How to Choose an Asset Allocation for a Financial Goal

Maya has to decide something she has never decided before

Maya has come into a significant sum at once. It is real, it is hers, and the first question anybody asks her is how it should be invested — a question she has no experience of, because until now the amounts involved were small enough that getting it wrong did not matter.

She has been offered two kinds of answer. One is a rule of thumb involving her age. The other is a questionnaire that asked how she would feel about a decline and produced a label. Neither is useless, and neither is close to sufficient, because neither of them asked the only question that actually determines the answer: what is this money for, and what happens if it is worth considerably less than this on the day it is needed?

Capacity comes before tolerance

There are two different things people mean by risk, and conflating them is the most common error in this whole subject.

Capacity for loss is a fact about the goal. It asks what a serious fall would do to the plan — whether the payment could be delayed, whether it could be smaller, whether there is another source. Money that funds a deposit on a flat Maya intends to buy in two years has very little capacity: the fall cannot be waited out, and the purchase either shrinks or does not happen. Money with no assigned purpose and no deadline has a great deal.

Tolerance for loss is a fact about the person. It asks what she would actually do partway through a decline. It matters not because feelings are a good guide to investing but because a plan that gets abandoned in the middle delivers none of the outcome it was designed for. A portfolio Maya sells out of at the worst moment is worse than a more conservative one she would have held.

The order is important. Capacity sets the ceiling and tolerance may lower it, but tolerance can never raise it. Feeling relaxed about volatility does not create the time a two-year goal does not have.

Making the loss concrete

The step that changes people's answers is translating percentages into the thing they actually represent.

A decline of some fraction is an abstraction. The same decline expressed as an amount of Maya's money, alongside what it would mean for the specific goal — this many months later, this much smaller, this thing not happening — is a question she can answer honestly. Most people discover at this point that their tolerance is lower than the questionnaire suggested, and a few discover the opposite.

It is worth doing this at a severity that feels excessive. Indian equity has had falls that would strike most people as implausible until they are shown the record; the distribution of long-run outcomes sets out how deep and how long. The test worth running is against something in that range rather than against a mild year, because a plan that only survives mild years has not been tested.

The second half of the question is the one people skip: not just how she would feel, but what she would do. Feeling awful and continuing to contribute is a perfectly good outcome. Feeling awful and selling is the one the allocation has to be designed around.

Give every part of the money a job

Rather than choosing one split for everything, it is more useful for Maya to divide the money by what it is for and let each part have its own answer.

Anything she may need within the next couple of years, or that would be catastrophic to have less of, belongs somewhere it will not move much. The return on that money is close to irrelevant; its job is to be there. Anything with a long horizon and genuine flexibility can carry growth assets and should, because the alternative — safety applied to money that will not be touched for twenty years — has its own cost, which is that inflation eats it slowly and invisibly.

Between those sits the money whose job is stability: not the emergency reserve, but the ballast that lets her rebalance without selling growth assets at a bad time, and that reduces how far the whole thing falls. This is the part most people either omit entirely or oversize.

The simple goal-based portfolio page sets out how to assemble this with the fewest possible components, which is the other half of the job.

What a concentrated holding does to all of this

Maya has one complication that the standard advice will not mention, and for her it may matter more than the allocation itself.

Part of what she owns came from her employer, and it is a large enough share of the total to change the risk of the whole portfolio. A holding like that is not diversified by the rest of the portfolio in the way a fund would be, and its risks are correlated with her income in a way nothing else she owns is: if the company does badly, the holding falls and her job becomes less secure at the same time.

The allocation question cannot be answered while ignoring it. Whatever split she chooses, the concentrated position has to be counted at its actual size within the growth allocation, not treated as a separate thing she happens to have. That will usually reveal that she is carrying more equity risk than she intended, and a good deal of it in one name.

Writing the maintenance rule

An allocation is not a decision Maya makes once. It is a decision she has to keep, and the keeping needs its own rules, decided now.

She needs the target percentages, a band around them wide enough that ordinary movement does not demand action, a date each year to look, and a statement of where new money goes by default. She also needs to say in advance whether risk should fall as any of these goals approach, which is the glide path question.

And she needs one line about what would legitimately change the target. The answer is: a change in the goal, the timing, or her capacity. Not a change in what markets have recently done. An allocation revised because one asset has been performing well is not an allocation; it is a performance-chasing rule with a policy document attached.

What to take away

The right allocation is the most growth the goal can carry without creating a loss the household cannot survive financially or behaviourally — and those two constraints are checked separately, capacity first.

Split the money by purpose rather than choosing one number for all of it. Express the worst case in rupees and in consequences before agreeing to it. Count any concentrated holding at its real weight. Then write down the target, the band, the review date and what would justify a change, because the allocation only does its work if it is still there in three years.