The Hidden Cost of Rebalancing: How Often Should You Trade?

How Often Should You Rebalance a Factor Portfolio?

You have built a momentum strategy that delivers a compelling Sharpe ratio in backtests. The signal updates monthly, so you rebalance monthly. But should you? What if rebalancing weekly captures more alpha? Or what if rebalancing quarterly sacrifices little return while cutting your trading costs in half? The question of rebalancing frequency sits at the intersection of signal decay, turnover, and transaction costs -- and the answer matters more than most investors realize.

Novy-Marx and Velikov (2016) tackled this question head-on in their comprehensive study of anomaly trading costs. Their findings are sobering: for many well-known factor strategies, the gross alpha that appears in academic papers is largely or entirely consumed by the transaction costs required to maintain the portfolio. The choice of rebalancing frequency is not a minor implementation detail. It is a first-order determinant of whether a strategy actually makes money.

The Turnover Tax on Factor Strategies

Every rebalancing event generates turnover. Stocks enter and exit the portfolio, position sizes shift, and each trade incurs costs -- commissions, bid-ask spreads, and market impact. The total cost scales with both the frequency of rebalancing and the amount of turnover per rebalance.

Novy-Marx and Velikov examined a broad taxonomy of 23 anomalies documented in the academic literature and estimated their net-of-cost returns using effective spread measures from Hasbrouck (2009). Their central finding: many anomalies that appear highly profitable on a gross basis become marginal or unprofitable once realistic trading costs are deducted. Short-lived anomalies -- those requiring frequent rebalancing to exploit -- are hit hardest because their high turnover multiplies the per-trade cost across many rebalancing cycles.

The mechanism is straightforward. If a strategy requires 200% annual turnover and incurs 50 basis points per trade in one-way costs, the annual cost drag is 2 x 0.50% = 2.00% of portfolio value. For a strategy with a gross alpha of 3%, this leaves only 1% net -- and that is before accounting for tracking error, implementation shortfall, or any estimation error in the alpha itself.

Rebalancing Frequency and Net Alpha: The Trade-Off

The relationship between rebalancing frequency and net performance is not linear. More frequent rebalancing captures more of the raw signal but incurs proportionally higher costs. Less frequent rebalancing reduces costs but allows the portfolio to drift from its target, leading to alpha decay.

The following table illustrates the trade-off for a stylized factor strategy with varying rebalancing frequencies, based on the patterns documented by Novy-Marx and Velikov (2016):

The pattern is clear. Weekly rebalancing captures the most gross alpha but generates so much turnover that net returns can approach zero. Quarterly rebalancing often emerges as a sweet spot for many equity factor strategies -- it captures a substantial share of the gross alpha while keeping turnover manageable. The optimal frequency depends on how quickly the underlying signal decays and how costly it is to trade the relevant securities.

Not All Anomalies Are Equal

A critical insight from Novy-Marx and Velikov's work is that anomalies differ dramatically in their sensitivity to trading costs. They group anomalies into three categories based on the cost of implementation:

Low-cost anomalies such as gross profitability and return on assets involve low turnover because the underlying firm characteristics change slowly. These strategies can be rebalanced infrequently -- annually or semi-annually -- with minimal loss of signal. Their net alphas remain economically significant even after accounting for trading costs.

Medium-cost anomalies like momentum and earnings surprises require more frequent rebalancing because the signals decay within months. They generate moderate turnover, and their net profitability is sensitive to the chosen rebalancing frequency and execution quality. Simple cost-mitigation techniques -- such as using buy/hold spread ranges rather than hard cutoffs -- can preserve much of the net alpha.

High-cost anomalies including short-term reversals and certain microstructure signals demand near-continuous trading. The turnover is extreme, and transaction costs consume nearly all of the gross return. These strategies are generally viable only for investors with direct market access and the lowest possible execution costs.

DeMiguel, Martin-Utrera, Nogales, and Uppal (2020) extended this line of research by showing that turnover-penalized portfolio optimization -- which explicitly incorporates trading costs into the objective function -- can substantially improve net Sharpe ratios relative to naive rebalancing rules.

Practical Cost Mitigation

Novy-Marx and Velikov propose several practical strategies to reduce the turnover tax without abandoning factor exposure:

Wider rebalancing bands. Instead of rebalancing to the exact target portfolio each period, use tolerance bands. A stock remains in the portfolio until it drifts sufficiently far from the target threshold. This approach alone can reduce turnover by 30-50% with only modest alpha decay.

Less frequent rebalancing. For anomalies with slow signal decay, simply trading less often is the most effective cost-reduction lever. Moving from monthly to quarterly rebalancing often preserves 70-80% of gross alpha while cutting costs by half or more.

Patience in execution. Spreading trades over multiple days reduces market impact, particularly for less liquid names. The cost savings from patient execution can add 50-100 basis points of net annual return for capacity-constrained strategies.

These mitigation strategies are not theoretical curiosities. They represent the difference between a strategy that survives contact with real markets and one that does not.

Implications for Factor Investing

The rebalancing frequency question has direct implications for how investors should evaluate factor strategies and the products built around them. A factor-investing product that rebalances monthly is not necessarily superior to one that rebalances quarterly, even if the monthly version captures a slightly higher gross alpha. The net return -- after all costs of turnover, tracking error management, and execution -- is what matters.

For practitioners constructing factor portfolios, the message from the research is clear: start with a realistic cost model, choose a rebalancing frequency that maximizes net (not gross) alpha, and embed cost awareness into every stage of portfolio construction. The hidden cost of rebalancing is only hidden if you choose not to look.