Liquidity and information Some financial markets are mostly about making informed bets. If you buy Tesla Inc. stock it is usually because you think that Tesla is good and will go up; if you sell Tesla stock short it is usually because you think Tesla is bad and will go down. Other financial markets are mostly about, ehhh, gotta put your money somewhere. If you buy a certificate of deposit from a bank, you are not betting that that bank's credit is under-appreciated and hoping to make money when it rallies. You're just putting your money somewhere to earn a little interest without thinking too much about it. Many interesting stories in finance are about one type of market turning into the other. On the one hand, a well-established story of the 2008 global financial crisis is that certain sorts of assets—AAA-rated mortgage-backed securities, asset-backed commercial paper, short-term loans to investment banks—were considered "information-insensitive," or "safe," where "safe" here means not that people carefully evaluated their credit quality and deemed it to be better than was implied by the interest rate they paid, but rather that people were happy to chuck money into them without thinking about their credit. And then one day people noticed that those assets were in fact not safe, that the risk of default was not zero. They started asking questions and discriminating and fleeing to quality, and things that were once boring and safe and stable became risky and volatile, with bad consequences for the financial system as a whole. On the other hand, one big story in modern finance is that people are increasingly just chucking their money into equity index funds without doing any analysis of the companies involved or choosing between good stocks and bad ones. People worry that stock markets, the classic information-sensitive markets, are becoming less sensitive to information, and that this will have bad effects for the allocation of capital. Here are a blog post and related article from Karthik Balakrishnan, Aytekin Ertan and Yun Lee about one such transition, titled "(When) Does Transparency Reduce Liquidity?" They look at trading in mortgage-backed securities before and after the European Central Bank introduced a regulation that "required banks to provide regular and detailed information regarding the individual loans comprising the MBSs that they had lodged with the ECB as collateral"; the regulation requires public disclosure and thus "substantially expands investors' information set about the fundamentals of the loans underlying ABSs." It used to be that people didn't know much about the credit quality of those MBS; now they know more. Or rather, now they can know more. If they want to know more they have to look at the voluminous boring loan-level information, and do the work to analyze it. It turns out that some people—sophisticated hedge funds, say—want to do that, and other people—pension funds, say—don't. The people who don't want to do that analysis just wanted to chuck their money somewhere with a bit of yield. Back when no one could do the work, that was a good trade. But now that some people can do the work, it's a bad trade for the people who don't. And so they cut back on trading MBS. The authors "find that the liquidity of treated MBSs declines by 14% post-regulation." They explain: As for those deterred by the shift of incentives, they include institutions such as pension funds and central banks, which, while sophisticated in comparison with retail investors, are generally thought to be at the unsophisticated end of the institutional investment spectrum. But increased transparency has changed the nature of the MBS market, making it less suitable for their purposes than had been the case. Previously, they would use the MBS market to park money for three to six months, earning a better return than that available from U.S. Treasury securities. It was, notwithstanding the negative publicity surrounding the role of MBSs in the 2008 financial crisis, a safe market and a way to make some return on institutions' cash. But in the wake of the LLD regime, it has been transformed into something closer to the equity market, with professional money managers, such as hedge funds, actively seeking to profit from it, hence their incentive to process and package newly-available MBS information for their own use. This is not like the financial-crisis story in which the safety of the investments was thrown into question by higher-than-expected defaults. This stuff is still "a safe market." But now different securities have different levels of safety, and trade at different prices, and might decline in price for reasons relating to their relative quality. If you just wanted an undifferentiated mass in which to park your money you can't get that here anymore, so you stop trading it. There is another nice nuance. The MBS involved here are tranched: Investors in super-senior tranches get paid off first and have very little risk of loss, investors in senior tranches get paid off next and are pretty