Who Decides What's in Your Index Fund? Inside the Quiet Power of Index Committees

Jun 16
6 min read

From sector classifications to the SpaceX IPO — how a handful of companies define what trillions of dollars must buy

KEY TAKEAWAYS

Index providers such as MSCI, FTSE Russell, S&P Dow Jones Indices, and MarketVector are commercial businesses that license their indexes to ETF issuers for a fee.
ETF issuers like Vanguard, VanEck, and Invesco do not choose individual stocks; they build funds that replicate an index someone else designed.
Different index methodologies — sector classification, growth/value scoring, exchange listing, or revenue thresholds — produce meaningfully different holdings even when funds sound similar.
SpaceX (SPCX) illustrates the difference: QQQ holders gained ~0.5% exposure within 15 days of the June 2026 IPO; VONG holders must wait until December 2026; VOO holders not before mid-2027; VGT and SMH holders never, due to sector and revenue rules.
If two funds track the same index, their pre-fee returns are identical — the only meaningful difference is the expense ratio. VONG (0.07%) and IWF (0.19%) hold the same stocks; the investor is choosing how much to pay, not which market to bet on.
The practical framework: first decide which index rulebook you want, then buy the cheapest fund that tracks it.

When an investor buys VONG, VOOG, VGT, or SMH, the natural assumption is that Vanguard or VanEck has chosen those holdings. In reality, none of these companies pick individual stocks at all. Each fund exists to replicate an index built by a separate company — and the rules that index company writes determine, almost mechanically, what ends up in the fund. Understanding who writes those rules, why, and how those rules occasionally collide — as they did this year with the SpaceX listing — reveals a layer of the market most investors never think about.

The Business Behind the Benchmark

Index providers make money by licensing their indexes to asset managers, typically for a fee scaled to the assets tracking that index. MSCI Inc. is itself a publicly traded company, spun off from Morgan Stanley in the late 2000s. FTSE Russell, which builds the Russell 1000 Growth and Value indexes underlying VONG and VONV, is a subsidiary of the London Stock Exchange Group. S&P Dow Jones Indices, which builds the S&P 500 and its growth and value variants underlying VOO and VOOG, is majority owned by the publicly traded S&P Global. Nasdaq Inc., publicly traded, builds the Nasdaq-100 underlying QQQ. The one notable exception is MarketVector, which builds the index underlying SMH and is a subsidiary of VanEck, a privately held asset manager.

In every case, the asset manager that runs the ETF — Vanguard, VanEck, Invesco — is simply the plumbing that turns an index provider's rulebook into a tradeable product.

Four Different Ways to Define “Qualifies”

Each index provider defines its universe differently, and these differences explain why funds that sound similar can hold very different things. VGT qualifies companies by sector classification: any company that the Global Industry Classification Standard (GICS) places in “Information Technology” — covering software, hardware, IT services, and semiconductors — is eligible, with no judgment about growth or value characteristics. VONG and VOOG qualify companies by style scoring: every company in the Russell 1000 or S&P 500 is ranked along growth and value factors and assigned a proportional weight in each style index. QQQ qualifies companies by exchange listing: membership requires simply being among the 100 largest non-financial companies listed on the Nasdaq exchange. And SMH qualifies companies by revenue concentration: a company must derive at least half its revenue from semiconductors to be eligible for the underlying index.

Inside the Growth and Value Formula

FTSE Russell's growth/value methodology is worth examining closely because it is recalculated every reconstitution, not set once and forgotten. Each company in the Russell 1000 is ranked on three measures: forecast earnings growth over the next two years, historical sales-per-share growth over the past five years, and price-to-book ratio. The two growth measures combine for half of a company's “composite value score,” and price-to-book makes up the other half. Companies with low scores are classified as growth; companies with high scores are classified as value; roughly 30% of the index's total value falls in between and is split proportionally between both the Russell 1000 Growth and Russell 1000 Value indexes. The growth and value allocations for any company always sum to 100% of its market value — together, the two style indexes reconstruct the entire parent index.

From Benchmark to Blueprint

Growth and value indexes were not originally built as investment products. They were created decades ago so that institutional money managers who specialized in a particular style could be measured against a benchmark matching that style — a growth manager judged against a growth benchmark, a value manager against a value benchmark. The ETF wrapper came later, repurposing tools built for performance comparison into products investors could buy directly.

