portfolio: nothing is lost, nothing raises an alarm, and recovering it depends entirely on someone knowing it existed.

There is a search tool, UDGAM (udgam.rbi.org.in), and it is worth knowing about — but it searches on a name, which means it finds what a bank has already reported and nothing held anywhere else. Reclaiming an unclaimed balance is a defined process; discovering that there is one to reclaim is not.

The useful diagnostic takes an afternoon and does not require anyone to change how the household divides its work.

• The cold-start test — without asking the person who manages the money, and without their laptop, could you list every bank, every scheme, every policy and every loan? Not the balances. Just that they exist, and where.
• The access question — incapacity is the harder case than death, because nomination and succession do nothing while someone is alive. Whether a power of attorney or a joint operating mandate is the right instrument is a separate question; having neither is a decision taken by default.
• The one-page index — institution, account type, and where the document lives. Not passwords, not balances, both of which go stale and neither of which is needed to find an asset. A checklist of what to hold is a better starting point than a plan.

The trade-off is real and worth stating rather than glossing. Documenting a portfolio reduces privacy inside a household and takes effort that produces no return in any year it is not needed. What it buys is that the plan survives the loss of the one person who understood it, which is the only circumstance in which any of it will be read.

Testing an interrupted schedule rather than an average one

Every figure above runs on a constant illustrative return, which is a modelling convenience and not a property of any portfolio. That matters more for an interrupted contribution schedule than for a smooth one: a gap does not merely remove instalments, it removes the instalments belonging to a particular stretch of market history, and the stretch it removes is not the average one.

FNOTrader's Mutual Funds app runs on the full history of published daily per-unit prices — net asset values, or NAVs — from the Association of Mutual Funds in India, around 34 million rows of them. It simulates a contribution schedule on any scheme and period and reports the internal rate of return for cashflows landing on irregular dates — XIRR — alongside invested against value, maximum drawdown, and rolling-return distributions across every available start date. Two tests follow directly from the arithmetic here: what a three-year gap at a chosen point would have cost across historical windows rather than at an assumed average, and whether a stepped-up contribution after the gap closes it from the worst of those windows rather than the mean. Why the distribution rather than the single window is argued separately.

Historical outcomes describe what happened, not what will happen; past performance does not indicate future results. FNOTrader is not a SEBI-registered investment adviser and nothing here is advice. What any of this implies for a particular person depends on facts an article does not have.

Common questions

Continue reading

More in Money by Life Stage · App: Mutual Funds · Definitions: glossary · Free tools: calculators · All: every article