safe, investors in junior tranches get paid off last and have significant risk of loss (and are compensated by significant upside). The authors find that the ECB regulations did not affect liquidity for the super-senior tranches, and that they actually improved liquidity for the junior tranches; the reduction in liquidity was all in the mid-level senior tranches. The authors write: In relation to the safest MBS assets, there is no more incentive now than before for sophisticated investors to put money and effort into processing the newly available data, because the private information created will offer little benefit. … At the other end of the risk spectrum, where the most risky MBSs are to be found, there has long been profit-seeking market behavior of the type now spreading across the spectrum in the wake of the LLD regime, in which professionals have always sought as much information as possible in order to avoid loss. Here, the availability of more information as a result of increased transparency will, we believe, lead to an increase rather than a decrease in liquidity. The super-senior tranches are substantively information-insensitive—knowing more about them really won't change your valuation—so adding information doesn't hurt. The junior tranches are obviously information-sensitive—no one ever just put money into them without thinking about it—so adding more information actually helps; everyone was trying to be smart anyway, so making them smarter increases trading. In the middle, the senior tranches used to be treated as information-insensitive because (1) they were pretty safe and (2) there wasn't much information. But adding information ruins that fiction and makes the market less attractive. There is no special reason to believe this is a bad thing; maybe pension funds shouldn't park their money in mortgage-backed securities without analyzing them. There's no special reason to believe it's a good thing either; it's not like they were losing a lot of money on those securities due to defaults (recently). "The welfare implication of our findings is not obvious," write the authors. Perhaps it's good for there to be places where people can just park money without worrying, and it's good for there to be places where people can compete to thoughtfully allocate capital, but it doesn't matter all that much for the world if any particular borderline market works one way or the other. It's just an interesting story to follow. In general you'd expect information to increase over time, both because regulation tends to increase transparency and also because, you know, computers and whatever make it easier to provide the information. So you'd expect markets to become more information-sensitive over time, and you might worry that that would be off-putting to ordinary investors. And you might expect a reaction in the form of products to make markets less information-sensitive, products like index funds that give investors tools to ignore all the information that they are bombarded with. Not everything is securities fraud Man, good for them: Mattel Inc.'s chief financial officer is leaving, and the company is restating some past earnings after completing an investigation into accounting issues raised in a whistleblower letter. The investigation found shortcomings in the toy maker's accounting and reporting procedures but concluded that the actions didn't amount to fraud. Shares of the maker of Barbie dolls and Hot Wheels cars rose more than 20% in post-market trading as the resolution, coupled with strong third-quarter earnings, removes a roadblock to the company's plans to refinance debt due next year. The whistleblower letter, disclosed in August, abruptly nixed plans to raise debt at the last minute. The investigation found that Mattel understated its net loss by $109 million in the third quarter of 2017 due to an error calculating its tax valuation allowance, and then understated fourth-quarter results that year by a similar amount. The error wasn't reported to the CEO at the time or the audit committee. We talked a little about this back in August when they pulled the bond deal for unspecified whistle-blower-related problems. At the time it was an absolutely shocking development because Mattel had already sold the bonds; the deal had priced and was about to close when it was pulled. But Mattel decided that the whistle-blower letter concerned accounting problems that could be material to bondholders, and so they pulled the deal, investigated, found that the accounting problems were real, fixed them, and will now, I suppose, do another bond deal with accurate information. It cannot have been easy to make the call to pull the bond deal, but it was the correct call, and really sort of an admirable example of a company doing the right thing. I mean the right thing would have been to get the accounting right in the first place, but that aside! Good work. They should get a reward, like now they should be allowed to commit one securities fraud without being sued for it, I don't know. High-frequency