Rebalancing means restoring a portfolio to its intended proportions after prices have moved them. The arithmetic is simple and the execution is uncomfortable, which is why it is so often skipped.
Drift is automatic and cumulative
A portfolio set to a chosen mix does not stay there. Assets that rise become a larger share, and those that fall become smaller, without any transaction taking place.
After a sustained move in one direction, the actual mix can be far from the intended one, and the risk carried is correspondingly different.
The investor has not changed their view or their circumstances. The portfolio changed on its own, which is the problem rebalancing exists to solve.
The required trade runs against recent experience
Restoring the target means selling from whatever has performed best and buying whatever has performed worst, which contradicts every instinct built by watching those results.
The trade is not a prediction that the winner will fall. It is a decision to keep the risk profile the investor originally chose when they had no view of what would happen next.
The alternative is not neutral either, since leaving the portfolio alone is an active decision to carry whatever concentration recent prices happened to produce.
Framing it as risk control rather than as market timing is what makes it possible to carry out, since the justification does not depend on being right about direction.
Discipline matters most at extremes
Drift is largest after big moves, which is exactly when the required trade feels most unreasonable and when conviction about the recent trend is strongest.
Investors who abandon the discipline in those moments end up with concentrated exposure at the point where concentration has already been rewarded.
The consequence appears later, since a portfolio that drifted into a heavy weighting carries more of the subsequent decline than the original plan intended.
Frequency involves a real trade-off
Rebalancing more often keeps the portfolio closer to target but generates more transactions, which costs money and may crystallise tax in a taxable account.
Rebalancing on a threshold rather than a calendar reduces the number of trades, acting only when the drift exceeds a defined tolerance.
Neither approach is obviously superior, and the choice depends more on cost and tax circumstances than on any property of markets.
Contributions can do the work quietly
An investor still adding money can direct new contributions toward whatever is underweight, which restores the mix without selling anything.
This avoids both transaction costs and tax events, and it removes the psychological difficulty because nothing successful is being sold.
The approach loses effectiveness as the portfolio grows relative to contributions, which is why it works well early and less well later.