Analytical Frameworks: Evaluating Banks & Insurers
Or, How I learned to stop worrying and love the RWAs
Preface and Objective: Financial Frameworks
When looking at the investable universe of equities, it’s helpful for most investors to separate them into financials and non-financial corporations. While almost all equities tend to behave the same and are more correlated to each other than other assets, financials are fundamentally different businesses than others. Many investors caution against including utilities and financials in a longer term portfolio because, although they are distinctly different they share one common thread:
In both utilities and financials, things are usually fine.
Unless they’re not,
and then they’re really not.
Additionally, investing in both entails a level of regulatory risk that cannot be predicted or assessed via comprehensive analysis of the business itself.
Perhaps most perverse to investors in other sectors is that, unlike, say, software or consumer goods, rapid growth can be a huge red flag and ultimately a negative in the long run when analyzing financials. Finding the sweet spot between growth, risk and profitability is essential when analyzing banks.
As many of you know, I am not a strict fundamentalist. Most of my equity investing is based on narratives, thematic trends, even short term positioning and sentiment. However, that’s not the case when it comes to banks. Since I’ve started investing I’ve had - perhaps a knack is too strong a word but, maybe, a fondness for financials.
They’re ultimately in the same business we are, taking risk, earning a spread or carry, trying not to blow up. It’s not too difficult to get into the motivations either.
The Reason for the Framework
As you might know, I had a thesis predicated on the eventual insolvency of Silicon Valley Bank that began with a short in August 2022.
I have thought about whether it would be worth it to try and extract the exact differences in how I analyzed it that made it work so well, but there are a ton of financial primers out there. They’re likely going to be more educational than this one, but at the very least this article will serve to illustrate to investors how to approach analyzing risks and upside in financial names by forming a foundation and then using it to assess the potential hidden risks that require a different viewpoint to grasp. I found one bank interesting and it lead me to develop a kind of structure for analyzing the rest of them. It works okay, too.
But, just to ensure I’m not speaking out of my a**, I have invited Dan Saedi, ex-Bridgewater Bank Guy, and a very skilled insurance expert who would rather stay anonymous, to present their own framework alongside my own. I also use charts from a great banking resource on twitter, @sec_digger.
On social media, I have been asked more than a few times how I ended up deciding to short SVB about 10 months before it failed. I think that if you go and look back in earnest, it becomes pretty obvious.
There were only a few things I did differently than anyone else that gave me the conviction to do so (and let’s be honest here, it really was a conviction issue more than an intelligence issue - same with Credit Suisse, almost everyone in the financial industry knew CS was a shitshow, but only a few had the foresight to short it into the ground).
When you see a bank growing fast during times of low interest rates, especially ZIRP, you begin to think about what kind of risks they are taking. When you think about a bank from the perspective of a spread trade + some idiosyncrasies, I’ve found that you can find out some things that are probably being ignored by the people who look at it every single day.
I needed to speak to a lot of experts, and then be cocky enough to say about one specific thing “no, you’re wrong”. That was scary, and it should give you an idea of how difficult it is to point out a risk lying in plain sight. Still, doing the work with the significant (if not fragmented) banking data out there was worth it. In the framework section below, I will occasionally revisit this work to use as an example.
But why was this opportunity in existence? Why hadn’t markets been efficient and priced in this risk?
Everyone, at the time, was concerned about credit risks (this was when I first began looking into SVB in Summer 2022). That’s because it’s very easy to be concerned about whatever caused the last crisis. But, as we discussed earlier, banks were cashed up, they’d built up buffers, they were being wary of the impending recession and yield curve inversion.
They were not blind to a credit event. If anything, they were prepared for one.
No, the real risk lay not in the loan book but in the security portfolio. The actual risk to banks was zero interest rate policy. Everyone knows - low interest rates encourage risk taking. It’s why they stimulate the economy and also stimulate fraud (you can read more about one of the most epic historical incidences of this, known as Poyais, here).
It was those banks who had reached not for credit risk but further out along the curve in order to maximize their profit that posed the risk.The biggest thing, probably, was not the interest rate increase. It was the expectation versus the reality. If the fed funds rate had followed the projected path, SVB (maybe) would have been fine. The key to recognizing that interest rate risk in general would be a threat to the banking sector was to monitor how rates were playing out against expectations - as expectations were going to end up driving hedging decisions. A bank that had listened to market expectations and made hedge decisions on the expectation of rates at 1.5% would be in trouble.
