Money Stuff: Banks Are Managing Their Stress

Money Stuff

Not so stressful


The severely adverse scenario is characterized by a severe global recession accompanied by a period of heightened stress in commercial real estate and corporate debt markets.

The U.S. unemployment rate climbs to a peak of 10 percent in the third quarter of 2021.

That is from the results of the bank stress tests that the U.S. Federal Reserve released on Thursday. The way the stress tests work is that every year the Fed prepares a "severely adverse scenario," a hypothetical economic catastrophe that would make life difficult for the big banks, and then asks the banks to model how they'd do in that catastrophe.[1] If they'd have enough capital even taking into account the catastrophe, then they have enough capital and that's good; if the catastrophe would bring them below minimum capital requirements then that's bad and they need to preserve or raise capital now. It is a good sensible way to make sure that, even in the good times, the banks are preparing for the bad times.

The problem is that the U.S. unemployment rate peaked at 14.7% in April, and was 13.3% in May. The stress test's catastrophic scenario—which the Fed announced on Feb. 6, a week or so before the coronavirus pandemic started crashing U.S. markets, oops—was significantly less catastrophic than reality.

Sort of? You can leaf through the summary of the severely adverse scenario (on pages 5 to 10 of the stress test report) and compare it to current economic reality, and it's a bit of a mixed bag. In the Fed's imagination, the interest rates on three-month Treasury bills and 10-year Treasury notes fall to "near zero" and 0.75%, respectively, early in the crisis; in reality those rates are about 0.13% and 0.65% now. In the stress test the Dow Jones Industrial Average loses half its value and the VIX, the volatility index, peaks at 70; in reality the Dow fell about 37% from peak to trough and recovered pretty sharply, but the VIX peaked at 82.69. In the stress test, yields on BBB-rated corporate bonds rise to nearly 7% before slowly falling back to about 4%; in reality they spiked to about 5.5% in March before quickly falling back under 3%.[2]

In any case, you can compare the banks' stressed results—the lowest their capital ratios would hypothetically be during the Fed's fictional scenario—with their actual results, the capital ratios they had at the end of the first quarter in March. The actual results are much better. JPMorgan Chase & Co. had a hypothetical stressed common equity tier 1 capital of 9.8% and hypothetical supplementary leverage ratio of 5.1%; its actual ratios at the end of March were 11.5% and 6.0%. Goldman Sachs Group Inc. was hypothetically 6.9% and 3.5%, actually 13.3% and 5.9%. Morgan Stanley: 11.1% and 4.5% versus 15.2% and 6.2%.[3] In reality, the banks are better capitalized in a worse crisis than the Fed expected them to be in a less severe hypothetical crisis.

Why? The biggest answer is surely that it's still early: The hypothetical stress-test crisis is not just about stocks dropping and unemployment increasing, it's about those bad conditions lasting for several years. The bankruptcies and mortgage defaults don't all happen at once; it takes time for the bad conditions to work their way through the system to the point that they hurt the banks. Defaults could still pick up; things could still get a lot worse for banks, and probably will.

But another important answer is that, when a crisis actually happens, people do something about it. They react, and try to make it better. In the case of the coronavirus crisis, the Fed and the U.S. government tried to mitigate the effect of a real disaster on economic and financial conditions. Unemployment is really high, but some of the consequences are mitigated by stimulus payments and increased unemployment benefits. Asset prices fell sharply, but then rose sharply as the Fed backstopped markets. Financing markets seized up, and then the Fed fixed them.

The banks themselves also acted to make things better, at least for themselves. One thing that often happens in a financial crisis is that banks' trading desks make a killing trading for clients in turbulent markets, which helps to make up for some of the money they lose on bad loans. And in fact many banks had blowout first quarters in their trading divisions: Clients wanted to trade and would pay a lot for liquidity, and banks took their money.

In a hypothetical stress test, you can't really account for any of this. If you're a bank, and the Fed asks you to model how you'd handle a huge financial crisis, you can't really write down "I would simply make a ton of money trading derivatives." It is too cute, too optimistic. But in reality, lots of banks just went and did that.

Similarly, you obviously can't write down "I would simply rely on the Fed to backstop asset prices and liquidity." That is super cheating. Much of the purpose of the stress tests is to make it so the Fed doesn't have to bail out the banking system; the point is to demonstrate that the banks can survive a financial crisis on their own without government support. But in reality, having a functioning financial system is better than not having that, so the Fed did intervene; keeping people in their homes is better than foreclosing on them, so the government supported incomes. So the banks are doing much better than you might expect with 13.3% unemployment.

