The 3000 Word Silicon Valley Bank Explainer That No One Asked For Is Here

Hi wine industry friends (and anyone else who comes across this blog.) You thought the tariff process we went through was complicated? Well welcome to a whole new level of complication: the Silicon Valley Bank failure.

If you’re wondering why this bank sounds familiar, it’s likely because the Wine Division releases an annual State of the Wine Industry report. It’s a cornerstone of data-driven trend reporting within an industry niche not typically known for its access to hard data.  The report usually kicks off a social-media flurry of industry-related thought pieces, comment wars, and meme activity and this year’s report, released in January, was no exception.

But right now, the headline story is about the bank’s collapse, a news story that goes well beyond the niche world of wine blogs and meme accounts. And of course, there are many, many, MANY pieces about what happened but a lot of them tend to be deep on jargon and light on detailed explanation.

So that’s what this piece is: an explainer that starts with a lession on bank balance sheets, touches on the accounting principles of bond valuation, and winds its way through what happened in the days before the FDIC closed the bank’s doors.

Scintillating? Maybe not. But it should give you an overview of what’s happening and why, how this bank is unique, and at the very end, a wee bit about how the Wine Division fits into all. (TLDR: it’s a casualty, not a cause.)

Happy reading!

– Christy Frank

Banking Balance Sheet 101

Commercial banks take cash deposits from individuals and businesses. They then use that cash to make loans to other individuals and businesses. They collect interest from the loan holders and pay interest to the deposit holders. They make money because the interest they take in is more than what they pay out. They’ll certainly have other income streams, but the main one will be based on this spread between these two interest rates.

In the language of accounting theory, those cash deposits are called short term liabilities. They are liabilities, because they’re something the bank owes someone. And they are short term, because if you go into the bank and ask for your money, it has to give it to you… pretty much right away.

All those deposits sit on the liability side of a bank’s balance sheet. On the other side of that sheet are the assets – including the portfolio of interest-bearing loans the bank holds. As the phrase “balance sheet” implies, these two sides need to, well, balance.

For most banks, most of the time, this balance is like a slow-moving seesaw. If the deposits are from a wide range of customers across a diverse set of industries, the risk that everyone will want access to their money at the same time is quite small. A further stabilizing force is that deposits under $250K are guaranteed by the FDIC – no matter what happens, that money is safe. And on the asset side, diversification also ensures that the loans come due on a regular, rolling basis, allowing for a constant stream of funds.

If this sounds like it might involve a lot of fancy math, you would be correct – which is why many banks have risk management departments working to make sure this all works smoothly.

In the case of Silicon Valley Bank (SVB), things didn’t exactly work so smoothly. Let’s explore…

Silicon Valley Bank’s Balance Sheet – Part 1

If we were to peer at the SVB balance sheet, we would see the bank had $173 billion in deposits on the liability side of its books. On the asset side, they had issued around $74 billion in loans to various companies. This left about $100 billion in deposits to be invested.

Now remember… to make money, banks need the interest on their loan portfolio to be higher than what they have to pay out on their deposits. $1 billion can’t just sit in a vault not drawing interest. That is not a money-making proposition.

So SVB invested that $100 billion in government-backed bonds. Were these assets low risk? Yes they were. Were they short-term? No they were not. Did this matter? It didn’t, until it did.

A Primer on How Bonds Work + Some More Accounting Theory

I know, I know…. you’re not reading this explainer to understand how bonds are priced or how they are valued on a balance sheet. But I’m going to tell you anyway. This is the first and potentially only time I’ll be able to put my London School of Economics degree in theoretical accounting and finance to use so I’m going with it. But trust me… it’s important.

Let’s say you’re a bank and you buy a $1000, 2% interest, ten-year government bonds. (You’re a very small bank.) You give the government $1000 (the face value of the bond) and the government gives you a piece of paper that says “IOU $1000 in ten years.” In each of those ten years, you collect 2% interest (or $20.) At the end of ten years, when the bond matures, you get your $1000 back and rip up the piece of paper. (Actual paper isn’t involved, but you get the idea.)

But meanwhile… more of that same bond is being traded on the bond market. Other banks and companies and individuals are buying and selling those $1000 face-value bonds. But as with any market, their value goes up and down. Some days you could sell yours for $1500 and other days, you might only get $500.

So what’s the main determinate of any bond’s value? Interest rates! I’ll spare you the details, but it is a truth universally (and mathematically) acknowledged, that if interest rates go down, the value of a bond goes up.

And if interest rates go up? The value of the bond goes down. This is very important to the mechanics of this story so I will repeat it: if interest rates go up, the value of a bond goes down.

Which way have interest rates been going? Up, up, up! Which means the value of those imaginary $1000 bonds trading on the bond market have been going down, down, down. Let’s say today, if you wanted to sell your bond, you would only be able to get $750 for it.

We’re now going to zoom back to the asset side of your very small bank’s balance sheet. You have to decide what your bond is worth. Which is it? The $1000 that you originally paid for it? Or the $750 that you could get if you sold it today?

