A bank nobody had heard of called Silicon Valley Bank (SVB) crashed recently. Since then, all hell has been breaking loose. If you look in the right nooks and crannies of the press the story is actually being covered reasonably well. But the usual oversimplifications and lack of appropriate context and background dominate the coverage provided by the media that most people actually follow. This gives me the opportunity to fill in some blanks and provide some the appropriate context and background.
Starting at the beginning, stocks are supposed to be risky, and bonds are supposed to be safe. The logic is simple. Many organs of the press report where the Dow's closes at the end of each day. That number has been bouncing all over the place. One day it's up. The next day it's down. The volatility of the Dow, and other indexes like the S&P 500 and the NASDAQ, reinforces the idea that it is easy to both make and lose money when investing in stocks.
Bonds, on the other hand are supposed to be safe as houses. You know what you're getting from the get-go, and you almost never lose money on the deal. Instead, you almost always get what you were promised when you signed up. And, in fact bonds rarely default (fail to pay investors what was promised on time and in full). But it is still possible to lose money on bonds even if they don't default. To understand why it is necessary to dive into how bonds work.
Bonds are characterized by an amount (often called the "face value" or the "denomination"), a "term" or duration, and an interest rate. So, for the purposes of explanation, let's talk about a $10,000 bond (the amount), whose duration is 10 years, and that has an interest rate of 1%. These numbers have been picked to make the math simple, but bonds of this exact type have actually been issued in the recent past.
The way things are supposed to go is that an investor hands over $10,000 in exchange for the bond. Some bonds pay interest quarterly. Some pay semiannually. Either way, our investor receives $100 (1% of $10,000) in the first year, another $100 in the second year, and so on. At the end of the tenth year the $10,000 is returned. In all, the investor puts in $10,000 and gets back $11,000.
And, as I said, bonds rarely default. Even "Junk" bonds, bonds issued by institutions who have problems of one kind or another, usually pay off. So, what could possibly go wrong? It turns out that if the investor holds the bond to maturity, 10 years in our example, usually nothing. The interest gets paid on time and in full. The investor gets his initial investment back, on time and in full.
But what happens if events similar to what has happened in the past several months happen? It turns out that the "Fed", the Federal Reserve Bank of the United States, jacked up interest rates. I am going to skip the details of how the Fed manipulates interest rates because it's complicated. Just trust me, it can, and it does.
The Fed has twin responsibilities. It is supposed to manipulate things so that inflation stays low, and the economy grows steadily at a relatively even rate. The standard tool for doing this is to manipulate interest rates.
If the Fed moves interest rates lower, it is supposed to make it easier for businesses to grow and expand. That lowers unemployment and increases economic growth, perhaps increasing inflation. If it moves interest rates higher, then that is supposed to decrease business activity with the likely side effect that unemployment goes up, hopefully decreasing inflation.
There is a lot wrong with this simplistic scenario. If you troll through my past blog posts, you will see me diving into all this in more detail and pointing out how poorly this works. Nevertheless, this is the standard tool the Fed has traditionally used. And, for various reasons, which I am again going to skip, the Fed has been goosing the economy by keeping interest rates near zero for most of the last decade.
But supply chain problems and other issues that coincided with the end of the critical phase of the COVID epidemic led to a consensus that the economy had gotten overheated. This resulted in rampant inflation. Unemployment was extremely low, so it would there would be little or no harm if it went up a bit. This was the analysis pushed by the business community. There was, however, little dissent from this view coming from other sectors. Not surprisingly, the Fed adopted this view and started raising rates very rapidly.
Typically, the Fed meets once per month to decide whether or not an interest rate adjustment is warranted. Most months they decide to do nothing. But for several months in a row the Fed decided to raise interest rates by 75 basis points. A basis point is 1/100th of a percent, so that amounts to three quarters of a percent change. That's a big change. To increase the interest rate by a large amount month after month is unprecedented.
Large increases in the interest rate coming month after month was supposed to quickly cool the economy off. A cool economy was supposed to decrease inflation to the Fed's target of 2%. Increasing interest rates quickly and dramatically was supposed to slam the brakes on spending by both businesses and consumers. That, in turn was supposed to drive inflation down.
But there was little change in spending behavior even after several months of drastic action. As a result, inflation, while declining, stayed high. So, the Fed kept jacking rates up. Recently, some signs of a slowdown finally started appearing, so the Fed only increased rates by 25 basis points the last time around.
