Money Stuff

Mortgage Relief and Fear Gauges

Also RINs, last look, an odd investment dispute, hacking and unicorns.

Consumer relief.

Liz Hoffman and Serena Ng went and answered one of the great weird questions of modern finance, which is: How does Goldman Sachs Group Inc. do consumer mortgage relief, given that it doesn't have a consumer mortgage business? You may recall that when Goldman announced its $5.1 billion settlement for crisis-era mortgage misdeeds last year, $1.8 billion of that number was not a cash payment to the government but rather "consumer relief." That usually means modifying mortgages so delinquent borrowers can stay in their homes. But ... what mortgages? Whose homes? "Here's some good news if you have a mortgage with Goldman Sachs," I wrote at the time, because you didn't. 

But now there's an answer! It is, as you'd expect, sort of dumb and baroque and not how you'd design a system from first principles.

Over the past year-and-a-half, the Wall Street giant has become the largest buyer of severely delinquent home loans from mortgage giant Fannie Mae. The firm has acquired nearly two-thirds of $9.6 billion in loans the agency has auctioned, representing unpaid loan balances of $5.7 billion, a Wall Street Journal review of government records shows.

The naive way to think about "consumer relief" is like, a bank lends a person $100, and the person runs into difficulty repaying, and the bank says "well okay we'll call it $80," and the person repays $80, and the bank has lost $20. "Consumer relief" is a relief to the consumers, and a cost to the bank. But that's not quite how the cash flows go here. "Goldman has paid between 50 and 90 cents on the dollar for the loans ... with an eye toward restructuring them by reducing interest rates, lengthening the term of the loan, or forgiving some of the debt outright." But if you buy a loan for 50 cents on the dollar, you can afford to forgive 40 percent of it and still make money. The "loan workout process can take one to two years, and buyers can make between five and 15 cents on the dollar above what they originally paid."

Those are just the normal economics of the loan-modification market. But for Goldman the calculation is a bit different: If it buys a $100 loan for $50, writes off $40 of principal, and resells it for $60, it doesn't just make $10 in profit, it also digs itself out from $40 of consumer-relief liability to the government. That's apparently why it buys such a large share of Fannie Mae's delinquent loans: "Because Goldman is getting credit toward fulfilling the terms of its settlement, it can afford to pay more."

This seems ... fine? (Disclosure: I used to work at Goldman, building derivatives, though not the bad kind.) It is consumer relief, after all. Without the Goldman settlement, someone else would be buying Fannie Mae's delinquent loans and restructuring them for a profit, but the settlement encourages Goldman to do it at less of a profit (that is, with more actual relief to the consumer, or with higher payments to Fannie Mae, or both). The point of the $1.8 billion of consumer relief in the Goldman settlement was presumably to provide $1.8 billion of consumer relief -- not to cost Goldman $1.8 billion.

But it won't cost Goldman $1.8 billion. It seems like it will make money for Goldman. "The bank’s immediate goal is to get credit for the settlement, though the longer-term goal is to make money over time." If you like your bank penalties to be, you know, penalties, then this profitable consumer relief may annoy you.

Also look at those (hypothetical) cash flows. If Fannie Mae buys a loan for 100 cents on the dollar, it becomes severely delinquent, Fannie sells it for 50 cents on the dollar to Goldman, and Goldman writes the loan down to 60 cents and resells it at 60 cents, then the homeowner has gotten 40 cents of relief. Who paid for that relief? The obvious answer is "Fannie Mae," which lost 50 cents on the dollar, of which 40 cents went to the homeowner (for relief) and 10 cents went to Goldman (for profit). Fannie Mae is of course owned, or quasi-owned or whatever, by the government. (Its profits and losses go to Treasury, anyway.) The government punished Goldman by demanding that it provide $1.8 billion of consumer relief. But it seems a little like the government might be paying for that relief, and Goldman might be making money on it. 

