Somebody Chooses the Index

Buying an index fund feels like stepping out of an argument.

You are not going to guess which company wins. You are not going to read filings on a Sunday. You are going to own the market, pay very little for the privilege, and go do something else with your weekend. That decision is a good one, and most of the evidence is on its side.

It just is not quite the decision people think they are making. Buying an index fund is not a decision to stop picking. It is a decision to let somebody else pick, and to stop being told who.

The S&P 500 Is Not the Five Hundred Biggest

Start with the sentence almost everyone would get wrong.

The S&P 500 is not a ranking. It is not the five hundred largest American companies, sorted by size, refreshed automatically when the order changes. If it were, it would need no staff.

It has staff. Constituent selection sits with the S&P Dow Jones Indices Index Committee — a small group that meets, deliberates, and decides. There are published criteria, and the criteria are real. But the criteria produce a list of companies that are allowed to be considered. The committee decides which of them actually go in, and when.

Those are two different things, and the gap between them is where all the interesting behavior lives.

What the Rules Actually Do

The published requirements for the S&P 500 cover the things you would expect a serious index to care about:

Size. A minimum market capitalization, reviewed and raised over time as the market grows. The threshold today is in the tens of billions of dollars, and a company that qualified five years ago might not qualify now on the same numbers.

Profitability. Positive earnings in the most recent quarter, and positive earnings summed across the four most recent quarters. This one quietly excludes a whole category of famous companies — the fast-growing and unprofitable — for as long as they stay unprofitable.

Public float. Enough of the shares have to actually be available to trade. A company that is mostly held by its founders or a parent company is not investable in the way an index needs it to be.

Liquidity. Enough daily trading that a fund can buy a position without moving the price against itself.

Listing and seasoning. A U.S. exchange listing, and a waiting period after an IPO before the company is considered at all.

Read that list again and notice what it is doing. It is not measuring which companies are best, or cheapest, or most likely to grow. It is measuring which companies a very large fund can hold without difficulty. Those overlap, but they are not the same test, and the index was built for the second one.

Passing Is Not Entering

Here is the part that surprises people.

Meeting every published criterion does not get a company into the index. It gets the company onto the list of companies that could go in. Selection remains at the committee's discretion.

The most visible reason is balance. The committee looks at how much of the index each sector represents, compared with the broad U.S. market. If one sector is already heavier in the index than it is in the market at large, a perfectly qualified company from that sector can be passed over in favor of a candidate from a sector that is underweight.

So a company can be large enough, profitable enough, liquid enough, listed in the right place, seasoned long enough — and still be waiting, because what the index needed that month was an industrial rather than another software company.

There is also the plain fact that a seat has to open. The index has a fixed shape, so joining generally means somebody else stopped qualifying.

And the way companies leave is worth a moment of its own, because it is not what people picture. The dramatic version — a famous company collapses and is thrown out — is the rarest version. Far more often a company leaves for reasons that have nothing to do with failure: it gets acquired, it merges into something else, it splits itself in two, or it simply drifts below the size and liquidity the index requires while the bar keeps rising underneath it.

That last one is quiet and slightly unfair-sounding. A company can be doing roughly what it always did, earning money, employing people, and still fall out — not because it shrank, but because the market around it grew and the threshold moved. Nothing was decided about its quality.

This has a consequence for how the index reads across decades. An index does not record what happened to the companies that were in it. It records what happened to the companies that stayed in it, with the others quietly handed off along the way. That is not a flaw — a list that never removed anything would be measuring something else entirely — but it does mean the long line on the chart is the performance of a maintained roster, not of a fixed set of businesses.

This Happens on Ordinary Wednesdays

None of this is theoretical or rare. It runs on a schedule most people never see.

S&P announces changes publicly, after the close, on plain days. In August 2026 it announced that Reddit would join the S&P 500. In early September it announced that Bloom Energy, Illumina, and Everpure would join, with other companies moving into the S&P 100, MidCap 400, and SmallCap 600.

Those announcements are short, procedural documents. They are also, for a few days, among the most consequential pieces of paper in the market — because of what has to happen next.

Why the Announcement Moves the Price

An index fund does not have the option to think about it.

If a fund promises to track the S&P 500, and the S&P 500 is about to contain a company it does not own, the fund has to buy that company. Not because anyone at the fund formed a view, but because tracking is the product. The same logic runs in reverse for the company leaving.