So it is likely that the Fed's stress test is both not harsh enough, in its economic scenario, and too harsh, in its assumption about how that scenario will affect banks.

But the fact that the stress test imagines an economic crisis that is much nicer than reality is still a little embarrassing, and the Fed can't really say "everything is fine even in the terrible downside case of 10% unemployment, the banks are doing great." So it also produced some new stress-test results (well, not quite a full stress test but a "sensitivity analysis") assuming various scenarios about the recovery from the Covid crisis ("a rapid V-shaped recovery," "a slower, more U-shaped recovery," and "a W-shaped double dip recession"). The banks are much less well capitalized in those scenarios than they are either (1) now or (2) in the original stress tests, though mostly still okay, and the Fed is asking the banks to reconsider stress and capital based on current reality. Also stop share buybacks:

In aggregate, loan losses for the 34 banks ranged from $560 billion to $700 billion in the sensitivity analysis and aggregate capital ratios declined from 12.0 percent in the fourth quarter of 2019 to between 9.5 percent and 7.7 percent under the hypothetical downside scenarios. Under the U- and W-shaped scenarios, most firms remain well capitalized but several would approach minimum capital levels. The sensitivity analysis does not incorporate the potential effects of government stimulus payments and expanded unemployment insurance.

In light of these results, the Board took several actions following its stress tests to ensure large banks remain resilient despite the economic uncertainty from the coronavirus event. For the third quarter of this year, the Board is requiring large banks to preserve capital by suspending share repurchases, capping dividend payments, and allowing dividends according to a formula based on recent income. The Board is also requiring banks to re-evaluate their longer-term capital plans.

All large banks will be required to resubmit and update their capital plans later this year to reflect current stresses, which will help firms re-assess their capital needs and maintain strong capital planning practices during this period of uncertainty. The Board will conduct additional analysis each quarter to determine if adjustments to this response are appropriate.

That seems right, really. You can't rely on a stress test that is less stressful than reality.

Bond buying

Speaking of the Fed supporting asset prices, the Federal Reserve's program for buying corporate bonds is quietly one of the wildest stories of 2020. It was announced in March, when things were very bad generally and people were worried that the investment-grade corporate bond market would seize up and it would be impossible for big real companies to get financing. So the Fed said: Don't worry, we'll take care of it, in a very broad and general way. The Fed was willing to lend money directly to companies, to buy their bonds in the secondary market, to buy exchange-traded funds that owned bonds, really whatever it took.

As far as I can tell, this announcement immediately and completely solved the problem. Everyone said "oh, okay, the Fed is backstopping the investment-grade market, we're good now," and they went back to enthusiastically buying bonds. Rates fell to record lows and issuance came at a record pace, even as the Fed bought zero bonds. The Fed accomplished its entire goal just by announcing the program.

You might think that would be the end of the story. This is exactly how it's supposed to go. "If you've got a bazooka, and people know you've got it, you may not have to take it out," Ben Bernanke memorably said; that is the point of announcing huge open-ended programs like this. You don't want to take it out; the Fed is not especially in the business of buying or owning corporate bonds, and it would be much easier for everyone if it didn't. If saying "we'll buy as many corporate bonds as we need to" completely fixed the corporate bond market, as it did, then the number of bonds you need to buy is zero, and the number of bonds you want to buy is zero, so the number of bonds you should buy is zero.

But then the Fed went and bought bonds. First it bought some bond ETFs, but more recently it started buying some actual bonds. Specifically it bought $207 million of individual bonds on its first day of buying, June 16, according to disclosures that the Fed made on Sunday.[4] U.S. investment-grade corporate bond volume is about $26 billion a day, so the Fed is about 0.8% of the market.

Also it paid about $221 million for that $207 million face value of bonds, buying every single one of them at a premium: Interest rates on corporate bonds right now are lower than they were when any of these bonds were issued, so the Fed had to pay more than 100 cents on the dollar for each of them. Philip Morris International Inc., the cigarette company, has a 1.125% bond due in 2023 that is currently trading at about a 0.55% yield. The Fed bought $4.5 million face amount of that bond (roughly a third of the day's volume on June 16), for about $4.57 million. It is hard to see what economic problem the Fed is solving by making sure that the interest rate on Philip Morris's bonds stays below 1%.