What??? You don’t know the answer? Well hold on tight… because we’re about to go down a jargon hole and explore the world of Generally Accepted Accounting Principles.

Hold-to-Maturity vs Marked-to-Market: on which side is your bond buttered?

The answer to the question (what’s the balance sheet value of my bond?) is: IT DEPENDS!

If you intend to hold your imaginary bond to maturity, you get to classify it as, wait for it, “Hold-to-Maturity” and value it at its $1000 face value – no matter how high or low it’s trading on the bond market. Over the course of its lifetime, it could swing from a low of $5 to a high of $3000. (There’s pretty much no possibility a government-backed bond will swing that much, but I like to go for impact in my examples.) But it doesn’t matter, because you still get your $1000 once the bond reaches the end of its life.

But let’s say your plan isn’t to hold your bond – it’s to actively trade it, selling it and buying it back, essentially placing bets on whether it will go up or down. If this is your intent, then financial accounting rules require you to revalue it on your balance sheet, or “mark it to market,” every single day. One day you might value it at $1000. The next day, $1500. The day after, $850. (The actual day-to-day swings would likely be much smaller, but hey, impact!) This is logical – you might sell that bond at any time, at any price, so your balance sheet needs to reflect that risk.

The bonds used to balance a bank’s deposits are typically classified as Hold-to-Maturity. Also logical, right? Sure it is… as long as reality doesn’t intrude on intent.

Silicon Valley Bank’s Balance Sheet – Part 2

Remember those $173 billion in cash deposits on the liability side of Silicon Valley Bank’s balance sheet? They were drawn mainly from a slate of Venture Capital (VC) firms that invest in tech-driven startup companies as well as the startup companies that they were investing in. Startups get advice from their VC investors and in this case, the advice was “put your money in SVB.” A portrait of diversification this is not, but as a niche strategy to become the “go-to financial partner for investors in the innovation ecosystem and beyond” it was working very well. Deposits increased from $49 billion in 2018 to $173 billion in 2022 as the industry went into overdrive during the pandemic.

Business seemed to be chugging along nicely, but the proverbial train was poised to go off the rails. Actually, since we’re in the startup world, let’s shift from train analogies to airplanes.

Trains, Planes, and Exit Strategies

Startup businesses always talk about their “runway” – how many months they can operate before they run out of cash. As they burn through cash – and approach the end of their runway – they need to raise more funds or they’ll go out of business. This additional cash usually comes from VC investors and is part of an on-going fundraising strategy.

But VC firms don’t want to pave endless runways. Eventually they expect their investment cargo to take off, exit the airstrip, and provide the original investors with a return on that investment. Typically the startup company will be sold to another firm or “go public” and get a listing on a stock exchange through an IPO. Either way, a chunk of the returns from an exit goes back into the VC fund – and into their bank accounts. And everyone flies off into the sunset!

Except…

For reasons that I’m not going to get into now, but are tied to interest rate increases in fairly predictable ways, those exit doors were starting to close. Companies weren’t so keen to buy startups and IPOs weren’t happening. So more runways needed to be paved for longer periods of time – and the VC firms needed cash to do this paving.

Silicon Valley Bank’s Balance Sheet – Part 3 (or Run, VCs! Run!)

Let’s head back to the SVB balance sheet, shall we?

VC firms needed to use their cash deposits to fund their current portfolio of startup companies for longer than expected. And a lot of that cash was part of the $173 billion in deposits at SVB. Or more specifically, that cash was now sitting in bonds owned by SVB – bonds that were classified as Hold-to-Maturity.

Meanwhile, out in the open bond market, due to recent interest rate increases, those bonds were trading at much lower values. (Remember: interest rates go up, bond values go down!) So what happens when SVB has to sell some of those bonds to generate that cash? Well those “unrealized losses” had to be realized. And reported. Out loud. In public. The bank had sold $21 billion in bonds resulting in a $1.8 billion loss. This was announced on Wednesday, March 8^th.

While the remaining bond portfolio didn’t need to be marked-to-market, some quick mathematical calculations could show that the remaining $80 billon (if my math is right, which it probably isn’t) wasn’t worth $80 billion in the open market.

Suddenly, the balance sheet doesn’t balance. Or more accurately, suddenly the world knew that the balance sheet didn’t balance. There weren’t enough assets on that side of the balance sheet to balance the deposits on the liability side. SVB had to somehow raise funds to fill that gap. On Thursday, March 9^th, they announced they would do this by selling $2.25 billion in stock.

Word began to fly through the startup “innovation ecosystem” that SVB was in trouble. VC investors had once recommended startup companies in their portfolios open accounts with SVB. Now they were urging them to get their money out of those accounts – and fast. (And remember, this is a tech-based innovation ecosystem. That “ecosystem” is really, really small. And “fast” is really, really fast.) To add to the fun, remember that the FDIC only insures deposit accounts up to $250K – and nearly all of SVB’s deposit accounts were more than $250K.

