Could Illegal Short Selling Improve the Market?

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

The relentless advance of modern technology opens up whole new realms of financial malfeasance but also probably closes some off. One thing that seems to be getting more difficult is violating Rule 105 of Regulation M, which is the rule forbidding this thing:

  • a company announces a stock offering,
  • you sell that company's stock short just before the offering prices,
  • you buy shares in the offering to cover your short.

This is naughty for reasons that I guess seem intuitive? You're selling at the market price, but the company can't do that; its offering probably comes at a discount to the market price. And then you're covering at the offering price, so you're making a risk-free-ish profit. And since your short sales tend to drive down the market price (and thus the offering price), your profit is at the issuer's expense. Here is the Securities and Exchange Commission's explanation:

A fundamental goal of Rule 105 of Regulation M is protecting the independent pricing mechanisms of the securities markets so that offering prices result from the natural forces of supply and demand unencumbered by artificial forces. The Rule is particularly concerned with short selling that could artificially depress market prices. Generally, the offering prices of follow-on and secondary offerings are set at a discount to a stock's closing price just prior to pricing. A person who expects to receive offering shares may attempt to profit by aggressively short-selling the security just prior to the pricing of the offering, thereby depressing the offering price, and then purchasing lower-priced securities in the offering.

Sure, fine, more on that in a moment. The news today is that the SEC settled with 22 firms, and sued another one, for violating Rule 105. The targets are an entertaining mixed bag; most are wee but some are big names, and some seem to have had Rule 105 violations as, like, a core strategy, while others seem to be just hedge funds that trade a lot and accidentally got caught up in this.*

The fact that even the accidents got caught is sort of telling. The settlements are just lists of each time each firm shorted a stock and then bought shares of that stock in a near-simultaneous public offering. They don't claim to be comprehensive, necessarily, but you get the sense that they are. From the SEC's announcement:

"The benchmark of an effective enforcement program is zero tolerance for any securities law violations, including violations that do not require manipulative intent," said Andrew J. Ceresney, Co-Director of the SEC's Division of Enforcement. "Through this new program of streamlined investigations and resolutions of Rule 105 violations, we are sending the clear message that firms must pay the price for violations while also conserving agency resources."

The SEC now appears to be in possession of computers, and programs for those computers, and data from brokers and exchanges. And gosh are they excited about it! The real goal here for the SEC must be to automate some kinds of misconduct out of existence. Not every kind: Insider trading, for instance, can't reliably be stopped by an algorithm, since you can always give your inside information to your aunt's neighbor's son's roommate's grandmother, have her trade for you, and then meet you in a parking lot to hand you a bag of cash representing your share of the profits. Hard to trace that. But the universe of things that can be solved algorithmically keeps expanding,** and surely short-selling violations -- which you need to set up by borrowing shares through a broker and so which should be easy to track -- are on the list.

Which is great for, um, stamping out Rule 105 violations. I carry a soft spot in my heart for Rule 105 violations because ... well, first, I'm a sucker for all flavors of obscure technical is-it-bad-or-is-it-not financial stuff. But second, there's actually a non-trivial case that this isn't so bad. One way to think about these trades is the SEC's way: Hedge funds are shorting stock to drive down the price and then cover at a profit to themselves.

Another way to think about them, though, is that the hedge funds are pre-distributing the offering. The way stock offerings work is that you do a big offering for a bunch of shares, find the clearing price, and sell everyone shares all at once at the clearing price. If you want to sell a million shares, and get 300,000 orders at $28, 400,000 at $27, and 500,000 at $26, then you sell a million shares at $26. For reasons of law and custom, you can't sell the 300,000 shares at $28, even though you had buyers willing to pay that price.

