Money Stuff

Insider Trading and Theoretical Indexes

Also trading floors, Altaba branding, memes and Berkshire dividends.

Leon Cooperman.

We talked a while back about the Securities and Exchange Commission's insider trading case against Lee Cooperman, which boils down to a pretty simple factual question. Everyone agrees that Cooperman talked with an executive at Atlas Pipeline Partners, L.P., and that he subsequently bought some Atlas stock, which went up. There may be some dispute about whether what they talked about was material, or nonpublic, but those are not the sorts of disputes that insider-trading defendants frequently win.

No, the important factual question is whether Cooperman agreed not to trade on the information. Cooperman was not an insider at Atlas, and the insider who gave him his information wasn't doing so in violation of any duty to the company. He was just an executive, talking to a big investor, which is kind of his job. It's only illegal insider trading if Cooperman had a duty to the insider not to trade, and the only way he'd have that duty is if he agreed to keep the information confidential and not trade on it. The SEC claims that he did. Cooperman denies that, and says that in fact he regularly begins conversations with corporate executives by asking them not to lock him up by giving him material nonpublic information.

I wasn't there, but I have to say that Cooperman's version of events seems more plausible than the SEC's. It is pretty common for investors to talk to executives with the understanding that anything that they hear is fair game for trading; it is pretty rare for investors to talk to executives, explicitly agree to keep the information confidential, and then go trade on it immediately. And when investors do agree to receive nonpublic information that will lock them up from trading, it is pretty customary to confirm that in writing. The SEC's theory seems to be that Cooperman agreed not to trade, but that there's no record of that other than the executive's memory. Meh? The last high-profile case like this -- against Mark Cuban -- ended in a loss for the SEC, as a jury believed that Cuban had not agreed to be locked up.

But who knows; we haven't seen the SEC's evidence yet. And Cooperman's lawyers and the SEC have gotten into a fun little fight about whether the SEC's theory is even adequate to get to trial. The dispute is over the fact that Cooperman and the executive had at least three conversations in the time leading up to the disputed trades; the SEC seems to think that Cooperman got material nonpublic information in each of these conversations, but it doesn't say when he promised to keep it confidential. In fact, the SEC "originally alleged that the Atlas executive provided Cooperman with confidential information on July 19, 2010 and 'only thereafter sought and received assurances from Mr. Cooperman that he would not use the information,'" though the complaint ended up being vaguer. The implication seems to be that the executive shared material nonpublic information with Cooperman, thought better of it, and called Cooperman to say "hey you're not going to trade on that are you?"

Does that matter? Cooperman argues that "any agreement made after information has already been shared cannot give rise to an insider trading claim." The SEC responds that it doesn't matter if the agreement comes before or after the disclosure; it matters if it comes before or after the trade. "The relevant question is: At the time of the trade, did the recipient of material, nonpublic information have an agreement to maintain the information in confidence?" What do you think? The SEC's theory seems more correct to me. After all, if the executive really did tell Cooperman a bunch of secret stuff, and then called him up to say "hey you're not going to trade on that" -- Cooperman could always say "sure I am!" Under both Cooperman's and the SEC's theory, he'd then be free to trade. So why promise not to? Insider trading law is pretty dumb, isn't it?

The Dow.

Every time the Dow Jones Industrial Average approaches some round number, there are articles about what it means, and then there are articles about how dumb the Dow is. The Dow is dumb! Calculating a price-weighted average of a small arbitrary group of stocks is no way to go through life, son, and no one should care if that average reaches 20,000. But you know something else? If you calculated the Dow in an even dumber way, it would already be at 20,000:

The metric in question is called the Dow’s “theoretical” high, and it’s computed using the intraday highs of each of the blue-chip index’s 30 components. It doesn’t matter if General Electric hits its high at the open, Procter & Gamble reaches its zenith at noon, and International Business Machines peaks just before the close. Each one counts when the day’s theoretical high is tabulated.

By this outdated measure, the Dow crossed the 20000 threshold for the first time on Dec. 13. The actual intraday high that day was 19954, but the theoretical high was 20048.

I love that it's called the "theoretical" high, as though there is a complex theory behind it. (The theory is: It was easier to calculate without computers.) So congrats, everyone, on Theoretical Dow 20,000! But why stop there? Why not pick an even dumber calculation method? For instance, the Quantum Metaphysical Dow is another measure of stock prices, which I just made up, and which is computed by taking the regular Dow Jones Industrial Average and multiplying it by two. By that method of calculating the index, the Dow has already crossed 39,000! We're all rich!

Elsewhere in round numbers and intraday highs, Bill Gross said that a move above 2.6 percent on the 10-year Treasury would be "the key to interest rate levels and perhaps stock price levels in 2017," and then Jeffrey Gundlach said that "a couple of second-tier bond managers talking about 2.6% as a key technical level on US 10-year are ignoring the fact that 10-year made intraday high [on 12/15] of 2.64%," and a great cry of "oh no he didn't" went up across the land. 