So if you’ve watched It’s a Wonderful Life, you know what a bank run looks like and this was that. However, Silicon Valley doesn’t have a George Bailey to explain to the good-hearted VC townspeople why they should leave their money in the bank. So the spiral continued and by Friday, March 10^th, the FDIC had taken over SVB and all deposits above $250,000 were potentially SOL.

Awkward Pause: What Exactly is the FDIC??

There’s no elegant way to work this into the story, so let’s just take a break to explain. The Federal Depository Insurance Corporation was created in 1933, after the great bank runs of the Great Depression. It’s exactly what it says it is – an independent federal government “corporation” that insures bank deposits.

For all my wine folks reading, think of it as wine marketing board, similar to Wines of Austria or Wines of Australia. All the wineries pay into a funding pool which is used to promote and grow the profile of the county’s wines. Except in the case of the FDIC, it’s banks. And the funding pool is used to make payments to the depositors at failed banks. Because banks fail more than we realize – they just tend to be smaller, with most of their deposit accounts under $250K. The fact that we don’t hear about them is silent proof that the FDIC is doing what it was established to do.

The Aftermath: A Bailout or Not a Bailout?

OK, back to our story. On Sunday, March 12^th, the FDIC made it clear it would guarantee all SVB’s deposits – including those over $250K. That collective sigh of relief you might have heard that evening? It was an entire network of companies, their employees and their vendors that wouldn’t have to scramble to make payroll, pay their suppliers, and keep their lights on.

Is this a bail out? As with so much in financial accounting: IT DEPENDS!!

Yes, uninsured deposits are now being insured, but that comes out of the FDIC insurance pool. The $74ish billion in loans on the asset side of the balance sheet still exist and the companies that owe that money will still need to make their regular payments. The $80ish billion in government bonds, also on the asset side, will be used to pay back the deposits (although I’m not entirely sure how this works and that’s too much minutia for even me.)

What’s different from the situation in 2008/2009 is that SVB is gone. Poof! The FDIC stepped in to guarantee the deposits, but the owners of the bank are not being protected. Its equity – including the shares that traded on the stock market – no longer exist. If you were a mutual fund or a hedge fund or an ill-advised grandma holding SVB shares, those are now worth $0. Period. End of story. And any debt that the bank issued on itself, that’s gone as well.

So is it a bailout? Kind of. But if it is, it’s a much more limited sort of bailout than what happened in 2008/2009 when the failed banks were kept alive in ways that could be the subject of an additional 3000 words.

The Aftermath: To Panic or Not to Panic?

Now here’s the thing. A $1.8 billion asset write-down (a fancy term for “a loss”) on a $21 billion bond portfolio is indeed a lot of money. But remember – SVB still had about $80 billion in remaining bonds as well as an additional $74 billion in other loans. VC investors and startup investors are theoretically quite savvy in their understanding of how these things work. It was nowhere near insolvent. But a panic is a panic. And no one wants to risk being the last rat on ship that seems to be sinking, even if it has plenty of life vests on board.

Will the panic spread to other banks? Logically, it shouldn’t. Most banks’ deposits are much more diversified than SVB’s. Most banks haven’t built their strategy around being the go-to for a single, tight-knit clubhouse of an industry. And (hopefully) most banks are doing a better job of stress-testing their asset portfolios to ensure they can handle the bigger swings that come with higher interest rates. And while the recent increases in interest rates makes it more difficult for any business that benefits from cheap financing (i.e. almost all businesses), the startup world was especially at risk (which is a story for another time.)

(This story is moving as quickly as I can type so I’m not even going to try to keep up. Keep your eye on the news – and this is important: check the date and time stamp of everything you read. It might be old news by the time you read it!)

So What’s the “So What?”

I’m tired of writing, so now that you have a good understanding of how this all works, you can dig into all the speculation about near- and longer-term impacts on the innovation ecosystem of VC funds and tech bros, startups in general, risks to the regional banks of California, and regulatory impacts.

But let’s talk about how SVB relates to the wine industry, since there’s a good chance that’s why you started reading this piece in the first place. (Is this what they call “burying the lede?)

SVB’s Wine Division was established in 1994. Over those nearly 30 years, it made about $4 billion in loans to hundreds of wineries. This represented about 2% of the bank’s total loan portfolio, so if you’re doing the math, you’ve already realized this made the division a very small part of the overall SVB business.

But within certain niches of the wine industry, the Wine Division was an important partner, providing loans and lines of credit to a complicated, inventory-intensive, equipment-heavy industry that’s not well-understood by or terribly attractive to more traditional banks. The head of the division, Rob McMillian, also put together an annual State of the Wine Industry report which generated data-driven insights for a segment of the industry that tends to be light on data.

As I write this, the FDIC has promised that all customer deposits will be protected, even those above the $250,000 threshold, so the immediate threats to winery payroll and other operations is less intense than it was when I first started typing on Friday, March 10^th.

But longer-term? It’s unclear whether the SVB Wine Division could have offered the sort of financing and partnership it offered – if it wasn’t part of the larger SVB. And if the conditions that allowed the larger SVB to exist, well, no longer exist? What does this mean for the overall sustainability of this niche of the wine industry?

Conversations for another time… because I need a drink.