Money Stuff: US Debt Rates Itself

Does US government debt even have a credit rating? I mean, it does. It has three, or probably more, but three big ones. Since 2011, the US government has had an AA+ rating at Standard & Poor’s; since August 2023, it has had an AA+ rating at Fitch. Until last week, it had an Aaa rating at Moody’s Ratings, but on Friday that too was cut to Aa1. AAA (or Aaa) is the best rating; AA+ (or Aa1) is the second-best rating.

But what is a credit rating? In some sense, there are companies — “ratings agencies” or “nationally recognized statistical ratings organizations” — that employ people who think about the creditworthiness of companies and governments and asset-backed securities, and who come up with ratings that tell investors how creditworthy their bonds are. If you are an investor, you might trust the ratings agencies to tell you what bonds to buy. If you want to buy risky bonds, you can buy bonds with “junk” ratings (or more politely “high yield,” BB+/Ba1 or lower). If you want to buy safe bonds, you can buy bonds with “investment grade” ratings (BBB-/Baa3 or higher). If you want to buy only the very safest bonds, you can stick to bonds with AAA/Aaa ratings.

But the actual importance of ratings agencies seems to be more formalized than this. It’s not that you want to buy BBB rated bonds or whatever; it’s that you have to. Historically, certain types of investors were legally required to buy bonds with certain ratings; most notably, money market funds were limited to buying mostly securities with the top short-term ratings. After the 2008 financial crisis, US regulators scrubbed most of those requirements from the rules, worried that the ratings agencies were too conflicted for their judgments to have legal status.

But they still have quasi-legal status. Lots of investors have mandates requiring them to buy, for instance, only investment-grade bonds: An investment-grade bond mutual fund or index might be required to contain only bonds with investment-grade ratings, as determined by ratings agencies. Individual investors will buy investment-grade bond funds because they want their money to be safe, and those bond funds will then be required — by their fund documents and disclosures to investors — to go out and buy bonds with investment-grade ratings, even if the funds’ managers happen to think that one particular BB rated bond is really safe. Or a contract will say “you can only post collateral that is rated AA- or better,” so investors will need to buy AA- rated bonds to post as collateral. Not everything is like this — if you’re an individual investor, or a bond-fund manager with a universal mandate, you can buy whatever you want — but quite a lot of things are.

And so when a ratings agency downgrades a bond or company, that has consequences, not because investors assume that the agency’s decision is correct but because it is binding. If you run an investment-grade bond mutual fund and a bond that is rated only by Moody’s goes from Baa3 to Ba1, you probably won’t buy any more of that bond. [1] You won’t go read the Moody’s downgrade report and evaluate their evidence and come to a different conclusion and say “nope, still investment grade.” You are constrained to buy investment-grade bonds, as determined by the ratings agencies, so their decisions directly affect yours.

And so when a company is downgraded from investment-grade to junk, its financing costs go up. It can’t borrow from specialized investment-grade investors any more, so it will have to borrow from a different group of specialized high-yield investors. As their name implies, they charge more.

Similarly, in asset-backed securities, there seems to be a lot of demand for AAA/Aaa ratings. Banks structure securities to have big AAA-rated tranches, so that they can sell those tranches to investors who want only the safest securities; you have to pay more interest to the people buying the AA+ tranches than the ones buying the AAA tranches.

Hypothetically you could imagine US government debt working like this. You could imagine that the US government might want an AAA rating as much as, and for the same reasons as, an asset-backed securities issuer. You could imagine that a lot of financial plumbing would rely on AAA ratings, that a lot of funds would only be allowed to hold AAA-rated bonds, or that a lot of collateral arrangements might only accept AAA-rated bonds as collateral. And all of that plumbing was merrily making use of the world’s largest supply of risk-free bonds, US Treasuries, and then Moody’s downgraded them on Friday. Perhaps financial plumbing would be thrown into disarray, demand for Treasuries would collapse — because they are no longer risk-free for plumbing purposes — and the US’s borrowing costs would go way up.

