top of page

Thoughts on the Collapse of Silicon Valley Bank (2Q23 Quarterly Newsletter)

April 15, 2023

The truth that has been brought home to the millions of us who are bank customers by the Silicon Valley Bank experience is the fragility of at least some banks when, for whatever reason, their safety and soundness is called into question. In theory, any bank is susceptible to a run (where large numbers of depositors, being privy to rumors, suggestions, whispers, tips, even false flag operations by short sellers, race to move their funds away - and, by the way, in so doing enact a classic self-fulfilling prophecy). With deposits leaving, and no offsetting deposits arriving, a bank’s balance sheet promptly turns catawampus, meaning liquidity erodes, equity diminishes or is eliminated, and without successful rescue efforts which fend off the regulators (but tank existing investors) a regulatory takeover ensues.

So much for theory. In fact, the great majority of banks, large and small, are safe and sound. Silicon Valley Bank (“SVB”), for reasons having to do with its culture, community and customers, had not been, at least for some time, “safe and sound” in the conventional banking sense of the terms.

The reason? Theirs was (at least this is how it is seen from an outsider's perspective), in the eyes of its employees, customers, venders, stockholders, financial journalists and the community in which it was embedded, a special case, an exception to the general rules governing traditional banking. Well, maybe not so much an exception to the rules (for so far as I know, in the years and months leading up to the catastrophe SVB pretty much stayed within formal banking parameters), but exceptions to the norms that constitute the common wisdom of bank operations.

Consider some of these “exceptions to the general rules/norms” (in no particular order of importance as I have no idea the extent to which each was contributory to the SVB story, just that each was).

First. Concentration of its assets (primarily loans, cash, securities) and its liabilities (primarily customer deposits) in one generic business world (early stage and mature tech, venture capital funds). I once invested in and sat on the board of a new community bank that was thought to have great promise. It actually flourished until the Great Recession of ‘08/’09/’10 caused the bank to close at 5:00 PM on a certain day and open the next day with a new name as the branch of a large regional bank. This radical change was the consequence of a decision made by our regulators that we were no longer safe and sound. And the reason? We had a loan portfolio substantially concentrated in real estate. Our borrowers, largely real estate investors and developers, were themselves feeling the pain of the recession and we had, therefore, a substantial number of delinquencies and write offs. (Note that had our depositor universe been concentrated among real estate companies we would have had double trouble, as was the case with SVB – tech companies were their primary depositors and borrowers.) Our regulators noted the bank’s condition and concluded, rightly, they had no choice but to step in and “solve” the problem - which they did on behalf of themselves (no FDIC claims) and our customers (no risk of losses above the insured deposit amounts) at a cost to our shareholders of a 100% loss of equity. In this case, the Feds arrived before any rumors, whispers (and tweets) circulated and so…no run.

Substantial concentration of loans to and deposits from one class of borrower brings with it special risks, a risk most banks with a diversified base of borrower and loan types don’t run. SVB was THE tech sector bank. It took pride in that role. When Silicon Valley prospered, SVB prospered, mightily. When Silicon Valley business slowed, when venture funding waned and when it became apparent that the cash reserves of many companies were going to have to stretch longer than planned, and in some cases much longer, the stage was set for a huge (albeit not at first obvious) run. “Have you heard that SVB may be having some problems?” (In fact, there had been some serious regulatory scrutiny during 2022.) “We have millions of dollars in deposits there which we cannot put at risk. Even if what we are hearing is wrong, prudence dictates we take no chances. Move the money.” (Customers withdrew $42 billion on one day in early March, leaving the bank with a negative $1 billion dollar cash balance.) Parenthetically, it seems there was a third area of concentration about which there is still not much known and so one about which I have only one comment. Banks have depositors, typically individuals and business. In the aggregate, even the smallest bank has a large number of such customers. But in the case of SVB, many of its business and individual depositors were subject, to some greater or lesser extent, to the influence exerted by a relatively small number (a concentrated number) of venture funds and investment banks who were their financiers and advisors. So decisions to withdraw funds seems to have been made by a small number of “influencers” rather than by the materially larger number of depositors they represented.

