(Bloomberg Opinion) -- Structured whatsits.

The way a lot of bank structured notes work is that you give the bank $100, and in a few years, if the S&P 500 index or whatever is up, the bank gives you back more than $100, based on some multiple of the index’s performance. If the index is down, though, the bank gives you back your $100. Heads you win, tails you don’t lose: You get a guaranteed return of your principal plus some extra money if the stock market does well. The bank has sold you a bond (or a certificate of deposit), and instead of paying you interest on that bond, it uses the interest to buy an option on the S&P 500 or whatever. You get the combination of (1) a bond that doesn’t pay interest and (2) a “free” option.

This product is built to be extremely attractive for the bank: The amount it saves in interest is significantly more than the option is worth—often by $2 or $3 in a $100 structured note—and it also charges you a hefty commission (perhaps another $3) for buying it. (The bank normally hedges the option and so doesn’t actually care if the market goes up or down.) 

But it is also built to look extremely attractive for you. Anyone could sell this stuff. If you don’t think too hard about it, it looks like magic: The bank offers you the opportunity to participate in the stock market if it goes up, but with no risk of losing your money. What’s not to like? Don’t answer “you could construct this yourself by buying bonds and options, or even bonds and stocks, and save the huge markups”; that is of course correct, but the whole point of the product is that it is sold to people who don’t say that. 

A lot of people really dislike structured notes. The Securities and Exchange Commission “has been focusing on the products, which have been criticized for their high fees and lack of transparency.” There is a widespread sense that structured notes are fundamentally a trick. But they are a subtle kind of trick. You don’t have to lie to sell them. You just have to describe them accurately, explain how the investor does well in every scenario, not emphasize the fees too much, and pitch them to investors in the sweet spot of sophistication: sophisticated enough to like structure and complication and the notion of “you will do well in every scenario,” but not so sophisticated that they ask questions like “well wouldn’t I do better in this scenario if I just bought stock” or “well what if I just bought the options myself instead?” The trick is not deception; the trick is in understanding what payoff curves look attractive and then designing a product with those curves, in using derivatives to tell a story that is based in fact but so enthralling that customers cannot resist. The trick is building a thing that makes customers feel smart, rather than worrying about whether the thing is smart. 

Yesterday Wells Fargo Advisors LLC agreed to pay $4.1 million to settle SEC charges relating to its structured notes business. The problem was not that it sold the structured notes, which often had “a selling commission of up to 3% paid to WFA, plus 2% to 3% of structuring and hedging costs received by WFA affiliates,” meaning that “a customer who purchased an MLI at a $100 par value would receive an MLI with a value between $94 and $95 as of the day of purchase.” (“MLI,” for “market-linked investment,” is one of many terms used to describe structured notes.)

Instead, the problem is that once Wells Fargo had sold the notes, it didn’t want to leave well enough alone, and some of its advisors would push structured-notes investors to trade them in for new and improved structured notes.

Here is the pitch for that. If you bought a $100 structured note, and the S&P 500 index went up over the next year, your structured note might now be worth $110. Of that $110, $100 is guaranteed: You’ll always get back, at maturity, at least as much as you put in. But the other $10 is still at risk: If the stock market goes back down, you might only end up with $100.

So your Wells Fargo broker would come to you and say: Look, why not take some risk off the table? Trade in your old $100 structured note, which is worth $110, for $110 face amount of new structured notes. The new ones are a lot like the old ones except that they’re based on today’s market price, meaning that if the market goes down you are guaranteed to get back your $110, not just the old $100.

Certain WFA representatives frequently justified the MLI exchanges by claiming that customers were “locking in gains” on their original MLI investments. The rationale behind soliciting customers to redeem early was that customers could capture their gains to date (or most of their gains when considering the markdown) rather than risk a decline in the performance of the reference asset that could result in the customer receiving less at maturity. Certain WFA representatives further solicited customers to invest the early redemption proceeds in a new MLI that would have a higher principal amount that was either fully or partially protected. Certain WFA representatives had great success selling customers on a “lock in gains” strategy, as it appealed to customers’ desire to protect their principal and avoid loss of gains in a market downturn.

That is a good pitch. The customer bought the structured note to get stock upside plus safety; this is a way to go back and appeal to those desires again. There is only one catch, which is that Wells Fargo is going to eat up most of your money in fees, markups and markdowns if you do this:

By way of illustration, if a customer makes a $100 par value investment in an MLI, it would be worth $95 on the date of purchase after taking into account the markup by WFA’s affiliate. If the reference asset appreciates significantly, it may correspond to an increase in the value of the MLI from $95 to $110. If WFA were to solicit its customer to engage in an MLI exchange at this point, a WFA affiliate would typically repurchase the investment from the customer at approximately $108 after a 2% markdown, resulting in a realized profit of 8%. If the customer then reinvests the $108 proceeds into a new MLI, the new MLI would be worth approximately 95% of par on the purchase date after costs, or $102.60. The customer exchanges an MLI valued at $110 for a similar product valued at $102.60, and more than 80% of the original MLI’s growth in value is consumed by transaction costs. If a customer regularly makes such exchanges, the customer is left with minimal long-term returns net of the significant costs.

