Managing a hedge fund requires a veritable arsenal of trading tools. Among them are the trusty duo of leverage and short-selling. The arms merchants in these pitched financial battles are the prime brokerages – the bank divisions that cater to the needs of nearly all hedge funds. Apparently, it’s good work if you can find it. Hedgeworld cites a new study  that pegs the industry at “$8 billion to $10 billion annually”.
Lending money is an ancient business. But what’s not so old is the business of facilitating short-sales (short-selling began in the 18th century). And the business of short-selling is about to change dramatically with the entry of what seems to be a flood of traditional managers into the 130/30 space. This will inevitably put new pressures on the prime brokerages and force them to address a common concern: fee transparency.
Some have wondered if there would be enough stock to short and whether the end was nigh. Last fall Goldman Sachs hypothesized that the end wasn’t imminently nigh, but might not be far off (we were less sold on their concerns – see posting ). Goldman’s numbers and those cited in the Hedgworld story are quite similar. Both say there is about $4.5 to 5 trillion of stock available to short in the world (about 10% of the world’s lendable stock and 3% of the world’s total equity supply).
The fee paid by investors to borrow shares they want to sell-short is notoriously difficult to forecast. Naturally “hard to borrow” stocks will come with a higher borrow fee as the prime broker spends his entire lunch hour calling all of his buddies at other securities lending desks to locate the requested stock. In some cases, the stock available for borrowing can be quite small if a few major holders of the name decide they don’t want their stock to be lent out (i.e. they don’t want the downward price pressure that would come from the short-selling). Borrow supply is a key determinant to the fee paid. But the exact calculation is quite opaque.
With greater non-hedge fund institutional participation in short-selling (via 130/30 strategies), there will inevitably be more pressure to increase fee transparency. But for now the problem seems to be one of hedge fund manager apathy:
“Hedge funds, whether it makes economic sense or not, are not cost-sensitive when it comes to the lending costs prime brokers are charging them. According to the report, 73% of managers said they felt they were ‘fairly charged’ by their prime brokers. The most likely explanation is that managers depend a great deal on their prime brokers. According to the report, prime brokers are the second most important relationship for a manager after investors. Hedge funds need prime brokers not just for margin and securities lending but also for research, access to initial public offerings and capital introduction, Mr. (Josh) Galper (of Vodia Group, the author a new report on prime brokerages) said. That’s why many funds are careful not to do anything that might jeopardize their prime brokerage relationship, according to the report.”
Translation: the prime brokers have the hedgies by the cojones.
But how long will this cozy relationship last with the growing number of pension funds shorting via 130/30 mandates (and facing different fiduciary standards)? According to the Hedgeworld story:
“In order for the opaque and illiquid lending market to become more efficient, hedge funds will have to put more price pressure on their prime brokers. The report found that most managers polled for the study do not see such development sooner than three to five years from now. One manager did not foresee any change for another 10 years.”
Not content to wait 10 years, we did some research and found this paper  by Pedro A. C. Saffi Kari Sigurdsson of the London Business School (published earlier this year). It sheds some helpful light on the determinants of security lending fees. According to the paper, high borrow fees act much like a constraint on short-selling. So besides being expensive for the investor, they can put upward pressure on stock prices (or more accurately – remove downward pressure on prices) and lead to generaly inefficient markets.
The study said that borrow fees range from under 20 bps to over 4% (annualized), with an average of 1% per annum for US securities.
Thirty percent of stocks in the study had borrow fees over 100bps (“specials”) and fully 5% had fees so high the study’s authors described them as having a “short constraint” even though they were technically shortable.
So why the huge range? Like anything, borrow prices are set based on supply and demand. According to the study, supply (as a percentage of firm size) is highly variable.
The authors find that a number of variables can limit the borrow supply of a particular name. These include:
- Small capitalization
- Government ownership
- Employee ownership
- Low book to market ratio
- High leverage
- Ownership by predominantly short-term investors
This is helpful background on the determinants of supply. But how does demand for short positions impact the borrow fee? The study does not actually examine the role of shorting demand on borrow fees. But it seems to us that an increased demand for shorting is more complex than a similar decrease in borrow supply. An increase in the demand to short a shock would likely be coincident with downward pressure on the share price itself. And as share price falls, then the demand for shorting might actually decrease (as marginal short-sellers might feel they had missed the boat). Then again, as one guru pointed out to us today, the desire to short a stock does not necessarily signal a negative absolute view on the name by the market as a whole. For example, a market neutral fund might push up demand for shorting without actually reflecting a dim view of the stock’s current price.
But in the end, it seems the artificial constraints placed on the supply of shorts by the factors listed above pervert markets and can create opportunities for prime brokerages to set their fees under cover of the fog of war. Although the fees may be totally rational, as the smoke clears, all sides might end up looking at the prime brokerages and asking for discounts on their next round of weapons purchases.
Addendum: In the meantime, the SEC has removed one of the tenets  of the Geneva Convention in this battle – the so-called “uptick rule” that (mercifully, some have argued) prevented shorting as a stock declined in value. At least that’s one less “short constraint” for academics to worry about.