Second. Its uses of cash, or can there be too much of a good thing? Wine is a good thing (but you know what happens when you have too much of it). Strawberries are a good thing (but I once spent three days in bed with hives covering my body after eating a couple of cartons of them). Rain is a good thing (until the floods come). And cash is a good thing (but SVB’s cash position, too much of a good thing as it turned out, was a major contributor to failure). Why?

Late in what we now know was the era of low interest rates, SVB found itself with billions of dollars in cash arising from a flood of deposits owned by its tech, and particularly its venture funded, customers. Normally, banks invest in a variety of loans which then constitute the bulk of their assets. Overly simply and ignoring for the moment the various other ways in which banks might make money, it is from interest earned on their loan portfolios that banks derive the bulk of their revenue. Interest income from loan portfolios provides, in turn, funds to pay operating expenses and interest on deposits. In most cases, not all, and in most times, not all, the margin between money in and money out is positive.

Finding good borrowers can be a tedious process. Underwriting loans even to good borrowers can be a tedious process. From our vantage point today we can understand why, as deposits outran loan opportunities, SVB sought out other investments for its billions of dollars in cash. (I am going to leave aside the question of the quality of the SVB loan portfolio except to say that, in some cases at least, conventional credit standards seem to have been bypassed in the case of many of its borrowers. In these cases, “collateral” consisted in the borrowing company’s actual or perceived ability to generate increasing revenue, which could include expected VC proceeds/pending capital call receivables. Nothing wrong with that, always assuming that the lender has the technical expertise to underwrite the risks, which SVB had.)

Conservative bankers may shudder, but remember, this is Silicon Valley. Most companies backed by SVB had proved worthwhile credit risks over years, even decades. Why wouldn’t banking practices which have proven so successful and so profitable not continue to be so?

So what did the bank do with most of that cash (the amount left after holding some cash for withdrawals and some for loan commitments pending)? It bought Treasury securities.

On 12-31-22 SVB had about $211 billion in assets, largely securities ($120 billion) and loans ($74 billion). Included among the assets were about $14 billion in cash and equivalents, reserved to cover possible withdrawals. Red flag. Total deposits were about $173 billion - of which about $80 billion were in non-interest-bearing accounts. Granted that some of this non-interest-bearing money was for depositor’s current use, but a material amount in those accounts was the result of build up over the years when no or nominal interest was paid by banks. So it didn’t much matter to depositors that these accounts paid no interest. It didn’t much matter, that is, until interest rates began to rise.

We will never know to what extent a careful reader of SVB’s year-end financial statement, after noting that some significant portion of its cash assets were in non-interest-bearing accounts, concluded that some/much of that money was vulnerable, as interest rates were rising, to being moved away from the bank to money market funds or bonds. That careful reader might also have noted that cash and equivalents available to cover withdrawals were only around $14 billion. In fact, any significant withdrawals, for whatever reasons, would require the bank to sell securities to meet withdrawal demands. Not necessarily a bad thing, just a bad thing in light of the character of SVB’s securities portfolio.

Third. For years interest rates have been low to nominal. Home mortgages dipping below 3%. Commercial real estate mortgages in the 3’s. Business loans in the 3’s. And, of course, bond yields reflected this. Yields on five-year Treasury bonds in 2021, for example, ranged from 0.36% to 1.34%.

In later years, as cash accumulated on the SVB balance sheet in amounts in excess of what it could comfortably lend, the bank bought Treasury securities. Timing was, in retrospect, unfortunate. The bulk of the buys occurred toward what we now know was the end of the low interest rate era. Had interest rates stayed low, no problem. But should interest rates begin to rise (as they in fact did) any sale of those Treasuries bought at low interest rates would have serious consequences.

Sidebar for those not altogether clear why this should be so. I buy a bond yielding 1.5% for $1,000 (known as par). My dollar yield is $15 per year. Time passes and a new issue of the same bond carries a 3% yield. Should I wish to sell my bond, the buyer has a fancy calculation to make. She thinks, rightly, she should pay less than $1,000 (as she will receive only $15 yearly vs the $30 yearly receivable if she bought the new issue), but she also knows that if my bond is bought at a discount it will, nevertheless, be redeemed at par when the bond comes due and so generate a capital gain, calculated from the discounted price paid to par. How to determine the yield to maturity (the coupon yield and the ultimate gain if bought at a discount - or loss if bought at a premium) is not as easy as it looks. But, the moral to the story: if you buy a bond when interest rates are low you can expect the price of the bond to drop if interest rates rise (unless maturity is close at hand, which SVB’s were not).