But the consensus among Fed watchers was that the Fed was going to continue with its "higher, ever higher" strategy. The only item that lacked consensus was how fast the Fed would raise rates. Would it go back to 75 basis points per month, or would it stick with a slower rate of 25 basis points per month. So, what's all this got to do with SVB? Interesting question.
It has to do with how you can lose money on bonds that don't default. Let's say that one year into the life of our 10-year bond we decide to sell it. Well, if interest rates are still at about 1% then the sale price will be $10,000, more or less. No harm. No foul. But what happens if instead in the interim the Fed has jacked rates up by a lot, and done it very quickly? Which, by the way, is exactly what they have done recently.
Then why should someone buy our bond for $10,000 more or less? It only pays a paltry 1% and they can buy a new 10-year bond that pays 4%. Instead of getting $100 per year in interest payments they can get $400. In those circumstances they wouldn't touch our bond with a ten-foot pole.
But what if we were to "discount" our bond, say by asking only $8,000 for it? That way they would eventually pick up an extra $2,000 when the bond got redeemed for the full $10,000 amount. $8,000 might not be the right price. But there is a price that would attract a buyer.
The business of figuring out exactly what that price would be is complicated. Fortunately, there is a "secondary" market in bonds. You can just look there to find out how much a particular bond needs to be discounted to in order to sell. The secondary market provides a "current market price" for our bond.
SVB did not get into trouble by selling a bunch of bonds and taking a bath on them. The situation was slightly more complicated.
As a result of the crash of 2008 a lot of institutions were required to periodically "mark to market" all of their assets. This stopped them from carrying "Zombie" assets on their books. These were assets that used to be worth a lot but were now worth far less. Zombie assets could make everything look fine when, in fact, it was not.
I don't know whether SVB was required to mark their assets to market, but investors and large customers became aware that SVB had a bunch of problematic assets on their books. They forced SVB to mark them to market and report a big loss.
At this point we don't have a problem. SVB could spread the actual losses over a period of years by not selling the bonds right away. So the problem could have been managed. They also did what they were supposed to do in a situation like this. They immediately set out to increase their capitol.
But when it comes to finance, it often resembles a game of musical chairs. In this case the loser of the game would end up having to eat the losses. So, a bunch of investors on a group chat decided it wasn't going to be them. It was going to be someone else. They all decided to immediately pull their money out of SVB. That way none of them would be stuck with the check.
Banking and finance are now just files on computers. A "bond" used to be a piece of paper that someone kept in their safe. Now it is information in a computer file. So are account balances. And now it only takes a matter of seconds to move money around. The time it takes is not affected by whether the amount is question is $1 thousand, $1 million, or $1 billion. The money moves equally fast in every instance.
A bunch of depositors pulling a few thousand apiece out of SVB wouldn't have made any difference. But a bunch of investors pulling out millions and billions made the difference between solvency and insolvency. SVB went from "just fine" to "dead man walking" in less than 48 hours. And SVB had another problem.
SVB specialized in funding for Venture Capitalists (VCs) and for other high-risk sectors of the tech industry. Tech has been hit hard in the last few months. It turns out that the economy was under enough pressure that they stopped seeing the growth rates that investors and VCs expected. Stock prices of even solid, well-established tech companies have declined substantially in the last six months.
In far too many cases, the companies SCB was investing in were not big and well-established. They were start-ups. As a result, they were likely hit even harder than the big, well-established firms. So, it was likely that the non-bond part of SVB's portfolio was also in a lot of trouble.
There is currently no hard evidence of this. But the fact that the FDIC, the part of the government responsible for cleaning up the mess that the failure of SVB left in its wake, has not been able to sell SVB off, either as a whole or in pieces, is a good indication of problems here too.
An over-reliance on a single market segment, or a number of closely related market segments, has been a red flag for forever. SVB should have not been allowed to do that. But the public does not care about the minutia of bank regulation while bankers and Wall Street does. So, regulators were pressured to look the other way. And they were not allowed to put regulations in place to outlaw this sort of behavior.
In that sense, SVB was unique, or nearly so. But as soon as SVB went under people started looking for other similarly situated banks. Signature Bank immediately went to the top of that particular list. It was over-invested in Cryptocurrency, another high-risk market segment. And it didn't take long for it too to go down.
When things are going well, high-risk market segments can create high profits. But when pressure is applied, they tend to go down farther and faster than more boring market segments do. It should come as no surprise to learn that both SVB and Signature Bank were Wall Street darlings.
Thank goodness, an over-concentration on high-risk market segments is not as common in the banking business as it once was. But the problems with their bond portfolio that SVB had is much more common. That's because of how banks actually work.