So an announcement creates something unusual: a large, price-insensitive buyer, arriving on a date everyone already knows. Every professional in the market can see it coming. The stock frequently moves on the announcement rather than on the day of the actual change, because the anticipation is where the trading is.

This is worth understanding precisely because it undercuts a tidy story. The index is often described as a neutral mirror held up to the market. But the mirror has a committee, the committee's decisions force trillions of dollars to move, and the movement happens whether or not anything changed about the underlying business that week. Reddit did not become a different company on the day it was announced.

"The Market Was Up" Depends Which Market

Once you see that indices are built rather than found, a second thing comes loose.

There is no single object called the American stock market. There are several constructed representations of it, and they disagree — not about the facts, but about what to include and how much each piece should count.

One index holds about five hundred companies. Another aims at the whole investable market and holds several thousand. The Dow Jones Industrial Average holds thirty, chosen by a committee as well, and weights them by share price rather than by company size — an old convention that means a company with a high share price moves the average more than a larger company with a lower one.

So "the market was up today" is a sentence about a particular construction. Most of the time the constructions agree enough that the distinction does not matter. The times they disagree are exactly the times someone is trying to tell you something with the number they picked.

And the habit does not stop at the stock market. The Consumer Price Index is an index in exactly the same sense — a basket somebody assembled, with weights somebody assigned, maintained by an agency that publishes its methodology and revises it. So when a monthly inflation print lands cooler than expected, the thing that came in cooler is that basket. Whether it also cooled in your kitchen depends on how much your spending looks like the basket's.

That is worth holding on to, because the names we give to economic conditions are built on those constructed numbers rather than underneath them. When several of them start pointing the same direction at once, the label that gets reached for — soft landing, stagflation — is a description applied to a pattern across several maintained indices. It is not a reading taken off a single instrument, and no such instrument exists.

The Edges of This

This is not an argument against index funds. Broad, low-cost index funds remain one of the few genuinely good deals available to an ordinary saver, and nothing here changes the arithmetic that makes them work — most of which is about the cost you pay to hold them. The point is to know what you bought, not to stop buying it.

The committee is not doing anything improper. Discretion is disclosed, the criteria are published, and the changes are announced in advance to everyone at once. A committee is arguably better than a rigid formula, because a formula cannot notice when it is about to do something stupid.

This is not a claim that you can profit from it. The pattern around index additions is well known to every professional in the market, which is usually what stops something from being an opportunity. Knowing why a price moved is different from being early to it.

None of this is investment advice, and nothing here says anything about whether any particular company belongs in any particular index.

A Note From the Far Side of an Ocean

I write about American markets from Korea, and here the distance makes something plainer than it might be from inside.

When people here say they are buying "the American market," they almost always mean one specific index, through one specific kind of fund. The phrase does not feel like a choice. It feels like the default, the way saying "the sea" does not feel like choosing a particular sea.

But a committee in New York decides what sits inside that phrase. Their decisions about sector balance, about whether a company has been listed long enough, about which of several qualified candidates gets the open seat — those decisions land in an account in Seoul held by somebody who has never heard the committee's name and never will.

That is not a complaint. It is closer to the opposite: it is a reminder that choosing not to pick stocks is itself a choice about who picks for you, and that the choice is a reasonable one precisely because it is made by people whose job is to be boring and consistent about it.

The Shape of It

The useful version of this is small.

An index is a product. Somebody designed it, somebody maintains it, and the design has opinions in it — about size, about profitability, about how much of one industry is too much. Those opinions are published, which is more than can be said for most things that decide what happens to your money.

You do not need to audit them. But it changes how you read the number. The level of an index is not a measurement of the economy the way a thermometer measures a room. It is the current value of a specific, maintained, human-edited list of companies, and it will tell you exactly what it contains if you ask.

That puts it in the same family as the other things worth knowing before you need them — what a fund's structure actually does to your returns, and which assets have historically outrun rising prices rather than merely kept up. None of them require a forecast. All of them are written down somewhere, by someone who was required to write them down.

Somebody chooses the index. It is worth knowing that, not because you would choose differently, but because "I don't pick stocks" and "someone else picks them for me" are two descriptions of the same afternoon, and only one of them is complete.

Data as of September 12, 2026. Sources: S&P Dow Jones Indices, S&P U.S. Indices Methodology and index announcements of August 13 and September 4, 2026.

Related Posts:

Visuals on this post are AI-generated. The author works with AI as a research and drafting assistant; topics, judgments, and final edits are the author's own. This post is observation, not investment advice. See full Disclaimer for details.

Search This Blog

IJin Insight