In fact, the bond market will take care of (re)pricing bonds on a daily (actually, minute by minute) basis, taking into account coupon yields, current yields, yields to maturity, credit quality and so on. And to the extent any entity owns bonds, their fair market value will fluctuate accordingly. Such periodic changes in valuation will be reflected on the entity’s financial statement.

Unless you are a bank which happens to not be a Systemically Important Financial Institution (SIFI), as was SVB.

To simplify. The Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank) became law following the onset of the Recession of ‘08/’09/’10 and beyond. It was a complex set of regulations designed to strengthen financial institutions in the hopes of preventing another meltdown. For our purpose, there was one feature that bears on the SVB story. All banks were required to mark their securities to market to reflect changes in market value as they occurred vs the traditional practice, insofar as bonds were concerned, of carrying them at par. However, after much intense lobbying (and with bipartisan support) this provision of Dodd-Frank was modified during the last administration to allow banks deemed to be non-SIFI to revert to the traditional practice of carrying bonds expected to be held to maturity at par. For the significance of this modification, read on.

In the face of the whispers and rumors (and perhaps some cold-blooded financial statement analysis), depositors began to withdraw their funds in amounts which quickly blew through SVB’s cash and equivalents. In order to meet withdrawal demands SVB was forced to sell bonds, primarily Treasures, at a loss. What had been an unrealized loss, but not a capital impairment under Dodd-Frank, became a $1.2 billion realized loss, a material diminishment of bank capital - enough, with the run gaining steam, to bring in the Feds.

There are other ingredients in this story but these three will suffice for openers. Postmortems (in the books to be written and the MBA case studies to be assigned) will flesh out the details. All the elements (bank history, management successes and failures, the role certain prominent VC’s played in stimulating the run, the Silicon Valley business climate past and present, the actions of regulators before and during the crisis and, most importantly, the “special case” culture of SVB, all this and more will be parsed for the purposes of assigning blame (of course) and for what changes need to be made in the financial services regulatory environment.

And speaking of blame, given the current political tendency to weaponize whatever can be weaponized, I note that during the current Senate hearings in regard to the SVB failure one Senator has said that the blame lies with the regulators who, instead of minding the regulatory store, were distracted by their focus on global warming. I know for a fact that this is not so. To the extent the regulators were distracted it was because they spent too much time scanning the heavens for UFO sightings.

Chris Weil

This communication may contain privileged and confidential information; people other than the addressee should not review, distribute or duplicate it without permission. Nothing in this communication constitutes a solicitation by us for the purchase or sale of any securities. We do not accept account orders or instructions by e-mail, and will not be responsible for carrying out e-mailed orders or instructions. We provide reports as an accommodation to help you monitor your investment activity; securities pricing may not reflect reliable values. In the event of a discrepancy, the information in your confirmations of daily activity and monthly statement of account shall govern. While the information in this communication comes from sources believed to be reliable as of today, we make no representation as to its accuracy and completeness and provide no assurances as to future returns or performance. We may own positions in securities mentioned in this communication. Investing involves risks, including the possible loss of the principal amount invested. There can be no assurance that recommended investments will be successful in meeting their objectives. Investment in mutual funds is also subject to market risk, investment style risk, investment adviser risk, market sector risk, equity securities risk, and portfolio turnover risks. More information about these risks and other risks can be found in the fund prospectus. You may obtain a prospectus for CWC's mutual funds by calling us toll-free at 888.550.9266 or visiting The prospectus should be read carefully before investing. CWC's mutual funds are distributed by Arbor Court Capital, LLC, Member FINRA/SIPC. Nothing herein should be construed as legal or tax advice. You should consult an attorney or tax professional regarding your specific legal or tax situation. Christopher Weil & Company, Inc. may be contacted at 800.355.9345 or

451 views0 comments


bottom of page