Fifteen numbers stand in for most of the world's public equity. Where they came from, how they are built, what they actually measure — and where each one quietly misleads.
An index is a summary, and every summary is an argument about what matters. The Dow argues that thirty large American companies tell you enough. The S&P 500 argues for five hundred, weighted by size. The Nikkei still argues, as it has since 1950, that a share priced at ¥40,000 should move the average more than one priced at ¥800 — regardless of how large either company is.
None of these is wrong exactly. They were built for different purposes at different times, and the design decisions embedded in each one determine what it can and cannot tell you. A reader who knows that the Dow is price-weighted understands immediately why a single high-priced constituent can swing it, and why comparing it to the S&P 500 on a given day is comparing two different instruments.
This guide covers the fifteen indexes that account for most of the world's traded equity value. Each has its own detailed guide linked below.
A stock index is a single number derived from the prices of a defined set of securities, recalculated continuously through the trading day. Three decisions define it: which securities are in it, how much each one counts, and what happens when the set changes.
Selection. Some indexes are rule-based — the largest companies by market value on a given exchange, refreshed quarterly. Others are committee-selected, which is how the S&P 500 works and why it is not simply the 500 biggest US companies. Committee discretion introduces judgment, which critics call opacity and defenders call the ability to avoid mechanical absurdities.
Weighting. This is where the real differences live, and it is the section most guides skip.
Market-capitalization weighting sizes each constituent by its total market value, usually adjusted for free float so that shares locked up by founders or governments do not inflate a company's influence. Most modern indexes work this way — the S&P 500, FTSE 100, DAX, Hang Seng, Nifty 50. It has an appealing property: the index behaves like the aggregate portfolio of everyone holding those shares.
Price weighting sizes each constituent by its share price alone. Company size is irrelevant. This is a nineteenth-century convention that survives in exactly two significant places — the Dow Jones Industrial Average and the Nikkei 225 — because both are old, both are famous, and changing the methodology now would break a century of continuity. It produces genuine oddities: a company can double in value and barely move the index, while a stock split mechanically reduces a constituent's influence without anything having changed.
Equal weighting gives every constituent the same share, rebalanced periodically. It removes concentration risk and introduces a small-company tilt. Most major national indexes are not equal-weighted, though equal-weighted versions of the S&P 500 exist and are widely used as a check on how much of a move came from the largest handful of names.
Two facts from that chart are worth pausing on. First, the United States appears twice at the top and its two exchanges together are larger than the next eight combined — as of April 2026, US-listed companies were worth more than the next nine national markets put together. Second, Nasdaq passing the NYSE is genuinely new. For most of the modern era the NYSE was the world's largest exchange by listed value, and the reversal is a direct consequence of where market value has concentrated over the past decade.
This is the single most important thing to understand about the modern S&P 500, and it applies in varying degrees to most cap-weighted indexes. A benchmark advertised as five hundred companies increasingly behaves like a concentrated position in a handful of them. It is not a flaw in the methodology — cap weighting is doing exactly what it is designed to do — but it does mean that "diversified across 500 companies" describes the constituent list rather than the risk.
Fifteen benchmarks, grouped by region. Each links to a full guide covering its history, methodology, constituents and quirks.
The default benchmark for global equity. Committee-selected rather than purely rule-based, and now more concentrated than at any point in its modern history.
The oldest continuously computed equity index still in use, and structurally the strangest. Thirty companies, weighted by share price, chosen by committee.
Everything listed on Nasdaq, which makes it broad in name and technology-dominated in practice. Distinct from the Nasdaq-100, which excludes financials.
Canada's benchmark, and an unusually direct read on financials, energy and materials — a sector mix that behaves very differently from the US market.
Named for the largest London-listed companies but earning most of its revenue abroad — which makes it a poor proxy for the British economy and a good one for sterling weakness.
Expanded from 30 to 40 constituents in 2021 after the Wirecard collapse prompted a governance overhaul. Unusually, the headline version is a total-return index.
Heavily weighted toward luxury goods, energy and pharmaceuticals — a concentration in global consumer brands unmatched by any other major national index.
A blue-chip index spanning the eurozone rather than any one country, created alongside the single currency and used as the region's derivatives benchmark.
Twenty companies, of which three pharmaceutical and food giants have historically carried an outsized share — one of the most concentrated developed-market benchmarks.
The other surviving price-weighted major index, and the one whose 1989 peak took more than three decades to reclaim — the defining cautionary tale in index investing.
The broader and methodologically sounder Japanese benchmark, preferred by institutions precisely because it is not price weighted.
Hong Kong's benchmark and the main listed gateway to mainland Chinese companies, expanding by design from 33 constituents at launch toward a target of 100.
Every stock on the Shanghai exchange, including state-controlled giants whose free float is small relative to their total size — a structural quirk with real consequences.
India's two benchmarks, tracking a market that has grown fast enough to put both of its exchanges into the global top ten.
Dominated by banks and mining, making it one of the most direct equity proxies for commodity demand anywhere in the developed world.
Three limits worth carrying into any use of them.
An index is not the economy. It measures listed equity, which excludes private companies, state enterprises, and in many countries the majority of employment. The FTSE 100 and the British economy have moved in opposite directions for extended periods without either being wrong.
Survivorship is built in. Constituents that fail are removed and replaced. Long-run index charts therefore describe the performance of a continuously refreshed set of successful companies, not of a fixed portfolio anyone could have held.
Headline levels hide dividends. Most quoted index levels are price returns, excluding dividends. Over long periods the gap is enormous — and it is why the DAX, quoted as a total-return index, is not directly comparable to the price-return figures quoted for most of its peers. Comparing them without adjusting is one of the most common errors in market commentary.
None of the national indexes here. A global benchmark such as the MSCI ACWI or FTSE All-World covers thousands of companies across developed and emerging markets. Because of the concentration shown above, however, even those are heavily weighted toward the same large US technology companies.
Familiarity and continuity. It has been published since 1896, it appears in headlines, and the general public recognises it. Professional investors overwhelmingly benchmark against the S&P 500 instead.
Not in the index itself, which is a calculation. You invest through funds that track it, and tracking is imperfect — fees, sampling, rebalancing timing and dividend treatment all introduce differences between the fund's return and the index's.
Index-tracking funds must buy it, which creates mechanical demand around the effective date. The effect is well documented and largely anticipated by the market, so the price move often happens on announcement rather than on inclusion.
Share classes, mostly. The S&P 500 holds 503 securities across 500 companies. Others simply keep a historic name after changing size — the DAX is still called the DAX with 40 members.