Big Data and Expected Returns

Also Wells Fargo, Valeant, Zcash, various frauds, and Money Stuff's Simon logo.


Here's a keynote speech that Commissioner Kara Stein of the Securities and Exchange Commission gave to the "Big Data in Finance Conference" last week, titled "A Vision for Data at the SEC." It's about how the SEC does and should use data to catch crimes and prevent crashes and so forth. But it seems to me that Stein raises and then passes over the most interesting questions:

The securities laws, in large part, can be understood as a set of rules about information.  What information is useful for public dissemination?  How should we ensure that it is accurate and reliable?  Does fairness require everyone to have the information at the same time?  How do you protect nonpublic information?  What information is necessary for price discovery?  The federal securities laws speak to all of these questions.

In this sense, “big data” is a continuation of an old theme.  In another sense, the developments in data over the last 10 to 15 years represent a wholly new phenomenon, in the same way that satellite imaging is completely different from surveying a landscape from the top of a hill.  At that scale, patterns become evident that would have been impossible to piece together by considering one plot at a time.

The way I like to think about this is: If I know that my company is about to be acquired by another company at a 50 percent premium, and no one else knows that, then that is Information with a capital I. It is obviously useful, and it would be equally useful to everyone. It is what the securities laws call "material" information. And the federal securities laws speak to a bunch of questions -- Stein's questions -- about that information. There are rules about when the company has to disclose it publicly (Form 8-K, etc.); there are rules prohibiting the company from disclosing it selectively (Regulation FD); there are rules about who can trade on it before it's disclosed (insider trading). There is a whole structure of securities law built up around the idea that you will know materiality when you see it.

What is interesting about "big data" is how it has changed what information is useful, and to which investors. In the old days, if a company knew that sales were up across all its stores nationwide, that would be Information, and we'd know what to do with it. If an investor went to one store, sat in the parking lot, and noticed that a lot of people were going in, that would be an interesting data point -- it might inform a hunch that sales were going up -- but it would not be of any particular interest to anyone beyond that investor. But now investors can look at satellite images of all of the parking lots of all of the company's stores, every day -- or just have their computers look at the pictures and count the cars -- and get a much better sense of how sales are going. Of course not every investor can do that; you need the satellites, and the computers.

Stein's speech never uses the word "material," but it seems to me that the interesting questions for the securities laws in the age of big data will involve figuring out what materiality means. Which forms of tiny incremental advantage -- co-location at the stock exchange, slightly faster access to news releases, electronic surveys of stock analysts -- matter, and how do we decide which are fair? If every piece of data can be useful to some investor somewhere -- if it can give that investor's model the tiny incremental advantage that it needs -- then how do we decide what is material? And if everything is material, how do the answers to Stein's questions change?

Elsewhere, "the Analysis and Detection Center of the SEC's Market Abuse Unit culls through billions of rows of trading data going back 15 years to identify individuals who have made repeated, well-timed trades ahead of corporate news." 

Private equity.

There's a fairly popular model of finance in which risk-free things earn the risk-free rate, and risky things earn the risk-free rate plus a relevant risk premium, and those things -- the risk-free rate, the risk premium -- change, and the returns you can expect on your thing vary based on macroeconomic conditions and the thing's inherent risk. And then there's another fairly popular model where the returns you can expect on your thing are 20 percent. It just seems like a good number, you know? And so for instance investors more or less expected a 20 percent return on equity from Goldman Sachs Group Inc. for many years, and then regulation dramatically limited Goldman's activities and required it to have much more equity at the same time that lower interest rates meant that expected returns on all sorts of asset classes went down. And so Goldman had to go out to investors and be like "hey it's not 20 percent anymore, sorry," and it was awkward.

Anyway, eight years into the modern low-interest-rate environment, private equity is going lower and longer:

They are offering investors vehicles that will run for 14 years or more, rather than the traditional 10 years, and offer 15 per cent returns or less, lower than the 20 per cent in a typical fund.

You can read this as a concession to economic conditions, but on the other hand, 20 percent really is a nice number. And did you know that Blackstone Group LP hasn't done a public-company leveraged buyout in three years? That seems like a noteworthy drought. It ended yesterday with the Team Health Holdings Inc. buyout.

Oh, Wells Fargo.

"Wells Fargo & Co has agreed to pay $50 million to settle a racketeering lawsuit accusing it of overcharging hundreds of thousands of homeowners for appraisals ordered after they defaulted on their mortgage loans," and I guess the trick is to get all of your awful-sounding settlements done in a relatively short window, so that people are still mad at you for the fake accounts and forget to get mad at you for the appraisal gouging.

The claim here is pretty simple:

Mortgage agreements allow banks to charge homeowners for the appraisals if they default on their mortgage loans, but Wells Fargo added large mark-ups to the amounts its third-party vendors charged, the 2012 lawsuit said.

