Rebalancing calculator
Target dollars minus current dollars is not a plan. This gives the trade list under a no-trade band, the version where new money does the work without a sale, and the tax bill each one realises, next to the drift each one corrects.
Worked example — Money to move, default 4-fund drift: $14,714. Under a 5/25 band: 2 of 4 holdings trade, for $646 of tax and 5.9 pts of drift corrected
Enter your portfolio and targets
Values, targets, and cost basis for each holding. The trade list, the tax cost, and the drift each method corrects update live across all three plans.
Default 4-fund drift: The page's own realistic drifted starting portfolio.
Advanced options 3
2 of 4 holdings sit inside their band and are left alone. The band that binds for each is the smaller of 5 points and 25% of its target.
$21,200 of new money would reach every target with no sale at all, and therefore no tax.
| Holding | Now | Target | Drift | Trade | After | Tax cost |
|---|---|---|---|---|---|---|
| US equity | 60.3% | 50.0% | +10.3 pts | Sell $7,357 | 53.2% | $646 |
| International equity | 21.4% | 25.0% | -3.6 pts | Inside the band | 21.4% | - |
| Bonds | 14.1% | 20.0% | -5.9 pts | Buy $7,357 | 21.3% | - |
| REIT | 4.2% | 5.0% | -0.8 pts | Inside the band | 4.2% | - |
Gains are pro-rated against average cost, which is what a broker reports by default. Selling specific lots chosen for a high basis would realise less, sometimes much less, and no calculator can compute that without your lot detail. State taxes, the net investment income tax, wash sales, and any carried-forward losses are not modelled.
Is rebalancing expected to add anything?
At your target weights, an average pairwise correlation of 0.35, and the volatilities you entered, the approximation is 0.52% a year: half the difference between the weighted average of the holdings' variances (0.02362, a variance rather than a rate) and the variance of the portfolio (0.01320, which is 11.49% of volatility squared). Positive here, so rebalancing is expected to help slightly. It is a second-order effect and it is smaller than the tax bill above in most taxable accounts. The approximation collapses toward zero as correlations approach 1 and assumes the holdings have no persistent return differences, which is exactly when it is least reliable.
What the 5/25 rule triggers at different targets
The band that binds is the smaller of 5 percentage points and 25% of the target, so a small sleeve drifting by a point is proportionally a bigger miss than the same drift in a large holding.
| Target weight | Band width |
|---|---|
| 5% | 1.25 pts |
| 10% | 2.50 pts |
| 20% | 5.00 pts |
| 30% | 5.00 pts |
| 50% | 5.00 pts |
| 60% | 5.00 pts |
Below 20% target, the 25%-of-target rule binds and the band narrows with the target. Above 20%, the flat 5-point band binds instead and stops narrowing.
Does rebalancing actually improve returns?
On this page's default portfolio, at an assumed average pairwise correlation of 0.35, the rebalancing premium comes out to 0.52% a year: half the difference between the weighted average of the holdings' variances and the variance of the portfolio itself. It is a second-order effect, positive only when correlations are low, and it disappears entirely as correlations approach 1. In most taxable accounts it is smaller than the tax bill a trade to fix the drift would trigger, which is why rebalancing is a risk-control decision more than a return strategy.
How the band and contribution modes work
The 5/25 rule triggers a trade when a holding is more than 5 points from its target, or more than 25% of its target away, whichever binds first. The band is tested on the weight a holding would carry after new money lands, not on today's weight, because a ten-point overweight can fall back inside a five-point band on the strength of a contribution alone. Drift itself is reported as half the sum of the absolute weight deviations, since every point overweight is matched by a point underweight somewhere else, and counting both would count the same misallocated dollar twice. Contribution-only mode allocates new money by water-filling the largest dollar gaps down to a common level, which minimises total weight deviation and never emits a sale.
What JMM thinks: rebalancing is risk control, not a return strategy
The rebalancing bonus is asserted constantly and computed almost never; the number above is computed from your own holdings rather than repeated as a claim. That leaves rebalancing doing what it is actually for: stopping a portfolio from quietly becoming a different portfolio. Rebalance on a band rather than a calendar, do it with new contributions whenever the contribution is large enough, and in a taxable account let the tax number decide. If correcting three points of drift costs several hundred dollars in realised gains, the drift is cheaper than the fix and you should leave it; in a tax-sheltered account there is no such argument. Gains are pro-rated against average cost, which is what a broker reports by default; selling specific high-basis lots would realise less, and no calculator can compute that without your lot detail. To check what your target weights imply for risk before committing to them, use the portfolio volatility calculator.
The next question after this one
Portfolio volatility calculatorCalculatorWhat the drift did to the risk before you correct it, and which holding was carrying it.Open nextOther questions
All 130 calculatorsHow JMM sources and checks its numbersWhat stays free, and what early access would add
What Pro adds here
Per-lot tax optimisation, multi-account asset location, and rebalancing across taxable and sheltered accounts at once are Pro features. The launch list sends one email at launch.