Hedge funds shorting throughout the day can sell at whatever price anyone is willing to pay at any point during the day. (I mean, they could except for Rule 105.) That is at least part of why they get higher prices on their short sales than they pay in the offering. The fact that these trades are illegal means that there aren't that many of them, so the offering prices as though they didn't exist and the hedge funds make an easy profit. But if this trade were legal then you might see more of it, and the profits on it would compress. If this trade became common, then you could see offerings where the shares were fully distributed before the deal "priced": Company X offers 1 million shares to price Monday night; hedge funds borrow and short 1 million shares (to "real" buyers at market prices) during the day on Monday; when the deal prices it's just sold to the hedge funds at a price that makes it worth their while,*** and there's no further market activity.

Part of why stock offerings come at a discount to the last trade is that no one is entirely sure how the stock will trade after the deal prices and there's new supply in the market, so the buyers need to be paid to take on that uncertainty. If all the supply is pre-distributed by short sellers, then that's less of a problem: You already know how the extra supply will affect trading, because it's been created by short sellers. Which might lead to deals pricing at less of a discount to market prices.

So the point is that if this short selling into a public offering were allowed, it might end up being good, rather than bad, for issuing companies. But I guess the other points are that it's not allowed and that the SEC has gotten pretty good at stopping it. So we'll probably never find out.

* Shorting a lot of shares the day the offering prices, and then covering roughly the same number of shares, looks like a strategy. Especially if you do it over and over again. So there is a Mr. Michael Anthony Stango , who seems to have pre-sold a whole lot of his purchases:

In June 1, 2011 and June 2, 2011, Respondents sold short 216,700 shares of Arch Coal Inc. ("ACI") during the restricted period at an average price of $27.6856 per share. On June 3, 2011, ACI announced the pricing of a follow-on offering of its common stock at $27.00 per share. Respondents received an allocation of 211,500 shares in that offering. ...
From December 13, 2010 through December 15, 2010, Respondents' sold short 19,000 shares of Cloud Peak Energy Inc. ("CLD") during the restricted period at an average price of $20.3095 per share. On December 15, 2010, CLD announced the pricing of a follow-on offering of its common stock at $19.50 per share. Respondents' received an allocation of 23,200 shares in that offering. ...
On March 1, 2011, Respondents' sold short 12,500 shares of EOG Resources Inc. ("EOG") during the restricted period at a price of $107.94 per share. On March 1, 2011, EOG announced the pricing of a follow-on offering of its common stock at $105.50 per share. Respondents' received an allocation of 20,200 shares in that offering. ...

Etc. etc.; there are thirteen paragraphs of this. On the other hand there are bigger names like D.E. Shaw , which seems to have violated the rule by accident:

From March 30 through March 31, 2011, D. E. Shaw sold short 400 shares of Kraton Performance Polymers Inc. ("KRA") during the restricted period at a weighted average price of $ 39.7775 per share. On March 31, 2011, KRA announced the pricing of a follow-on offering of its common stock at $37.75 per share. D. E. Shaw received an allocation of 100,000 shares in that offering. The difference between D. E. Shaw's proceeds from the restricted period short sales of KRA shares and the price paid for the 400 shares received in the offering was $811.00.

There are a few more examples, but D.E. Shaw's shorts are always less than a tenth of their allocations. This seems like an algorithm that wasn't aware of the offering, not a conscious scheme.

Somewhere in between is Ontario Teachers' , another big name, with four violations between July 2010 and January 2011 in which they short 31,900 shares and buy 25,000, short 50,000 and buy 25,000, short 9,600 and buy 15,000, and short 40,192 and buy 50,000. So that sounds ... kinda intentional? Maybe they didn't know this was illegal in the U.S.?

** Among my favorites: the SEC's effort to catch accounting fraud through quantitative textual analysis . It's pretty odd.

*** Presumably a price below their average price. Not necessarily a price below their last-sale price (and even less necessarily at a price significantly below the last sale, or below a manipulated-into-the-close last sale). The dynamics here are not 100% obvious: The hedge funds have leverage (pulling an offering looks and is bad for the company), but then so does the company (those hedge funds are exposed to a lot of short risk if the company pulls the offering).

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Matthew S Levine at