That shift matters because of scale. As trillions of dollars now flow passively through these indexes, the act of being added to or removed from a major index can move a stock's price by itself — passive funds are mechanically required to buy or sell at reconstitution, independent of any change in the underlying business. Index reconstitution dates have become some of the highest-volume trading days of the year for exactly this reason.

The SpaceX Test Case

SpaceX's stock market debut in June 2026 produced a real-time illustration of how differently these committees can act on the same company. FTSE Russell added SpaceX to the Russell 1000 and Russell Top 200 under fast-entry rules introduced earlier in the year, effective June 26. MSCI added SpaceX to its standard and large-cap indexes effective June 29. Nasdaq adjusted its own rules to allow SpaceX into the Nasdaq-100 after just 15 days of trading. S&P Dow Jones Indices, by contrast, declined to fast-track SpaceX into the S&P 500, citing its long-standing requirement of four consecutive quarters of profitability — a bar SpaceX has not yet cleared.

The scale of the resulting mechanical flows was substantial: industry estimates suggested S&P 500 funds alone would need to absorb close to a fifth of SpaceX's public float upon any future inclusion, with Russell 1000 and Nasdaq-100 funds together absorbing roughly another quarter. Because only a small fraction of SpaceX's shares are freely tradable — most founder shares remain locked up for a year — passive funds were on track to hold a substantial share of that free float within just a few weeks of listing, almost entirely as a function of these classification decisions rather than ordinary investor demand.

Do You Own SpaceX? It Depends Which Fund You Hold

The SpaceX IPO makes the rulebook differences immediately concrete for investors. SpaceX listed on June 12, 2026, at $135 per share, valuing the company at roughly $1.75 trillion — the largest IPO in history. Whether an investor owns it, and how much, depends entirely on which fund they hold:

QQQ (Nasdaq-100, Invesco): Roughly 0.47%–0.70% exposure, added within 15 trading days of IPO — late June or early July 2026. Nasdaq applies a modified float weight of approximately 3x the available float for qualifying large IPOs.
VTI (Total Market, Vanguard / CRSP): Roughly 0.07%–0.11% exposure, added within approximately 5 trading days. Small weighting reflects SpaceX’s low float — only about 3–5% of total shares are publicly tradeable.
VONG / IWF (Russell 1000 Growth): No SpaceX until the December 2026 style reconstitution at the earliest.
VOO / SPY / IVV (S&P 500): No SpaceX before mid-2027 at the earliest. S&P declined to fast-track, maintaining its requirement for four consecutive quarters of GAAP profitability.
VGT / SMH: No SpaceX exposure regardless of timing. SpaceX is classified as aerospace and defense under GICS — not Information Technology — and derives no meaningful revenue from semiconductors, so it qualifies for neither fund.

Two investors who both describe themselves as holding “a tech index fund” could end up a year or more apart in when they first own SpaceX — or never own it at all — purely as a function of which rulebook their fund follows.

Same Index, Same Returns — Compare Expense Ratios

One more implication follows directly from how this system works: if two funds track the same index, their returns before fees are functionally identical. The index is the product — the ETF issuer is just the delivery mechanism. SPY, IVV, and VOO all track the S&P 500. Their holdings are the same, their reconstitution dates are the same, and their pre-fee performance is the same. The only meaningful difference is cost: SPY charges 0.09% annually, while IVV and VOO each charge 0.03%. On a $100,000 investment held for 30 years, that 0.06% gap compounds into a material difference in final wealth.

The same logic applies across Russell 1000 Growth funds: VONG (Vanguard, 0.07%) and IWF (iShares, 0.19%) track the same FTSE Russell index and hold the same stocks. An investor choosing between them is not making a different bet on the market — they are simply choosing how much to pay for the same exposure. The one exception is when an index is proprietary to a single issuer: SMH’s underlying MarketVector Semiconductor index is only licensed to VanEck, so there is no competing cheaper alternative. But for every major index where multiple issuers compete, the practical guidance is straightforward: decide which index (which rulebook) you want, then buy the cheapest fund that tracks it.