top of page

Index investing: Is replication or sampling best?

Updated: Nov 13





By LARRY SWEDROE


Investors who choose to invest in the stock market through index funds can do so through one of two main strategies — replication or sampling. In other words, they can either use funds that replicate the index or funds that engage in what's known as representative sampling.


Owning each stock in proportion to the underlying index is known as “full replication”. While full replication eliminates, or at least minimises, any tracking variance, the strategy requires adjustments to all the portfolio’s positions each time an index adds or removes a stock. Since many of the required adjustments are small and involve less illiquid stocks, creating the potential for large trading costs that reduce the benefits of replication, to mitigate these costs some funds update their portfolio weights only periodically (such as monthly) instead of immediately adjusting positions each time an index is reconstituted. In addition, they may not update each position, especially those with a smaller initial index weight or a smaller change in index weight. These practices can reduce trading costs, leading to higher returns, but can also create the potential for tracking errors that violate the fund’s mandate to follow the index.


Unlike full replicators, representative samplers select a subset of securities from the index using characteristics (such as dividends or earnings yields) other than index weights while maintaining the goal of matching index returns. Sampling creates the potential for increased tracking errors. However, because the strategy requires holding fewer stocks, it may reduce trading costs. For example, because they do not hold the entire index, samplers might be able to avoid the most illiquid stocks or avoid forced trading following index reconstitutions.


In addition, the sampling manager’s skill and luck at identifying reliable indicators of expected returns could help or hurt fund performance.


Travis Dyer and Nicholas Guest, authors of the March 2022 study Disparate Investment Outcomes of Index Funds: Replication vs Sampling, examined the performance of equity index funds using the two strategies to determine if sampling added value. Their study covered the period 2010-2020 and 3,365 fund-year observations. Following is a summary of their findings:


  • 52 percent of index funds were replicators and 37 percent were samplers. The other 11 percent were “hybrids”, which typically use replication but may implement sampling under certain conditions (such as periods of market illiquidity).


  • The popularity of the sampling approach was about the same at the beginning and end of the sample period.


  • Funds focusing on domestic stocks were less likely to use sampling compared to their foreign fund counterparts.


  • Funds were more likely to be samplers when they followed an index based on a particular style or sector than when the index was based on market cap.


  • Samplers trade more — the selection of stocks using variables other than index weights more than offsets any reduction in trading arising from holding fewer positions. Turnover was often substantially higher (by 3 to 4 times). They also had higher expenses (0.14 percent) and earned lower returns (replicators outperformed samplers by about 60 basis points per year).


  • Higher expenses and fees, greater transaction costs and poor stock picking each contributed to samplers’ return underperformance — only about 25 percent of the return difference was explained by higher expenses, leaving 75 percent to be explained by other factors.


  • Sampling funds produced marginally higher return volatility and significantly lower risk-adjusted returns — their lower returns were not the result of taking less risk.


  • The results were robust to subsamples of S&P 500 indexers and other market-cap-based indexers.


  • Their results were strongest among funds following indices with fewer constituent stocks, entirely disappearing for samplers following indices with 1,000 or more stocks.


Their findings led Dyer and Guest to conclude: “Our analyses suggest that sampling is a relatively active strategy that is detrimental to index investors’ outcomes. Moreover, the cost, performance, and flow differences we document have economically large wealth implications for investors, many of whom largely ‘set and forget’ their index investments for lengthy periods of time.”


The authors added that, given their findings, it was counterintuitive that investors’ funds increasingly flowed to samplers, suggesting that most retail investors and even many finance professionals are likely unaware of the evidence.



Investor takeaways

While index investors are encouraged to focus on such factors as expenses and fees, the evidence presented demonstrates they should also be focused on the replication versus sampling dimension of index investing. Unfortunately, individual investors may have little awareness or understanding of the extent to which fund expenses and returns are driven by the mechanics, including replication versus sampling, underlying their fund managers’ efforts to track an index. You no longer have that excuse.




© The Evidence-Based Investor MMXXIV. All rights reserved. Unauthorised use and/ or duplication of this material without express and written permission is strictly prohibited.

bottom of page