Truth, Yachts and Accounting

Also Ponzi data, MBA preferences, complacency and a unicorn-eater.

Radical truth.

"The hero comes back from this mysterious adventure with the power to bestow boons on his fellow man," wrote Joseph Campbell, and Ray Dalio read that and was inspired. Fortunately he has been on a mysterious hedge-fund adventure for the last few decades, and he's back with a boon for you. Several boons, actually: He's turning the "Principles" on which he runs Bridgewater Associates into a book, but also a computer program:

Dalio says he’s thinking about open-sourcing the computer code Bridgewater has developed, an artificial intelligence program dubbed PrincipleOS that does everything from summarize meetings to analyze reasons a person might be feeling anger, confusion, or embarrassment when interacting with a colleague. The programs could automate about three-quarters of Bridgewater’s management decisions within the next five years, according to the company.

Ah. Yes. That's from this Bloomberg Markets story about Dalio and his hedge fund, Bridgewater Associates, which is a lifelong experiment in programming Ray Dalio's mind into a computer. First Dalio took his investment process, systematized it, and taught a computer how to do it; now he is taking his interpersonal management style, systematizing it, and teaching a computer how to do it.

There is an odd redundancy here: If the computer is doing the investing, why employ the people? Why program the computer to manage them? Why should the computer care if they are angry or embarrassed, if it is making the investment decisions based on cold logic and objective facts? There are reasonable answers to these questions: Some of Bridgewater's humans are developing new investing theses to add to the algorithms; others are doing client relations. ("Its client services group employs 200 people—one for every 1.75 clients.") Still it is a little odd for a firm to be so focused on developing the personalities of its human employees, and also so focused on automating them out of any meaningful decision-making.

The way I think about Bridgewater is that it is (1) a sort of New-Agey personal-development cult yoked to (2) a massive asset manager. The central thesis of Bridgewater is that thing (1) is important for thing (2): By implementing Dalio's "Principles," and insisting on radical transparency and brutal self-criticism, the firm becomes better at managing assets. (Or, at programming computers to manage assets.) They seem to have persuaded themselves, and their customers, of that thesis:

To most investors, Bridgewater’s mantra of radical truth and radical transparency is the reason for its success. “The culture is key,” says James Grossman, CIO of Pennsylvania Public School Employees’ Retirement System. “It drives them to do things better. Their secret sauce is that they are constantly looking for the truth about how markets work and behave.”

But is it true? Is focusing on radical truth actually good for investment returns? And does Bridgewater bring the sort of radical reflection to that question that it brings to questions like who said what about whom in a meeting? It seems intuitive enough that the path to good investment returns would run through relentless honesty and self-examination, but has anyone analyzed it and tested it empirically? In investing, Bridgewater's managers "start with an investment thesis—say, geopolitical risk leads to a rise in oil prices—and use copious back-testing to ensure the rule is timeless and universal." Is this one timeless and universal?

Since the beginning of 2012, Bridgewater’s Pure Alpha II has posted an annualized return of 2.5 percent, according to a document reviewed by Bloomberg Markets, a far cry from its historic average of 12 percent. It’s down 2.8 percent this year through July. (A smaller Bridgewater hedge fund, Pure Alpha Major Markets, has fared better, as has the company’s long-only product.)

Even a hardcore fan is concerned about performance. “Their returns have been unspectacular recently, and it makes you wonder if this is the beginning of the end,” says Michael Rosen, chief investment officer of Angeles Investment Advisors, who’s steered clients to Bridgewater since the early 2000s and prizes its research reports over others in the industry. “There’s only been one market cycle since the financial crisis, and so if this performance continues in the next cycle, then there may be cause for concern.”

What if market regimes have changed, and now the way to get good investment returns is by laziness, groupthink, fibbing, and trying not to think about things that make you uncomfortable? I mean, I don't know why that would be true. (Indexing?) But wouldn't it be funny if it was? And what would Bridgewater do about it?

Also here is co-Chief Executive Officer David McCormick on another of Bridgewater's great themes, This Is Not for Everyone:

“It’s like watching the Navy SEAL TV ads—the uniforms, the fast roping from the helicopter into the water. It’s very appealing,” says the co-CEO. “But when it’s 3 a.m. and you’re treading water in the middle of the freezing ocean, well, that’s SEAL School.”

When it's 3 a.m. and you're treading water in the middle of the freezing ocean and a computer is asking you to rate the experience and explain why you're feeling upset, well, that's Bridgewater.

Where are the customers' yachts?

Hahahaha Goldman is foreclosing on them:

The story behind the boat begins with a 2014 loan to a prized Goldman client, billionaire Texas oilman William Kallop. It ends with Goldman suing its own client and the U.S. Marshals last month swooping down on a West Palm Beach marina to impound the yacht—which boasts of a movie theater, Jacuzzi and helipad.