Trading floors.

We have long since passed the inflection point where, when I read about an exchange closing its physical trading floor, my reaction is now "oh wow they still had a trading floor?" rather than "I can't believe that iconic trading floor is closing." Did you know that the Amex still had -- still has! -- a physical trading floor where actual human beings show up to work every day to trade stocks? I mean, it's called NYSE MKT -- the New York Stock Exchange bought it in 2008, and it's part of Intercontinental Exchange Inc. now -- but it is a direct descendant of the American Stock Exchange, previously the New York Curb Exchange, formed by "traders excluded from the NYSE" who "initially did business on the sidewalks of downtown Manhattan." The trading floor will close this year. I hope that the traders will keep showing up every day, and the security guards won't let them in, and they'll congregate outside and trade stocks on the sidewalk, and they'll get back to their roots. The human traders went soft, once they moved indoors, which gave the computers an opening to replace them. But once trading moves back outside, the humans will have a big advantage, especially if it rains.


I really enjoyed this strangely misguided article about Yahoo! Inc.'s plan to change its name to "Altaba" after it sells all of its Yahoo businesses to Verizon and ends up as just a pot for holding shares in Alibaba Group Holdings Inc. and Yahoo Japan Corp. All these brand experts weigh in on the new name, as though there is branding to be done:

"Anchoring the new name on the valuable link with Alibaba makes sense, as it is a fresh break from Yahoo's troubled recent past. If true that the first part of the name is about 'alternative,' it remains to be seen what is alternative about the new firm and how its strategy will relate to Alibaba."

For Jon Wilkins, executive chairman of agency Karmarama, the name Altaba is quite corporate. "Brands should focus on the impact on the user, rather than in this incidence (where) it looks like a corporate affairs decision, as opposed to a decision based on the user," he told CNBC by phone.

Yes right that's true but it's because there are no users. Altaba will be just a pot of shares, an investment company that will own stock in Alibaba and Yahoo Japan and not do anything else. "What is alternative" about it is that it will be a way to get exposure to Alibaba without buying Alibaba shares (presumably at a discount). "How its strategy will relate to Alibaba" is that it will sort of hang around looking pitiful until maybe one day Alibaba buys it. "The name Altaba is quite corporate" because Altaba will be a pure creature of corporate finance; no consumer will ever have any interactions with Altaba, so its unlovely name won't matter.

Meme stock market.

The bond market is a prediction market, and what's nice about it is that you are automatically paid off for correct predictions about credit quality. If you buy a bond, you'll get interest payments, and the bond will pay off at maturity, unless the issuer defaults. Your job is to predict which issuers won't default, and if you are good at it, the simple process of the bonds paying off will reward you. It is more complicated than that, and in practice many people make their money in bonds by buying and selling to each other without waiting for maturity. But that basic mechanism underlies the whole thing: You predict which bonds will actually be paid, and buy them, and then if you're right you get paid.

The stock market is like that, only instead of paying interest stocks sometimes pay dividends, and instead of maturing stocks are sometimes bought back by their issuers, and often neither of those things happen and a company's stock just floats around for decades. So the "buying and selling to each other" dominates, and the actual or potential cash flows from the issuer of the stock are just background theoretical support for the secondary speculation. If you look at the trading in, say, Inc. stock, it's weird to say "oh yes these investors are just calculating the expected value of future dividends and using that to price the stock." But that theoretical support does matter. If a company went public with a charter saying that it could never pay dividends, could never buy back stock, could never merge, and that if it ever went out of business or sold its assets it would donate all of its money to charity -- then I think the stock would be worthless? I hope? I suppose someone should try it and find out.

At least in the abstract, these markets are pleasingly self-fulfilling. You buy stock in a company because it gives you ownership of the company; if the company does well, you will own more stuff. You buy bonds from a company because they are a promise of repayment; if the company does well, you will get paid back with interest. What distinguishes "prediction markets," and derivatives markets generally, from these more primary markets is that they lack this quality. They are just agreements between two people about how they will pay each other if future events occur. Often those events are pretty easy to define. If Microsoft Corp. stock closes above $65, my $65 call option will pay off in a predictable way. If Donald Trump is elected president, my "Trump to win presidency" prediction-market contract will pay off. 

Sometimes they are much harder to define. There are occasionally efforts to build shadow stock markets -- prediction markets -- for private technology companies, the idea being that you buy an instrument on Uber, and if Uber ends up being worth $X, the contract pays you some fraction of $X. But what does "ends up being worth $X" mean? What if Uber is sold in a private transaction in which different share classes are paid off differently? What if it goes public but spins out some of its businesses simultaneously? What if it goes public after the contract expires? These are contingencies that a share of common stock just sort of takes care of automatically -- it is a residual claim on the equity of the company -- but that a prediction contract has to actually spell out. 

