How Indices Are Built: Market-Cap Weighting, MSCI World, FTSE All-World, S&P 500 and Concentration

What you will learn

• The building blocks of an index rulebook: universe, selection, weighting and review.
• How to calculate free-float market-cap weights step by step.
• The key differences between MSCI World, FTSE All-World and S&P 500.
• How to measure concentration with the "effective number of stocks".

Explanation

What an index is and how exchanges publish them was introduced in the course "Stocks & Fundamental Analysis". Here we look at the rulebook, because an ETF can only be as good as the index it follows. Every index provider publishes a methodology with four parts:

1. Universe: which countries, exchanges and security types are eligible.
2. Selection: size, liquidity and sometimes profitability filters.
3. Weighting: how much each company counts.
4. Review: how often the list and weights are updated. Changes at reviews create turnover inside the fund.

Most broad indices use free-float market-cap weighting. Market capitalisation (market cap) = share price × number of shares. The free-float factor is the share of stock that can actually be traded, excluding for example large founder or state holdings. The weight of company i is:

weight_i = free-float market cap_i ÷ sum of all free-float market caps

Advantages: low turnover (weights move with prices automatically) and the index mirrors the investable market. Disadvantage: companies that have risen strongly get larger weights, so the index can become concentrated. Alternatives are equal weighting (every stock the same weight, regular rebalancing needed), price weighting (used by the Dow Jones Industrial Average) and factor indices.

The three most-used equity indices for ETF investors (figures change; check the current factsheet):

A "world" index is not geographically equal. According to MSCI's factsheet of 30 September 2026, the United States made up about 73% of the MSCI World and the ten largest companies about 28%. Small and micro caps are not included in any of the three; they are covered by separate "small cap" or "IMI" (investable market index) versions.

To measure concentration, use the effective number of stocks: N_eff = 1 ÷ (sum of squared weights). An index of 4 equal stocks has N_eff = 4. If one stock dominates, N_eff falls towards 1, even though the index still "holds" many names.

Worked example

Hypothetical example: a mini index with four companies. Prices in euros, shares in billions. Costs are not relevant here.

1. Total free-float cap = €200 + €50 + €48 + €30 = €328bn.
2. Weights: A = 200 ÷ 328 ≈ 60.98%; B = 50 ÷ 328 ≈ 15.24%; C = 48 ÷ 328 ≈ 14.63%; D = 30 ÷ 328 ≈ 9.15%.
3. N_eff = 1 ÷ (0.6098² + 0.1524² + 0.1463² + 0.0915²) ≈ 1 ÷ 0.4248 ≈ 2.35. Four names, but the risk behaves like about 2.4 equal stocks.
4. Suppose over one year A returns +10%, B −5%, C +2%, D +20%. Cap-weighted return ≈ 0.6098 × 10% + 0.1524 × (−5%) + 0.1463 × 2% + 0.0915 × 20% ≈ +7.46%. Equal-weighted return = (10 − 5 + 2 + 20) ÷ 4 = +6.75%.

The single large company A drives most of the result. That is concentration in practice.

Common mistake

Practice activity

On paper: an index has three companies. X has a market cap of €300bn and 100% free float; Y has €150bn and 60% free float; Z has €60bn and 100% free float.

1. Calculate each free-float weight.
2. Calculate the effective number of stocks.

Key takeaways

• An index is a rulebook: universe, selection, weighting and review.
• Free-float market-cap weight = a company's free-float cap ÷ the total; large winners automatically gain weight.
• MSCI World covers developed markets only, FTSE All-World adds emerging markets, S&P 500 is US only and committee-selected.
• Use the factsheet and N_eff = 1 ÷ Σ(weight²) to see real concentration.

Next step

In the next lesson you will see how an ETF actually tracks its index, physically or synthetically, and what happens to dividends.