trading arms race I laughed at this story that Brandon Richardson, founder of three-person Chicago trading firm Emergent Trading, told the Wall Street Journal: After Emergent entered the Eurodollar market, it discovered there was an advantage to quoting prices for more contracts than any other market maker. The benefit: If another trader bought or sold Eurodollar futures, executing against quotes from different market-making firms, the market maker with the biggest quote would get notified 10 millionths to 20 millionths of a second before its next-biggest competitor, Mr. Richardson said. That is enough time for a quick trader to infer that the price of Eurodollar futures is moving up or down, and to use that knowledge to profit before anyone else. … Emergent tweaked its algorithms to ensure it always had the largest quote—but its rival did the same. So if Emergent posted a quote for 2,000 contracts, the other firm would post a quote for 2,010 contracts a split-second later, for instance. The two firms raced until they hit a maximum size limit, then dropped down and started over, multiple times a second, Mr. Richardson said. Why did they drop down? Why not just hit the maximum and stay there? What is the benefit to you, when your rival is quoting the max contracts, to quote fewer? And if you do drop and quote fewer, what is the benefit to your rival of dropping? Why wouldn't she just stay at the max, since you are now below the max so she is ahead of you? Anyway the point is that they somehow got into this dumb infinite loop, and it basically overwhelmed the Chicago Mercantile Exchanges's messaging infrastructure and made trading Eurodollar futures more difficult, and now the CME has new rules about not doing this anymore. I guess the point is that if you structure a market to reward some easily gameable behavior, the behavior will be gamed, and the game might end up being dumb. Repo riches Oh man, here is a story about how high-rolling former proprietary traders at big banks who have struck out on their own to run powerful secretive hedge funds profited from market dislocations last month to place bold bets and make a killing in volatile high-risk trading. "Exiles from Wall Street's old proprietary trading desks were among those who cashed in on September's spiking repo rates, and now they're lining up for the next big move," it begins. It cites Ardea Investment Management as an example of this bold new breed of ex-prop traders whose willingness to take risks and be greedy when others were fearful paid off in a big way. How big? Unburdened by the type of regulations that hamstring the big banks, Ardea's traders pounced, snapping up Eurodollar options whose value rose as money-market volatility increased. The move helped earn the Ardea Real Outcome Fund a total return of 0.3% in September as Treasuries due in one-to-three years lost 0.12% and bills maturing within three months gained just 0.18%. Ahahahahahaha yes, yes, yes, a total return of 0.3%! That's 3.6% annualized! The thing is that if repo rates are 2%, and one day they go up to a shocking and unheard-of 10%, and you are able to fully capture that value, you make an extra 800 basis points annualized, but for that day, which means, like, 2 basis points of actual money. You can do better if you get cheap leverage, as some big banks apparently did, but for most hedge funds the repo is the leverage, so, um? I guess you can do better with options, as these guys did. But the basic point is that wild once-in-a-lifetime volatility in overnight secured interest rates is only so volatile. Unsolicited bond ratings I confess that I did not know this: After the financial crisis, Washington focused on the credit-ratings firms and the conflict of interest that made them "essential cogs in the wheel of financial destruction," according to the federal government's report on the crisis. But the government didn't eliminate the conflict, where the firms are paid by the entities whose bonds they rate. Instead, the Securities and Exchange Commission decided that enabling ratings firms to publish unsolicited ratings on securities they weren't hired to analyze would be the best solution. The agency crafted a rule to give them access to deal data to publish such ratings. I suppose that it is a little embarrassing for me, as a financial journalist, that I was not aware that the SEC's fix for bond-rating conflicts of interest was to encourage unsolicited ratings. In my defense, no one else seems to have noticed either. It's not like there … are … unsolicited ratings: The SEC declined to answer questions about the program, but said in response to a public records request that, after a "thorough search" of its records, it "did not locate or identify" any examples of unsolicited ratings published by ratings firms under the program. Moody's Corp., S&P Global Inc., Fitch Ratings Inc., and Kroll Bond Rating Agency Inc. all said their respective firms haven't produced any unsolicited ratings under the SEC's rule. DBRS Inc. and Morningstar Inc., which