There is a long unspoken tradition in finance of thinking that if someone defaults on you, all bets are off. It is your right, even your duty, to gouge him everywhere you can, because he has broken the contract and put your money at risk. You should feel free to turn his default into a profit opportunity for you. You see this in Wells Fargo's appraisal markups, or in the Royal Bank of Scotland's controversial Global Restructuring Group, or even in JPMorgan and Citigroup's gleeful profiteering in the Lehman Brothers bankruptcy.

It's easy to see why. For one thing, you really are mad! You loaned him money in good faith, he didn't pay you back, and now you have to go to all the trouble and expense and worry of trying to get paid. Of course you want compensation for that worry. For another thing, he is not a particularly sympathetic character: He has broken his word, and no one is likely to take his side against you. He's a defaulter, a deadbeat, a failure, the enemy.

One big thing that has changed in banking since 2008 is that people no longer think that defaulters are the enemy. They think that banks are the enemy. This will have continuing awkward implications.

Oh, Valeant.

Is it a little surprising that "U.S. prosecutors are focusing on Valeant Pharmaceuticals International Inc.’s former CEO and CFO as they build a fraud case against the company that could yield charges within weeks," and that those charges would revolve around "potential accounting fraud charges related to the company’s hidden ties to Philidor Rx Services LLC, a specialty pharmacy company that Valeant secretly controlled"? I mean, on the one hand, no bad news about Valeant can really be that surprising, and the Philidor stuff really was super weird, and Valeant really did discover and disclose accounting improprieties. On the other hand, the Philidor stuff has never really struck me as all that bad? It's what started the unravelling of market confidence in Valeant, but that unraveling seems to have much more to do with doubts about Valeant's acquisition-hungry business model, its pricing practices, its dealings with insurers, its leverage, etc. -- real business doubts -- than with questions about the accounting. But I mean, you are also supposed to get the accounting right. Anyway, the stock was down 12.3 percent on the news, closing at $17.84 yesterday.

By the way, every time there is bad news out of Valeant, someone asks me to revisit the rough math I did a while back calculating that Pershing Square Capital Management, L.P.'s investment in Valeant (1) breaks even at some comically high price (I got about $161.21) and (2) has "another, sadder breakeven at a stock price of about $17.82," where Pershing Square loses the entire $4.08 billion it's invested so far and has to stump up more money, because 9.1 million shares of its investment are actually in the form of put/call option combinations with large negative values. The main thing that has changed in Pershing Square's position -- from this Schedule 13D/A filed in June -- is that it (1) sold the $95 call options it owned and replaced them with (longer-dated) $60 calls and (2) bought in the $60 puts it had sold and replaced them with longer-dated $60 puts. The net expense for those trades was about $24.5 million, which is pretty much rounding error. (You could add it to the $4,084.4 million that I calculated Pershing Square has already spent on its Valeant trades, but that doesn't count, like, legal bills, so it's a bit of false precision.) But if you take it seriously, then Pershing Square has spent an extra $24.5 million, so to get back to being down "only" the $4.08 billion that we talked about last time, it would have to see the stock go up to at least $18.61 by January 2019. That is not especially meaningful math; there is no magic to the $4.08 billion. Someone could probably calculate the stock price that would make Pershing Square's Valeant investment profitable overall -- that would earn it back the $4.1 billion it has sunk into Valeant so far -- but that seems even less meaningful. 

Blockchain blockchain blockchain.

It does seem a little weird that a currency whose central feature is that it creates an immutable auditable public record of all transactions ever became popular for buying drugs? I mean, "each Bitcoin user has an address, made up of letters and numbers, and the authorities are often able to link an address to a real person using sophisticated data analysis." And then they've got all your transactions. I suppose the drug dealers are no different from the big banks in this regard, using blockchain technology not so much because it is well suited to their particular needs but just because it has a general aura of coolness around it. 

Anyway now there's Zcash, which is like bitcoin but more anonymous, and everyone is very excited about it. Except the cops: "The privacy features of Zcash could make it harder for the currency to win support from regulators and bankers." I have to say, if we were living in a fully credit-card-based society, and someone invented the idea of paper cash, we'd probably lock him up forever. Anyway, here is my Bloomberg View colleague Elaine Ou on Zcash, which briefly made its founders billionaires.

Elsewhere in drug financing, here is a Bloomberg News article about how banks won't lend even to legal pot companies, illustrated with some pretty adorable animated marijuana leaves. 

Affinity fraud.

Here is the story of a guy named Avery Stone, who started a Hong Kong cash-advance company called Global Merchant Funding Ltd., raised $32.5 million from investors, lost it, and then fled the country. There was a brief stopover in Westchester to allegedly steal $400,000 from his father ("He's a crook, ok? You can quote me on that," says the father. "He’s lying low, because he’s a low-life."), and then Stone vanished, leaving behind a bunch of disgruntled investors many of whom are pretty high-powered international bankers in Hong Kong. One is John LeFevre, also known as @GSElevator, though he got his money out in 2013, before the problems:

“You get on the bus and it’s all rich, smart guys, and we all think someone is driving, but everyone’s just taking a nap in the back,” said LeFevre, who cited frequent name-dropping of big investors as a means of bringing in new money.

