Why Sampling-Based Index Funds Underperform: How Replication Method Costs You 50–70 Basis Points Annually

The Index Fund You Think You're Buying Might Not Be the One You Own

You've decided to go passive. You're buying an index fund to track the S&P 500, FTSE 100, or some broad market benchmark. Simple. Low cost. No fuss. That's the whole idea, right?

Here's what many investors don't realise: not all index funds actually hold the index. Some hold only a carefully selected subset of stocks and call it "replication." And that choice—one the fund company makes, usually without much fanfare—can quietly cost you significant returns over a decade.

Related reading: Why Where You Buy Index Funds Matters More Than Which One: The Account Type Strategy That Compounds Over Decades Why 80% of New ETFs Launched in 2026 Are Actively Managed: The Shift From Passive Index Tracking

Two Very Different Ways to Track an Index

There are really only two main approaches an index fund manager can use.

Full Replication: The Straightforward Approach

Full replication mimics the index with exactness. If you're tracking the S&P 500, you own all 500 stocks in the same weights as the index. If the index is reconstituted, you adjust all your holdings accordingly. It's transparent, mechanical, and predictable.

Representative Sampling: The "Efficient" Alternative

Representative sampling holds a subset of the index. Instead of buying all 500 stocks (or worse, all 3,000+ components in a total market fund), the manager selects a representative sample—often 80% to 90% of holdings by value—and calls it done. The logic is simple: why pay transaction costs to own the tiniest, most illiquid positions if they barely move the needle?

That logic sounds reasonable. Fund managers like the stratified sampling approach because it's easier, faster, and usually involve fewer transactions than fully replicating an index. But what actually happens in practice tells a very different story.

The Data: Sampling Underperforms by 50–70 Basis Points Per Year

A comprehensive academic study analysed 3,365 U.S.-based equity index funds and ETFs from 2010 to 2020. Among the studied funds, 52% use physical replication, 37% use representative sampling, and the remainder are "hybrids" that typically employ physical replication but may implement sampling under certain conditions.

The results are stark. Samplers trade 3-4 times more, have 30-50% higher expenses and fees, and earn 50-70 basis points lower annual returns, which is substantial given index funds' mandate to limit tracking error to a few basis points.

To put that in perspective: The differences in costs and returns translate into about 60 basis points lower returns for samplers per year on a net return basis. On a $100,000 portfolio, that's roughly $600 per year. Over 20 years, with compounding, the gap widens dramatically.

Why Does Sampling Fail? It's Not Just Higher Expenses

You might think sampling underperforms simply because the funds charge higher fees. But the problem runs deeper.

Only about 25% of the return difference appears to be explained by higher expenses, leaving 75% to be explained by other factors such as additional turnover and poor stock picking. Let that sink in. Three-quarters of the underperformance has nothing to do with stated fees.

Here's what's actually happening:

1. Samplers Trade Far More Than Expected

When you hold a subset rather than the whole index, you're making a bet on which stocks matter most. That bet requires constant tweaking. The selection of stocks using variables other than index weights more than offsets any reduction in trading arising from holding fewer positions, with turnover often substantially higher by 3 to 4 times.

Every trade creates friction: bid-ask spreads, market impact costs, and in taxable accounts, potential capital gains realisation. It adds up fast.

2. Poor Stock Picking Within the Sample

Here's the uncomfortable truth: fund managers selecting which 80% of the index to own aren't any better at picking winners than active investors. Sometimes their sample includes tomorrow's laggards. Sometimes they omit tomorrow's winners. Samplers' underperformance is not purely driven by higher expenses and transaction costs, but also poor stock picking.

3. Rebalancing Challenges Create Drift

The sample is not the index, so its return can drift slightly from the benchmark in either direction, resulting in larger tracking error than a fully replicated fund would have.

Where Sampling Actually Makes Sense (But Remains Risky)

There are real situations where sampling is defensible—but not in the way it's typically applied.

Not all markets are as liquid as the companies in the S&P 500, and some indexes include thousands of constituents, making full replication impractical for ETFs tracking indexes in more complex or less liquid corners of the market.

Fixed-income ETFs nearly all use a sampling approach, because oftentimes, the fund's underlying index holds thousands of bonds, many of which haven't traded in years. That's legitimate. A bond sampling strategy can actually improve returns by avoiding illiquid micro-positions.

But for equity funds tracking broad, liquid indices like the S&P 500, Russell 3000, or the FTSE 100? Broad liquid equity funds like VOO and VTI fully replicate, delivering near-perfect tracking. The cost savings from sampling don't materialise—they evaporate into hidden underperformance.

How to Check What You Actually Own

Most fund companies are transparent about this, at least on paper. Fund companies almost universally (99%) divulge which of two indexing approaches they use: full replication or representative sampling.

But the information isn't always prominent. Look at your fund's prospectus. Search for "replication," "sampling," or "representative sample." If it says the fund holds a subset or uses "optimisation," that's sampling. If it says it holds all or substantially all constituents, that's replication.

For large, liquid equity indices, ask yourself: Why would the fund need to sample? If it's not an explicit cost-saving story for bonds or emerging markets, it's likely just underperformance waiting to happen.

The Practical Takeaway

For most individual investors in English-speaking markets, the choice is straightforward:

• For large-cap equity indices (S&P 500, Russell 1000, FTSE 100, ASX 200): Choose funds that fully replicate. The cost difference is negligible, and the tracking accuracy is measurably better.
• For bond indices: Sampling is often the smarter choice—provided the fund's historical tracking error is low (under 10 basis points annually).
• For emerging markets or small-cap equities: Check the fund's actual tracking error. If it's better than 10–15 basis points, sampling might be working. If it's drifting higher, full replication alternatives are worth comparing.

The real lesson: costs and fees aren't the only way index funds leak performance. The structural choice of how to build the portfolio matters just as much. Representative sampling is detrimental to index investors across the broad, liquid equity markets where most people invest.

You're paying for simplicity and low cost. Make sure you're actually getting them.

Disclaimer

This article is for informational and educational purposes only and does not constitute financial advice. This is not investment advice, and you should not rely on it to make financial decisions. The specific performance figures cited come from academic research on historical fund data and should not be interpreted as predictions of future results. Interest rates, market conditions, fund structures, and regulatory requirements vary by country and change over time. Consult a qualified financial advisor before making any investment decisions, and always verify current fund prospectuses and performance records with the fund provider or your financial adviser before investing.