The account was opened on a Sunday in September. Somewhere between the PAN upload and the bank mandate there were eight multiple-choice questions, and they took under two minutes. How would you feel if your investments fell twenty per cent in a month — concerned, uncomfortable, or would you see an opportunity. How long do you plan to invest for — under three years, three to seven, or more than seven. Then a coloured band appeared with the words Moderately Aggressive and, beneath it, a suggested allocation of seventy per cent in equity. Nobody in the household has looked at that screen since. The seventy per cent is still there four years later, applied to a considerably larger amount, and it is the single most consequential number in the family’s finances. It was produced by a form that was structurally unable to ask the two questions that decide the answer.
You join a gym and fill in a health form. Do you feel breathless climbing stairs, have you exercised before, how do you rate your fitness out of five. A trainer reads it and writes a programme. Two things were never on the form. What your left knee will actually take, which is a structural fact a doctor could establish and your feelings about stairs cannot. And what you are training for — whether there is a wedding in eleven weeks or nothing in particular — because that decides how hard the programme has to be, and it is not a matter of how fit you feel.
A profiling form measures willingness. What a fall does to a household is decided by its income, its fixed commitments and the date the money is needed — none of which the form asked. What the plan needs is a third quantity again, computed from the amount, the date and the monthly contribution. Three questions, three different kinds of answer, and one coloured band on a screen.
Three different questions wearing one label
Allocation questions get muddled because three separate things are all called "how much risk should I take". They have different sources, they move at different speeds, and — this is the part that matters — they are not three opinions to be averaged.
| The question | What actually sets the answer | How it behaves |
|---|---|---|
| [[Risk tolerance]] — how much you are willing to sit through without acting | Temperament, upbringing, and what the market has done recently. This is what a questionnaire measures, because it is the only one of the three that can be measured by asking | Least stable of the three, and it moves the wrong way: it is highest after a long rise, which is precisely when it is about to be tested, and lowest at the bottom, when acting on it is most expensive |
| [[Risk capacity]] — whether a fall can be lived through without selling into it | Income and how secure it is, fixed commitments, the size of the buffer, and the date the money is needed. None of it is about how you feel | Arithmetic. It does not move with your mood, and it can be computed on a Tuesday by somebody who has never met you |
| [[Goal-required return]] — the return the plan needs in order to arrive | The target amount, the date, what is already saved, and what you can add each month. Four numbers and a compounding calculation | A consequence rather than a choice. It changes only when one of those four numbers changes, and it is completely indifferent to what any allocation can deliver |
The same household, tested three ways
A single income of ₹1,45,000 a month after tax. A home loan taking ₹42,000 of it. One month of expenses in the bank beyond that. The goal is the balance due on a flat: ₹30,00,000, on a date fixed by an agreement roughly five years away. ₹8,00,000 is already set aside for it and ₹25,000 a month can be added. The rates below are illustrative and are before tax and costs — put your own in and the arithmetic runs the same way.
What a questionnaire cannot see, however well written
- It is answered in one emotional state, about a different one. "What would you do if your portfolio fell thirty per cent" is a forecast about a stranger. It is almost always answered by somebody calm, and calm is not the state in which the behaviour occurs. Your own transaction history from the last real fall is stronger evidence than any answer you can give today, and another lesson in this track is about reading it.
- It is answered after the market has done something. Willingness tracks recent returns closely enough that the same person, asked in two different years, produces two different bands. The form has no way to know which year it is being filled in, and its output carries no date.
- It sees one account. The form is completed at one platform and its recommendation applies to the money on that platform. A household’s provident fund, PPF, deposits and property sit outside it, so "seventy per cent equity" applied to a quarter of the household’s assets is a completely different allocation from the one the label describes. A separate lesson in the previous module works through what that does.
- It usually has no date in it. A horizon band — "three to seven years" — is not a date, and the difference between year three and year seven is the difference between an absorbable fall and a fatal one. That single fact does more to set capacity than every temperament question put together.
- Its output is a category, and a category cannot be checked. "Moderately aggressive" is not comparable to anything, cannot be wrong, and cannot be tested against a bank balance. The question that can be checked is: what rupee fall does this allocation permit, on what date would that fall be unrecoverable, and what would the household do the following morning.
- Six months of expenses in a liquid fund, no loan.
- A 35% fall costs about a year and a half of savings and nothing else.
- The date is not a date; it is a decade-wide window that can move.
- The goal-required return is high, because the target is very large against what is saved so far — and twenty-six years of contributions is what makes it reachable at all.
