How Market Indexes Are Actually Calculated
Investors quote index levels every day without knowing that different indexes use fundamentally different arithmetic, which means "the market" can look up on one index and down on another during the identical session. The mechanics are not a footnote, they determine what an index fund actually owns.
The core principle: weighting is a design choice
An index is a formula for summarizing the price behavior of a basket of securities into a single number, and every part of that formula, which securities are included, how each is weighted, how often the basket is rebalanced, is a deliberate design decision made by the index provider. There is no single "correct" way to measure a market's performance; there are several defensible methods that answer subtly different questions and can diverge meaningfully over any given period, sometimes by several percentage points in a single year even when tracking the exact same underlying group of companies.
The choice of weighting method is the single most consequential decision in index design, because it determines which companies drive the index's movement. Understanding this is not academic: it explains why the fund you hold might behave very differently from a headline index with a similar-sounding name.
Beyond weighting, index providers also make consequential decisions about eligibility criteria, how large a company must be to qualify, whether it must be profitable, whether foreign-domiciled companies are included, and about rebalancing frequency, how often the basket of holdings is reviewed and adjusted to reflect changes in the underlying market. Two indexes covering nominally the same market segment, large-capitalization domestic stocks, for instance, can hold meaningfully different sets of companies and produce meaningfully different returns over any given period purely from differences in these secondary rules, even when both use the same market-cap weighting method.
The three main weighting methods
A price-weighted index averages the raw per-share prices of its components, so a company whose stock trades at a high dollar price influences the index far more than a company whose stock trades at a low dollar price, entirely independent of which company is actually larger by total value. This method is a historical artifact from an era before computers made more sophisticated calculations easy, and it survives today mainly in a few well-known legacy indexes.
Market-capitalization-weighted indexes weight each company by its total market value, share price multiplied by shares outstanding, so a company's influence on the index scales with its actual economic size. This is by far the dominant method used in modern broad-market and sector indexes, because it matches how a buy-and-hold investor who simply bought a proportional slice of the entire market would naturally be weighted.
Equal-weighted indexes give every constituent the same weight regardless of size, which mechanically tilts the index toward smaller companies relative to a cap-weighted version of the same universe, and requires periodic rebalancing to maintain equal weights as prices drift apart between rebalancing dates.
A further variant, float-adjusted market-cap weighting, refines the basic cap-weighted approach by excluding shares that are not actually available for public trading, such as shares held by founders, governments, or other companies in a controlling stake. This matters because a company's total market capitalization can meaningfully overstate the value actually accessible to ordinary investors if a large portion of its shares are locked up and never trade; float adjustment corrects for this so the index weight better reflects the shares an investor could realistically buy.
The math: two indexes, two different winners
Worked example one: price-weighted versus cap-weighted. Consider an index of two companies. Company A trades at 200 dollars per share with 10 million shares outstanding, a 2 billion dollar market cap. Company B trades at 20 dollars per share with 500 million shares outstanding, a 10 billion dollar market cap, five times larger than Company A despite its much lower share price. In a simple price-weighted average, the two-stock index level is (200 + 20) / 2 = 110. If Company B rises 10 percent to 22 dollars and Company A is unchanged, the new index level is (200 + 22) / 2 = 111, a rise of only (111 − 110) / 110 = 0.9 percent, even though the actually larger company just added roughly 1 billion dollars of value. A cap-weighted version of the same index, by contrast, would rise by close to the full 10 percent weighted by Company B's dominant share of total market value, correctly reflecting that most of the market's actual dollar value just increased.
Worked example two: concentration in a cap-weighted index. Suppose a broad market index has a total value of 40 trillion dollars, and its ten largest companies together are worth 12 trillion dollars. Those ten companies represent 12,000,000,000,000 / 40,000,000,000,000 = 30 percent of the entire index's weight, even though they might be only ten names out of several thousand held. If those ten companies as a group fall 15 percent while the remaining thousands of companies are flat, the index as a whole falls by roughly 0.30 x 15 percent = 4.5 percent, illustrating how a cap-weighted "broad market" fund's return can be driven overwhelmingly by a small handful of mega-cap names.
Worked example three: equal weighting's opposite tilt. Take a simplified five-stock index where one company is worth 800 billion dollars and the other four are worth 50 billion dollars each, a total of 1,000 billion dollars. In a cap-weighted version, the largest company alone represents 800 / 1,000 = 80 percent of the index, while the four smaller companies share the remaining 20 percent. In an equal-weighted version of the identical five stocks, each company instead represents exactly 1 / 5 = 20 percent, meaning the largest company's influence drops from 80 percent to 20 percent purely from the weighting choice, while each smaller company's influence rises from roughly 1.7 percent to a full 20 percent. The underlying five companies have not changed at all; only the lens through which their combined performance is measured has changed, and that lens alone can produce dramatically different index returns depending on which of the five companies performs best or worst in a given period, a gap that widens further as real-world indexes span not five companies but hundreds or thousands.
