Portfolio Rebalancing: All at Once or Little by Little?

Rebalancing is important for keeping an investor’s portfolio aligned with its goals. Determining how often to rebalance involves considering the tradeoffs between rebalancing costs and drift from the target asset allocation.¹ Once you’ve decided to rebalance, though, does it matter how quickly you trade back to target? What is the impact of trading immediately versus over multiple days?

To explore these questions, we construct a set of hypothetical 60% equity/40% fixed income (60/40) portfolios that hold the Russell 3000 Index for equity exposure and the Bloomberg US Aggregate Bond Index for fixed income exposure. The portfolio weights change daily based on the performance of the indices. We vary the rebalancing approaches across the portfolios along three dimensions: rebalance tolerance band size, rebalance end point, and, most importantly for this study, rebalance trading speed.

We rebalance each portfolio using a tolerance band approach set at the following tolerance band sizes: 1%, 2%, or 3%. A tolerance band is a weight threshold around a target weight that, when breached, triggers a rebalance. For instance, a 2% band for a 60/40 portfolio means that when the equity sleeve moves above 62% or below 58% (or equivalently, when the fixed income sleeve moves below 38% or above 42%), a rebalance is initiated. When the portfolios rebalance, they rebalance to one of two end points: back to the asset class target weights (60/40) or to the nearest edge of the tolerance band that was breached.²

In terms of rebalance trading speed, the portfolios use one of four approaches: immediate rebalance, rebalance over five trading days, rebalance over 10 trading days, or rebalance over 20 trading days. This results in 24 portfolios (3 tolerance band sizes x 2 rebalance end points x 4 trading speeds) that we examine from January 1989 through December 2025.³ The primary focus of our study is the impact of trading at different speeds on returns and turnover.

In Exhibit 1 below, we compare the annualized compound return across the portfolios. Within each portfolio group (grouped by rebalance end point and tolerance band size), there is not a clear performance winner among rebalancing immediately versus over five, 10, or 20 trading days—the annualized returns are similar, and no trading speed consistently outperforms the others. Across all groups, the average difference in annualized compound return between the best- and worst-performing trading speed portfolios was only 4 basis points. Moreover, the difference in average monthly return between portfolio pairs in each group is not statistically reliable in most cases.

In contrast with returns, we see meaningful differences in average annualized one-way turnover across the four trading speed approaches as shown in Exhibit 2. Turnover decreases as trading speed slows down (in other words, as the rebalance trading period lengthens) across all portfolio groups, with larger differences for tighter bands. In Panel B, for example, for the group that rebalances to a 1% band, average annualized one-way turnover decreases from 7.1% for immediate rebalance to 4.7% for rebalance over 20 trading days for the period 1989 through 2025. A rebalancing approach that trades over a longer period allows more time for natural market movement to potentially move the portfolio back toward its target weights, reducing the amount of trading needed, and lowers the risk of unnecessary back-and-forth trading in volatile markets.

These results suggest that the decision of how quickly to rebalance back to target should not be based on expectations of return differences. In our study, rebalance trading speeds over a wide range—from immediate to up to 20 trading days—resulted in similar returns but different levels of turnover, particularly when using tighter rebalance tolerance bands. For investors able to spread trading out in a cost-effective manner, applying a slower trading approach when rebalancing may lead to lower portfolio turnover without a significant impact on returns.

Hypothetical Portfolio Construction and Trading Methodology Information

Each hypothetical 60/40 portfolio holds the Russell 3000 Index in the equity sleeve and the Bloomberg US Aggregate Bond Index in the fixed income sleeve. For our analysis, trading decisions are determined based on end-of-day weights and, if a trade is triggered, trading starts the following day. For portfolios that rebalance back to asset class target weights, when a rebalance is initiated, the portfolio trades back to 60% equity/40% fixed income during the rebalancing event. For portfolios that rebalance back to tolerance band, when a rebalance is initiated, the portfolio trades back to the nearest edge of the band that was breached. For example, when the equity weight in a portfolio with a 2% band and that rebalances back to band ends a trading day at 62.5%, the following day a rebalance is initiated that will move the portfolio back to 62%, which is the nearest edge of the band that was breached.

For the three trading approaches that rebalance over multiple days, the number of days traded for a given rebalancing event may be less than the planned number of days due to market movement. During a rebalancing event, the amount traded each day is determined by the distance to the rebalance end point and the planned number of trading days remaining, with the distance based on beginning-of-day weights. Consider day one of a rebalancing event for a 60/40 portfolio with a rebalance to target x 1% band x rebalance over 5 trading days. The portfolio ended the prior day with an equity weight of 61.5%. The trading approach will trade the portfolio 0.30% back toward 60% equity/40% fixed income (1.50% distance to rebalance end point divided by 5 days remaining = 0.30% trade for the day). If market returns move the portfolio further from the rebalance end point, the trading approach will account for this by making larger trades on the days remaining in the event. If the market moves the portfolio closer to the rebalance end point, the trading approach will make smaller trades on the days remaining. If the end-of-day weights on a day during the event are within 10 bps of the rebalance end point, the trading event will end early.

All performance results of the hypothetical “60/40 Portfolios” are based on performance of indices with model/backtested asset allocations; the performance was achieved with the benefit of hindsight; it does not represent actual investment strategies. The model’s performance does not reflect advisory fees or other expenses associated with the management of an actual portfolio. There are limitations inherent in model allocations. In particular, model performance may not reflect the impact that economic and market factors may have had on the advisor’s decision-making if the advisor were actually managing client money.

Past performance is no guarantee of future results. Indices are not available for direct investment; therefore, their performance does not reflect the expenses associated with the management of an actual portfolio. Frank Russell Company is the source and owner of the trademarks, service marks, and copyrights related to the Russell Indexes. Bloomberg data provided by Bloomberg Finance LP.