Index Fund

Reviewed 12 August 2026 · Sourced from the SEC, William Sharpe's 1991 “Arithmetic of Active Management,” S&P Dow Jones Indices' SPIVA scorecards and published index methodology, FTSE Russell, and the Economic Report of the President

What an index fund actually is

You have finished the free course, the emergency fund is not zero anymore, and there is $150 a month nothing else has a claim on. You search for a fund. Two results sit next to each other holding almost the same list of American companies. One charges four hundredths of a percent a year. The other charges eighty-five hundredths — roughly twenty times as much. Nothing explains why, and no bill for either number will ever arrive, so it is hard to believe it matters. It matters more than almost anything else you will decide here.

An index fund is a fund whose stated objective is to track a specified market index rather than to beat it. The SEC's investor education material puts it directly: index funds “follow a passive investment strategy that is designed to achieve approximately the same return as a particular index before fees,” and such a fund “will attempt to achieve its investment objective primarily by investing in the securities of companies that are included in a selected index.”

Notice what the manager's job becomes. Not deciding which companies are undervalued, not timing anything — the job is minimizing the difference between the fund's return and the index's return, by buying what the rulebook says in the weights it says, cheaply enough that the gap stays small. Success is measured as an absence.

The words doing the quiet work in that SEC sentence are before fees. An index fund matches the index before its own costs come out — so after costs it is designed to land slightly below the index, forever. Not a flaw somebody could fix. The size of that shortfall is the only part you get to choose.

Index fund and ETF are two different questions

These two terms get used as if they were the same word. They are not even on the same axis.

“Index fund” describes the strategy — what the fund does with your money. “ETF” describes the structure — how shares are created, priced and traded. Neither one implies the other.

All four squares are occupied. So “I bought an ETF” tells you nothing about whether anybody is picking stocks, and “I bought an index fund” tells you nothing about how you will buy or sell it. The wrapper mechanics — how the price stays near what the fund holds, premiums and discounts, the bid-ask spread — are a page of their own: read ETF. This page stays on the strategy.

How a small investor actually buys one

• Minimums. A mutual fund can set a minimum initial investment in its prospectus. An ETF has no minimum beyond the price of one share.
• Fractional shares. Many brokers let you buy part of a share, which is how $150 becomes a purchase instead of a rounding problem. Read the SEC's bulletin first: “Not every brokerage firm offers fractional share investing,” you “generally cannot transfer fractional shares to another brokerage firm,” and some firms “do not guarantee the liquidity of fractional shares.” That transfer limit surprises people — switching brokers can mean your fractions get sold, a taxable event in a regular brokerage account.
• Automatic contributions. Recurring purchases are usually simpler in a mutual fund, which transacts at one price a day. Brokers increasingly offer recurring ETF buys too, and this varies entirely by firm.
• Where it sits. A fund is not an account. The same index fund can live in a workplace plan, a Roth IRA or a taxable brokerage account, and the tax treatment differs completely in each. In a taxable account you track cost basis; inside a Roth IRA that question never arises.

The index is a rulebook somebody maintains

An index fund tracks an index. So the honest next question is what an index is, and who decides what goes in it.

An index is a published set of rules, written and maintained by a commercial provider, for which securities are included and how much each counts. In U.S. equities the big four are S&P Dow Jones Indices (the S&P 500 family), FTSE Russell (the Russell 1000, 2000 and 3000), CRSP — the Center for Research in Security Prices at the University of Chicago Booth School of Business, behind widely tracked total-market indexes — and MSCI internationally. Each publishes its methodology as a document you can download.

How much of each company you end up owning

Most broad stock indexes weight by float-adjusted market capitalization. S&P's methodology for its U.S. indices states that “the indices are weighted by float-adjusted market capitalization,” applying an Investable Weight Factor — the share of a company's stock actually available to public investors, excluding blocks held by founders, governments and other strategic holders. For S&P Composite 1500 eligibility a company must “have an IWF of at least 0.1.” So your fund owns companies in proportion to their publicly tradable value, and the largest company gets the largest slice automatically. See market capitalization.

The part that surprises people: a committee

The S&P 500 is not a mechanical list of the 500 biggest companies. S&P's own methodology says “constituent selection is at the discretion of the Index Committee and is based on the eligibility criteria,” and that S&P reserves “the right to make exceptions when applying the methodology if the need arises.”

The rules also change on a schedule. S&P updates share counts and float factors quarterly, with rebalancings effective “after the close of business on the third Friday of March, June, September, and December.” FTSE Russell rebuilds its U.S. family in an event called reconstitution, which per FTSE Russell “is changing from an annual to a semi-annual schedule in 2026”: eligibility was fixed on rank day, Thursday 30 April 2026, and the rebuilt indexes took effect after the U.S. market close on 26 June 2026.

