fundamentals

What Is an Index Fund?

"Tracks the market" is technically accurate and practically useless. Here's how index funds actually replicate a benchmark, why active management faces a mathematical headwind after costs, and what concentration you take on when you buy one.

Engineered Finance 21 May 2026 · 8 min read

The standard explanation — "an index fund tracks the market" — is accurate and nearly useless. It doesn't tell you how the tracking works, why index funds consistently outperform most active alternatives over long horizons, or what you actually own when you buy one.

The engineering framing is more precise: an index fund is a replication system. The objective is to match a target benchmark's return with minimum tracking error and minimum cost. How that replication is implemented, and what it costs, determines whether the fund is actually useful.

Start with the index itself, because most people have the wrong mental model of what they are replicating.

What a Market Index Actually Is

A market index is not a basket of equally-weighted companies. The S&P 500 holds 500 companies, but the effective exposure is heavily concentrated in the largest ones. Each company's weight equals its market capitalization divided by the total market cap of all 500 companies combined.

In practice, the ten largest holdings can represent 30-35% of the entire index. The 250th company might account for 0.05% of your exposure. The 1st company might account for 7%. Buying "the market" means owning a lot of the biggest companies and progressively less of everything below them.

Market-cap weighting has one useful mechanical property: it is self-adjusting. Companies that grow larger become more heavily weighted automatically, without any trading required. When a company shrinks, so does its weight. This keeps transaction costs low and makes the index easy to replicate, but it also means the index tilts toward whatever has already appreciated the most.

How a Fund Replicates an Index

There are two replication approaches. Full replication holds every security in the index at its exact weight, continuously rebalanced as market caps shift and the index composition changes. Sampling holds a representative subset: the largest positions in exact proportion, smaller positions approximated. For the S&P 500, full replication is tractable; indices with thousands of securities or illiquid constituents often rely on sampling.

The measure of replication quality is tracking error: the standard deviation of the difference between the fund's daily return and the index return. For a well-managed large-cap index fund, annual tracking error is typically 0.01-0.05%. The expense ratio creates a predictable and expected performance gap below the index; genuine tracking error beyond that is the quality metric to watch when comparing funds.

One mechanism that partially offsets expenses is securities lending. Index funds often lend their holdings to short sellers in exchange for a fee. For large funds holding heavily-shorted securities, that income can cover a meaningful fraction of the expense ratio, sometimes reducing the effective cost below the stated rate.

Deep Dive — Why Active Management Must Underperform in Aggregate +

Sharpe's Arithmetic

William Sharpe demonstrated in a 1991 paper that before costs, active management as a whole must earn exactly the market return. The reasoning is straightforward: all investors together own the entire market. Passive investors hold the market portfolio by definition. Active investors hold everything else — which is also the market portfolio in aggregate. Before costs, the average active dollar must earn the same return as the average passive dollar.

After costs, active management must underperform. The formal argument:

Let M = gross market return Let P = passive investors' aggregate return = M (by construction) Let A = active investors' aggregate return P + A = M (all investors together own the market) ∴ A = M (before costs) After costs: P_net = M − ER_passive A_net = M − ER_active − transaction costs − tax drag Since ER_passive << ER_active, and transaction costs > 0: A_net < P_net (in aggregate, necessarily)

This is not a claim that no individual active manager can outperform. It is a mathematical statement that they cannot all outperform simultaneously. For every active dollar that beats the index, another active dollar underperforms by an offsetting amount, before costs. After costs, the average active dollar must lag.

The Persistence Problem

The follow-on question is whether a subset of active managers outperforms persistently enough to be identifiable in advance. SPIVA (S&P Indices Versus Active) data shows that over 20-year periods, roughly 85-90% of active large-cap US equity funds underperform their benchmark after fees. The 10-15% that outperform in one period show limited persistence in the next. Identifying the future outperformers from the full population of active funds is the challenge, and the fees paid while searching are guaranteed regardless of outcome.

Where Active Management Has a Stronger Case

Sharpe's arithmetic applies most cleanly to markets with many participants and efficient price discovery. In less liquid or less followed markets — small-cap international, emerging markets, high-yield credit — active managers may have better access to information advantages. Whether that advantage survives after fees is an empirical question that varies by asset class and time period. The case for indexing is strongest in large-cap US equities, where market efficiency is high and active manager fees represent a large fraction of the available outperformance.

The Cost That Compounds

The expense ratio is a guaranteed first-order deduction from gross return, applied every year regardless of performance. It does not give up when markets are down. At 7% gross and a 0.03% expense ratio, net return is 6.97%. At 0.75%, net return is 6.25%. A 0.72 percentage point difference.

That gap compounds. On a $50,000 starting balance with $6,000 in annual contributions over 30 years, the difference between 0.03% and 0.75% accumulates to roughly $100,000 in final portfolio value. The tool below lets you adjust the inputs; the cumulative impact is almost always larger than it appears from the annual percentage alone.