When you put money into a bank the bank doesn't sock it away in a vault. Instead, it loans it out to other people. The interest on their loan portfolio is how banks cover operating costs and the cost of whatever interest they pay on CDs, for instance. It is also how they generate a profit for their shareholders. This sounds risky, but if done properly it usually isn't.
Banks are aggregators. As part of how they do business they mix your money in with everybody else's. If on a given day more money comes in as deposits than goes out as withdrawals, loans, etc., then everything is fine. There is more than enough money to go around. However, if more goes out than comes in, this is a potential problem. But if the net is small, and over time it gets balanced out by more money coming in on other days, then it is an easily managed problem.
As a safety measure, banks are required to set a certain percentage of deposits aside as a "reserve". Is this money stacks of bills in a vault? Again, no. They are, however, required to invest this reserve money in "safe" securities. The regulators determine what securities qualify as safe. But they always include U.S. Treasury bonds in the list of securities that qualify. So, banks own a bunch of them as part of their reserve requirement. Well, guess what. These are the very bonds the Fed was jacking the rates up on.
So, it's not just SVB that was subject to this problem. And it's not just around SVB that musical chairs got played. The collapse of SVB caused people in the know to look at banks near and far. They soon zeroed in a bank called First Republic Bank. It's another of those banks that I had previously never heard of. But the jackals started circling within a day or so of the collapse of SVB.
First Republic is still in trouble. But apparently it wasn't in as bad a shape as SVB, because the Feds swooped in and started propping it up rather than shutting it down. As of this writing First Republic is still in business. If the pressure lightens up it will survive. If the pressure intensifies, it likely won't. The irony is that in a normal environment it would have sailed along nicely, and people like me would still have never heard of it.
And it's now not just U.S. banks. A giant Swiss bank called Credit Suisse was soon in trouble. They were old enough and big enough that their name was familiar to me from long before the present crisis. But first a digression because the story of Swiss banks is an interesting one. And it turns out to be relevant.
Up until the '30s they were the kinds of banks you would expect to find in a country the size of Switzerland. But Swiss banks smartly leveraged Swiss neutrality laws to their advantage in the run-up to Word War II. They took in deposits from people being persecuted like the Jews. They also were happy to deal with the persecutors.
As more and more countries cut ties with the Nazis, Germany started doing business through Swiss banks. Swiss banks grew enormously during this period. In the aftermath of World War II lots of money was stranded in Swiss banks because its owners, Nazis, Jews, and others, were no longer around to collect it.
Swiss banks quickly moved on to playing the same "collect money from both sides" game during the Cold War. Only the players changed. As opportunities there eventually diminished with the end of the cold war, they shifted to doing business with drug lords, authoritarian dictators, and the like.
But all good things eventually come to an end. In the last ten or twenty years the Swiss have been forced to open up their banking system to outside scrutiny. And other countries like Panama have also gotten into the business of hiding and laundering money, so the Swiss lost their near monopoly.
This has diminished the advantage of Swiss bankers. But by now they were hooked on the power and prestige of being major players. They have since resorted to getting into the business of playing the Wall Street game where they get involved in risky ventures in order to juice their balance sheet. When the game of musical chairs went international, a development that only took a few days, Credit Swisse was the fattest target.
The Swiss government brokered a deal where Union Bank of Switzerland (UBS), the other Swiss behemoth bank, took them over. The Swiss government was so concerned that they didn't let a little thing like the law get in the way. They simply changed the law that made such mergers illegal overnight to permit this particular merger to go through. UBS has its own problems. But the Swiss government decided that putting all the problems into one basket made them easier to manage.
Is the banking system now back in good order? It's too soon to tell. If we can go a couple of weeks without anything else popping up, then we are likely out of the woods. But if another big bank starts making headlines for being in trouble, or if First Republic goes under, or if the Swiss merger goes south, then we are in for more trouble.
There is a lot more I could get into. How we got into this mess. What should be done to fix it. What will be done to fix it. But I am going to confine myself to one additional subject.
There is talk of raising the limit on how much of an account balance the FDIC insures. The current limit is $250,000. Almost all of the money on deposit at SVB was uninsured because it was in an account that had far more than $250,000 in it. Most banks have a far lower percentage of their total deposits that are outside of the insurance umbrella in this way.
Theoretically, the FDIC could have left the owners of those accounts hanging out to dry. The law permitted the FDIC to say, "here's your $250,000 - sorry about the rest". The whole reason there was a run on SVB was because a bunch of big depositors did not want to take a haircut. And who can blame them?
