Money Stuff

Overbooking and Cross-Selling

Also fake stock tips, crumbling quotes, bond indexing, Yale, Harvard, duration and convexity.

I blame the index funds.

Yesterday United Airlines provided a nice demonstration of the proposition that capitalism is built on a foundation of violence, when it summoned agents of the state to beat up a customer who insisted that United provide the service he had paid for. "Come and see the violence inherent in the system," he might have yelled, had the police not knocked him out. ("He fell," commented the police.)

United then went on to demonstrate that if you are a major airline in 2017, you don't have to be very good at public relations, putting out a series of blasé statements whose main message was "whatever, we are an airline, you will come crawling back." It was interesting to see people on Twitter talk about boycotting United over this incident, as though that was a possible course of action. Consider the revealed preferences: The man at the center of the incident, who was violently attacked for sitting in the seat that he had paid for, tried to run back onto the plane. He's not boycotting United! He just wanted to get home. 

We talk a lot around here about the theory that increasingly concentrated cross-ownership of the airline industry by institutional investors has reduced competition among airlines, and I suppose you could read this incident as proof that United is so insulated from competitive pressures that it can afford to beat up its customers without losing any market share. But really this story seems more like the result of competition -- but competition solely on price, not on service. If airlines compete solely on price, some passengers will get beaten up. "Investors seem impressed by the sadistic commitment to cost control," comments Matt Klein. "By auctioning off overbooked seats, economist James Heins estimates that $100 billion has been saved by the airline industry and its customers in the 30-plus years since the practice was introduced." Ryanair would introduce Beating Class if it could save money.

Anyway blah blah blah United should have run a fair auction and only removed people voluntarily at agreed-on rates of compensation, says the Economist, but of course from United's perspective that's not true. Why pay more to rescind a passenger's ticket, when you could just call in the cops? United could have thrown its surplus passengers off the plane mid-flight and you'd still be back on United tomorrow if its fares were $5 cheaper than Delta's. Helaine Olen points out that "the decline of customer service is part of the political anger out there." "The Corporation Does Not Always Have To Win," argues Albert Burneko, optimistically. And here is Izabella Kaminska on "Flight seats as fractional reserve banking."

Wells Fargo.

The report of Wells Fargo & Co.'s board's investigation into its fake-accounts scandal doesn't really add that much to our understanding of the scandal. The regulatory lawsuits and settlements told the story of the fake accounts, which is interesting; the internal investigation mainly tells the story of executives failing to spot the fake accounts, which is relatively dull. There's not much action involved in not noticing something. You just sit there, don't notice it, and collect your bonus. (Former Chief Executive Officer John Stumpf, and former head of retail banking Carrie Tolstedt, will have to give back another $75 million of bonuses based on this investigation.) Though Wells Fargo executives did sometimes go to heroic lengths to not notice the scandal:

The failure to frame the issue properly extended to senior management’s view that firing 1% of the Community Bank workforce every year for sales integrity violations was acceptable. For example, in November 2013, in the wake of the first Los Angeles Times article on sales practice issues, John Stumpf asked for data on the number of terminations associated with sales integrity violations. When the data showed that 1% of employees had been terminated for such violations, Stumpf, Tolstedt and other Community Bank leaders received the figure positively, believing it proved that a vast majority of individuals were behaving appropriately. ... Raphaelson, when presented with data showing ethics-related terminations of a similar magnitude in 2013, wrote that it was “mind boggling to me it’s so low – I think it shows our [employees] are significantly more ethical than the general population (no data whatsoever to back that up, just impressionistic comment!).”

If the report has one main explanation for how the scandal happened, it's that Wells Fargo viewed its branches as essentially retail stores that sold financial products, and was comfortable with the high levels of turnover and sales-focused culture that are normal in retail businesses:

As a result of the retail focus, many Wells Fargo branch employees were relatively inexperienced, and many witnesses stated that together with the high-pressure environment this contributed to employees not doing things “the right way.” In addition, witnesses said that inexperienced bankers frequently were promoted based on sales success and became inexperienced managers who understood that success was measured by sales performance. More generally, witnesses consistently stated that promotions at all levels in the Community Bank were regularly — though not exclusively or in every region — based on sales performance.

One reaction to "inexperienced managers who understood that success was measured by sales performance" is -- well, how else would you measure it? You can imagine the banker at your local community bank who measures her success by the smiles of her customers and the satisfaction of knowing that she has helped them achieve their financial dreams, but how do you put that in a spreadsheet and compare it across thousands of branches in different regions? A giant company with thousands of employees is going to quantify its performance metrics. Performance metrics that contribute directly to the bottom line (selling stuff) will get more emphasis than ones that don't (helping customers achieve financial dreams). Locally, banks might be in the advice business, but at a large enough scale they can only really be in the sales business. And people expecting advice will be confused.

