Equity Loans: How to Sell What You Do Not Own

The problem of delivering shares one does not have is solved by equity lending. Executing a share loan requires access, such as any U.S. brokerage would offer, to equity loans. An equity loan is best thought of as a temporary swap of legal ownership. The lender transfers ownership of some shares to the borrower, who is then free to pass the ownership on to someone else. At the same time as the shares are transferred, the borrower transfers ownership of collateral, usually cash. In the U.S., the standard collateral is cash amounting to 102% of the value of the shares, to be adjusted daily as their value fluctuates.

So, to borrow 20,000 shares of a company trading at $50 a share, the borrower would remit $1.02 million to the lender and would pay in or get back collateral as the shares rise or fall respectively. The loan is closed out when the borrower transfers the 20,000 shares back to the lender, who simultaneously returns the collateral.

But who gets the interest on the collateral? This is the key to pricing the loan. The lender invests the collateral and collects the interest, and the standard investment is a minimal-risk overnight instrument earning close to the Federal Funds rate (this being the inter-bank lending market where repayment is almost certain). If the lender simply kept all interest, the loan’s cost and the lender’s gross revenue would be as follows. Assuming the loan lasts n days, the lender would receive n days of interest on the collateral at the Fed Funds rate, that is: 1.02 x security value x Fed Funds rate x (n/360).

For the example earlier, this would be $992 for 7 days (assuming the Fed Funds rate is 5% and the price stays at $50 a share.) But the lending market is not so kind to lenders; they have to give back most of this interest.

The loan contract will specify a rebate rate, which is the interest the lender must repay to the borrower. This rebate is usually almost everything. Securities that are not scarce in the lending market – which includes most stocks on a given day and almost always the large ones – have the same rebate rate, known as the general rate.

The general rate is around 20 basis points below the going overnight rate (one basis point being 0.01%), so the rebate in the loan described above would not be 0%, but more like 4.8%. This implies a cost to the borrower and gross revenue to the lender of just $40. It is important to note that these are wholesale terms for large loans; retail brokerage customers get a much smaller rebate because of the proportionately larger fixed costs they incur on each loan and the intermediary layers they encounter. But even the big wholesale customers see their rebates shrink as the security gets scarce.

When supply of a stock is small enough relative to its borrowing demand, lenders negotiate lower rebates. A stock in this situation is said to be “on special” and the shortfall in its rebate is called its specialness. Why would a stock go on special? For one thing, it probably started there.

Stocks are notoriously hard to borrow in the first days after their initial public offerings. In our research we find IPOs to be invariably on special at first, with an average specialness of more than 300 basis points. Specialness gradually settles down to mature-stock levels, in particular after the insider-selling restrictions end at 180 days. Once a stock emerges from the initial shortage it is usually not scarce unless a new event associated with widespread shorting occurs.

The major event associated with borrowing scarcity is merger arbitrage. When one company proposes to absorb another, arbitrageurs such as hedge funds often wish to buy the target and short-sell the acquirer. This generally pays off if the merger goes through and loses if it does not. An example is At Home’s acquisition of Excite, which made At Home very expensive to borrow. Another example was the proposed merger of Office Depot with Staples, which made Staples extremely expensive to borrow, compounding the losses to arbitrageurs when the deal fell through. Specialness in these situations can sometimes reach 90%. This means the arbitrageur’s profit net of borrowing costs can be much lower than it first appears.

It also means lenders can make a lot of money. The lender can potentially earn the gross return of the hedge funds’ speculative position by selling its stake in the acquirer and buying the target; or it could simply lend the shares and thereby accrue some of the trade’s gross return with none of its risk. This source of income, often called hypothecating income, can offset the expenses of managing certain portfolios.

Portfolios in the best situation to benefit from lending are index funds, especially those that contain many small stocks.

Indexing brings two related benefits. The index’s low turnover implies low turnover for the fund, so the fund can lend with little fear of needing the shares back soon, which would happen if the fund needed to sell. Also, the fund’s mandate to track – rather than outperform – the index encourages it to hold even those index constituents that others are shorting, and not to withhold shares from the lending market on the theory that short-selling would hurt the price. The virtue of small stocks is simply that they are more likely to grow scarce, other things being equal.

Consider, for example, Vanguard’s Extended Market Index Fund, a $5 billion, 3,055-stock fund tracking the Wilshire 4500 index. The company’s 2000 annual report shows security lending income of $10.5 million (about 20 basis points) and total expenses of $13 million (about 25 basis points). So the income from lending shares offsets 80% of expenses. The report also shows that only 3% of the fund’s year-end value was out on loan ($164 million of $5.3 billion).