- Drawdown capacity is wide, so tolerance is genuinely the binding question here.
- One month of buffer, a loan taking 29% of take-home pay.
- A 35% fall in year four costs the flat, or a loan taken at a worse rate to cover it.
- The date is written in an agreement and carries a penalty.
- The goal-required return is about 8% — lower than the allocation is reaching for.
- Capacity is narrow, so it binds, and tolerance does not get a vote.
- 1Write the date and the amount
Not "retirement" and not "a flat" — a year and a rupee figure. Everything downstream is computed from these two, and a goal without them cannot be tested at all. Where the date is genuinely a window rather than a date, write the window, because its width is itself a form of capacity.
- 2Compute what the plan needs, at three rates rather than one
What is already saved, compounded to the date, plus the monthly contribution compounded to the date, against the target. Run it at three rates. If the answer at the lowest of the three still arrives, the plan does not need much from the market; if it only arrives at the highest, the plan is relying on the assumption rather than on the contribution, and another lesson in this track is about what that does.
- 3Test a fall against the date, in rupees
Take the allocation you hold, apply a fall of the size the asset has actually produced before to the equity part, and answer three questions about the next morning. Does income still cover the commitments. Is any of the money needed before a recovery could plausibly have happened. Who else has to be told. Capacity is the answer to those three and to nothing else.
- 4Then use your temperament as the tie-breaker
Where capacity permits a range — and for a long goal it usually permits a wide one — the allocation you can actually hold through a bad year is better than the one that is theoretically optimal and gets abandoned in March. That is where willingness belongs: choosing inside the range, not setting it.
- 5Date the page, and keep the old ones
Tolerance moves, so a number written in a calm month is evidence that it was written in a calm month. Two dated pages, one from a good year and one from a bad one, tell you more about your own tolerance than any questionnaire, and they are the only version of it that was recorded while it was actually being tested.
The form says moderately aggressive and the registration is in five years
The household above notices the four-year-old profiling screen for the first time. ₹8,00,000 is set aside for a ₹30,00,000 payment due in five years, ₹25,000 a month is going in, and the allocation is seventy per cent equity because a form said so in September four years ago.
A household needs ₹30,00,000 in five years for a payment whose date is fixed by an agreement. It has ₹8,00,000 saved and can add ₹25,000 a month, which together need about 8% a year to arrive. A profiling form places the household at 70% equity. Which of the three tests should decide the allocation?
Gym join karte waqt form bharte ho: umar, wazan, "lakshya kya hai — fat loss ya muscle?" Aath sawaal, do minute, aur trainer ek plan de deta hai. Kisi ne yeh nahi poocha ki ghutna kab tak theek hai, aur agli shaadi kab hai. Risk profiling form bilkul waisa hai. Shaant mahine ki Sunday ko bhare gaye aath sawaalon se nikla "moderately aggressive" aur 70% equity — aur ghar chaar saal usi pe chal raha hai. Form ne teen cheezon mein se sirf ek naapi: aapka mizaaj — aur woh wahi cheez hai jo market ke saath badalti hai. Do cheezein jo naapi hi nahi gayi: paisa kab chahiye (paanch saal baad registration hai toh 70% equity ka koi jawab nahi), aur girawat rupaye mein kitni padegi ("35% gir sakta hai" alag baat hai, "₹7,70,000 gir sakta hai aur registration usi saal hai" alag). Isiliye do bilkul alag ghar wahi form bhar ke wahi jawab paate hain. Teen test khud karo, ek page pe: tareekh aur rakam likho jab paisa chahiye; plan kya maangta hai teen rate pe nikaalo, ek pe nahi; girawat ko us tareekh ke saamne rupaye mein rakho; aur mizaaj ko sirf tie-breaker banao, faisla lene wala nahi. Page pe tareekh daalo aur purane page phenko mat — jo form shaant mahine mein bhara tha, woh bure March mein aapke liye jawab nahi dega.
- Three different questions get called "how much risk" — willingness, ability to absorb a fall, and the return the plan needs — and a questionnaire can only ask the first.
- Tolerance is a preference, capacity is a constraint, and the required return is arithmetic; when they disagree, the constraint binds.
- A questionnaire has no date in it, sees one account, is answered in a calm month, and outputs a label that cannot be checked against anything.
- The rule runs both ways: a nervous answer that produces an allocation too small to reach the goal is the same mistake with the sign flipped.
- Where the required return exceeds what capacity permits, the quantities that can move are the contribution, the target and the date — not the allocation.
Mark it done to track your progress through the curriculum.