Which index you pick is the active part

These are all index funds: a total U.S. market fund holding thousands of companies; an S&P 500 fund holding, per the methodology, “500 constituent companies”; a single-sector fund; a single-country fund; and a thematic fund tracking an index somebody built last year around one idea. The word “index” does identical work in all five names while describing wildly different amounts of risk.

A thematic index fund is a concentrated bet with an index fund's reassuring name. The rules are published, nobody is picking, the expense ratio may be modest — and you may still own thirty companies in one industry. Everything the next section says about indexing is about tracking the broad market. It says nothing about whether one theme will do well. This is where the label gets abused, and the abuse is legal, because it is technically accurate.

Where the cost advantage comes from, and why it compounds

The cost gap is not a promotion. It comes from work that does not have to be done.

• No research staff. An active fund pays analysts to study companies, managers to decide, traders to execute. An index fund reads a rulebook, and that payroll difference lands in the expense ratio.
• Low turnover. An index fund trades when the index changes, not when somebody changes an opinion. The SEC makes the link explicitly: “Passive management usually translates into less trading of the fund's portfolio (fewer transaction costs), more favorable income tax consequences (lower realized capital gains), and lower fees than actively managed funds.”

Cost is the only input you can know in advance

You cannot know next decade's return. Nobody can, and any page that says otherwise is selling something. But the expense ratio is printed in the prospectus before you buy — close to the only number in this subject that behaves that way, and the one lever you can pull with certainty about what pulling it does.

And you never see it charged. It is deducted from fund assets, so nothing appears on your statement asking approval — only a return slightly lower than it would have been, every day you hold the fund. Exactly the shape of cost compound interest magnifies.

Why the gap is double the fees

Because a dollar taken out in year three does not just cost you that dollar. It costs you everything it would have earned over the remaining twenty-seven years. Fee drag compounds against you in exactly the way returns compound for you — same curve, same math, pointed the other direction. That symmetry is the most useful idea on this page, and it is why 0.85% does not stay small.

Read the last two columns against each other in every row: the shortfall is right around twice the extra fees, every time. Not specific to these inputs — that is what a small drag does over thirty years.

The arithmetic behind indexing, and what the evidence says

Most pages assert that index funds beat active managers. Two things make that checkable rather than a slogan: a piece of logic, and a standing scorecard.

Sharpe's arithmetic, 1991

William Sharpe laid it out in The Financial Analysts' Journal, Vol. 47, No. 1, January/February 1991, pp. 7 to 9, as “The Arithmetic of Active Management.” A passive investor is one who “holds every security from the market, with each represented in the same manner as in the market”; an active investor is “one who is not passive,” whose portfolio “will differ from that of the passive managers.” Two conclusions follow. Before costs, “the return on the average actively managed dollar will equal the return on the average passively managed dollar.” After costs, it “will be less.”

That is an accounting identity, not a claim about skill. It also does not say no manager can outperform. Some do, and some must, because the group averages out: every dollar of outperformance somebody earns is a dollar somebody else lost. What it establishes is that outperformance is a zero-sum pool with costs subtracted from it.

SPIVA, and exactly what it counts

The standing empirical record is SPIVA — S&P Indices Versus Active — published for over two decades by S&P Dow Jones Indices. The SPIVA U.S. Scorecard, Year-End 2025, published 3 March 2026, reports this for actively managed large-cap U.S. equity funds measured against the S&P 500:

Two things make those numbers mean something. Returns are net of fees, excluding loads. And the denominator is survivorship-corrected: SPIVA uses “the entire investment opportunity set, made up of all funds in that particular category at the outset of the period,” so a fund closed or merged mid-period still counts instead of quietly vanishing. That matters, because the attrition is severe — of the large-cap funds alive 20 years earlier, only about 35% were still operating at the end.

What SPIVA does not tell you

• It measures funds, not your outcome. A percentage of funds is not a percentage of dollars, and not a statement about the fund you would have picked.
• It is backward-looking, and benchmark dependent. Which benchmark a fund is measured against is a methodological choice, SPIVA's approach to it has been challenged in the practitioner literature, and different windows give materially different figures. The companion Persistence Scorecard exists because past winners rarely stay winners.
• S&P Dow Jones Indices is an index provider. It licenses the indexes index funds track and earns more as indexing grows. The data is the best standing record available; the conflict of interest is real, and worth stating rather than ignoring.

Tracking difference, tracking error, and where the gap comes from

An index fund does not equal its index. It approximates it, and the two words for the gap mean different things.