People are worried that the fear gauge doesn't reflect fear fearfully enough.

An underrated piece of poetry in finance is that the opposite of fear is complacency, and you should fear complacency. So if you have a thing that measures fear, and it shows fear, then that means there's fear. But if it doesn't show fear, then that means there's complacency, and that means that there are people who fear  the complacency. Down is down, but up is also down. There is no good measurement. Financial news is structurally pessimistic, even though stocks mostly go up.


Taken at face value, a stubbornly low VIX is flashing signs of complacency, even as some investors fear turbulence ahead. That’s not to say the VIX isn’t functioning how it’s supposed to. It’s a function of realized volatility, not necessarily a sentiment gauge, a read on what traders are willing to pay for options ...

A whole financial-markets worldview is packed into that short passage. You might naively think that a low reading on the "fear gauge," as people sometimes call the CBOE Volatility Index, would be a good thing. But connoisseurs know that a low VIX means not that all is well, but that investors are too complacent. We should fear the lack of fear. Also, the VIX is not a function of realized volatility: It's technically a function of implied volatility, a calculation based the prices of stock index options, which are determined by what traders think volatility will be in the near future. But it has long been the woke thing to say that the VIX is really just a measure of realized (historical) volatility -- I have said it myself -- because that happens to be empirically true. The VIX does not do a great job of predicting volatility over the next 30 days, but it does a fantastic job of predicting volatility over the last 30 days. (That is a less useful job.) In a sense that does tell you something about complacency: The market's governing assumption is that tomorrow's volatility will be a lot like yesterday's. And yesterday's was low.

Which is weird, right? Like ... a lot has happened in the last few months, you know? The world feels volatile. The stock market has not been. There does seem to be a story here, about how geopolitical and existential volatility no longer translates into asset prices. But the VIX can't explain that story; it can only mutely reflect it. And just noticing it doesn't tell you much about the future. "Some investors fear turbulence ahead," but some investors feared turbulence behind, and that turbulence totally happened, and the market shrugged.

Anyway if you fear that the fear gauge isn't measuring fear fearfully enough, there are some alternative fear gauges for your consideration. The most aesthetically pleasing of them is probably the CBOE SKEW Index, which measures option skew. (The least aesthetically pleasing of them is probably the value of the dollar, which is just not a satisfying measure of fear even if it happens to be correct.) Theoretically, the VIX measures volatility expectations, but volatility could mean prices going up as well as down. SKEW measures skew expectations, meaning roughly whether the options market thinks a move down is more likely than a move up. And the skew index has been going up. I hope by 2020 we will be talking about the kurtosis index. There will always be some moment of the distribution where you can find fear. Elsewhere: "Markets calm in the face of March madness." 


The way the U.S. energy market works right now is that oil refiners like CVR Energy Inc. have a legal obligation to blend their products with renewable fuels like ethanol. But many refiners can't do that, so instead they are allowed to buy renewable-fuel credits (called RINs, for "renewable identification numbers") from fuel blenders who can. This is a controversial system: Many refiners, including CVR, think that the renewable blending shouldn't be their obligation, and that the Environmental Protection Agency should just require the fuel blenders to do the blending and leave the refiners out of it. If that happened -- if fuel blenders blended in ethanol because they were directly required to, rather than because they could generate RINs to sell at high prices to refiners -- then presumably the RIN market would collapse.

CVR Energy is mostly owned by Carl Icahn. President Donald Trump, and his EPA Administrator Scott Pruitt, seem to listen to Icahn. ("Within days of his victory, the president-elect ended a Trump Tower interview with Scott Pruitt, his future EPA chief, by directing him two blocks uptown to meet with Icahn.") Icahn has pushed the administration to shift the renewables point of obligation from refiners to blenders, which will save CVR Energy a lot of money because it will no longer have to buy RINs. Icahn sees no problem with this. Other people disagree. ("This looks more like what you'd see in a banana republic," one says.) I think it's a hard question: On the one hand, people should be allowed to lobby the government for policies that would help their businesses (particularly if they think those policies are correct), and government officials should listen to the arguments of people who are deeply involved in a particular business (particularly if those arguments are correct). On the other hand there's something unseemly about a president informally putting his refinery-owning buddy in charge of refinery policy.

