Free Float

Index providers start with total shares outstanding and strip out holdings judged unavailable to ordinary investors: government and sovereign fund stakes, founder and family blocks, cross-holdings by other listed corporates, strategic partners, employee share schemes subject to vesting, and shares locked up after an IPO. Treasury stock is removed too, but it is only one item on a much longer list. What remains is the free float, and dividing it by total shares outstanding gives the float factor.

Float-adjusted market capitalisation is then price multiplied by shares outstanding multiplied by that float factor. A constituent's index weight is its float-adjusted capitalisation divided by the sum of float-adjusted capitalisations across all constituents. Providers typically round the factor into bands and review it on a fixed schedule so that small ownership changes do not trigger constant, costly rebalancing for the funds tracking the index.

Float is not static, and changes in it move passive money. When an IPO lock-up expires, when a state sells down a privatised holding, or when a founder places a block with institutions, the float factor rises, the constituent's index weight rises, and index-tracking funds must buy even though the company's fundamentals are unchanged. Secondary offerings and share buybacks work in the same mechanical way, in opposite directions.

In emerging and frontier markets the adjustment often extends beyond ownership. Where statutory or sectoral caps limit aggregate foreign holdings, providers apply an additional inclusion factor reflecting what a foreign investor can actually own, which is why some very large domestic companies carry modest global index weights. Phased inclusion of a market or share class, applied through a partial inclusion factor, is the same tool used deliberately to stage capital flows.

A small float concentrates trading in a thin pool of shares. Bid-ask spreads widen, market impact per unit of order size rises, and realised volatility tends to be higher than the company's business risk alone would justify. Borrow for short selling is scarce and expensive, which weakens the arbitrage that normally corrects overpricing, so low-float names can sustain valuations that informed traders cannot profitably fight. Squeezes and abrupt repricings around index events are the visible symptom.

Float therefore connects directly to market efficiency arguments at Level I. Efficient pricing depends on active investors being able to trade cheaply on information; where float is small, coverage thin and borrowing constrained, the practical conditions for that mechanism are weaker, and observed anomalies are correspondingly more persistent.

Expect to compute a float-adjusted weight from a table giving price, shares outstanding and a stated float percentage, and to recompute weights after a change in float or price. Be able to place float-adjusted capitalisation weighting alongside price weighting, equal weighting and fundamental weighting, and to state its consequences: it is still capitalisation weighting, so it inherits the momentum tilt of holding more of what has risen and less of what has fallen, and it remains concentrated in the largest names.

Its distinguishing merit is investability. Because weights reflect only tradable shares, the index can be replicated at scale without bidding for stock that will never come to market, which keeps tracking error and rebalancing costs manageable. Candidates should also be ready to explain why two indexes covering the same market can differ purely through divergent float and foreign-ownership rules.