You really could imagine that, because when Fitch downgraded the US in 2023, people did. (Many of these sorts of quasi-regulatory requirements are of the form “the middle rating of the three big agencies is controlling,” so the Fitch downgrade, coming second, was in some sense more decisive than S&P’s first move or Moody’s third.) But, because US Treasuries really are such a large supply of risk-free bonds, and because they are obligations of the US government, it would be a little weird if things really worked this way, if a Fitch (or Moody’s) downgrade could really prevent a lot of investors from holding Treasuries. We talked about it at the time, and I looked at various rules, and as far as I could tell it didn’t matter. I wrote:

The collateral rules for the Federal Reserve’s discount window require investment-grade ratings (BBB- or better) for US dollar corporate bonds, and AAA ratings for foreign-currency corporate bonds. But US Treasuries are just US Treasuries; they are always eligible and there are no ratings criteria. The Bank of England also accepts US government bonds categorically as Level A collateral for its liquidity insurance schemes. The US Commodity Futures Trading Commission’s rules for derivatives margin also say that US government securities are eligible collateral with no ratings criteria.

Or for bank capital, the rule is that a bank “must assign a zero percent risk weight to an exposure to the U.S. government, its central bank, or a U.S. government agency.” For insurance capital, the National Association of Insurance Commissioners sets standards for risk-based capital based in part on ratings; but there is “no [risk-based capital] requirement for bonds guaranteed by the full faith and credit of the United States … because it is assumed that there is no default risk associated with U.S. Government issued securities.” Nothing about ratings.

Later I quoted a Goldman Sachs Group Inc. research note doing a similar but more comprehensive survey and finding that, “because Treasury securities are such an important asset class, most investment mandates and regulatory regimes refer to them specifically, rather than AAA-rated government debt.”

That is: The credit rating of US government debt, for most purposes for which people would use a credit rating, is “US government debt.” The credit ratings assigned by S&P or Fitch or Moody’s are not, as far as I can tell, binding on any investors; the thing that is binding is the particular legal status of Treasuries as US sovereign debt. As I wrote in 2023:

If I were a ratings agency my rating on US government debt would not be a semi-arbitrary collection of As. My rating would be “this is US government debt.” For good and for ill, people mostly know what that means!

Now, that is too glib, and Moody’s job is not only to issue binding rulings about what debt is and isn’t safe, but also to justify its quasi-regulatory status by doing correct and reasoned credit analysis. So they have to come up with some rating for the US government, and I suppose there is some pressure to react to fiscal profligacy or other embarrassments by saying “actually it’s Aa1 now.” The Moody’s downgrade is newsworthy because it reflects market sentiment — people are worried about the budget deficit, etc. — more than because it affects it.

Of course if you are a completely unconstrained investor — an individual, or an unconstrained bond manager, or I guess a foreign sovereign — and you owned a lot of Treasuries because they had the very top rating at Moody’s, and now Moody’s has come along and said “actually they’re a bit riskier,” you are free to sell. If your credit analysis is heavily influenced by Moody’s, and if you want only the very safest bonds that Moody’s rates, you can dump Treasuries and probably buy a lot of AAA tranches of structured credit stuff. I am sure someone will, but that really is an unusual investing niche.

Anyway here are my Bloomberg colleagues John Authers, Joe Weisenthal and Tracy Alloway on the decision. Here is Moody’s press release, which “reflects the increase over more than a decade in government debt and interest payment ratios to levels that are significantly higher than similarly rated sovereigns,” but notes that “the US retains exceptional credit strengths such as the size, resilience and dynamism of its economy and the role of the US dollar as global reserve currency.” You probably didn’t need Moody’s to tell you that.

If you are a customer of a large reputable investment bank, there are three main ways for you to lose all your money trading foreign exchange derivatives [2] :

1. You make a simple bet and lose for simple reasons. You make a straightforward bet on some market move, and the market moves the other way. You bet a million dollars that the dollar will appreciate against the Swiss franc, say, [3]  and instead the dollar depreciates and you lose your million dollars.
2. You make a complicated bet and lose for complicated reasons. The bank comes to you to pitch some ferociously complicated derivatives bet, and you make it, and you lose the bet for ferociously complicated reasons. “We will bet you a million dollars that at most 17 of the following 31 events occur,” the bank will say, but the title of the presentation will be like “Super Dollar Go Up Bet,” and the bank will show you some payoff charts and disclosures, and broadly speaking the charts will show you that you are mostly betting that the dollar will appreciate against the franc, though there are a few weird edge cases where the dollar will appreciate and you will lose the bet anyway. You think the dollar will go up, so you make the bet, and the dollar does appreciate against the franc, but you fall into a weird edge case and lose the bet and feel aggrieved. “Why did you sell me that thing,” you will ask the bank. “You called it a Super Dollar Go Up Bet, and the dollar went up, but I fell into a weird edge case and lost all my money.”
3. You make a complicated bet and lose for simple reasons. [4] The bank comes to you to propose some ferociously complicated derivatives bet, and you make it, and you lose the bet for simple intuitive reasons. The bank shows you the “Super Dollar Go Up Bet,” and you take it, and the dollar goes down and you lose all your money. The bet had various edge cases, but they didn’t matter, because the main thing you were betting on didn’t happen.