Goldman’s nautical trophy is a strange but inevitable outcome of Wall Street’s latest gold rush: lending to wealthy clients, the loans backed by everything from Warhols to wine.

What are banks for? That story starts with "Goldman Sachs Group Inc. owns hundreds of billions of dollars of stocks, bonds and commodities," implying that the yacht fits oddly in that portfolio, and I suppose it does. (Goldman is trying to sell it.) But really the social purpose of a bank is to be a collection of senior claims on weird assets: weird assets, because simple fungible liquid assets can easily get market financing rather than relying on banks, and senior claims, because banks transmute risky economic activities (yacht-buying) into "risk-free" claims (checking accounts), and taking senior claims on the weird stuff is a key way to reduce risk. (The yacht is listed at $39.9 million; the unpaid balance on the loan is about $28 million.)

From a banking perspective, it's weird for a bank to own hundreds of billions of dollars of stocks, bonds and commodities. (It's less weird for Goldman, which is only barely and reluctantly a bank.) We have gotten used to that, to the point that regulators now want banks' balance sheets to be full of liquid assets, and lending against exotic assets is tinged with suspicion. Just outright owning liquid securities is a big part of what modern banks do, but it is not a particularly bank-y activity. Lending against illiquid weird stuff is. The fact that Wall Street is expanding in this business suggests that it's becoming more traditional and boring.

(Oh, disclosure, I used to work at Goldman, though not alas in the yacht-financing department.)

People are worried about non-GAAP accounting.

Some companies think that some non-GAAP measure of their earnings is more useful, or at least more flattering, than the relevant measure of their earnings under U.S. generally accepted accounting principles. So they report that non-GAAP measure, along with (as required by Securities and Exchange Commission rules) the GAAP measure. Some people think that non-GAAP measures are always bad and deceptive, and that GAAP measures are always more useful, so they worry about this practice. But it's not clear why they should care: If the market agrees with them that the GAAP numbers are more useful, then everyone will just look at the GAAP numbers and not the non-GAAP ones. The problem fixes itself.

But in many cases the market does pay attention to the non-GAAP measure, because analysts and investors think that the non-GAAP number is more useful than the GAAP number in representing some economic reality of the company's business. (I mean, you could have a more cynical explanation -- "because analysts and investors are dumb and do whatever management tells them"? -- but I tend to be efficient-markets-ish about these things.) And so, for instance, analysts will estimate earnings based on the non-GAAP measure, and if you want to know whether the company beat or missed expectations, then you'll have to look at its non-GAAP earnings measure to find out.

But here is a weird story about Activision Blizzard Inc., which used to report a non-GAAP earnings measure that backed out revenue deferrals. The market found that measure useful, apparently, and analysts estimated Activision's earnings based on that non-GAAP measure. But the SEC found that measure misleading, and told Activision to stop reporting it. (It could still report non-GAAP earnings measure, but not the one that backed out deferrals.)

This was a problem: The market liked the non-GAAP measure, and analysts estimated it, and now Activision was not allowed to tell the market how it was performing on the measure that the market used. Activision pushed back, saying that the non-GAAP measure "is consistent with the way the company is measured by investment analysts and industry data sources," but the SEC was apparently unmoved.

What the company did next, however, is troubling.

The company now reports GAAP per-share earnings and revenue, and what it calls non-GAAP redefined EPS, which includes the revenue deferrals, as the SEC requires. It then provides a number it describes as reflecting the impact of GAAP deferrals, and it emails reporters to ask them to subtract the impact from the non-GAAP number to give another number that is more comparable with the analyst consensus. 

Ha, well, I mean, what is the problem? Activision wanted to report earnings before X. The market wanted to know about earnings before X. Analysts were estimating earnings before X. The SEC, though, did not want Activision to report earnings before X. So it reported earnings, and it reported X, and it let the market do the (extremely simple!) math. And then, because it didn't trust reporters to do that (extremely simple!) math, it emailed them directly with some math instructions. It is one thing to worry that companies report non-GAAP numbers to mislead investors. It is another thing to worry that companies report GAAP numbers and tell investors how to calculate the non-GAAP numbers themselves. If the investors are doing the math themselves, how are they misled?

Ponzi scheme analytics.

Here you go:

67 individuals were sentenced in 2016 for Ponzi-related violations.

  • The median sentence for schemes valued less than $5,000,000 was 33 months (2.75 years). These individuals on average will spend 39 months in prison for each $1 million dollars stolen.
  • Individuals sentenced in schemes valued over $100,000,000 received a median 168 months (14 years) and will spend on average 21 days in prison for each $1 million dollars stolen.
  • By comparison, the average federal sentence for robbery in 2015 was 117 months with a median loss of $2,898. Therefore, robbers are sentenced to 40,372 months in prison for each $1 million stolen. Similarly, the average 2016 federal sentence for theft was 22 months with a median loss of $137,828. Therefore thieves were sentenced to 159 months per $1 million stolen.