Anyway the details of the measurement sound just horrifying:

Regional bank-wide sales-reporting processes included frequent rankings against individual, branch and regional sales goals, and against one another. Witnesses frequently cited daily and monthly “Motivator” reports as a source of pressure. These reports contained monthly, quarterly and year-to-date sales goals, and highlighted sales rankings down to the retail bank district level. Circulation of the reports — and their focus on sales-based rankings — ramped up pressure on managers, such that some “lived and died by” the Motivator results.

If a giant corporation calls anything a "Motivator," you know it is going to be terrible.

Fake stock tips.

Look, I certainly don't want to defend the passel of goofballs who were fined by the Securities and Exchange Commission yesterday for running "various alleged stock promotion schemes that left investors with the impression they were reading independent, unbiased analyses on investing websites while writers were being secretly compensated for touting company stocks." Don't do that, that is bad. At the same time, though, if you read an article by "Wonderful Wizard" on touting microcap biopharmaceutical company Galena Biopharma, Inc., and Wonderful Wizard mentioned “I am not receiving compensation for [this article] (other than from Seeking Alpha),” and you bought Galena stock ... why did you do that? (Actually Galena's stock more than doubled within six months, so good for you!)

I mean, yes, it turns out that in fact Galena had paid The DreamTeam Group, LLC $25,000 to tout its stock, and that DreamTeam had gone out and solicited Wonderful Wizard to write "an extra good article" on Galena, and that it had paid Wonderful Wizard $300 to write that Seeking Alpha article. (DreamTeam: impressive margins!) So you were misled about Wonderful Wizard's incentives in touting a biopharma penny stock. He was doing it for dirty dirty money, not for the pure love of biopharma penny stocks. Your trust in Wonderful Wizard was misplaced! But my question is more like ... why did you trust Wonderful Wizard in the first place? What made you so confident that Wonderful Wizard was an unbiased source of useful advice? Along with the fines, the SEC released an Investor Alert telling investors to watch out because some seemingly independent anonymous penny-stock recommendations may in fact be paid-for anonymous penny-stock recommendations, but that seems like sort of the wrong thing to watch out for. What are you doing relying on any anonymous penny-stock recommendations? 

Crumbling quotes.

Here's a white paper from Allison Bishop at IEX Group LLC about the "crumbling quote signal" that it uses to protect discretionary pegged orders. It begins with a description of how IEX built its speed bump to protect orders, and then evolved from there:

And yet, even with this speed bump in place, we began to see an increase in adverse selection experienced by resting orders on IEX. Trades were often executing at prices just before IEX observed a change in the NBBO that would have been favorable to the pegged resting order. But how could this happen? The speed bump was still doing its job - it was preventing someone from quickly reacting to an NBBO change and picking off a resting order on IEX, but it was not preventing someone from anticipating an NBBO change. We hypothesized that traders were building probabilistic models to predict price changes far enough in advance to circumvent the protection of the speed bump. Naturally their predictions would not be perfectly accurate, but they could likely gain an edge by predicting some price changes say 1 or 2 ms before the actual change solidified. 

A lot of what people in market-structure debates refer to as "front running" is actually this: Computerized traders figure out that "they could likely gain an edge by predicting some price changes say 1 or 2 ms" in advance. They build probabilistic models to predict price changes, and act on those models, and get them right more often than wrong, but in any case always faster than people without those models. And the slower people complain about being "front-run." It is a maddening abuse of terms.

That's not IEX's fault, of course. The rest of the paper is about how IEX can use some of the same predictive algorithms to thwart the fast traders' predictive algorithms. ("We can fight math with math!") Which is fine: People can build predictive algorithms to take advantage of price changes, and other people can build other predictive algorithms to avoid being taken advantage of by price changes, and so forth, in saecula saeculorum. (Some people think that stock exchanges, as opposed to investment firms and brokers, shouldn't be in the business of building those algorithms, but that ship has mostly sailed.) But the point is that this is not a moral fight, not a story of innocent investors being protected from evil front-runners. It's just that when prices change, someone has to get there first, and people will fight over that tiny edge. 

Bonds and indexing.

Here is a white paper from Pacific Investment Management Co. about why you shouldn't invest in bond index funds. Pimco rejects the Sharpe equality (that the average returns to active management, before fees, have to equal the average returns to the market portfolio, that is, the index) in the case of bonds, for a number of reasons, including that many bond investors are non-economic actors (and therefore provide active investors an opportunity to make above-market returns), and that you have to trade bonds: 

Sharpe's arithmetic implicitly assumes passive investors buy and hold and don't trade securities. In reality, most bond indexes are rebalanced monthly, requiring both active and passive investors to trade, if only because bonds mature, new bonds are issued, and index inclusion and exclusion rules create movement in and out of the index.

