Direct answer
The borrower pays a lending fee, split between the fund - meaning the holders - and the lending agent, often a company in the manager's own group. Income from reinvested cash collateral is usually split too. The holders' share improves the fund's return against its index; the agent's share does not. For it the fund carries the risk that a borrower fails and the collateral covers less than the repurchase, plus whatever the reinvested collateral carries. Both halves sit in the fund's documents, not in the ongoing charge.
- The condition that matters most
- The split and the terms are fund-specific and change. There is no standard share, no standard collateral rule and no standard limit on how much of a portfolio may be out on loan at once, so the only figures that describe your fund are the ones in your fund's prospectus and annual report.
- What this does not tell you
- This explains a mechanism and names nothing. It does not tell you whether lending is worth the exposure, does not compare providers, states no typical split and no typical revenue, and prices no risk. It says nothing about the rules in any jurisdiction, which differ and change, and nothing about tax.
- Last reviewed
The number two trackers are compared on does not contain this
Two funds track the same index. One publishes an ongoing charge of 0.20% and the other 0.07%, and the comparison looks settled. It is not, and one of the reasons is that both funds may be earning money from an activity the ongoing charge is defined not to include, on terms that differ between them, in exchange for a risk that also differs between them.
The activity is securities lending. The fund lends out the shares or bonds it holds, to somebody who wants to borrow them, and is paid for doing it.
This article is about who receives that payment, and what the fund is exposed to in order to earn it. It is not about whether the arrangement is good or bad. It is about the fact that a holder comparing two funds on one published number is comparing them on a figure that contains neither half of it.
What the arrangement actually is
A fund holds securities. Somebody else — most often a market maker managing inventory, or an investor who has sold short and must deliver something they do not own — needs to hold those same securities for a while. So the fund lends them.
Three things happen at once, and it is worth separating them because they create three different exposures.
The securities leave. Legal title passes to the borrower. The fund keeps the economic interest by contract: dividends or coupons that fall due are passed back as a manufactured payment, and the fund keeps reporting the position as held. That last part is why this is invisible from the outside. A holdings list does not distinguish a share the fund is holding from a share the fund has lent and expects back.
Collateral arrives. The borrower posts something — cash, government bonds, other equities — worth more than the securities they took. The excess is a margin against the price moving before anyone notices there is a problem, and it is marked to market as prices change.
A fee is paid. The borrower pays for the loan, usually as a rate on the value of what they borrowed. The rate is not one number: it depends on how hard that particular security is to find. Most of a large index is easy to borrow and earns very little. A small number of positions are scarce and earn a great deal more.
Where the money goes
The fee is revenue, and it is split.
One share goes to the fund. This is the part that reaches you: it is income inside the portfolio, and it improves the fund’s return relative to the index it tracks. In a fund whose charge is small, a lending programme can offset a meaningful part of that charge — which is exactly why a fund can sometimes report a return closer to its index than its own charge would suggest is possible.
The other share goes to the lending agent — the entity that finds borrowers, negotiates terms, takes and manages the collateral, and handles the mechanics. That work is real and somebody has to be paid for it.
Here is the part worth holding onto: the lending agent is frequently a company in the same group as the fund manager. When it is, the manager’s group is earning revenue from the fund’s assets through a channel that is not the management fee, and the size of that revenue is a function of how much lending the fund does. That is not an accusation of anything. It is a structure, and a structure with an incentive in it is a thing a holder is entitled to know about before they decide it does not matter.
There is a second pot where cash is posted as collateral. Cash sitting still earns nothing, so it is generally reinvested, and the income from that reinvestment is typically split as well. This is a second revenue stream from the same arrangement, and it carries a risk the lending fee does not — see below.
What you are exposed to for it
Four exposures, and they are genuinely different from each other. None of them is a reason not to hold the fund; all of them are things the ongoing charge does not tell you about.
