The expense ratio is the number the industry taught investors to compare, and the comparison worked. Fees on mainstream index products have been competed down to a few basis points, and in a handful of cases to nothing at all.
A fund charging nothing still has costs, which raises a question most investors never ask: what else is the portfolio earning? For a large share of index funds, part of the answer is that the shares themselves are out on loan.
A real business inside a passive product
The mechanics are old and unexciting. A fund holding a large static position lends stock to a borrower — typically someone shorting it — against collateral, and earns a fee. The position stays in the fund, the economics of ownership are unchanged, and the borrower posts more than the value of what they took.
At scale this generates meaningful revenue, and it is why several products can charge nothing and still operate. Lending income offsets the cost of running the fund, and in some cases exceeds it.
What varies, and varies more than most investors would expect, is who keeps the proceeds. Some managers return substantially all of it to the fund. Others retain a share as the agent arranging the loan, and the split is disclosed in documents that are technically public and practically unread. Two funds tracking the same index at the same headline fee can deliver different returns for this reason alone.
The risk is genuine but narrow, and it is worth being precise rather than alarming. The borrower posting collateral is not the exposure; the exposure is what the collateral is reinvested into, and the historical episodes where lending programmes hurt investors were reinvestment failures rather than borrower defaults. Conservative programmes hold cash collateral in government instruments and accept a lower return for it, and the difference between a conservative and an aggressive programme does not appear in any figure an investor routinely sees.
Scale has made the question larger. As passive ownership has concentrated, the same managers holding an increasing share of the market are also the largest lenders in it, which means a practice that was once marginal now runs through a substantial portion of tradeable float.
The thematic end of the market is where the practice earns most and is examined least. A broad market fund lends a large book at very low rates because everyone holds the same liquid names; a narrow sector or small-cap product holds stock that is genuinely hard to borrow, and the fees on it are a multiple of that. Some of the products that survived the great winnowing of thematic launches are viable partly for this reason, which is a fact about the fund rather than about the theme it is sold on.
There is a governance edge as well. Shares on loan generally carry their votes with them to the borrower, so a lending programme run through a proxy season quietly transfers voting power away from the long-term holder — a fact that sits awkwardly beside the stewardship language in the same manager's marketing.
None of this argues against index funds, and it would be a poor reading to take it that way. Lending revenue is one reason the fees came down at all, and a well-run programme is close to free money for the holder. The point is narrower: a cost comparison that stops at the expense ratio is comparing one term of a longer arrangement, and the maturing retail investor base that learned to compare fees has not generally been told there was anything else to look at.
The disclosure is improving slowly, and the funds with the most conservative programmes are the ones talking about it, which is the usual pattern. For an investor the practical step is unglamorous: read the annual report for the lending income line and the split, and treat two identical-looking fee figures as a starting point rather than an answer.
