How to Build a Simple Goal-Based Investment Portfolio

Neha keeps being sold components

Neha has been earning for about a year and has begun, in a scattered sort of way, to invest. There is a fund somebody at work recommended, a policy an uncle arranged, a tax-saving thing she bought in March because the deadline was approaching, and money in her salary account that has never been given a purpose.

None of that is a portfolio. It is a set of purchases, each defensible at the moment it was made, none of them chosen in relation to any of the others or to anything she actually wants. If you asked what any individual holding was for, the honest answer would be that it was available.

Building a portfolio means reversing that. The goal comes first, the structure comes from the goal, and the products are the last thing chosen rather than the first.

Start at the far end

The first step is not an investment decision at all. It is arithmetic.

The work is to name the goal, put a rough amount on it in today's money, and put a date on it. Then that amount has to be inflated to what it will cost when it arrives — which is the subject of estimating the future cost of a goal and choosing an inflation assumption, and is more consequential than it looks. Then she subtracts anything she already has assigned to it, and what remains is what her contributions have to produce.

At that point one of two things is true. Either the required contribution is affordable, in which case the plan is real and the rest of this page is about how to structure it. Or it is not, in which case she has learned something important before investing a rupee, and the honest responses are to lengthen the date, reduce the goal, find another source, or accept that this particular version of it is not happening.

The response that is not available is raising the assumed return until the arithmetic works. That changes the spreadsheet and nothing else — she is not going to earn more because she wrote down a larger number, and a plan built on a return she has no reason to expect will fail quietly, several years in, when there is no time left to fix it.

Three jobs, and only three

A portfolio serving a goal needs to do three things, and almost every sensible structure is some arrangement of them.

It needs money available for spending and for contributions without anything having to be sold at a bad moment — call that liquidity. It needs a stable component, whose job is that money required soon does not depend on a recovery, and which doubles as the thing Neha rebalances against. And it needs a growth component, which is the only part expected to outpace inflation over a long horizon and which is accepted as volatile in exchange.

That is the whole design. What proportion each job gets is the allocation question, covered in how to choose an asset allocation. What matters here is that every holding should be answerable to one of the three, and a holding that cannot be assigned to one is a holding that has not been justified.

It is worth trying on whatever is currently owned. The tax-saving product and the policy from her uncle are the two most likely to have no answer, and finding that out is more useful than any addition she could make.

Fewer components than feels right

The instinct when building a portfolio is that more holdings means more diversification. It usually does not.

Two funds that hold substantially the same companies do not diversify each other; they duplicate, and the second one adds paperwork rather than protection. Real diversification comes from holding things that behave differently, and beyond a small number of genuinely distinct components the additions stop doing anything. Meanwhile every extra holding is another set of statements, another nomination to keep current, another thing to value at review time, and one more reason the annual review does not happen.

The operational argument is the stronger one and gets the least attention. A structure Neha can describe from memory is one she will maintain. A structure of fourteen holdings across four platforms is one she will stop looking at within two years, and an unmaintained portfolio drifts — which undoes the thing the structure was for.

A portfolio should be the smallest number of genuinely different components that can do the three jobs. For most single goals that is a very short list.

Write the rules before you need them

The structure is not finished when the money is allocated. It is finished when Neha knows what to do with the next rupee and with the last one.

Where new contributions go by default is one, and it is the mechanism in rebalancing with new investments first. A review date, and a band that says when drift has become large enough to act on, is another. Whether the risk should be coming down as the goal approaches, and on what schedule, is the third. And the fourth is how the money actually gets out at the end — which holding is sold, in what order, how long it takes to reach the account, and what tax falls due.

That last one is the most commonly missing and the most embarrassing to discover late. A goal-based portfolio that cannot be liquidated in time for the goal has failed at the only thing it existed to do.

The month's work

Take the existing holdings and assign each to a job, or to nothing. The ones assigned to nothing are not necessarily to be sold immediately — exit costs and lock-ins are real and some products punish early departure — but they should stop receiving new money.

Then pick the goal that matters most, do the arithmetic honestly, and build the smallest structure that serves it. Write the four rules down on one page. Then leave it alone until the review date, which is the hardest part and the one where most of the value is.

What to take away

A portfolio is a structure that delivers a specific amount on a specific date, not a collection of things that seemed like good ideas. Start from the goal, inflate it, subtract what exists, and find out whether the contribution is affordable before choosing a single product.

Give every holding one of three jobs — liquidity, stability, growth — and remove anything that cannot claim one. Use as few components as will do the work, because the portfolio you can maintain beats the portfolio you designed. And write down where new money goes, when you review, how risk falls, and how the money comes out, because a structure without those four rules is not finished.