The Unintended Consequences of Rebalancing
Campbell R. Harvey (Duke University), et al.
January 2025
Institutional investors engage in trillions of dollars of regular portfolio rebalancing, often based on calendar schedules or deviations from allocation targets. We document that such rebalancing has a market impact and generates predictable price patterns. When stocks are overweight, funds sell stocks and buy bonds, leading to a decrease in equity returns of 17 basis points over the next day. Our results are robust to controls for momentum, reversals, and macroeconomic information. Importantly, we estimate that current rebalancing practices cost investors about $16 billion annually-or $200 per U.S. household. Moreover, the predictability of these trades enables certain market participants to profit by front-running the orders of large institutional funds. While rebalancing remains a fundamental tool for investors, our findings highlight the costs associated with prevailing strategies and emphasize the need for innovative approaches to mitigate these costs.

Index Rebalancing and Stock Market Composition: Do Index Funds Incur Adverse Selection Costs?
Marco Sammon (Harvard Business School) and John J. Shim (U. of Notre Dame)
January 2025
We find that index funds incur adverse selection costs from responding to changes in the composition of the stock market. This is because indices rebalance directly in response to composition changes to maintain a value-weighted portfolio, both on the extensive margin (IPOs/delistings or additions/deletions) and intensive margin (issuance/buybacks). This rebalancing approach successfully tracks the market as it evolves, but effectively buys at high prices and sells at low prices. Using several long-short portfolios, we estimate that both intensive-margin and extensive-margin rebalancing trades lead to around a -4% return per year, though only the intensive-margin portfolio’s return is robust to factor exposures. Despite representing less than 10% of index funds’ AUM, these rebalancing portfolios do poorly enough to drag down overall index fund returns. We estimate that a “sleepy” strategy that is less responsive to changes in the stock market’s composition improves fund returns by 20 to 80 bps per year, and is monotonically increasing in how sluggishly it responds to compositional changes. We argue this is because sleepy rebalancing avoids the short- and medium-term adverse selection associated with relatively quickly taking the other side of firms’ primary and secondary market activity. Our findings highlight a large cost incurred by passive investors that simply wish to track the total stock market. And, our proposed simple alternative stock index design boosts returns by an order of magnitude more than typical index fund expense ratios.

Household Stock Market Participation: Learning from Pension Fund Asset Allocation
Ulf Nielsson (Copenhagen Business School), et al.
February 2025
The global shift towards defined contribution pension schemes significantly increases households’ exposure to equity markets through their pension savings. Does this shift also influence households’ non-pension investment choices, and if so, why? We present evidence demonstrating that after an increase in the stock market exposure of their pension plans, households are more likely to subsequently invest in equities through their non-pension savings, a finding with implications for the depth of equity markets. We interpret our results within the framework of a simple model, illustrating how equity exposure via pension savings can lower barriers to stock market participation for non-pension investments.

Asset Allocation Mandates and Price Reactions to New Information
Ruosen Yang (Monash University)
October 2024
This study investigates how asset allocation mandates influence stock price reactions to new information. By examining mutual funds with such mandates (rebalancing funds) and using earnings surprises as the new information, I find that stocks with higher rebalancing funds ownership exhibit smaller price reactions to earnings surprises. Specifically, a one-standard-deviation increase in rebalancing fund ownership corresponds to a 12% reduction in price reaction. This effect only exists when the level of earnings surprises and rebalancing fund ownership is sufficiently high and is stronger when the magnitude of aggregate earnings surprises is larger. The results align with the rebalancing mechanism where funds need to rebalance their portfolios when price reactions cause deviations from the target allocation.

Optimizing Portfolio Performance through Clustering and Sharpe Ratio-Based Optimization: A Comparative Backtesting Approach
Keon Vin Park (Seoul National University)
January 2025
Optimizing portfolio performance is a fundamental challenge in financial modeling, requiring the integration of advanced clustering techniques and data-driven optimization strategies. This paper introduces a comparative backtesting approach that combines clustering-based portfolio segmentation and Sharpe ratio-based optimization to enhance investment decision-making. First, we segment a diverse set of financial assets into clusters based on their historical log-returns using K-Means clustering. This segmentation enables the grouping of assets with similar return characteristics, facilitating targeted portfolio construction. Next, for each cluster, we apply a Sharpe ratio-based optimization model to derive optimal weights that maximize risk-adjusted returns. Unlike traditional mean-variance optimization, this approach directly incorporates the trade-off between returns and volatility, resulting in a more balanced allocation of resources within each cluster. The proposed framework is evaluated through a backtesting study using historical data spanning multiple asset classes. Optimized portfolios for each cluster are constructed and their cumulative returns are compared over time against a traditional equal-weighted benchmark portfolio.

Learn To Use R For Portfolio Analysis
Quantitative Investment Portfolio Analytics In R:
An Introduction To R For Modeling Portfolio Risk and Return
By James Picerno