Greetings from the LHH! Well, it is Sunday of spring break and I ventured onto campus for a minute to pick up a book from my office and you could see the stream of parent cars returning their young scholars to the dorms. My seniors will graduate two months from tomorrow. Can I tell you, it never gets old, you never quite get used to it. I pick them up when they are juniors and we spend a rather intense two years together, my team trying to work with them, to teach them, coach them, and generally try to get them to get out of their own way, and then, like rockets, they are gone. And the next fall there are these strange kids sitting in their seats and we start all over again. I guess if I ever do get used to it, it will be a good sign that it is time to go.
You know what else never ceases to surprise me? How beautiful Durham is. Photo above is of Mill Pond in Durham. The pond is still partially frozen, but we’ve had temps in the low fifties the last few days (after the Nor-easter that blew through on Tuesday) and everything is melting. I remember playing broomball (no Quidditch) on Mill Pond when the wife was a student here at UNH and we were kids. The seasons always bring new things to see and do. I’m hoping I can get out on the paddle board this week given the warm weather. We’ll see!
You know what else is amazing? Sheet cake. This recipe for white Texas sheet cake is super simple and OMG good. We had Sunday dinner today with my father and his wife and I made this for dessert.
So I wanted to circle back to my discussion of Silicon Valley Bank (SVB) and what seems to be a possible banking crisis. Since I wrote to you on Wednesday, Credit Suisse, a much larger bank than SVB or Signature (banks that failed last week) was heading toward failure and had to be bailed out with a $50B loan from the Swiss government, and now has been bought by another mega-bank, UBS in an effort to stem any further contagion and reassure depositors (that’s us) that banks aren’t going to keep failing. I said on Wednesday that I wanted to talk about how interest rate hikes are causing stress across the board on many banks, and that we may be in for more of a challenging ride than just watching some tech-bros lose their money.
To understand what happened to SVB in particular, and is happening to banks in general, you have to understand a bit about bonds. Many of you have some sort of loan - car loan, student loan, or home loan. Governments and businesses take out loans, too. When the loans are relatively small - you know, less than $20M or so - governments and businesses will work with a commercial bank. Commercial banks specialize in lending money to governments and businesses, whereas retail banks specialize in lending to individuals. Most banks have both services - SVB, Signature, and Credit Suisse all did. But when a government or a business wants to get a really big loan to do something like build a new factory or hospital (business), or a new school or road (government), and the loan is really big - like $100M or more - they will go to the bond market to get the money. Most commercial banks don’t want to have so much of their capital tied up in one place - they usually like to be diversified - so they don’t like to make such big loans. A bond issue allows many investors to buy a piece of the loan.
So a bond offering is just a big loan that is divided up into little pieces. The Oyster River School District (ORSD), which is the school district Durham is a part of, recently needed to build a new junior high school. They needed about $50M, which is more than a local bank would lend, so they issued bonds to finance the school. The bonds had a maturity of 25 years and an interest rate of about 2.5%. Bonds like this are typically sold to the public in increments of $100, $500, or $1,000. So you could potentially buy $1,000 of the $50M issue. I don’t know all the details of this particular bond issue, and bonds can be issued with a variety of details about how the payment structure will look, so I will talk about what I call a “vanilla” bond.
A vanilla bond is a basic bond. All bonds are some variation on the vanilla bond, like chocolate chip ice cream is a variation on vanilla ice cream. Like any loan, all bonds pay interest and principal on some schedule. A vanilla bond pays a semi-annual (that’s every six months - I always get bi-annual and semi-annual mixed up) coupon payment for the life of the bond and then they pay back the principal in one lump sum at the end. So if the ORSD bond was a vanilla bond, they might have issued 50,000 bonds at a price of $1,000 each (50,000 x $1,000 = $50M). Investors would buy as many of the $1,000 bonds as they wanted for their portfolio. If you were to buy one, your annual coupon payment would be 2.5% x $1,000 = $25, but it would be paid out in two semi-annual payments of $12.50 each. So for 25 years you would receive a payment of $12.50 every six months, and then at the end of 25 years you would get your $1,000 back. It’s a nice way of locking in an interest payment of 2.5% per year for a long time. And if you hold the bond to maturity, you get $1,000, which is nice. You can also typically sell your bond at any time, which is also nice. But the price you will get when you go to sell your bond depends on what rate other bonds like yours are currently getting. If the interest rates other bonds are getting falls, your bond will sell for more money. For example, if bonds similar to the ORSD bond are now being issued with 2% coupon rates, your bond will sell for more than $1,000. That’s because if you buy the 2% bond, you will only get $10 every six months, instead of $12.50. I don’t know about you, but I would rather get more money than less money. On the flip side, if new bonds are being issued with a coupon payment of 3%, your bond won’t look as good because people will be able to get $15 every six months from the new bonds. As a result, the price you can get for your bond will go down.
SVB bought a whole bunch of US Treasury bonds with maturity dates that were far out in the future when these bonds were yielding very low interest rates, and the rates went up dramatically over the last year, which is making most of the bonds they are holding worth much less. Let’s talk about US Treasury bonds.