But it is even weirder than that, as my Bloomberg Opinion colleague Brian Chappatta explained recently. The Fed has a program to buy bonds in the secondary market, and another program to buy bond indexes, meaning mainly exchange-traded funds. It has bought more ETFs than bonds, and it started earlier on the ETFs. Why did it do that? Well, one reason is that if you are trying to push down corporate interest rates generally, you might want to buy a broad ETF rather than picking specific bonds. But there was also a legal reason. As I wrote in May:

Part of the reason that the Fed is not yet buying individual bonds is that, as a footnote in the Investment Strategy points out, "corporate bonds cannot be purchased in the Facility until Eligible Issuers have completed a certification process." The term sheet for the program lists the certification requirements, which are annoying. Bond issuers have to certify that they have significant U.S. operations and investment-grade-ish ratings, but also that they have "not received specific support pursuant to the CARES Act or any subsequent federal legislation" and that they "satisfy the conflicts of interest requirements of section 4019 of the CARES Act." The Fed's specific bond purchases have political strings attached; the Fed can only buy bonds of companies that certify they are politically acceptable under the pandemic bailout programs. Or it can buy ETFs, which are just generic piles of bonds and cut out the political questions.

Well! The Fed is cleverer than I am, and came up with a way to buy individual bonds and also cut out the political questions. As Chappatta explains, quoting BMO Capital Markets analysts, the CARES Act (which authorized the Fed's corporate bond purchases) "requires purchases of only companies with 'significant operations in and a majority of its employees based in the United States' unless the purchases are 'securities based on an index or that are based on a diversified pool of securities.'" So it could buy diversified ETFs ("based on an index" etc.) without certification, or it could buy individual bonds of companies that certified they were based in the U.S. and not getting other government support.

But it went a third way: It built its own index of, essentially, all the bonds, and then announced that it would buy bonds to try to match that index. It is starting slow, I guess—the index has almost 800 names; the Fed started with about 60 positions—but I suppose that after a few weeks of buying it will own chunks of a broadly diversified list of bonds, some of which have significant U.S. operations and some of which don't. But: why?

40% of everything is securities fraud

Everything, I like to say, is securities fraud: If a public company does a bad thing, or if a bad thing happens to it, shareholders can sue it for fraud. It didn't disclose the bad thing as soon as it happened, or if it did it didn't adequately warn about the bad thing in advance, so shareholders were deceived, and when it did disclose the bad thing the stock went down, so the shareholders sue.

This is an anecdotal, over-broad claim; I have not actually studied it rigorously. I have not, you know, counted up every time a bad thing happened to a public company and checked if it was sued for fraud. But these guys have!

One often hears that "whenever there is a stock drop, there will be a lawsuit." This of course is an overstatement. But how much of an overstatement is it?  In this post, we answer that question in some detail.  We identify all net-of-market stock drops of specified sizes in 2017 for common shares listed on the New York Stock Exchange and the NASDAQ, and we determine what fraction of those stock drops led to securities class actions. We then look at the distribution of net-of-market stock drops that led to lawsuits. We choose 2017 because it is long enough ago for nearly all potential lawsuits to have materialized, and recent enough to capture current practices among plaintiffs' counsel in deciding to file lawsuits.

That is from Michael Klausner, Sam Blake Curry and Jason Hegland of Stanford Law School. For their largest category of companies—those with market capitalizations above $2.8 billion—they find that stock drops of 20% or more lead to lawsuits 40% of the time; stock drops of 10% to 15% lead to lawsuits 13% of the time. (Also, almost 70% of securities fraud lawsuits filed against big companies involve stock drops of zero to 10%.) Most bad news doesn't lead to lawsuits:

These findings suggest that the decision to file a lawsuit is a lot more than a knee-jerk response to a stock drop. Many large stock drops do not result in lawsuits. On the other hand, do 40% of stock drops larger than 20% in large companies reflect underlying securities fraud, as opposed to simply bad news? That seems unlikely.

Also, I suppose that every time a company's stock drops by 10% or more, a bad thing has happened, but the reverse isn't true: Sometimes a bad thing will happen and the company's stock won't drop when it's disclosed. Sometimes this is because companies are keenly aware that they'll be sued if their stock drops, so, when they have bad news, they try to combine it with good news in the same press release, so there won't be a stock drop to sue over. 