What the historical record shows
Cap-weighted indexes have historically shown a persistent tendency toward rising concentration during strong bull markets, since the companies that perform best mechanically grow their weight in the index, which then increases the index's sensitivity to those same companies' subsequent performance, a self-reinforcing pattern visible across multiple market cycles. Equal-weighted versions of the same universe have, over some multi-decade stretches, modestly outperformed their cap-weighted counterparts, largely attributable to a structural tilt toward smaller companies and a disciplined form of forced rebalancing that trims winners and adds to laggards, though this has come with higher turnover, higher trading costs, and periods of underperformance during exactly the concentrated bull markets when a handful of mega-caps drove most of the return. Neither method has proven reliably superior in every environment, which is itself the more useful lesson than declaring a permanent winner.
Bond index history adds its own instructive wrinkle. Because most bond indexes weight by amount of debt outstanding, sovereign and corporate debt indexes have, across documented periods, shown rising weight toward the largest issuers of new debt precisely during periods when those issuers were adding leverage most aggressively, a structural quirk with no clean analogy in equity indexes, where growing larger by market value typically reflects growing profits and business success rather than growing indebtedness. Bond index investors have generally still benefited from this construction over the long run, since diversification and liquidity in the largest issuers carry real value, but it is a different justification for the weighting method than the one that applies to equities.
How this affects the fund in your portfolio
When you buy a fund tracking a broad market index, you are not buying diversified exposure to thousands of roughly equal companies; in a cap-weighted fund you are buying a portfolio whose fate is disproportionately tied to whichever handful of companies currently sit at the top. That is not necessarily a flaw, it mirrors how capital is actually allocated across a real economy, but investors seeking to deliberately reduce concentration in mega-cap names sometimes add a small allocation to an equal-weighted or small-cap fund alongside a core cap-weighted holding, accepting the higher turnover and cost in exchange for a different risk profile. For bond exposure, checking a fund's issuer concentration matters even more directly, since weighting by debt outstanding can quietly concentrate a fund in the largest borrowers.
The practical takeaway for most investors is that reading a fund's actual prospectus and holdings list, a five-minute exercise, tells you more about what you truly own than the index's name ever will. A fund named after a broad market category can still be dominated by a handful of familiar mega-cap names, and a fund named after a narrower category can sometimes turn out more diversified than expected once float adjustment, eligibility rules, and rebalancing frequency are all accounted for. None of this is a criticism of index investing itself, which remains a sound, low-cost core strategy for the overwhelming majority of investors; it is simply a reminder that the word "index" describes a category of methods, not a single, uniform thing.
Actionable breakdown
- Before buying any index fund, check:
- Whether it is price-weighted, cap-weighted, or equal-weighted.
- Its top ten holdings and their combined weight.
- For bond index funds, check:
- Issuer concentration by debt outstanding.
- Whether the largest weights are the most indebted issuers.
- Do not assume:
- Two funds with similar names hold similar things.
- An index automatically means broad diversification.
Common pitfalls
Investors often assume that owning an index fund automatically means broad, evenly spread diversification, when a cap-weighted index can be heavily concentrated in a small number of mega-cap companies. Another mistake is comparing performance across two differently constructed indexes and drawing conclusions about "the market" as though both measured the same thing, when the underlying methodology alone can account for a meaningful part of the return gap. A third is failing to notice that as a company grows, its weight in a cap-weighted index grows too, meaning your exposure to that single name increases automatically, compounding your concentration exactly as the company becomes a larger share of your total portfolio, without any decision on your part. A fourth is assuming a bond index fund is automatically diversified across issuers, when debt-weighted construction can concentrate it in the largest borrowers. A fifth is ignoring float adjustment entirely, assuming a company's full market capitalization, rather than its publicly tradable float, is what determines its index weight.
The bottom line
Index construction, price-weighted, cap-weighted, or equal-weighted, changes what an index actually measures, so check the methodology and the top holdings before assuming any two indexes, or the funds tracking them, are interchangeable, and treat a fund's actual holdings list as more informative than its name.
Related reading: funds and ETFs, passive index investing, the single-index model, diversification and portfolio risk.