A fund can have a large tracking difference and almost no tracking error — landing 0.85% behind every single year, reliably, because that is its expense ratio. Another can have a tiny average difference and a jumpy one. The first is a cost problem, the second an execution problem, and sales material rarely distinguishes them.

The honest sources of the gap

• The expense ratio. Usually the largest and most predictable piece. The index does not pay operating costs; your fund does — including the fee it pays the provider to license the index, since an index is somebody's intellectual property.
• Sampling versus full replication. A total-market fund may hold every eligible name, or a representative subset chosen to behave like the whole. Sampling exists because the smallest members of a broad index are tiny and thinly traded, so buying them exactly is expensive. It also adds a gap full replication does not have.
• Cash drag. The index is a calculation and is always fully invested. A real fund holds a little cash for contributions, redemptions and dividends not yet reinvested. In a rising market that cash lags; in a falling one it helps slightly.
• Reconstitution and rebalancing. When the rulebook adds and drops names on a published date, every fund tracking that index trades the same direction at roughly the same time. That costs money, which is why the S&P and Russell calendars get watched closely.
• Securities lending, which pushes the other way. A fund may lend portfolio securities for a fee. Per the SEC staff summary, “the borrower posts collateral with the fund” in “an amount at least equal to the value of the borrowed securities, marked to market daily,” and the fund's income “may come from fees paid to the fund by the borrowers and/or from the reinvestment of the cash collateral.” That is fund income, so it offsets costs — occasionally enough that a fund tracks more closely than its expense ratio alone would suggest. It is not free: it carries counterparty risk and risk in how the cash collateral gets reinvested.

An index fund is designed to land slightly below its index after costs. One landing meaningfully further below than its expense ratio explains is telling you something about how it is run.

What an index fund does not protect you from

This is the section that matters most for anybody reading this site, and the one most pages skip.

Diversifying across a broad index does one specific, valuable thing: it removes single-company risk. Own a few thousand companies, and if one commits fraud or goes bankrupt the damage to you is a rounding error. That is most of the case for owning a broad fund instead of five stocks you read about.

It does nothing at all about market risk. When the whole market falls, a fund built to match the whole market falls with it. That is not a malfunction. That is the fund performing exactly as designed.

One more figure from the same reports, for scale: “household and nonprofit net worth declined 20 percent between December 2007 and December 2008, or by about $13 trillion.”

Concentration: the whole market may be less spread out than it sounds

Weight is assigned in proportion to float-adjusted market value. So when a small number of companies grow into a large share of total market value, they become a large share of your fund. Nobody decided that; the weighting rule did what it says. A “total market” fund holding thousands of companies can still carry much of its value in a handful of the biggest names, often in related businesses. You own many companies. You may not own many independent outcomes, which is what diversification is for.

And “diversified” guarantees less in law than it sounds. Under 15 U.S.C. § 80a-5(b)(1), a fund is a “diversified company” if “at least 75 per centum of the value of its total assets” sits in holdings where no single issuer exceeds “5 per centum of the value of the total assets.” Read the boundary: the other 25% is not constrained by that test at all.

What is confirmed, what is convention

This page mixes regulatory and academic facts, published index rules, our own arithmetic, and a few things we could not trace.

What trips people up

• Treating “index fund” as a synonym for “safe.” It removes single-company risk and leaves market risk completely intact. A broad fund can lose roughly half its value, as the arithmetic above shows it did into March 2009, and the reader who did not know that in advance is the reader who sells at the bottom.
• Assuming an index fund and an ETF are the same thing, or opposites. One is a strategy, the other a structure. Check both separately: what is it trying to do, and how do shares of it trade.
• Counting a thematic or single-sector index fund as diversification. Published rules and no stock-picking is all “index fund” promises. Thirty companies in one industry is a concentrated bet however transparently the list was assembled.
• Comparing two funds on last year's return instead of the expense ratio. Two funds tracking the same index differ mostly by cost, and cost is the number you can know before you buy. An index fund's return last year is mostly a fact about the index.
• Waiting to see the fee so you can judge it. It never appears. It comes out of fund assets, which is exactly why 0.85% feels like nothing and compounds into roughly double the fees paid.
• Reading SPIVA as a promise. It reports what happened to a category of funds over specific windows, after the fact. It does not say your fund will beat the average, and it says nothing about whether a broad index will go up.
• Getting the account wrong before getting the fund right. The same index fund is taxed completely differently in a Roth IRA, a workplace plan and a taxable brokerage account, where you also track cost basis. Stage 4 · Invest covers that order first, and it is the bigger decision.

The figures on this page are checked against the source that publishes them, and dated. Published rates move after the release named above — the linked source always carries the current number. This page explains a term; it does not recommend a product.