recently merged, said they didn't produce any unsolicited ratings in 2019 and aren't expecting to do so in the future. Also it is kind of an insane fix? It "only applies to 'structured' debt" like mortgage-backed securities and collateralized loan obligations. Issuers of those securities tend to be repeat players, big banks that regularly package the bonds, and the classic concern is that they will ratings-shop by picking the agencies that give them the most generous ratings, and that the agencies will respond to incentives by lowering their standards. But the unsolicited-ratings approach opens up new possibilities! I mean the new possibility is blackmail. You go around slapping low unsolicited ratings on every deal where you don't get hired, and then when the issuers complain you say "hey well perhaps we would have a different opinion if we were working with you more closely, hint hint hint," and then you get hired to provide paid solicited ratings and they're much more generous. Part of me doesn't understand why that hasn't happened a little bit. Why not launch an unsolicited-ratings business in one area that you want to get into and go around doing a little bit of blackmailing until you gain some market share? I suppose one answer is that it would be a little too obvious and you'd get in trouble, though "ratings agencies are mostly trying to get their ratings right and it's not worth it to do the hard serious work of rating a bond unless you're paid for it" is also a pretty good answer. Stock market predictions It seems pretty obvious to me that if some big money manager says that some presidential candidate he doesn't like will be bad for the stock market, what he means is just "I don't like this presidential candidate" and not "I have actual insight into what the stock market will do contingent on this candidate being elected." Here Bloomberg's Sarah Ponczek and Vildana Hajric point out that the historical performance of those predictions is pretty bad: "Strategists predicting the impact of a presidential election are worse than the pollsters," Matt Maley, equity strategist at Miller Tabak + Co., said by phone. "It's one of the best contrary indicators out there." They round up all the bad predictions that the market would crash if Trump, and before him Obama, were elected. I do not expect this to change anything really; the basic problem is that (1) rich money managers have political opinions, (2) they tend to express those opinions in the language of market predictions, and (3) market predictions by rich money managers tend to be treated as news. WePlay Ah: We Co., the beleaguered parent of WeWork, has been quietly building an electronic-gaming business. After branching into co-living with WeLive, fitness with Rise By We and education with WeGrow, the money-losing startup is seeking a Play By We trademark, according to an application published by the U.K.'s Intellectual Property Office last week. It has also hired a handful of staffers, according to people familiar with the matter, who asked not to be identified because it's private. … According to its trademark application, Play By We may offer entertainment services such as "providing online competitive, professional video games," including conducting contests, games, tournaments and exhibitions. The application also suggests the venture could provide facilities for video-gaming events, conferences, meetings and trade exhibitions, and, naturally, shared office space. Yeah you'd want to do that quietly, given that WeWork is getting rid of all its silly side businesses and this seems like it could be a pretty silly side business? But my real concern is that I don't understand why some We businesses are WeVerb and others are Verb By We. Presumably it was a branding nuance that made sense to someone back when every possible verb was a potential We side business, but it seems weird now. Elsewhere in, uh, amusing self-awareness: Bill Ackman sees a risk of SoftBank having to write off all of its investment in WeWork, he told the Robin Hood investor conference in New York on Tuesday. "I think WeWork has a pretty high probability of being a zero for the equity, as well as for the debt," the billionaire hedge fund manager said. "As someone who has put good money after bad, I think this looks like putting good money after bad, and SoftBank should have walked away." Things happen Can PG&E Survive the California Wildfires? PG&E Trade Punishes Hedge Funds as California Burns. Aramco IPO to Come at 'Right Time' and When Crown Prince Decides. Deutsche Bank Suffers Another Quarterly Loss as Restructuring Bites. The U.S. Treasury Department Is Exploring a 50-Year Bond for the First Time. Alan Howard Opts for Trading Over Management to Boost Returns. Facebook's Libra faces eurozone backlash. Nikkei loses $29m in alleged money transfer fraud. Luger's is bad now. Popular cat names. OK Boomer. If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Or you can subscribe to Money Stuff and other great Bloomberg newsletters here. Thanks! | 
Post a Comment