Borrower default. The borrower fails before returning the securities. The fund keeps the collateral and buys the securities back in the market. If the price moved against the fund between the last collateral call and the liquidation, the collateral does not cover the repurchase, and the gap is the fund’s. This is the exposure the excess collateral exists to absorb, and it is smallest when the security is liquid and prices move gradually — which is to say, it is largest exactly when markets are worst.
Collateral quality. Everything above assumes the collateral can be sold promptly for roughly what it was marked at. Collateral that is itself illiquid, or correlated with the thing it is standing in for, is worth less at the moment it is needed than at the moment it was accepted. What a fund is willing to accept, and how much excess it demands, is a policy decision that varies between funds and is written down in the fund’s documents.
Cash collateral reinvestment. This is a different animal from the other three, and the reason it deserves its own paragraph is that it is a risk the fund takes on its own account rather than one imposed by the borrower. Cash posted as collateral must be returned in full when the loan closes. If it has been reinvested in something that has fallen in value or cannot be sold quickly, the fund must make up the difference whether or not the borrower did anything wrong. A programme that lends conservatively can still be exposed here, because the exposure is in what was done with the collateral, not in who borrowed.
Indemnities are a promise, not a guarantee. Many programmes carry an indemnity from the agent covering borrower default. An indemnity is only worth the balance sheet standing behind it, and it is a claim on a counterparty in precisely the conditions that would make a large borrower fail. It reduces the exposure. It does not remove it.
Why none of this reaches the charge you compare
The ongoing charge figure answers one question: what proportion of the fund’s assets is taken each year to run it. Lending revenue is not taken from the fund’s assets — it is paid into them. The agent’s share is not taken from the fund’s assets either; it is deducted from the revenue before the fund’s share arrives. Neither one is a charge in the sense the figure is measuring, so neither one appears in it.
That is not a loophole. It is the definition working as intended. But it has a consequence a holder should be clear about:
Two funds tracking the same index with the same ongoing charge can deliver different returns, in either direction, because of what their lending programmes earn and how that revenue is divided — and can have carried quite different risks in doing so.
The quantity that does capture it is the fund’s tracking difference: how far the fund’s actual return sat from the index’s, over a stated period. Lending revenue pushes tracking difference in the fund’s favour and charges push it the other way, so the published difference already has both in it. It is a backward-looking figure about a period that has ended, and it does not decompose itself — a good tracking difference does not tell you whether the fund was cheap or was lending heavily, and that is precisely why the decomposition has to come from the documents rather than from the number.
What to look up, and where
This site does not hold the figures for your fund, and no general article can: they are fund-specific, they are set by the manager, and they change. The prospectus or key information document, and the annual report, are where a fund states them. What is worth finding:
- The split. What share of gross lending revenue the fund keeps, and what share the agent takes. It is stated as a percentage and it is not the same everywhere.
- Who the agent is, and whether it is in the same corporate group as the manager.
- How much may be on loan at one time, as a proportion of the portfolio, and how much actually was over the reported period. A permitted maximum and an average in use are different facts and both are informative.
- What collateral is accepted, in what excess, and whether cash collateral is reinvested — and if so, into what.
- Whether an indemnity exists, what it covers, and who provides it.
- The revenue itself, over the last reported period, against the fund’s ongoing charge. That comparison is the one that tells you whether this mechanism is a rounding error in your fund or a material part of what you receive.
What this page will not do
It will not tell you that a lending fund is better or worse than a non-lending one. A fund that lends heavily and splits the revenue generously with holders is earning them money for a risk they may be entirely comfortable with. A fund that does not lend at all has removed an exposure and given up an income. Which of those suits a holder depends on things this page has no access to.
It will not state a typical split, a typical revenue or a typical proportion on loan. Those numbers exist per fund and per period, and a figure quoted here as representative would be a fact invented to make the article feel more complete.
And it will not name a fund, a manager or an agent. The mechanism is the subject. Who does it well is a question that requires evidence about specific funds, and this article is not that evidence.