Elsewhere in disclosing bad things, here's a Stanford paper by David Larcker, Bradford Lynch, Brian Tayan and Daniel Taylor about "The Spread of Covid-19 Disclosure":

Coronavirus references were relatively scarce in the early months of the year. By January 31, only 0.7 percent of companies had referenced the virus in 10-Ks, 10-Qs or 8-Ks; by February 29, only 22 percent. As the severity of the pandemic became more apparent and shelter-in-place programs instituted across the United States and Europe, disclosure increased exponentially: 41 percent by March 15, 64 percent March 31, 73 percent by April 15, and 86 percent by April 30. By the end of our measurement period, virtually every company (99.6 percent) made some level of disclosure about the pandemic. ...

When the data is analyzed by month, we see that the emphasis of disclosure changed as companies came to different realizations about how the virus would impact their business. In the early months, supply-chain impacts were the most common issue disclosed; by May, disclaimers to forward-looking statements became the most commonly issued disclosure, as companies realized that the full effects of the pandemic could not be easily quantified. Also, we see a significant increase in disclosure on cash positions, as market fears about liquidity and solvency increased. Cash was the seventh most frequently disclosed issue in February; by May, it rose to second (see Exhibit 4). These trends are indicative of a pandemic that was originally seen as contained to China but later one that spread to materially impact the sales, operations, and financial position of most U.S. businesses.

You know how people sometimes do epidemiology by tracking Google searches? I wonder if you could do it by tracking corporate 8-Ks. If every bad thing is potentially securities fraud, then every bad thing is going to find its way into securities filings.

Things happen

"One trader at a major firm bought an album by Mumford & Sons, the English folk-rock band known for its twangy banjo picking, to have something to talk about with a White manager." Chesapeake Pushed Into Bankruptcy by Plunging Energy Prices. Chesapeake's Collapse Is Latest in Long Line of Shale Disasters. Wirecard Scandal Puts Spotlight on Auditor Ernst & Young. Germany to overhaul accounting regulation after Wirecard collapse. Credit Suisse Review of Funds Prompted by SoftBank's Multiple Roles. Hedge Funds Are Rushing to Get Out of Bearish U.S. Stock Bets. Private Equity's Trillion-Dollar Piggy Bank Holds Little for Struggling Companies. Too-big-to-fail banks mostly a thing of the past, say regulators. China Weighs Letting Big Banks Broker Deals on Wall Street Threat. Crypto lender Genesis reports uptick in stablecoin borrow rates amid 'yield farming' craze. Nikola Sells $5,000 Reservations for a Truck With No Prototype. Casinos Consider Cashless Gambling to Fight Coronavirus. The sheikh with so many homes, he forgets where they are: One in Spain had no visitors in 17 years, another in France has 'supplies flown in from Gulf daily' and a Berkshire mansion piped with Evian. Only 5% of Americans in Survey Say Things Are Going Very Well.

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!

[1] There is also a "baseline scenario," which is less bad than the "severely adverse" one. For 2020, it is so embarrassingly gentle that the Fed barely discusses it.

[2] You can get more comparisons on page 8 of the Fed's "Assessment of Bank Capital during the Recent Coronavirus Event," which charts the stress-test scenario versus reality and several new hypothetical scenarios.

[3] JPMorgan is on page 61 of the stress test results; its actual numbers are on page 3 here. Goldman is on page 58 of the stress test results; its actual numbers are on page 71 here. Morgan Stanley is on page 64; its actual numbers are on page 4 here. In each case I am comparing the "stressed capital ratio," "minimum" from the upper left table on the stress-test results page to the actual CET1 and SLR.

[4] Here is the homepage for those disclosures. What you want is the June 28, 2020 "Transaction-specific Disclosures XLSX" for the Secondary Market Corporate Credit Facility; in that spreadsheet, you want the tab "Position Summary-Bond." (The PDF has some summary and discussion.)


Like Money Stuff? | Get unlimited access to, where you'll find trusted, data-based journalism in 120 countries around the world and expert analysis from exclusive daily newsletters.

Before it's here, it's on the Bloomberg Terminal. Find out more about how the Terminal delivers information and analysis that financial professionals can't find anywhere else. Learn more.



Popular posts from this blog

Mulan DID NOT make $250 million and the future of film releases

Stars Unite for Table Reading of Fast Times At Ridgemont High

How to Dance Across Medium with Fantastic Writers