One customer paid $900,000 in commissions and markups to flip 110 structured notes from 2009 through 2012, while making $300,000 in investment returns. (That worked out to a 1.5 percent annualized yield, which actually … wasn’t bad for insured certificate of deposits in 2009 through 2012?)

Anyway you are not supposed to do this, I guess; Wells Fargo settled without admitting or denying the charges, and the problem seems to be that the brokers “did not fully understand all of the costs associated with MLI exchanges,” and “never evaluated the effect such costs would have on customers’ ability to achieve their investment objectives.”

“It is important that brokers do their homework before they recommend that their retail customers buy or sell complex structured products,” says the SEC. Ehh. I am happy to believe that the brokers didn’t understand what they were selling, but I also tend to think that if they had understood it, some of them would have sold it anyway. It made them a lot of money! And the pitch to investors was so appealing—without being quite right—but also without being quite misleading. Nothing here is all that much worse than selling the structured notes in the first place; the flipping pitch is just a concentrated essence of the regular structured-notes pitch. 


Oversimplifying slightly, here is how the Federal Reserve’s stress tests work:

  1. Figure out how much capital a bank has.
  2. Subtract how much capital it would lose if there was a recession.
  3. Subtract how much capital it has asked to pay out to shareholders this year.
  4. Compare the result (1 minus 2 minus 3) with the required amount of capital.
  5. If it’s more, then the bank passes and can pay out what it asked to.
  6. If it’s less, then the bank fails and can’t pay out anything. (Well, can’t raise its dividend or do new buybacks.)

This is …  inefficient? Item 3, there, is a guess, based on another guess about Item 2. The bank probably wants to pay out as much as it can, to optimize so that it gets as close as possible, in the post-payout-and-recession world, to the minimum required capital. But computing how much capital it would have in the post-recession world is an uncertain process, and the bank’s answer may differ from the Fed’s. If Bank A thinks that it would have $10 billion of excess capital in a recession, it might ask to pay out $8 billion to shareholders (to be conservative); if the Fed concludes that actually it will only have $7 billion, then Bank A will “fail” the stress test and not be able to pay out any of it. If Bank B thinks that it would have $3 billion of excess capital in a recession, it might ask to pay out $1 billion; if the Fed concludes that actually it will only have $2 billion, then Bank B will nonetheless pass and be able to make its payout to shareholders. But Bank A is better capitalized than Bank B! The failure is all about its failure to guess the Fed’s models, which is a real enough failure I suppose, but not exactly a failure of solvency.

You could just get rid of the guesswork. Like here is a much simpler process:

  1. Figure out how much capital a bank has.
  2. Subtract how much capital it would lose if there was a recession.
  3. Compare the result (1 minus 2) with the required amount of capital.
  4. If it’s more, then the bank can pay out to shareholders anything up to the excess.
  5. If it’s less, then the bank fails and needs to raise capital.

You have replaced an inscrutable all-or-nothing function—guess how much stressed capital the Fed thinks you have, and if you go a penny over then you fail—with a simple transparent continuous function: The Fed tells you how much stressed capital it thinks you have, and then you can pay out anything up to that amount.

Anyway I guess that is sort of happening:

The Fed is moving toward eliminating passing and failing grades from the annual exams, which test whether banks could continue lending during a severe recession. In future years, perhaps as soon as 2019, a bank’s public grade will instead take the form of a capital ratio that the firm must meet during the following year.

The changes, set in motion during the Obama administration and continuing under President Donald Trump, reflect the Fed’s uneasiness with surprising bankers and financial markets with bad news, such as Citigroup Inc.’s unexpected stress-test failure in 2014. …

For banks, the changes promise less risk to their reputations and less urgency to invest in passing the tests, which can cost firms hundreds of millions of dollars in compliance costs.

The Fed has been steering “a gradual course away from the potential for public shaming,” said Michael Alix, a former Fed official who works as a consultant advising banks at PricewaterhouseCoopers LLP.

It is certainly more straightforward. And I get that publicly undermining confidence in banks is not a great idea for the Fed. But the current inefficient pass/fail process has some obvious charms. For one thing, it encourages conservatism: In a pass-fail system, every bank is going to ask to return less capital than it thinks it will be allowed to, to avoid the risk of failure; in a continuous system, every bank is going to optimize its capital return to return exactly as much as it's allowed to.

For another thing, it focuses the mind: There is that “urgency to invest in passing the tests” when your reputation is on the line. Some of that might be wasted effort! It’s not like the stress-test scenarios will play out exactly; it’s not even clear that, if they did, you’d go do what you planned to do in your stress-test preparation. (It can feel a little like Fed stress testing is a standalone compliance function at banks, rather than an integrated part of how actual businesses think about risks.) But getting banks to think seriously about bad cases, and plan for them, and imposing serious real-world consequences on hypothetical worst-case failures, seems like a useful discipline. And it’s different from the usual process of just optimizing capital levels to meet all the minimum requirements. If you turn the stress tests into purely another capital requirement—another exercise in minimizing capital subject to a constraint—then you lose some of that benefit.