You might think that the ideal targets for financial fraud would be dumb unsophisticated people, the real key seems to be finding people who think that they're smart, sophisticated and cynical. Ideally you'd combine those two attributes and target people who are unsophisticated but think they're sophisticated, which is why conspiracy theorists make such good marks. But they tend not to have much money. Targeting people who think they're financially sophisticated, and who are financially sophisticated, requires you to up your game a bit, but the rewards are also much higher. Obviously none of this is any sort of advice.


Speaking of which, the story of Andrew Caspersen -- a wealthy Groton, Princeton, Harvard Law graduate who worked in private equity and who pleaded guilty to scamming millions of dollars out of investors so he could gamble unsuccessfully on index options -- is very sad, but I am not sure how sympathetic the sentencing judge will be to his lawyer's argument "that it was an extreme compulsion to trade stock options that drove his client with an Ivy League pedigree to lie and steal from his friends, family and a hedge fund foundation." I mean, I guess stock-option-trading addiction is a thing? But it seems like the sort of addiction that you'd pretty much have to pick up at Groton. Or Princeton:

The stock trading addiction began at Princeton, Mr. Shechtman said, but was amplified by a $2.7 million distribution that Mr. Caspersen received from a family trust in 1999, when he was at Harvard.

Over the following years Mr. Caspersen received another distribution, which he lost as well. Soon, he turned to friends and family for investments, getting a million dollars here, several million dollars there. 

I suppose if people kept writing me multi-million-dollar checks, I would eventually develop an addiction to spending them, but fortunately for me I have never been faced with that particular health risk.

Elsewhere, Iftikar Ahmed -- who was charged with stealing money from his venture capital firm and fled to India -- is having a rough time of it. And: "Singapore Puts Banker on Trial in Case Linked to 1MDB Money Laundering."


We talked yesterday about Goldman Sachs Group Inc.'s Structured Investment Marketplace and Online Network, or as it is affectionately known, Simon. I said that "if I had designed Simon, its logo would be a cute cartoon squid named Simon, with a little sailor hat and a big smile and just the tiniest bit of blood dripping from its tentacles," and you can probably guess what happened next. From Elle on Twitter:


I like it a lot. (Here, sadly, is Simon's actual mascot.) In other Simon news, reader Don Vollum pointed out the parallels to the 1978 electronic game Simon. Specifically, Simon (the game) has a slogan that would work perfectly for Simon (the structured notes platform): "Simon's a computer, Simon has a brain, you either do what Simon says or else go down the drain." Honestly it is a little shocking that Goldman did not hire me to do Simon's branding.

People are worried about unicorns.

"Flat is the new up," says the co-founder and chief executive officer of Postmates, which just closed a $140 million financing round at the same $400 million valuation at which it raised money last year. I guess if that's true, that would be a reason to worry about unicorns.

People are worried about bond market liquidity.

I've got nothing really, but in two weeks the Brookings Institution will have an event featuring Federal Reserve Vice Chairman Stanley Fischer, titled "Do we have a liquidity problem post-crisis." So mark your calendar. Meanwhile: "Corporate bond market set for record year of debt issuance."

Me elsewhere.

I wrote about Bess Levin at Dealbreaker.

Things happen.

CenturyLink, a Network Provider, to Acquire Level 3, a Rival. Big Mergers May Face a Higher Bar in Washington. Citigroup Says CFTC Is Probing Banks’ Rate-Swap Businesses. Mark Carney to Lead Bank of England Through June 2019. Kuroda admits BoJ will not meet inflation target before term ends. Goldman Sachs Said Outsourcing Dark Pool Operation to Nasdaq. EU markets watchdog warns over UK ‘conflict of interest.’ This Investment Bank Sold All UK Assets Before Brexit Vote. EU Kicks Off Corporate-Tax Overhaul. Clock ticking on Monte dei Paschi alternative rescue plan. Gross eager to repeat magic that made him investing guru. Election Turns Up Market Fear Dial. Donald Trump Used Legally Dubious Method to Avoid Paying Taxes. Venezuelans Give Up on Counting Piles of Cash and Start Weighing Them. Alipay Mobilizes for World-Wide Expansion. Asset Bubbles From Stocks to Bonds to Iron Ore Threaten China. Career Bankers Alone Can’t Solve the Financial Industry’s Problems. "The key tool for implementing optimal opacity while fine tuning stigma is the haircut for bonds offered as collateral in lending facilities." America Is Building More Three-Car Garages Than One-Bedroom Apartments. Isn't this just about anchoring? Cashew Prices Are About to Go Nuts. The secret to staying young is broccoli and cabbage. Why science says pudgy dads are more attractive. Burglar survey. Cocktail forgery. "[Campaign] has been no exception, using cutting edge technologies like Meerkat, Vines, Infographics and even Emojis to engage with their supporters on their smartphones."

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.