The US Treasury is the branch of the government that is responsible for paying all of the US Government’s bills, including collecting taxes through the IRS, and getting financing when our taxes won’t cover our Federal spending (this is the budget deficit you hear about). As I have written about in the past, we have an enormous national debt (currently $31 Trillion according to the Treasury), which is financed by the US Treasury issuing - you guessed it! - bonds.
Below is a table of interest rates for US Treasury Bonds for this past Friday (St. Patrick’s Day) and one year ago on the same day.
The US Treasury issues new bonds and pays off old bonds every day. They issue them in a range of maturities, from as little as 30 days to as long as 30 years. The bonds that are less than a year are called T-Bills, but they are still bonds. I’ve left them off of the above table because they are not “vanilla” bonds and I’m already losing most of you so I won’t take the time to explain them.
If you look at the row above for 3/17/2022 or the orange line below, you can see that the shorter maturities have lower yields and the longer maturities have higher yields. This is called the yield curve, and it is normal for the yield curve to slope upward, which it was doing on St. Patrick’s Day last year. If you look at the row for this past Friday, or the blue line below, you can see that the yield curve right now is generally downward sloping. This is an inverted yield curve, also known as a hot mess.
Why is the yield curve inverted? Well, mostly it has to do with inflation expectations. There are three major factors that influence the interest rate on a bond:
Default Risk - the risk the bond issuer will not fulfill the bond contract and pay all of the promised interest and principal.
Duration Risk - the farther out the maturity of the bond is, the more chance you have that the market rates will rise while you own the bond, resulting in the price falling.
Inflation Risk - since the bond is a contract, you are stuck with whatever rate you agreed to. If inflation suddenly jumps, the interest you agreed to is going to be worth less.
When any one or more of these risks rise, the rate the market requires on new bonds rises. So if you are holding an old bond with a lower rate, what happens to the price you can get if you want to sell it? The price goes down.
US Treasury bonds (“Treasuries” or “T-Bonds”) are often referred to as “risk-free”. We even use them in financial models to proxy for the “risk-free rate” - meaning the rate without risk. In reality, the only risk element that is zero for Treasuries is default risk. Why? Because, like the Lannisters (from Game of Thrones), the US Treasury always pays its debts.
Since the founding of the United States, the Treasury has a perfect record of always paying its debts. It was Hamilton that actually saw the value of this and started us down this road. By always paying our debts, year after year, decade after decade, century after century, we have created a realistic expectation that we will keep on doing it. No other country has that kind of track record. (insert mic drop here.) So the US Treasury bond is regarded as the “risk free rate”, but only in the sense that it will never default.
Looking again at the 2022 line on the table, you can see that even the US Treasury pays higher interest rates for longer term bonds. The yield (or interest rate - I keep using those interchangeably, and they are, for the most part) that investors demand rises with the longer maturities because of both uncertainty about price (duration risk) and uncertainty about inflation. Between 2022 and 2023, the duration relationship didn’t change. What did change was inflation expectations. The yield curve inverted mostly (IMHO) because investors are worried about inflation in the near-term, but not especially in the long-term. The fact that the curve now slopes downward indicates that investors think inflation will be high for a few years, but eventually go back down to more normal 2-3%.
Now, if you hold the T-bonds to maturity, changes in the market rate (i.e., changes in the yield that similar bonds are offering) doesn’t matter. You will get your contracted rate of return, and you will get your principal back. The fact that the price for which you could sell the bond at any given time doesn’t matter.
Let’s say you're a bank. And you are taking in deposits and instead of lending them out, you decide to buy long-term T-bonds. Let’s say it’s St. Patrick’s Day, 2022 for example. You offer your depositors 1% interest on their savings accounts. They are excited because most banks are offering much less than that. They give you lots of money. You turn around and buy “risk-free”, 30-year T-Bonds that have a yield of 2.5%. This gives you a great return. You pay the investors 1% and you keep the difference - 2.5% - 1% = 1.5%. It’s win-win all around! You look like a genius. More people give you their money and you buy more T-Bonds.
You know what’s even better? Because your investment portfolio is all (or mostly) T-Bonds, and not bonds from some little school district in New Hampshire or mortgages on Silicon Valley mansions, when the Federal Reserve sends its bank inspectors in, they mostly say, wow, you have a risk-free portfolio! You are a genius! To which your reply, with typical Silicon Valley humility, I know, right?
Furthermore, if you push your investments in long-term T-bonds into an account that you label “hold to maturity” or HTM (which bank regulations allow), you don’t have to go through a process called “mark-to-market” where you have to reveal on your balance sheet what the market price of the bonds you own would be if you tried to sell them today. (This is some crazy sh#t that I only learned about this past week. Hospitals that hold investments like T-bonds have to mark-to-market when they report their balance sheets. I didn’t realize banks had given themselves this kind of loop-hole.) If you don’t have to mark-to-market, you would just list your bond at its purchase price (as $1,000), which right now is a huge, stinking lie.
