Suppose you settle on a simple mix: 70% stocks, 30% bonds. You chose those numbers for a reason — in the last article you worked out how much of a crash you can stomach, and 70/30 keeps the worst likely day inside your line.
Then you do the healthy thing and ignore the portfolio for two years. Stocks have a great run. Without you touching anything, you now hold 80/20.
Nothing failed. That is just what markets do to a mix: the winner grows its own share. But quietly, your portfolio has become riskier than the one you designed. The crash you sized for 70/30 now lands on 80/20, and your “roughly 35% worst case” has drifted toward 40%. Drift is how people end up holding far more risk than they ever agreed to, one good year at a time.
Rebalancing is the fix: sell some of what grew, buy some of what shrank, return to 70/30. That’s the whole move.
Why it feels wrong (and why that’s fine)
Read the move again. You are selling the thing that has been winning and buying the thing that has been losing. Every instinct objects — winners feel like they’ll keep winning, and adding to the laggard feels like rewarding failure.
But notice what the mechanic actually does: it forces you to sell high and buy low, in small doses, on a schedule, without a forecast. Every other place in investing, that behavior requires either luck or courage. Here it falls out of arithmetic. The main purpose is risk control, not profit — over some periods rebalancing helps returns a little, over others it costs a little. What it does reliably is keep the portfolio you own equal to the portfolio you chose.
When to do it
Two respectable schools, both fine, pick one:
By the calendar. Once or twice a year, on a date you’ll remember — a birthday, the first weekend of January. Check the weights, restore the targets, close the laptop. Its virtue is that it cannot be gamed by mood.
By a band. Act only when an asset drifts more than some threshold from target, say five percentage points: 70/30 becomes 75/25, that’s your trigger. Between bands you deliberately do nothing. Slightly more responsive, requires slightly more attention.
What doesn’t work is “when it feels right.” Feelings correlate with headlines, headlines correlate with exactly the wrong moment, and a rebalance you skip during a crash is the one that mattered most. A crash is the one time the rule does something dramatic: it has you buying stocks in the middle of the panic, at prices future-you will brag about. Nobody does that on gut feel.
How to do it cheaply
Rebalancing can cost money — trading fees and, in a taxable account, capital-gains tax on what you sell. Three ways to soften it, in order of preference:
- Rebalance with new money. If you invest monthly, point the new contributions at whatever is under target. Small portfolios can often stay balanced this way for years without selling anything, which means no tax at all.
- Rebalance inside tax-sheltered accounts first. Trades in a retirement account typically trigger no tax; do the heavy lifting there.
- Don’t overdo the frequency. Monthly rebalancing is churn dressed up as discipline. The differences between annual, semiannual, and banded approaches are small; the difference between some rule and no rule is the whole game.
Write it down
A rebalancing policy fits in two sentences. For example: “Targets: 70% global stocks, 30% bonds. Every January, and any time a weight drifts 5 points off, I restore the targets — using new contributions first.”
That’s it. Ten minutes a year of mildly counterintuitive bookkeeping, and your portfolio stays the size of your courage instead of the size of the latest bubble. In the final article of this course, that sentence becomes one line of a slightly longer document: the single page that runs your money.