Rebalancing

Rebalancing begins with measurement. Each asset's current weight is its market value divided by total portfolio value; drift is that current weight minus the strategic target. Because the highest-returning assets grow fastest, an unmanaged portfolio steadily loads up on whatever has just performed well, and its risk profile migrates away from the one the investment policy statement authorised.

The trade itself falls out of the same arithmetic. The required position size is the target weight multiplied by total portfolio value; the order is that figure less the position's present market value, a sale if negative and a purchase if positive. Because total value is the denominator for every asset, one large move re-prices all weights at once, so candidates should recompute the full weight vector rather than adjusting a single line. Cash flows into or out of the portfolio, coupon and dividend receipts, and derivatives overlays can absorb part of the required adjustment without trading the underlying holdings.

Two disciplines dominate. Calendar rebalancing acts on fixed dates, which makes governance and monitoring costs predictable but leaves the portfolio exposed to drift between reviews and makes the trade date arbitrary. Percentage-of-portfolio rebalancing sets a corridor around each target and acts only when a weight breaches it, so attention is directed at genuine deviations. Corridors are commonly expressed either in percentage points of the total portfolio or as a proportion of the target weight, and the two conventions imply very different bands for small allocations.

Corridor width is an optimisation, not a convention. Higher transaction costs, illiquidity, and taxable gains argue for wider bands; higher volatility in the asset itself argues for narrower ones, since drift accumulates quickly. Greater investor risk tolerance and a high correlation between the asset and the rest of the portfolio both permit wider corridors, because the consequences of drift for total portfolio risk are smaller.

Systematic rebalancing sells strength and buys weakness, so its payoff profile resembles a short volatility position: it is rewarded when markets oscillate around a level and penalised when they trend persistently, because the rule keeps trimming the very exposure that continues to rise. Any so-called rebalancing premium is therefore contingent on mean reversion rather than guaranteed, and it must be netted against commissions, spreads, market impact and, for taxable investors, realised capital gains.

The durable justification is risk control. Rebalancing keeps volatility, drawdown sensitivity and factor exposures close to the levels the mandate contemplated, which matters most in stressed markets where correlations shift and the drifted portfolio is typically the riskier one. It also imposes behavioural discipline, forcing purchases precisely when conviction is lowest, which is why the rule is written down in advance.

Rebalancing policy is a standard clause in institutional investment policy statements, specifying the method, the corridor widths, the responsible party and the permitted instruments. Target-date and balanced funds rebalance mechanically toward a glide path; index funds trade at reconstitution dates; sovereign wealth funds and large endowments publish rebalancing regimes and often use futures overlays to restore beta cheaply while physical trades settle over time. Predictable large-scale rebalancing flows are themselves observable in end-of-quarter market activity.

At Level I, be able to compute current weights and deviations, identify which corridors have been breached, express the required trades in currency terms, and justify the choice between calendar and threshold discipline. Equally, be prepared to reason about corridor width from transaction costs, volatility, correlation, liquidity, tax status and risk tolerance, and to explain the concave payoff without overstating the return benefit.