Securities Lending Inside Funds: The Revenue You Did Not Know About

This is not a scandal and not a secret: it is disclosed in fund documentation, permitted and constrained by regulation, and practised by a large share of index funds and ETFs across markets. But it is genuinely invisible unless you look, it explains something the tracking article flagged — how a fund can lag its index by less than its own charges, or occasionally beat it — and it introduces a set of risks and a conflict of interest that a reader is entitled to understand. This article covers the mechanics, the revenue split, the risks and what mitigates them, and what to look for. It expresses no view on whether the practice is acceptable; that is a judgment, and readers can make it once they know how it works.

The mechanics: who borrows, why, and what the fund gets

A fund holding a large portfolio of securities owns something other market participants want to borrow. Borrowers are typically institutions needing the security temporarily — most commonly to support short selling, but also for hedging, market making, and settlement purposes where a delivery obligation must be met. The transaction is a loan of securities against collateral: the borrower delivers cash or acceptable securities worth more than the loaned stock, the fund lends the shares, and the borrower pays a fee for the period of the loan. The fund remains economically exposed to the loaned securities — it retains the price movement and is compensated for dividends through the lending arrangement — so lending does not change the fund's market exposure, which is the reason it can be done at all inside an index fund. Fees vary enormously by security: lending a widely held large-cap share earns very little because supply is abundant, while a security in heavy demand for short selling and short supply can command a substantial fee. That variation means lending revenue is concentrated in particular holdings rather than spread evenly, and it is why the revenue matters more in some fund types — small-cap, emerging-market, and specialist funds — than in broad large-cap ones. An agent usually runs the programme, either the fund manager itself or a specialist lending agent, and the revenue is split between the fund and whoever operates the programme. That split is the commercially interesting part: practice varies widely, with some providers passing the great majority to the fund and others retaining a much larger share, and the arrangement is disclosed in fund documentation rather than advertised. The fund's share reduces its net cost. A fund with a 0.20% charge earning 0.05% in net lending revenue costs its holders 0.15% in effect — which is why lending revenue shows up as tracking difference smaller than the headline charge, and why comparing two funds on tracking difference alone can mislead, exactly as that article warned. Regulatory frameworks constrain the practice: UCITS rules and equivalent regimes elsewhere impose collateral requirements, limits on how much may be lent, disclosure obligations covering revenue and its split, and requirements that the arrangement be in the fund's interest.

The risks, the mitigations, and the conflict

Three risks, stated plainly. Counterparty risk is the main one: if the borrower fails and does not return the securities, the fund keeps the collateral — and whether that leaves the fund whole depends on the collateral's value at that moment. Collateral risk follows from it: collateral that has fallen in value, or that is illiquid, or that is correlated with whatever caused the borrower to fail, is worth less than its posted figure suggested. The concern is the same one the replication article raised about swap collateral, appearing here in a different form — and it is why collateral eligibility rules and haircuts exist. And reinvestment risk applies where cash collateral is reinvested: the fund earns a return on it, and that reinvestment carries its own credit and liquidity exposure, which was a documented source of loss in the 2008 period for some lending programmes. Now the mitigations, which are real. Over-collateralisation is standard — collateral exceeding the loan value, marked daily and topped up as required. Collateral quality rules restrict what may be accepted. Limits on lending volume cap how much of the portfolio may be out on loan at once. Recall rights allow the fund to demand return of securities, which matters both for redemptions and for exercising votes. And indemnification is offered in some programmes, where the agent or manager guarantees to make the fund whole against borrower default — a meaningful protection whose value depends on the indemnifier's own strength, and whose presence or absence is worth identifying. Then the conflict, which deserves naming directly. The party deciding how much to lend and on what terms is frequently the same party keeping a share of the revenue. That is a structural conflict of interest, and it is managed rather than eliminated: by regulation requiring the arrangement to serve the fund's interest, by disclosure of the split, and by competitive pressure as cost-conscious buyers compare terms. Two further considerations a reader may care about. Voting: loaned securities are generally not voteable by the fund, so a fund lending heavily around a contested vote faces a choice between revenue and exercising the franchise — a real tension for anyone who values stewardship, and one that lending policies address with varying seriousness. And the short-selling connection: lending facilitates short selling, which some investors object to on principle and others regard as a legitimate and useful market function. This portal takes no side; the point is that a holder of a lending fund is, indirectly, part of that machinery, and may reasonably want to know.