Portfolio Rebalancing — And the Cheaper Way to Do It

Suppose a household decided on a split between equity and safer holdings, and then did nothing for four years while equity had a strong run.

The equity portion is now a larger share of the total than intended, purely because it grew faster. Nobody made that decision. The household is carrying more exposure than it agreed to, and it will find that out in the next poor stretch rather than now.

The same happens within equity. A small cap holding that did well becomes a larger slice of the equity portion, so the portfolio has quietly tilted towards the sharper-moving end, which our page on large, mid and small cap funds describes.

Rebalancing is simply correcting that back towards the intended proportions. It is maintenance rather than a market call.

The textbook version says sell what grew and buy what lagged. In India that instruction carries a bill.

Selling units is a redemption, so the gain becomes a taxable event, and exit load can apply where the units are inside the scheme\'s defined period. A switch between schemes is treated the same way, since it is a redemption and a purchase, as our pages on mutual fund taxation and exit load set out.

So rebalancing enthusiastically, several times a year, can cost more than the drift it corrects. That is the part missing from most explanations, and it changes how often the exercise is worth doing.

Rather than selling, change where new money goes.

If equity has grown beyond its intended share, direct the next several months of instalments and any lump sums towards the part that lagged. No redemption, no tax event, no exit load, and the proportions correct themselves over a period.

For a household still in its accumulation years, with monthly instalments running, this handles most of the drift most of the time. Actual selling then becomes something you do rarely, when the gap is large enough that redirection alone would take too long.

The same logic applies to reducing exposure as a goal date approaches, which is a planned shift rather than rebalancing, and our page on the Systematic Transfer Plan covers doing it in stages.

Once a year is enough for almost every household, and it should be at the same time each year rather than whenever markets prompt the thought.

• Start with what exists. A consolidated statement lists everything against your PAN, as our page on the consolidated account statement explains.
• Work out the actual split, including any shares held directly, since those add to the equity share.
• Compare against what you intended, which requires having written it down at some point.
• Correct with new money first, and only sell where the gap is genuinely large.

One practical note on doing it while still saving. If you are adding money monthly, the redirection approach means you are rebalancing continuously without ever calling it that, and the annual review becomes a check rather than an exercise. Households that have stopped adding, typically after retirement, have no such lever, which is why the selling version matters more for them and why the corpus split by date matters more still.

Some households set a band instead of an exact figure, and act only when the drift exceeds it. That prevents fiddling and it is a sensible way to keep the exercise rare.

A rebalancing exercise that only looks at mutual fund holdings is measuring half the portfolio in most households.

Shares held directly are equity and belong in the equity share. So does the equity portion of any hybrid scheme, and under NPS a part of the corpus is invested in market-linked funds whether or not the subscriber chose that, as our page on SIP for government employees notes.

On the other side, deposits, the provident fund balance and anything held in a low-risk category belong in the safer share. Property is usually left out of this calculation entirely, which is defensible given it cannot be rebalanced, but it should at least be known.

Add all of that up once and most households find their actual split is some distance from what they assumed, before any drift is even considered.

Three things it gets confused with, and all three are more expensive.

It is not switching to whatever did well. That is the opposite: rebalancing sells what grew and adds to what lagged, which feels wrong every single time and is the entire point.

It is not a market view. You are not predicting anything; you are restoring proportions you decided in advance.

It is not a substitute for the goal-date shift. Moving a corpus towards safety as its date approaches is a separate, planned exercise, and it happens whether or not the portfolio has drifted.

If you would like somebody to work out your current split across all holdings and tell you plainly whether it needs correcting, that is a short review and there is no charge. Get in touch.