Now, it didn’t take much of a leap to go from “interest rate risk is bad” to “interest rate risk is bad for banks who took too much interest rate risk” in conjunction with “also, interest rate risk is bad for tech” to “I should short Silicon Valley Bank”. Literally, I could stop this part of the article here. I did end up doing math (sad). I did learn a lot more about banking in general (actually kind of fun). It was great, but really if I had the conviction, I could have just shorted SVB there and then been fine. Unfortunately, I probably would have been short squeezed during one of the many rallies if that had occurred. Instead, I opted to do the research.
Channel checks had revealed a number of sweetheart deals with founders on mortgages and loans, cost of capital was low but most accounts were institutional, with high balances that they’d want to earn the risk free rate on eventually. Growth sectors had been hit hard by the increase in interest rates and decreases in available lending, and SVB banked many VCs. Intermediation for capital calls was a big business, but had fallen off. And finally, this all lead me to conclude that SVB’s depositor base - many that were above the $250k FDIC insured level - would be more likely to have to withdraw money in significant amounts over the next year. That did not necessarily equate to a bank run, but I found it very easily could. After examining SVB’s balance sheet - including both their held to maturity and available for sale portfolios - I found that their actually CET1% was closer to 1% after marking to market than the 12% that was obtained without doing so and reported to regulators (and therefore used by investors).
If I was the kind of person who cared about impressing you, my esteemed readership, I would probably portray myself here as a Michael Burry type (without the weird angsty stuff). I’d say something like “and this is the moment I knew the banks were fucked!”.
However, I’m not
and, also,
that’s not what happened at all.
I thought I’d done the math wrong. After all, we had regulators that were on top of this, right? “Of course, it’s the 16th largest bank in America”, I thought. I called anyone I knew who knew even a little bit about banking. They assured me of two things: 1) the math was right 2) the math didn’t matter, held to maturity portfolios were held to maturity. I was being chicken little on this one.
Someone said to me, “Well, if you’re gonna do that you’re gonna find that tons of banks are almost insolvent!”. But I wasn’t concerned about tons of banks. I didn’t think Bank of America had positioned itself to need to deliver on 20%+ of its demand deposits in the next year. I thought Silicon Valley Bank had, and that’s why I shorted them.
I figured, the risk/reward was good. The bank wasn’t well situated regardless. The unexciting scenario would have been they just didn’t do that great as a business over the next year and their crazy overvalued equity calmed down. So, I wasn’t Burry - I didn’t tell the establishment to go fuck itself or anything. I second guessed myself a million times throughout this process, I figured I wasn’t really in a position to second guess people who had made banking their entire lives work (I was correct, I probably still am not). But I could, essentially, trust myself to find asymmetry in a trade. And that’s what I did.
I still have the screenshot of the chart I sent to people from back in October-ish 2022:
If you’re sitting there, looking at that chart and going “looks bad, I have no clue what it means though” then I think this article will help. Understanding how banking works is probably never going to be a net negative if you are in finance, either.
That’s because the next crisis is probably not going to look like the last one (aside from the fact that it’ll probably involve the financial sector, if it’s not directly caused by it). So, if you understand the last crisis but not the underpinnings that caused it, you probably won’t be able to catch the next one. Create a foundation of knowledge, be on the lookout for things that simply don’t make sense from a risk perspective and, with the right tools, you’ll find something interesting.
The funniest part to me, probably, out of the whole SVB thing is the way it ended. Just like how I describe to people the simplicity of the thought process that got me to short SVB in the beginning (“it’s a tech bank that’s bad at interest rate management”), the ending note was pretty good too. “Bank has to sell securities”.
You know how little you have to understand about banking to understand how bad that is?
If you don’t, take only one thing away from this article: the lifeblood of a bank is its ability to raise capital (whether via debt or equity). When they are selling anything besides those two things to raise money, there’s a high probability they’re screwed. If they’re selling their held to maturity portfolio…well, I mean, you speak english, right?
I had added to what was an already big short at that point, but I still think I should have done more after that announcement.1
Anyway, they won’t all be that easy. The whole process really got me to look at the financial industry and regulation in a way that has made it somewhat approachable, or at least more manageable, for my own limited brainpower. And I think there are some valuable aspects to it.
So, I’m sharing two frameworks: one is for banks, the other (written with the help of a friend, because insurance is hell) is on insurers.