Similarly, Vanguard’s 1,896-stock Small-Cap Index Fund shows 10 basis points of lending revenue, against 20 basis points of expenses, with only 2% out on loan. The implication is that the fund’s shareholders achieve net returns much closer to those of their target indices than their expense ratios, taken in isolation, would suggest, while lending only a small fraction of shares – presumably just the specials. Once we move beyond small-cap index funds, though, this revenue mostly dries up. Within Vanguard we find only 0.4 basis points of lending revenue for the Mid-Cap Index Fund, and 1 basis point for the Windsor I Fund, which is actively managed.

Lending revenue is one way that investors in general, and not just short-sellers, benefit from equity lending. There is also the less tangible, but more general, benefit of increased market efficiency. By facilitating short sales, the lending market transforms negative information into negative demand, speeding its incorporation into prices.

For example, stocks that are not specials, and so are relatively easy to short, respond relatively less well to negative earnings announcements than do the specials, although their response to positive news is normal. So when scarcity in the lending market makes a stock hard to short, the market is more surprised by the publication of bad news than it is by good news.

The implication is that investors who already know or suspect bad news before its publication have relatively less success trading on it with the specials, and so incorporate less of it into the pre-publication price. By contrast, trading on good news is not a problem. Bad news will eventually have its effect; society’s benefit from equity lending is that this happens sooner, reducing the likelihood that some investors might buy at prices that other investors know to be high.

The research organization Robert Morris Associates estimates the value of US equities on loan as $104 billion at the end of 2000. The figure for European equities is $43 billion.

In spite of its size, this market is far from centralized. An equity loan arranged by a broker might be made up of shares from another account at the brokerage, or from a mutual fund lending its portfolio, or from a custodian bank lending from its client’s portfolios, as the clients permit. The whole variety of institutional accounts participate, and in contrast to the visible and publicly available market for equities, the market for equity loans is obscure and privately negotiated.

Every day, lenders have borrowing prices for all stocks but they do not report them, even to the Financial Times, and they do not systematically make them available to all investors. For example, our research found that at least three out of four IPOs can be borrowed on their respective first days, but retail investors hoping to short IPOs should expect much less success; their brokers will likely report that the trade is simply not possible.

There is a dimension of equity lending that is driven not by short-selling but by the seemingly technical distinction between legal and beneficial share ownership. If A loans B a share on Monday and gets it back on Tuesday, then B is the share’s legal owner at the close of Monday trading. That means B receives any benefit that accrues to legal owners.

For the loan not to interfere with A’s enjoyment of the benefits of ownership, however, B is generally required to reimburse to A whatever accrues to Monday’s legal owners, so A remains the share’s “beneficial” owner. The usual benefit would be a cash dividend, but it could be a stock dividend or some other distribution. If Monday was the record date of such a distribution, B owes it to A. (There is an exception to this rule in corporate votes. If Monday is the record date of a vote, B gets the right to make the vote, not A.)

These distribution issues may seem like technicalities but they can be the primary reason for equity loans. That is, investors borrow not because they shorted and need to deliver, but because they value the distributions they get more than the reimbursements they give.

To see how this could occur, consider dividend tax credits. Many countries (such as Canada, France, Germany and Italy) give some sort of credit to their tax-paying citizens for the dividend income they receive from domestic corporations. One way to get this credit is to borrow a domestic company’s shares for a dividend record date. The dividend coming in is taxable but qualifies for the tax credit and the reimbursement going out is a deductible expense.

The net effect (roughly) is zero cash flow at dividend time, with the dividend matching the reimbursement, plus zero taxable income, net of the deduction, plus the credit. The opacity of the lending market makes it hard to gauge the intensity of this sort of activity, but the incentive is large. An investor not eligible for the credit, such as a foreign citizen, cannot earn it directly but can get some of it indirectly, via the pricing of the loan.

The equity-lending market is obscure but vital. By facilitating short-selling it extends the opportunity to sell an equity beyond the group of investors already holding it. This is not only a direct service to investors interested in negative exposure to the equity, but also an indirect benefit to everybody because it enlarges the set of information that the equity’s price reflects.

Consumers may also benefit, without realizing it, from equity-lending income earned on their pension or mutual funds. This income can be so large that it offsets most of the funds’ expenses, and it might not result from short-selling.

Finally, in certain situations, such as dividend days for some countries’ equities, loans are popular simply for the legal ownership, potentially leading to tax credits, that they confer.