With that in mind, it looks like it would make sense to increase the amount. An argument against increasing the limit to infinity as some are proposing goes under the rubric of "moral hazard".
Let's stay that all deposits are insured. Then what's to stop someone from putting their money into a bank that they know has problems? The incentivizing of bad behavior like this leads to a moral hazard. Or so the argument goes. But the "how high should the insurance limit be" ship sailed a long time ago.
At this point I would not characterize our current banking crisis as a big one. Compared to the size of the economy (trillions) even a few billion is small beer. But in the last few decades we have had two banking crises that do qualify as big banking crises. There is the one that people still remember, the crash of 2008, and the one they don't.
The one that everyone now conveniently forgets about is the Savings and Loan Crisis of the late '80s and early '90s. It turns out that there are lots of different types of "banks". The easiest way to organize them is by looking at who regulates them.
What most people think of as a "bank" is actually a "commercial" bank. They often have the phrase "National Bank" in their name. These operate under a "charter" issued by the Federal Government. They are regulated by Federal agencies and insured by the FDIC. As a group they are the most heavily regulated and have the strictest operating requirements.
But then there are the state-chartered banks. These used to be regulated and insured by the state equivalent of the Feds and the FDIC. I don't know if that is still true. SVB was a state-chartered bank, but the Feds, including the FDIC, have been all over them.
Then there are Savings and Loans (S&Ls). Back in the day they couldn't make loans to businesses and couldn't offer checking accounts. The biggest "bank" failure in the U.S. is Washington Mutual. Technically, it was a "mutual saving bank", but it was regulated by the same people that regulated S&Ls.
The original idea was that S&Ls couldn't cause much trouble, so regulation was much lighter on them. And at one time that was true. But in one of the waves of deregulation that swept the U.S. they were deregulated to the point that they could offer checking accounts and make a much wider variety of loans.
What could possibly go wrong? Lots. S&Ls started doing all kinds of things. Besides doing stupid things some were run by outright crooks. The whole thing came crashing down during a ten-year period running from roughly 1986 to 1996. S&Ls went under by the scores and the Federal Government ended up picking up the pieces.
Some laws were changed but S&Ls were not forced to confirm to the rules national banks had to operate under. Standards and regulations remained much looser. And throughout the S&L Crisis almost all cases depositors were made whole even if they had far more than the amount on deposit that was covered by insurance. At this point it became de facto policy to cover all deposits no matter the amount.
The demise of Washington Mutual (or WaMu, as it was commonly called) was not part of the S&L Crisis. It managed to make it through the S&L Crisis unscathed, because at that time it was well run. Instead, it belonged to the events connected to the crash of 2008. By that time good management had been replaced by bad.
And one contributing factor to WaMu's demise was the light regulatory environment it operated under. In the case of WaMu, and all the other "banks" that went under in this event, depositors were made whole regardless of the size of their account balance.
And the crash of 2008 introduced a whole new group of players into the public consciousness, "investment" banks. Again, the actual game was to find an excuse for diminished regulation.
A law called Glass-Steagall had been passed in 1933 carving out investment banks as a group subject to a separate, more permissive, regulatory regime. They couldn't do "retail" banking, offering checking accounts, making loans to individuals and businesses. They were restricted to only doing business with Wall Street.
The argument was that the customers of investment banks were sophisticated people who were savvy about money and risk. As such, a heavy regulatory hand was not required. Unfortunately, Glass-Steagall was repealed in 1999 as part of yet another wave of deregulation. As a result, most of the names mentioned in the headlines surrounding the crash of 2008 were investment banks. Not surprisingly, they had done stupid things in pursuit of ever higher profits.
All depositors caught up in the S&L Crisis and in the crash of 2008 were made whole regardless of how much of their balance was or was not supposed to be covered by insurance. In fact, I can't think of a time in the last half century when a depositor has taken a haircut as a result of a "bank" failing.
So, the limit on how much of a deposit is insured by the Federal Government is a fiction. Changing it will make literally no difference. Except, perhaps, to fool some people into believing that something is being done when nothing is being done.
And finally, to return to my original subject, bonds. SVB was a publicly traded company. As such it issued stock and was nominally owned by its shareholders. They are in for a haircut. But SVB also issued SVB bonds. The holders of these bonds are likely to be reimbursed 100%. It may take a few months, but they will probably get all of their money back.
It is common to see stockholders losing a lot and bondholders not losing anything when a company goes bust. And that is part of the reasoning behind the idea that bonds are safe while stocks are not.