This is neither legal nor investing advice but most of the time, if you fall into the first category, you suck it up. Sometimes customers make bets that don’t work out, and that’s their problem.

Much of the time, if you fall into the second category — particularly if you are a high-net-worth individual or a small corporate or institutional client — you at least complain, perhaps you sue, and quite possibly you get your money back. The bank, in some loose approximate sense, “tricked” you. Sure sure sure sure sure it showed you accurate disclosures, but the example payoff graphs and the general tenor of the disclosures were, at least in hindsight, misleading. You thought you were betting that the dollar would go up, and the dollar did go up, and you didn’t get the bet you thought you were making, even if the fine print of the trade made the risks clear.

The second category is common enough, and makes people angry enough, that people have learned to recognize the pattern. “FX derivatives mis-selling,” it is called. “I bought a complicated derivative from a bank and then I lost all my money,” you will say, and everyone will be like “oh yes, those banks are so tricky, you should get your money back.”

And this bleeds into the third category. “I bought a complicated Dollar Go Up derivative from a bank and then I lost all my money,” you say, and everyone is like “oh yes, so tricky, you lost all your money even though the dollar went up, you should get it back,” and you are like “actually the dollar did go down though,” and they’re like “ehh you should get your money back anyway, so tricky.”

We talked once about a series of complicated equity derivatives trades that Goldman Sachs Group Inc. sold to the Libyan Investment Authority; the trades all went to zero and Libya sued. (Goldman ended up winning.) Two stylized facts about these trades are:

1. They were complicated, the LIA team that bought them might not have been especially sophisticated, and Goldman arguably did not explain all of the nuances in readily understandable ways.
2. They were big bets that a bunch of international bank stocks would go up, made in January 2008.

International bank stocks did not go up, in 2008. The subtle nuances of these trades were not what blew up the LIA; what blew them up is that they bet on bank stocks going up, and there was a banking crisis. (One bet was a complicated leveraged derivatives trade on Citigroup Inc., which had a 100% loss, but just buying regular Citigroup stock in January 2008 would have gotten you an 84% loss by November.)

This is an important risk for banks: If they sell you a weird derivative, and it works out in your favor, they will lose money. [5]  If they sell you a weird derivative, and it works out poorly for you for weird reasons, they will make money, but they will get complaints and refund requests and bad publicity and lawsuits. If they sell you a weird derivative, and it works out poorly for you for normal reasons — because the market moved against the bet you thought you were making — they will also get some complaints. “That derivative was weird,” you will say, and they will say “sure but the weirdness did not affect your outcome,” and you will say “yes but it was weird.”

Anyway I am not 100% sure from this story which category these trades were in:

UBS is in talks to compensate some clients for losses after they were sold complex foreign-exchange derivatives that wiped out much of their investments when U.S. President Donald Trump's tariffs sparked wild moves in currencies, sources familiar with the matter said. ...

Among those asking for compensation from UBS for the losses incurred are wealthier retail customers who argue they were sold complex products that they did not understand, the second source said, adding that they were only suitable for sophisticated investors.

In one example, a UBS client has lost more than 50% of an investment made in February into an FX derivative designed to bet on the direction of the dollar versus the Swiss francs, according to a document detailing the performance of the investment dated May 9 and seen by Reuters. …

UBS sold clients 'conditional target redemption forwards', a February prospectus of the product seen by Reuters and sources said.

These are exotic derivative FX products that allow clients to buy dollars and sell Swiss francs at a more favourable rate than the prevailing market rate, but can cause big losses if the rate moves past certain levels over a set period of time, according to the prospectus. Losses accumulate and can exceed an initial investment.

In the example of the product that caused more than 50% losses, the investor agreed to buy dollars and sell francs in $300,000 chunks if the market rate moved above or below certain thresholds, according to a prospectus for the investment dated February 10 and seen by Reuters.

But below a 'kick-in level' defined in the terms of the investment, the customer is forced to buy dollars at an exchange rate that locks in a loss, the term sheet showed.