Here’s the thing - the market went nuts (technical, finance term) over the last year because of inflation triggered by the profligate Federal COVID spending of the last three years. If you bought a one-year T-Bond at each of these maturities (i.e., 1 year, 2 years, etc.), the price you could have sold them for on the market one year later is listed in the row, “Value 1 year later”.
(If you would like to look at my calculations, my spreadsheet is here: https://docs.google.com/spreadsheets/d/1t1pHhmZEdek0ex3bw3jYXHjrScVV1beq/edit?usp=share_link&ouid=116556535244578085779&rtpof=true&sd=true )
If you were conservative and bought a 1-year bond, the value would be $1000. Why? Because one year later, the government would have sent you your second and last coupon, as well as a check for $1,000. So if you are holding a check for $1,000, how much money do you have? It’s not a trick question. You have $1,000. That’s truly about as risk-free as you can get. But if you bought any other maturity, you would have a bond with a price below $1,000. If you bought the 30-year T-Bond as I discussed in my example above, the value of your asset would be $803 - a roughly 20% loss. But! But…you really don’t have a loss unless you have to sell. This is important. If you hold to maturity (HTM), the price will eventually be $1,000. You just have to wait 29 more years and the price will automatically be $1,000. Please hear some sarcasm in this statement. You have to be able to wait 29 years. Here’s a visual of the losses:
So back to SVB, and then we’ll zoom out for the bigger picture.
SVB did what I described above - they took in deposits with no commitment from the depositors that they would keep the deposits in the bank, and they used those deposits to buy long-term treasuries. I do not know the exact maturity of their portfolio, but from what I am reading, it was long. When interest rates spiked over the last year, they did not move fast enough to sell off their long-term bonds and move into shorter maturity bonds. As a result, the value of their portfolio fell dramatically. When depositors got wind of this, they started pulling out their deposits. This is where the problem lay. SVB would have been fine as long as they didn’t have to sell the bonds. Eventually they would have been worth 100% of what they paid for them. They just had to wait either until the interest rates fell again, or the bonds reached maturity. Unfortunately, maturity would have been 10, 20, or 30 years away, and the depositors wanted their money today. SVB failed because its investments were worth less than the deposits they were obligated to pay back.
Something I didn’t do in Wednesday’s letter was include bank equity. Let me do that here. Assume you are starting a bank. To do that, you would have to contribute some money to get it started. Let’s say you have $10M in cash. Your balance sheet would be:
Then let’s assume you take in deposits of $100M. These would come in as cash. So then your balance sheet would look like:
If you’re not familiar with a balance sheet, both sides always have to be equal.
Then let’s say you use $90M of the deposited cash as well as your $10M to buy 30 year T-Bonds.
Then let’s assume the great inflation of this past year hit and your T-Bonds fall by 20% in value. If you can hide the fact that your Assets have fallen by 20% by sticking them in that special HTM account, then nothing changes. But if you are forced to mark-to-market because investors start asking for their money back, you wind up with negative equity (the $(10) means negative $10).
Here in lies the problem - you just went broke. Your stake in the company, your equity, is now negative. You are bankrupt. You can’t pay back your debts (you’re no Lannister) because you don’t have the assets to do so. Even though your investments were “risk free” according to bank regulations, they really were nothing of the sort.
This is the story of SVB as I understand it. If they had had more of their own equity in the bank, the bank might have survived. If they had taken less duration risk, they wouldn’t have taken as big a hit against their investments and they might have survived. But they didn’t have the equity and they did take the duration risk. And so the wheels came flying off last weekend.
Not every bank is this poorly capitalized, and not every bank has so much maturity risk, but every bank has some of both of these risks. If things go badly enough, no modern bank carries enough equity by my understanding. And borrowing short (deposits) and lending long is basically what banks do, as I said on Wednesday. But finding a balance between duration of your portfolio - i.e., not making all of your investments long-term - is an important way of being conservative. Having lots of capital is another way of being conservative. Banks are supposed to be conservative. Even better, banks are supposed to be boring.
Right now, I am afraid there are a lot more SVBs and Signature banks out there. The market is afraid, too. The fact that Credit Suisse had to sell itself is a sign that even big banks are at risk.
Sorry for the long post, but I said on Wednesday, “while SVB was overexposed to price risk, all banks are currently taking hits from increasing interest rates, so while this bank did some stuff to leave it especially exposed to interest rate changes, it could represent the tip of an iceberg of liquidity trouble in the banking industry overall, which could really hurt all of us.” I’ve tried to show why SVB failed, and why SVB is likely the tip of the iceberg. Inverted yield curves almost always signal a recession. But worse would be another banking crisis. Let’s hope things aren’t as bad as they look.
I hope this is useful. If you go this far, please let me know what you think.
I’ll be back Wednesday with links. As usual, willing good for all of you! Happy Sunday!
Friends: If you are enjoying this newsletter, won’t you share it with a friend? The price is right, after all.
Thanks for reading RWL Newsletter! Subscribe for free to receive new posts and support my work.