Or there are credit default swaps, which pay out if a company or country defaults on its debts. But what does that mean? The history of CDS is in large part a history of arguing over whether, and how, the contracts should pay out in the case of weird forms of quasi-default. (I suppose that's part of the history of bonds, too, but the CDS arguments are vaguer and harder and more fraught.) Or there is the longstanding desire to build house-price derivatives, which always run into the fact that it's hard to measure "house prices." But this is not a problem for people who buy and sell actual houses. If you sell a house, you know what price you got for it.

One thing that this means is that the stock market is a decent way of measuring the value of a companies, and the bond market is a decent way of measuring the creditworthiness of companies, but prediction and derivatives markets are not automatically good ways of measuring the thing that they're supposed to be trading. Sometimes CDS markets measure a company's probability of default; sometimes they just reflect quirks in CDS contract design. The thing that these markets measure is not just the underlying quantity, but also their own ability, or inability, to approach that underlying quantity.

Anyway here's a story about "How a group of Redditors is creating a fake stock market to figure out the value of memes." Basically, they seem to be creating a scoring mechanism to decide how much a meme is worth, and then a prediction market to predict how high various memes will score on that scoring mechanism. But the work of "valuing" the memes really seems to be done by the scoring mechanism, not by the market. 

People are worried about stock buybacks.

"Stock buybacks" to me is a synecdoche for corporate capital return more broadly, so let's talk about my Bloomberg Gadfly colleague Tara Lachapelle's call for Berkshire Hathaway Inc. to start paying a dividend. Berkshire paying a dividend! That's never happened before, except once, in 1967, sort of by accident. ("Buffett later said -- in his typical potty humor -- that he 'must have been in the bathroom' when that decision was made.") But times are changing, or will be soon:

I'm not talking about Buffett paying a dividend. I'm talking about a post-Buffett Berkshire, where there will be tremendous pressure to spend money as well as he has, finding the kind of fairly priced, "elephant-sized" acquisitions for which he and his business partner Charlie Munger, 93, are famous. A dividend may help the next CEO win favor with investors. It also lessens the burden and urgency of finding big acquisitions to make use of Berkshire's ever-expanding cash hoard, which is far less valuable sitting in a bank. 

The common complaint about stock buybacks and dividends is that they show that managers are out of better ideas, that they have given up on finding good new projects for their companies and so just give the money back to shareholders. I think that this is generally a much more nuanced question: Most companies have specific domains of expertise, and if they have more money than they can spend pursuing that expertise, why should they spend it pursuing other things? Facebook is very good at making a social-media website, but that website generates far more money than it costs. Should Facebook's social-media-website managers decide where to invest that money, or should, you know, BlackRock and Fidelity decide?

But Berkshire Hathaway! You don't invest in Berkshire because of its specific expertise at making widgets, or even insurance. You invest in Berkshire Hathaway because Warren Buffett is the greatest living investor, and you want him to make your investing decisions for you. If his existing businesses throw off more money than they need, then by gum, he should use it to buy more businesses. Better him than you. Giving you some of your money back, in the form of a dividend, would be pointless: What are you, or BlackRock or Fidelity, going to do with the money that is better than what Buffett would do with it?

(On the other hand, Berkshire has occasionally bought back stock, when it trades below "intrinsic business value." But that's just Buffett making one more investing decision.)

Of course a post-Buffett Berkshire won't have Buffett. The question, then, will be: Is Berkshire Hathaway a vehicle for the best post-Buffett investment managers to make investing decisions, or is it a weird insurance-and-industrial-parts conglomerate whose corporate-finance decisions are just like those of any other company? 

Things happen.

The Wall Street Trader Behind SoftBank’s $100 Billion Fund. ‘Cartel’ Currency Traders Charged in Market-Rigging Case. VW in Talks to Plead Guilty in U.S. Emissions Case. Supreme Court Considers Role of Free Speech in Explaining Credit Card Fees. (And Noah Feldman on the case.) Saudi Arabia plans bond deal to help finance budget deficit. M&A Boutiques in Risky Place Under Trump. Valeant Strikes Deals to Shed $2.1 Billion in Assets. Brookdale Senior Living in Early Deal Talks With Blackstone, Others. Great Debate Erupts Over Great-Rotation Thesis for Stocks, Bonds. Bernstein, asset allocation and Marxism. The Federal Trade Commission is mailing checks to 350,000 Herbalife distributors. SEC: Port Authority Omitted Risks to Investors in Roadway Projects. Legal Institutions and IPO Puzzles. Does the Second Amendment protect cryptography? "I think Washington is in many ways a more natural fit." Al D’Amato kicked off flight after starting rebellion against crew. launches futuristic pop-up 'Espresso Yourself' 3D-printing cafe to help singletons find love in their latte. Research: Cracking a Joke at Work Can Make You Seem More Competent.

If you'd like to get Money Stuff in handy e-mail form, right in your inbox, please subscribe at this link. Thanks! 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the author of this story:
    Matt Levine at

    To contact the editor responsible for this story:
    James Greiff at

    Before it's here, it's on the Bloomberg Terminal.