There’s this interesting article in the Journal today about how quickly and expensively short sellers managed to find shares of Facebook to borrow, including a graphic showing that like 150 million shares were shorted on Friday, at -10% to -40% rebates, which, what? Anyway you should read it; I for one was surprised that that much borrow came online so quickly and so share the Journal’s suspicion that some of the locates given by prime brokers were a little aggressive? Or something.
I’m less willing, though, to view the basic transaction the way the Journal does:
The role of the firms in enabling short sellers in Facebook’s stock shines a light on a long-standing Wall Street business that has the potential to create conflicts of interest. Even as one arm of a brokerage firm is getting paid to drum up interest in a stock, another part of the firm could be earning big profits by helping bet that the stock will fall in price.
“In general, Wall Street has conflicts of interest, and conflicts of interest are profitable,” said Daylian Cain, a Yale School of Management professor of business ethics. “It’s hard to navigate them when there are millions of dollars at stake.”
Two things are worth noting here. First of all, this is the sort of “conflict of interest” that comes inextricably from being a market intermediary: some of your clients want the one thing to happen, others want the other thing to happen, and the things are often – normally in secondary market trading – mutually exclusive. If your role is “to help your clients have the things they want to happen, happen,” then you have conflicts of interest up the wazoo, all the time, no matter what, because some of your clients don’t get to have the things that they want to happen, happen, exactly because your other clients want the opposite things to happen and they happen. The same is roughly true if your role is “to help the clients who want asset prices to increase, have asset prices increase,” so long as you also have clients who want asset prices to decrease; this is the version that gets the most attention.
But if your role is “to help your clients implement financial market positions that reflect their views of what will happen” then there are fewer conflicts. Someone wants to buy stock, you help them buy it. Someone wants to sell stock, you help them sell it. Someone wants to borrow stock and sell it short, you help them borrow it. If that’s your view then the “conflict of interest” would be, not providing market access to everyone, but refusing to lend short sellers otherwise borrowable stock because you have other, more important clients – IPO purchasers or issuers – who have a vested interest in having the stock go up and so you don’t want to facilitate shorting. On that view Morgan Stanley, which refused to lend shares in the early days, was the conflicted one – in fact, was practically a market manipulator, though it didn’t do much good:
But Morgan Stanley wouldn’t lend shares. The prime-brokerage sales staff cited long-standing practices at the firm prohibiting doing so when the bank is lead underwriter on an IPO, according to the people.
“Who do you think you’re protecting?” one hedge-fund manager said to a prime-brokerage manager as Facebook shares were falling. “We can get it everywhere else.”
Of course Morgan Stanley was a market manipulator on Friday, in the straightforward sense that it was buying lots of stock with the explicit goal of propping up Facebook’s stock price, though, again, didn’t do much good. But that – and the not lending – was fine. Securities offerings are a special case generally; regulators explicitly allow a certain amount of market manipulation – only up! – because it facilitates capital raising by companies. Which I guess is non-zero-sum and socially beneficial and whatnot, though maybe not for Facebook.
Second, though, looking for the conflict of interest at the banks/prime brokers is a bit misplaced. The conflict here is basically between people who are long stock and people who want to short it. The longs want it to go up, the shorts want it to go down. To implement those desires, the longs own the stock, and the shorts borrow the stock to sell it short. From whom do they borrow it? Well, sure, their prime brokers, whatever – but the prime brokers are just middlemen. Where do they get it from? In the case of a new IPO, or at least in the case of Facebook, they kind of get it from their imagination, but in general they get it from someone who owns the stock, and even in the IPO that’s sort of the in-expectations case. Where else could they possibly get it?
Two things to keep your eye on are (1) the guy who owns the stock doesn’t have to lend out the stock – many big investors don’t, and (2) the guy who owns the stock and lends it out is facilitating downward pressure on the stock. Presumably the guy who owns the stock wants it to go up, so why is he helping it go down?
The answer is mostly money: the borrower pays the prime broker to borrow the stock, and the prime broker passes some of that payment along to the lender, though of course they clip a bit of it and on expensive-to-borrow stocks that clipping business is, as the Journal suggests with respect to Facebook, pretty profitable.* Share lending is a market like any other, driven by supply and demand, and the clearing price for share lending is roughly the clearing price of a bet: the lender bets that the short sellers won’t reduce his price appreciation by more than the stock borrow fee, while the borrower bets that the stock will go down by more than the stock borrow fee. That betting market is, conceptually, no more problematic than any other betting market on the price of stocks – the stock market, for instance. Of course if there are long investors who for structural reasons can profit off the stock borrow fee but don’t much care if stock prices appreciate, then that market can be distorted – which suggests that those investors, not the prime broker middlemen, would be a good place to look for conflicts of interest.
Short Sellers Find Friends in Banks [WSJ]
* This oversimplifies but not in an essential way. It’s easier to think of the stock borrow market as primarily one where I lend you shares and you pay me the stock borrow fee, so that’s how I’m describing it. The reality is that I lend you shares and you give me cash as collateral and I pay you fed funds or whatever less that stock borrow fee. But now I have cash – in other words, share lending finances my position. If I’m a mutual fund this isn’t a big deal, but if I’m a hedge fund repoing my shares is a good way to finance them, which is a big part of why I do it – bigger than the couple of bps of borrow fee that I get on cheap-to-borrow stocks. But there’s no conceptual point there: if hedge funds don’t want to finance their shares by repoing them, they can equity finance them or whatever, it just costs more. So share lending saves them money. You could think of the “real” share lending fee to the lender as like (1) their otherwise cost of funding the position minus (2) the (possibly negative) net cost of repoing them, but still, it’s reducible to money.