But never mind that. If Icahn succeeds in shifting the point of obligation, that won't just save CVR Energy money. It will also drive down the price of RINs. And so of course Icahn is "betting on a decline in the market for renewable-fuel credits that he’s urging the president to overhaul." Exactly how he's doing that is somewhat disputed. "I’m not selling ’em, I’m not buying ’em," he told Bloomberg's Zachary Mider and Jennifer Dlouhy: CVR is naturally short RINs because it is obligated to buy them within a year after they sell fuel, and right now it's waiting to buy RINs in the hopes that prices will come down. But CVR also "took the rare step of selling large quantities of the credits on two occasions last year," including in December after Pruitt was appointed. It wasn't just naturally short RINs, it also got shorter as Icahn pushed for changes that would affect RIN prices. Here's Icahn's general counsel Jesse Lynn:

“There is nothing unusual or inappropriate about any RINs trading that may have been conducted by CVR,” Lynn wrote. “Carl Icahn at no point had the ability to influence EPA or White House policy with respect to RINs or the selection of the head of the EPA, but merely an opportunity to express his views, as did many others on both sides of the issue.”

I think we will spend a lot of time over the next few years debating the difference between an "ability to influence" and an "opportunity to express his views." 

Elsewhere in the Trump administration: Who will be the Federal Reserve Board's vice chairman for supervision? And: "Ex-Workers at S.E.C. Nominee’s Firm Urge Him to Denounce Travel Ban." And: "Edge or Liability? White House Ties May Cut Two Ways for Goldman."

Last look.

There are three basic ways to run a financial market:

  1. Dealers quote prices at which they're willing to buy and sell widgets, and if they say "$100 / $101," you can call them up and say "I'll buy at $101" and they have to sell to you at $101.
  2. Dealers quote prices at which they're willing to buy and sell widgets, and if they say "$100 / $101," you can call them up and say "I'd like to buy at $101," but then they can say "oh we actually just sold the last of the $101 widgets, but I have some lovely widgets for $102 if you're interested."
  3. Dealers sit quietly and you call them up and say "how much are widgets?" and they say "for you, chief, $105."

Options 2 and 3 are ... more fun ... for dealers? They are both pretty popular, in financial markets, and in the world. Option 1 is by far the best suited to fast electronic trading: If you're going to build an electronic stock exchange, it really helps if people can see stock quoted at $101 and just push a button to buy it, rather than enter into uncertain negotiations. But, charmingly, you can run an electronic business with option 2: You just give the dealers a "last look" where, for some brief period of time after the customer hits the button, the dealer can say "never mind." It is controversial:

Last look gives a market maker, such as a dealer, time to back out of a trade. Some argue that the practice has merits, allowing market makers to quote better prices. But a firm could unfairly learn a counterparty’s intentions without having to complete the transaction. The Bank of England has said the practice is vulnerable to misuse.

And so various banks and platforms are disclosing what their last look policies are, to head off controversy ("JPMorgan Chase & Co. for example has told clients it swears off some of the most contentious behaviors including hedging during the last look window") or at least to get out in front of it ("Goldman Sachs Group Inc. says it 'may trade prior to or alongside a counterparty’s transaction,' though will try to avoid 'undue market impact.'") 

A ... heist?

I always wonder why there aren't more stories like this in finance. Chinese billionaire Xie Zhikun says that he invested about $1 billion in London private-equity firm XIO Group, and that now it won't return his phone calls. XIO says he never invested the money. There are four possibilities:

  1. He did, and they are lying;
  2. He didn't, and he is lying;
  3. Some boring intermediate contractual-interpretation thing; or
  4. Some brilliant heist where a clever evil third party convinced him he was investing in XIO and then took his money.