My rough understanding of target redemption forwards is that they are a trade in which the customer wants to buy a foreign currency — here, buying US dollars with Swiss francs — over time, and the bank offers the customer a good deal with a small catch. “Instead of selling you $1 million for CHF 831,000 in a month,” the bank will say, “we will sell you $1 million for CHF 825,000” or whatever. “Except that if the dollar is trading below 0.825 francs, we will sell you two million dollars at that price.” That is, if the dollar goes up or stays flattish, the clients buy the amount of dollars that they wanted, at a fixed price a bit below today’s price. If the dollar goes down, though, the clients buy more dollars than they wanted, at that same fixed price. They get a discount in the good case, but they pay for it in the bad case. [6]

So there is a catch: If the dollar goes down, you buy extra dollars at a price that looked good when you started the trade (when the dollar was above the strike price) but looks bad now (when the dollar is below the strike price). Bad for the clients! On the other hand, it was “an FX derivative designed to bet on the direction of the dollar versus the Swiss francs,” and they bet … wrong? The simple thing they were doing was buying dollars, and buying dollars is fundamentally a bet that the dollar will go up. The dollar went down. The clients made quite complicated bets that the dollar would go up, but the dollar did pretty simply go down.

We talked on Thursday about the hack of Coinbase Global Inc., in which hackers got access to some customers’ personal data, but not to their crypto. Given the long history of hackers getting into crypto exchanges and stealing all the crypto, I counted this as a win for Coinbase, writing:

If you are really dedicated to hacking Coinbase, you can bribe a customer service employee, get some customer email addresses and use them for scams and social engineering projects. But you can’t bribe the person with the keys to give you Coinbase’s crypto. 

This was perhaps too optimistic. I am used to hacks of regular banks, where (1) it’s very hard for hackers to steal the money but (2) the consolation prize is personal data that hackers can use to do scams. But in crypto things are different. The Coinbase data included not just email addresses but home addresses, and if you give criminals a list of addresses of crypto customers, their first thought might not be “scam” or “social engineering” but rather “kidnapping.” Bloomberg’s Emily Nicolle reports:

Even before Coinbase Global Inc. disclosed that hackers had stolen the home addresses and account balances of its customers, Jethro Pijlman was seeing an uptick in interest from concerned clients with large crypto holdings who were looking for bodyguards and other forms of protection. …

“We’ve had more inquiries, more long-term clients, and more proactive requests from crypto investors who don’t want to be caught off guard,” said Pijlman, a managing director at Infinite Risks International. “They’re realizing that intelligent security measures are part of the cost of doing business at this level.”

People with crypto wealth face unique physical risks because public blockchain networks like Bitcoin and Ethereum allow tokens to be transferred instantly and anonymously. This means that if an individual is coerced into giving up the access credentials to their holdings, their assets can vanish within seconds with little chance of recovery. Conversely in traditional financial services, bank accounts can be frozen or seized by law enforcement, allowing more chances to get back lost money. ...

“Crypto traders are acutely concerned about their privacy during data leaks,” said Ronghui Gu, co-founder of blockchain security firm CertiK and associate professor of computer science at Columbia University. “Cryptocurrency can be transferred with just a private key, and is extremely difficult to recover,” he added. “This makes crypto traders prime targets for criminals.”

And here is a Wall Street Journal story along similar lines:

The brazen attack [on the family of a crypto executive in Paris] was the latest in a wave of violent abductions around the world, including several in the U.S., targeting crypto executives and their families. Victims have been pistol whipped, abducted, and—in two cases—had fingers severed.

The criminals’ goal: millions of dollars in ransom in cryptocurrency. 

The assaults are often called “wrench attacks” because they rely on simple tools for inflicting pain to coerce victims, rather than sophisticated tools for hacking them. 

Hacking has long been the primary risk for the crypto rich. But to thwart hackers, savvy cryptocurrency investors have increasingly taken their digital wallets offline in favor of physical devices, making remote theft more difficult. Real-world crypto crime bypasses those safeguards.

“A lot of people are getting to the hide-your-gold-under-the-mattress level of security,” said Jameson Lopp, the co-founder of bitcoin security company Casa. “But if you are a high-profile person…that’s when you have to worry about the physical attack.” 

I think sometimes about the term structure of crypto futures. Buying a Bitcoin for delivery in seven months costs about $4,000 (3.8%) more than buying a Bitcoin today. Some of that is time value of money — I could get interest on my dollars for the seven months, which is probably less true of the Bitcoin — but some of it is what I have half-jokingly referred to as storage costs. If I buy a Bitcoin future, I don’t have to put the Bitcoin anywhere for seven months; if I buy an actual Bitcoin, I do have to store it. It’s not like storing crude oil, in that I don’t need a big storage tank; the Bitcoin is electronic and storing it just means remembering the password. But it turns out that storing your Bitcoins is very expensive. You have to remember the password and pay bodyguards.