This is sort of a general story in the active versus passive debates: You might want "passive" investing, meaning something like investing indiscriminately in all economic activity in proportion to its size, without making any judgments, but that is an unattainable ideal. There will always be some set of criteria involved in making an index, and at the edge those criteria will always be a bit arbitrary. Also bonds mature.

Yale and indexing.

Another general story in the active versus passive debates is: If you can outperform consistently over a long period of time, net of fees, by investing with active managers, you should probably do that, and if not not. Maybe you can't, but somebody probably can. Yale University, for instance, probably can, and its annual endowment report is proud of it:

“What Buffett, Gladwell and other fee bashers miss is that the important metric is net returns, not gross fees,’’ the report said. “Weak or negative returns would result in low or no performance-related fees, but would be a terrible outcome for the university.’’

Yep! I feel like a key trick on both sides of the debate tends to be about convincing you to ignore observable data. Critics of Wall Street fees will sometimes use creative math about fees to distract from the fact that using active managers and private equity has made more money for a given pension, over a long period, than indexing would have. Or active managers will talk about an upcoming rotation that will embarrass index strategies, to distract from the fact that their active strategies have underperformed the market for a long period. There's always some subjective or speculative distraction to point to. But sure yeah if you have found a way to consistently outperform the market, net of fees, you should just do that.

Harvard and greed.

Here is Andrew Ross Sorkin on Duff McDonald's forthcoming book, "The Golden Passport," about Harvard Business School:

“The Harvard Business School became (and remains) so intoxicated with its own importance that it blithely assumed away one of the most important questions it could ask, which was whether the capitalist system it was uniquely positioned to help improve was designed properly for the long term,” Mr. McDonald writes in the book, to be released in two weeks.

It would be funny if Harvard Business School engaged in deep examination of whether capitalism might actually be bad, but really that seems like more of a Yale thing.

Blockchain blockchain blockchain blockchain.

"Fidelity Labs, the innovation arm of asset manager" Fidelity Investments Inc., is joining some blockchain initiative. As financial blockchain initiatives go this one seems fairly tame, but I mention it to you because you may remember Fidelity Labs from such previous hits as StockCity, "a Sims-like virtual reality game that turns your stock portfolio into a city," and Stocks Nearby, a tool that lets people buy stocks (and options!) on the go "if they have an idea inspired by a nearby business." They have a lot of fun over there at Fidelity Labs, is the point here. Now I guess they will find a way to let customers walk through a virtual city and buy all the stocks they see, but on the blockchain.

People are worried about duration.

They stopped for a while, but now they're worried again:

“Duration has never been this long in my career,” says Jeffrey Gundlach, the chief executive of Los Angeles-based asset manager DoubleLine Capital. “With rates near the lowest levels ever and duration at literally the highest level ever, it is the worst possible set-up versus history. You are [taking] more risk and getting less reward.”

People are worried about convexity.

I mean, negative convexity has long been the trendy worry in certain sectors of the bond market, but now it is coming to junk bonds:

BB-rated junk bonds are now characterized by so-called negative convexity, suggesting that junk bonds become more rate-sensitive as rates rise and less rate-sensitive as rates fall, according to Bank of America Merrill Lynch. That’s something that had never happened until last fall.

The upside-down condition is happening largely because many high-yield bonds now have provisions built into them to allow the issuer to redeem them early at a set price.

People are worried about the third derivative.

Nah, but give it time.

Things happen.

The SEC as Financial Stability Regulator. Yellen Says Fed's Focus Has Shifted to Holding Growth Gains. Activist Investor Jana Partners Takes Nearly 9% Stake in Whole Foods. Hedge Fund Seeks Spinoff of BHP Billiton’s U.S. Oil Business. China Is Playing a $9 Trillion Game of Chicken With Savers. "Male stock analysts tend to write more favorably about public companies headed by men than about companies led by women." Was North Korea Behind the Hacking of a Bank in India? US retreat from global financial system lets China in. Monetary policy's rising FX impact in the era of ultra-low rates. In the $75 Billion Videogame Industry, Hiring People Is a Last Resort. Coachella Isn’t All That Chill About Its Brand. "From my British point of view, what I want for my sons and daughters alike is for them to be decent human beings, economically independent, and happy-ish." A New, More Rigorous Study Confirms: The More You Use Facebook, the Worse You Feel. You Can Now Apply to Work at McDonald’s Via Snapchat. South Dakota man cited $190 for not putting leashes on pythons. Unsuspecting Woman Brings Therapy Dog to Local Furry Convention. Trump adviser Anthony Scaramucci: Jared Kushner is 'like Alexander Hamilton.'

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.