One thing that I tend to think about financial crimes is that the amount of money involved is not an especially good indicator of the moral culpability of the criminal, or of the sentence that she should get. Stealing a million dollars from poor retirees through outright intentional fraud is worse than losing $100 million at a public company through plausible but incorrect accounting choices. I guess the data bears that out.

Countercyclical indicators.

Of all the countercyclical indicators, the most famous is probably the percentage of business-school students who want to work in finance: The higher it is, the more doomed the financial industry is. So good news:

A new survey shows that jobs outside of Wall Street are becoming increasingly enticing to recent MBA graduates. According to data from Training the Street, bulge bracket banks were a top employment choice for just 19 percent of respondents. Not only is that a drop of 7 percent from last year, it’s also the lowest level in the eight-year history of the survey.

Elsewhere: "Banker Bonuses Are Rising, But Traders Aren't as Lucky."

People are worried that people aren't worried enough.

Oh, hey, that was a fun, like, 12 hours of worrying about nuclear war, but we are back to worrying about complacency again. Though now the worry is about complacency about nuclear war:

While stocks in Asia sold off initially, the main market measures in the United States fell less than half a percent on Wednesday and ended the day nearly unchanged, as investors stuck to focusing on buoyant economic fundamentals and ignoring the chaos of American politics.

But beneath the calm there were signs that investors — who have been conditioned since the presidential election in November to embrace risk instead of running from it — are becoming more cautious.

People are worried about unicorns.

A staple of television nature shows is watching the young of some species hatch and immediately race to safety before they are eaten by the predators waiting for them on the beach. This Wall Street Journal article is like that, but for the Enchanted Forest: New baby unicorns are born, and must immediately race to build a massive user base, because Facebook is waiting to eat them. Here is the story of "Houseparty, creator of a hot video app":

Facebook Inc., a dominant force in Silicon Valley, is stalking the company, part of the social network’s aggressive mimicking of smaller rivals. Facebook is being aided by an internal “early bird” warning system that identifies potential threats, according to people familiar with the technology.

This fall, Facebook plans to launch an app similar to Houseparty, internally called Bonfire, say people familiar with the project.

Every time a baby unicorn is born, a bell rings at Facebook, and it sends out a team of people to kill it. 

Elsewhere, here is the story about the lives of two teen tech founders. ("On a typical day, we wake up, get dressed, do email (my morning ritual is to get to inbox zero), answer work requests, eat, work, then go to bed.") They make enterprise software for salespeople, though, so Facebook probably won't eat them.

People are worried about bond market liquidity.

Well, loan market liquidity; here is "Investor Diversity and Liquidity in the Secondary Loan Market" from the New York Fed's Liberty Street Economics blog: 

While our evidence points to a strong relationship between investor diversity and loan liquidity, not all investors increase liquidity. On close inspection, we find that investor types that are believed to follow a buy-and-hold strategy—such as banks and insurance companies—have an adverse effect on loan liquidity. In contrast, investor types that are believed to be active traders—namely, private equity firms and funds—have a positive effect on loan liquidity. 

Our findings confirm the thesis that it takes a difference of opinions to have a horse race. The arrival of other investors, besides banks and insurance companies, in the syndicate loan market has increased loan trading and improved market liquidity and efficiency. 

This basic story -- "it takes a difference of opinions to have a horse race" -- is worth keeping in mind in a lot of markets. Is bond market liquidity troubled because regulation is causing banks to retreat from market making, or is it troubled because perpetual low interest rates and the decline of the single-name CDS market have reduced differences of opinion among bond investors? Are long-only real-money investors not trading with each other on equity dark pools because evil high-frequency traders are front-running their orders, or are they not trading with each other because they all want to buy the same stocks at the same time? Liquidity is not just a technical issue; it also requires some people who want to buy and other people who want to sell.

Things happen.

Labor Department Seeks 18-Month Delay in Fiduciary Rule. SEC Delays Decision on Chinese Buyout of Chicago Stock Exchange. Goldman to use 'personality test' for hiring decisions. Canadian Private-Equity Giant Catalyst Accused of Fraud by Whistleblowers. Chinese crackdown on dealmakers reflects Xi power play. The profitability of banks in a context of negative monetary policy rates: the cases of Sweden and Denmark. "Citigroup has appointed its most high-ranking, white, male managers to lead its internal diversity networks." Should You Kondo Your Kids? At $350 Million, Beverly Hillbillies Mansion Is Most Expensive in U.S. The Story of the DuckTales Theme, History’s Catchiest Single Minute of Music. Thai Navy’s Secret Weapon Against Marauding Monkeys: Vasectomies. People are drinking a lot. "Markham’s most famous show cow, who was recently honoured for her place in the city’s history with a controversial statue erected in the middle of a quiet subdivision, never actually lived in Markham."

If you'd like to get Money Stuff in handy email 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.