“You shouldn’t listen to YouTubers,” says a guy who makes YouTube videos telling people what initial coin offerings to invest in, and who gets paid “bounties” in the form of tokens in those ICOs. If people invest in the ICOs, the tokens are worth something, and he makes money. It is an alignment of incentives, of sorts:

The payouts are in some ways like stock options, giving promoters a stake in the issuer’s success, said Saransh Sharma, president of 4New, a startup with an ongoing bounty campaign. 4New, which wants to generate electricity from waste, restarted its campaign in April after the ad bans halted its marketing momentum.

“It’s really a very cost-effective mechanism for developing a brand,” Sharma said by phone. “Before you know it, there’s a snowball effect.”

Yes. “About 18 percent of cryptocurrency-related posts on Reddit, Twitter and online forum Bitcointalk.org now originate from bounty campaigns.” I sometimes feel like a bit of a chump writing about cryptocurrencies without getting paid in those cryptocurrencies.

But not really! Because hooooooooooooooooooo boy are some of these people going to get into trouble with the SEC. All of this is standard practice in the penny-stock world, where issuers regularly give promoters stock in exchange for the promoters’ efforts to pump the stock—and where the SEC regularly cracks down on those promoters when they don’t adequately disclose their affiliations or register as brokers. Same, probably, in crypto:

Based on the specifics of the offering, an ICO may be considered a security by the SEC, according to a primer posted on the regulator’s website. And anyone who promotes such an offering in exchange for compensation tied to the sale may be breaking the law if they don’t first register with the regulator, according to Richard Levin, Denver-based chair of fintech and regulation practice at law firm Polsinelli PC.

John McAfee, who said “in March that he charged $105,000 per tweet,” has recently quit promoting ICOs “after receiving unspecified ‘threats’ from the Securities and Exchange Commission.” I suspect that the “threats” were along the lines of a gentle reminder that securities laws exist, and that they probably apply to people who pump ICOs just as much as they apply to people who pump stocks.

Elsewhere, here is a story about someone who encoded the private key controlling one Bitcoin onto a strand of DNA. “It might also be one of the safest ways to store your keys today as not everyone has instant access to DNA-sequencing technology (yet),” great. I did not really read this story because, you know, bzzz bzzz bzzz, but for some reason I thought of Roger Fisher’s famous proposal for how nuclear weapons should have to be launched:

My suggestion was quite simple: Put that needed code number in a little capsule, and then implant that capsule right next to the heart of a volunteer. The volunteer would carry with him a big, heavy butcher knife as he accompanied the President. If ever the President wanted to fire nuclear weapons, the only way he could do so would be for him first, with his own hands, to kill one human being.

Obviously you do not need to go to those lengths to extract and decode DNA. (The Bitcoin thing was synthetic DNA anyway.) And obviously a knife-to-the-heart level of security would not actually be useful for any Bitcoin exchange: You want it to be hard for hackers to extract your Bitcoins, yes, but you also want it to be fairly easy for customers to get back their Bitcoins; requiring murder would not strike the right balance. (Also the hackers would probably have no qualms about butchering the guy with the key, if they got a hold of him.) Still, if there is not a dumb financial thriller about Bitcoin that features this idea as a plot element, I guess I will have to write it. 

Matthews South.

Today’s Wall Street Journal features a front-page article about Vijay Culas, a former Goldman Sachs Group Inc. structured equity capital markets banker who now runs his own firm, Matthews South, which advises companies on how to deal with structured equity capital markets bankers. Normally a story about reformed derivatives bankers would be extremely in my wheelhouse and we would talk about it all day, but this one is actually a little too in my wheelhouse. Culas was my boss at Goldman, and Matthews South employs many of my former colleagues, and I remain friendly with them, and I will just leave it at: They are good people and you should employ them for all your structured equity advisory needs. Though it pains me to leave a sentence as good as “Mr. Culas launched Matthews South, named for his Harvard University dorm” just hanging there. Not a lot of people know this, but this newsletter is actually named for my Harvard freshman dorm, Money Hall, no no no I am sorry I will stop, just read the article.

Things happen.

GE Exiting Health, Oil as CEO Races to Shrink Flagging Titan. Harvard Billionaires Bail Out Alma Mater From Poor Fund Returns. Supreme Court Sides With American Express on Merchant Fees. Scaramucci's SkyBridge sues Premium Point Investments. Xerox CEO Calls Fujifilm Lawsuit ‘Misguided.’ TCI, Axon Accuse India Real Estate Fund of $1.5 Billion Fraud. Malaysia’s Extravagant Ex-First Lady Lands in Graft Investigators’ Sights. Does CFPB Oversight Crimp Credit? Amazon Workers Facing Firing Can Appeal to a Jury of Their Co-Workers. How to Steal 50 Million Bees. Besuited Man Paddleboards to Work Across the Hudson River. Dog performs CPR.

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

To contact the author of this story: Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story: James Greiff at jgreiff@bloomberg.net

©2018 Bloomberg L.P.