There's no sense in me speculating on which one it is. (XIO gave Xie "a portrait of himself painted in tea," which is an odd thing to do for a non-investor, but perhaps they viewed him as a potential investor?) It is just such an elementary dispute, and I'd like to see more like it. Wouldn't it be fun to show up at the Blackstone Group LP and say "hey guys, I'd like my billion dollars back"? Wouldn't it be fun to start a hedge fund, take seed money from Julian Robertson or whoever, and then be like "Julian who? Nope, doesn't ring a bell. This was always my money."

Yes yes sure I know there are legal documents and proofs of wire transfers that can resolve most of these disputes pretty easily, but just go with it. Soon everything will be on the blockchain anyway. (Right?) We are nearing -- perhaps already past -- the close of an era, which has lasted for thousands of years, in which the financial system has been founded essentially on trust, and in which abuses of that trust can be consequential and mysterious. We should take some time to commemorate its passing.

Bank hacking.

The governing paradigm of the internet is advertising. Great juggernauts like Facebook and Google make billions of dollars on advertising, and their model -- that the economic purpose of all online activity is to advertise things to each other -- seems to infect everything. So as far as I can tell most hacking is about advertising. You spend months hacking into a bank, and you steal ... email addresses? So you can send millions of people spam emails touting pump-and-dump stock scams? Because of course if the purpose of the internet is advertising, then the purpose of internet crime should be criminal advertising. 

I tell you what, if I ever hacked into a bank, I would try to take money. I am old-fashioned like that. Anyway here's a story about how "Anonymous" and other hackers are planning to hack into central banks, and about the "critical weaknesses of global financial systems." That sounds bad! And yet there is something satisfying about central-bank hackers who are at least trying to take money. That is a sort of crime that I can understand.

Elsewhere in crime, and advertising: "The Financial Industry Regulatory Authority (FINRA) today issued an Investor Alert warning anyone involved in binary options trading through unregistered non-U.S. companies to be on guard for a one-two punch: losses followed up by potentially fraudulent pitches to help recoup those losses." I mean, once you have the mailing list of people who were ripped off by binary-options scams, you might as well try to run another scam on them, right? You know they're good targets. 

People are worried about unicorns.

"Human drivers were forced to take control of Uber’s self-driving cars about once per mile driven in early March during testing in Arizona." And here is a Bloomberg Businessweek story about Uber Technologies Inc.'s fight with Alphabet Inc.'s Waymo division about whether Uber stole Waymo's self-driving car technology. (At Otto, the self-driving truck startup that is at the center of the controversy, someone "distributed stickers—in OSHA orange—with a tongue-in-cheek slogan: 'Safety third.'")  

Things happen.

Wells Fargo Leaders Reaped Lavish Pay Even as Account Scandal Unfolded. WorldQuant’s Part-Time Virtual Army Do Battle for Jobs. Bob Diamond Starts U.K. Return With Panmure Gordon Takeover. Why Banking Leverage Requirements Are Not Enough. IMF under pressure in Washington over Greek bailout. Global financiers line up to engage with Le Pen. U.S. Cities Battle Each Other for Jobs With $45 Billion in Incentives. Trump May Not Want Immigrants, but Rust Belt Mayors Do. Bank Overdraft-Fee Revenue Bounces Back. Americans Shun Life Insurance, Forcing New Tactics at Prudential. New York Pension Fund Dismisses a Third Brokerage Amid Bribery Scandal. Fast-food robots. "Thinking like an economist." Martin Shkreli keeps buying up the personal domain names of journalists who write about him. Operation London Bridge. Zion CurtainStump the Bookseller. Moscow petting zoo sues ad company over 'erotic' raccoon shoot.

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:
    Tobin Harshaw at

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