Rebalancing: The Only Free Discipline in Investing

Set an allocation - say 60% equities, 40% bonds - and leave it alone for five years. It will not be 60/40 any more.

Whatever performed best now occupies a larger share. The portfolio has become riskier, and nobody chose that. It happened by arithmetic while nobody was looking.

Why drift is the problem

Suppose equities return 10% a year and bonds 3%. Start at 60/40 and after five years the mix is roughly 68/32 - without a single transaction.

That sounds minor and it is not. The portfolio's risk has moved materially. A 40% bond allocation was presumably chosen to limit how much a bad equity year would hurt. At 32%, the cushion is thinner.

And the drift always runs the same direction: toward whatever has risen most, which is also whatever has become most expensive.

So a portfolio left alone systematically increases exposure to the asset that has already run, and reduces it in whatever has lagged. It is a momentum strategy nobody signed up for.

What rebalancing does

Sell some of what has grown past its target. Buy what has fallen below it. Return to the intended weights.

The mechanical result is selling what has done well and buying what has done badly - which is emotionally backwards. That is precisely why a rule helps. Almost nobody does it reliably by judgement, because the moment to buy is always the moment it feels worst.

The benefit that is real

Risk control. This is the honest, dependable case, and it is sufficient on its own.

An allocation is a statement about how much loss you are prepared to sit through. Drift makes that statement untrue without telling you. Rebalancing keeps the portfolio matched to the risk you actually chose.

The value shows up in downturns. A portfolio that drifted to 75% equities before a serious decline falls considerably further than one held at 60%. The extra loss was not the market's doing - it was the drift.

The benefit that is oversold

You will often read that rebalancing improves returns - the "rebalancing bonus" - because you systematically sell high and buy low.

This can happen. It requires specific conditions: assets with similar long-run returns that fluctuate substantially and independently. Under those conditions, trimming winners and topping up laggards captures the oscillation.

Those conditions frequently do not hold. When one asset substantially outperforms another over a long stretch - as equities have over bonds across many periods - rebalancing means repeatedly selling the better performer. That reduces returns relative to leaving it alone.

Which is fine, because the reduced return came with reduced risk. But it should be stated properly. Rebalancing is not a way to earn more. It is a way to keep the risk you intended, and it sometimes costs return to do it.

Calendar versus threshold

Two approaches, and they are not equivalent.

Calendar rebalancing. Act on a fixed schedule - annually, say. Simple, easy to automate, requires no monitoring. Its weakness is that it can trade when the portfolio has barely moved and wait when it has moved a lot.

Threshold rebalancing. Act when an allocation drifts more than a set amount from target - commonly five percentage points, or a relative band. This responds to actual drift, so it trades when trading is warranted and not otherwise. It requires periodic checking.

Research generally finds threshold approaches somewhat more efficient, and finds the difference between reasonable methods smaller than the difference between doing it and not.

Combining them works well: check on a schedule, act only if a band has been breached. You get the discipline of a calendar without trading for its own sake.

The costs that erode it

Rebalancing is not free, and ignoring the frictions is how a good idea becomes a bad implementation.

Transaction costs. Every trade costs something. Rebalancing too often - monthly, say - can consume more than the discipline is worth.

Tax. In a taxable account, selling an appreciated holding realises a gain. This is frequently the largest cost of all and it is entirely avoidable in most cases.

Spreads, on anything less than highly liquid.

Doing it without the costs

Use new contributions. Direct incoming money toward whatever is underweight. This rebalances gradually with no selling at all, so no tax and minimal cost. For anyone still adding to a portfolio, this alone handles most drift.

Use dividends and interest. Rather than reinvesting income back into its source, direct it to underweight holdings. Same effect, no disposal.

Rebalance inside sheltered accounts first. If you hold assets across taxable and tax-advantaged accounts, do the selling where it is not a taxable event.

Use bands wide enough to matter. Five percentage points is a common choice. Rebalancing on a one-point drift generates cost for no meaningful risk benefit.

Do not rebalance into a permanently impaired asset. The discipline assumes assets that fluctuate. Mechanically topping up something in structural decline is not rebalancing - it is averaging into a loss.

The part that is genuinely hard

The rule is trivial. Following it is not.

Rebalancing tells you to buy the thing everyone is avoiding and sell the thing that has been working. In March of a bad year, it says to buy more equities. After a long bull run, it says to trim them.

Both instructions arrive at the moment they feel most wrong. That is not a flaw in the method - it is the method. If it felt comfortable, it would not be adding anything, because comfortable positioning is what drift produces on its own.

Writing the policy down in advance - target weights, bands, what triggers action - and following it is the whole discipline. The decision is made when you are calm and executed when you are not.

The bottom line

Portfolios drift toward whatever has done best, which raises risk without anyone choosing it. Rebalancing restores the allocation you actually decided on.

The return bonus is conditional and oversold. The risk control is neither. And the cheapest way to do it is not to sell anything at all - just point new money at whatever has fallen behind.

This article is educational and is not financial advice. The value of investments can fall as well as rise.