The expense ratio is also not the full cost in a taxable account. Portfolio turnover generates capital gain distributions: when a fund sells appreciated securities, shareholders receive a taxable distribution even if they sold nothing. Active funds typically run 50-100% annual turnover — large-cap index funds are often below 5%, since index composition changes infrequently. In the 24% federal bracket, substantial gain distributions from a high-turnover fund add meaningful drag that does not appear in the stated expense ratio.

Deep Dive — The Full Cost Stack: Expense Ratio, Tax Drag, and Market Impact +

The Compounding Math

Given starting balance B, annual contribution C, gross return r, and expense ratio e, net return is r_net = r − e. After n years with end-of-year contributions:

FV = B × (1 + r_net)^n + C × [(1 + r_net)^n − 1] / r_net

The expense ratio drag is not a flat fee — it is a percentage of an ever-larger balance. At 0.75%, year one on a $50,000 portfolio costs $375. Year 30 on a $500,000 portfolio costs $3,750. The fee accelerates as the portfolio grows, which is why the cumulative impact is so much larger than the annual percentage suggests.

Tax Drag on Turnover

In a taxable account, fund turnover creates taxable events. A fund with 80% annual turnover sells most of its portfolio each year, realizing gains. Those gains are distributed as either short-term (taxed at ordinary rates, up to 37%) or long-term capital gains (0%, 15%, or 20%). Estimating the additional drag:

Tax drag estimate = turnover rate × avg embedded gain % × marginal tax rate Example: 80% turnover × 3% avg gain × 24% rate = 0.58% additional drag per year

This drag compounds in the same way as the expense ratio. It also does not appear in any fund's marketing materials or Morningstar summary.

Total Cost Stack for a Taxable Account

Active fund (estimated): Expense ratio: 0.75% Tax drag on turnover: 0.50% Market impact, spreads: 0.10% ────────────────────────────── Total annual drag: 1.35% Index fund (estimated): Expense ratio: 0.03% Tax drag on turnover: 0.01% Market impact, spreads: 0.01% ────────────────────────────── Total annual drag: 0.05%

The full gap in a taxable account is often closer to 1.3 percentage points per year, not the 0.72 point difference in the stated expense ratios. For the active fund to break even after all costs in a taxable account, it would need to generate persistent gross outperformance of roughly 1.3% annually. The SPIVA data on how often that occurs over 20-year periods is not encouraging.

// Interactive

Adjust the expense ratios and time horizon to see how the cost differential compounds. Fund A defaults to a typical index fund; Fund B defaults to the low end of the active fund range.

What You Actually Own

Market-cap weighting creates concentration at the top. In the S&P 500, the ten largest companies by market cap typically account for 30-35% of total index weight. "Diversified across 500 companies" and "equally diversified across 500 companies" describe very different portfolios. You likely own the former, though there are equally-weighted funds available for many indexes.

However, this may not be a flaw to correct. Large companies are large because they have generated returns, and owning more of them is a mechanical consequence of tracking that market signal. But it is worth understanding the position you hold: an S&P 500 index fund is a bet on US large-cap equities, skewed toward its largest constituents at any given time.

The S&P 500 is also not the economy. It covers the 500 largest US public companies by market cap, excluding small-cap stocks, private companies, and all non-US markets. A total market index fund extends coverage to small and mid-cap names but remains US-only and market-cap weighted throughout. International exposure requires a separate allocation. None of these are flaws in the instrument; they are positions you hold deliberately once you understand the mechanics.

Where Index Funds Show Up

Once you understand the mechanics, the questions that follow are portfolio construction decisions: which index, which account type, which provider.

• 401k and IRA equity core — the S&P 500 or total market fund is the right default for the equity allocation in most tax-advantaged accounts; it establishes a low-cost baseline that active fund selection has to beat

• Taxable account equity — where tax efficiency matters most; low-turnover index funds have a meaningful structural advantage over high-turnover active funds through reduced capital gain distributions

• International allocation — total international index funds extend the same logic globally; a reasonable allocation is 20-40% of total equity depending on conviction about US vs. international valuations

• Factor tilts — small-cap value indices, dividend indices, and sector funds allow systematic factor exposures without individual stock selection

• Fixed income — bond index funds are reasonable, though the active-versus-passive case is less decisive in fixed income than in equities

Each of these gets its own treatment on this site. The mechanics are consistent throughout: define a target, track it at minimum cost, and let the math work over time.

Takeaways

• A market index is not an equal-weighted basket; market-cap weighting means the largest companies dominate the effective exposure

• Before costs, active management in aggregate must match passive management in aggregate; after costs, it must underperform — this is arithmetic, not opinion

• The expense ratio is a compounding headwind applied every year against an ever-larger balance; the cumulative impact is substantially larger than the annual percentage implies

• In taxable accounts, portfolio turnover creates additional tax drag that does not appear in the expense ratio; the full cost gap between low- and high-turnover funds is wider than stated fees suggest

• The S&P 500 is not the economy: it is a US-only, large-cap, market-cap-weighted portfolio, and treating it as a complete picture of diversification understates the positions you are implicitly taking

fundamentals