Similarly, I am perpetually baffled by the fact that MicroStrategy Inc. (1) is a publicly traded pot of Bitcoin and (2) trades at roughly twice the value of its Bitcoins. But presumably you won’t get kidnapped for your shares of MicroStrategy; perhaps that is worth paying a premium for.

I guess a stylized story would be:

1. Everyone wants to work in finance, because it is prestigious and you can make a lot of money.
2. The central skill of finance is identifying a good trade, an arbitrage, a mistake, an asset that the market undervalues, a place where you can buy a dollar for 50 cents.
3. You get enough finance people thinking about things for a while and eventually some of them will notice: Hey, everyone wants to get into finance. That suggests that finance is overvalued, and not-finance might be undervalued.
4. Therefore, the arbitrage opportunity is to get into not-finance.
5. In some crude sense “not-finance” could mean “building large language models,” sure sure sure, but it might be a better bet to get into an industry that is as not-finance as possible.
6. You know what it is!
7. It’s pest control!

And this shift is well underway; I should rename this newsletter Pest Control Stuff. We talk all the time around here about how the most desirable and prestigious jobs for the top students at top MBA programs, these days, are in pest control, and about how there is a whole apparatus for reframing pest control jobs as fancy financial jobs — private equity, search funds, AI rollups — so that ambitious young people can get both the financial rewards of going into lucrative unglamorous businesses and the psychic rewards of going into lucrative glamorous businesses.

Still, that apparatus isn’t perfect, and some people get rich by doing lucrative unglamorous businesses unglamorously. Here’s a Wall Street Journal article about “the stealthy wealthy”:

Finance and Silicon Valley offer glamorous, high-profile paths that can lead to significant wealth. But a vast universe of traditional routes focused on providing goods and services has become increasingly central to the accumulation of significant, if less obvious, wealth in the U.S. 

“We call it the stealthy wealthy,” said Owen Zidar, a Princeton University economist who has studied the group with University of Chicago economist Eric Zwick. …

Behind a paycheck, the largest source of income for the 1% highest earners in the U.S. isn’t being a partner at an investment bank or launching a one-in-a-million tech startup. It is owning a medium-size regional business. Many of them are distinctly boring and extremely lucrative, like auto dealerships, beverage distributors, grocery stores, dental practices and law firms, according to Zidar and Zwick. …

In his 37-year career, [Rockefeller Capital Management Chief Executive Officer Greg] Fleming has seen significant wealth created by carwashes, residential-lighting companies and companies that distribute parts for industrial appliances in need of repair. 

I assume that there are enormous regulatory and cultural impediments to doing a private equity rollup of auto dealerships, but I would be interested to hear from someone who’s trying. Elsewhere here’s a profile of Brad Jacobs, who seems to have fun [7] :

In 1989, Jacobs had another lightbulb moment. He was scanning research reports and came across one about environmental services. That’s when he realized he could bring consolidation to unglamorous industries that typically go unnoticed.

He started with garbage collectors, buying up mom-and-pop businesses under a company called United Waste. One of his go-to strategies is using technology to improve operations—for example, by rerouting garbage trucks to cover pickups more efficiently. Eight years later, he sold United Waste for around $2 billion. 

Right I mean, in the early 1990s, if you got into the garbage collection business, you probably did not have to compete with the most sophisticated private equity firms, [8]  and there was probably money to be made.

SEC to Look at Rules for Investing in Private Funds. Jamie Dimon’s Would-Be Successors Audition for the Top Job at JPMorgan. Lutnick Divests Cantor Fitzgerald to His Children, 26North. New CLO Managers Are Pouring Into the Market. 23andMe Sells Gene-Testing Business to DNA Drug Maker Regeneron. Nvidia seeks to build its business beyond Big Tech. International Paper Is on a Quest to Build a Better Cardboard Box. Will Anyone Take the Factory Jobs Trump Wants to Bring Back to America? Stricter US border controls prompt business travel rethink. “Are we going to have Russian oil tycoons owning Frisbee teams and world cups in Saudi Arabia for ultimate Frisbee?” “Alcohol used to occupy a much bigger share of people’s entertainment and joy.” 

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!