The Silicon Valley Bank failure strikes me as a colossal failure of bank regulation, and instructive on how rotten the whole edifice is. I write this post in an inquisitive spirit. I don't know the details of how SVB was regulated, and I hope some readers do and can chime in.
As reported so far by media, the collapse was breathtakingly simple. SVB paid a bit higher interest rates than the measly 0.01% (yes) that Chase offers. It attracted large deposits from venture capital backed firms in the valley. Crucially, only the first $250,000 are insured, so most of those deposits are uninsured. The deposits are financially savvy customers who know they have to get in line first should anything go wrong. SVB put much of that money into long-maturity bonds, hoping to reap the difference between slightly higher long-term interest rates and what it pays on deposits. But as we've known for hundreds of years, if interest rates rise, then the market value of those long-term bonds fall. Now if everyone comes asking for their money back, the assets are not worth enough to pay everyone back.
In sum, you have "duration mismatch" plus run-prone uninsured depositors. We teach this in the first week of an MBA or undergraduate banking class. This isn't crypto or derivatives or special purpose vehicles or anything fancy.
Where were the regulators? The Dodd Frank act added hundreds of thousands of pages of regulations, and an army of hundreds of regulators. The Fed enacts "stress tests" in case regular regulation fails. How can this massive architecture fail to spot basic duration mismatch and a massive run-prone deposit base? It's not hard to fix, either. Banks can quickly enter swap contracts to cheaply alter their exposure to interest rate risk without selling the whole asset portfolio.
Michael Cembalist assembled numbers. This wasn't hard to see.
Even Q3 2022 -- a long time ago -- SVB was a huge outlier in having next to no retail deposits (vertical axis, "sticky" because they are insured and regular people), and a huge asset base of loans and securities.
Michael then asks
.. how much duration risk did each bank take in its investment portfolio during the deposit surge, and how much was invested at the lows in Treasury and Agency yields? As a proxy for these questions now that rates have risen, we can examine the impact on Common Equity Tier 1 Capital ratios from an assumed immediate realization of unrealized securities losses ... That’s what is shown in the first chart: again, SVB was in a duration world of its own as of the end of 2022, which is remarkable given its funding profile shown earlier.
Again, in simpler terms. "Capital" is the value of assets (loans, securities) less debt (mostly deposits). But banks are allowed to put long-term assets into a "hold to maturity" bucket, and not count declines in the market value of those assets. That's great, unless people knock on the door and ask for their money now, in which case the bank has to sell the securities, and then it realizes the market value. Michael simply asked how much each bank was worth in Q42002 if it actually had to sell its assets. A bit less in each case -- except SVB (third from left) where the answer is essentially zero. And Michael just used public data. This is not a hard calculation for the Fed's team of dozens of regulators assigned to each large bank.
Perhaps the rules are at fault? If a regulator allows "hold to maturity" accounting, then, as above, they might think the bank is fine. But are regulators really so blind? Are the hundreds of thousands of pages of rules stopping them from making basic duration calculations that you can do in an afternoon? If so, a bonfire is in order.
This isn't the first time. Notice that when SBF was pillaging FTX customer funds for proprietary trading, the SEC did not say "we knew all about this but didn't have enough rules to stop it." The Bank of England just missed a collapse of pension funds who were doing exactly the same thing: borrowing against their long bonds to double up, and forgetting that occasionally markets go the wrong way and you have to sell to make margin calls. (That's week 2 of the MBA class.)
Ben Eisen and Andrew Ackerman in WSJ ask the right question (10 minutes before I started writing this post!) Where Were the Regulators as SVB Crashed?
“The aftermath of these two cases is evidence of a significant supervisory problem,” said Karen Petrou, managing partner of Federal Financial Analytics, a regulatory advisory firm for the banking industry. “That’s why we have fleets of bank examiners, and that’s what they’re supposed to be doing.”
The Federal Reserve was the primary federal regulator for both banks.
Notably, the risks at the two firms were lurking in plain sight. A rapid rise in assets and deposits was recorded on their balance sheets, and mounting losses on bond holdings were evident in notes to their financial statements.
moreover,
“Rapid growth should always be at least a yellow flag for supervisors,” said Daniel Tarullo, a former Federal Reserve governor who was the central bank’s point person on regulation following the financial crisis...
In addition, nearly 90% of SVB’s deposits were uninsured, making them more prone to flight in times of trouble since the Federal Deposit Insurance Corp. doesn’t stand behind them.
90% is a big number. Hard to miss. The article echoes some confusion about "liquidity"
SVB and Silvergate both had less onerous liquidity rules than the biggest banks. In the wake of the failures, regulators may take a fresh look at liquidity rules,...
This is absolutely not about liquidity. SBV would have been underwater if it sold all its securities at the bid prices. Also
Silvergate and SVB may have been particularly susceptible to the change in economic conditions because they concentrated their businesses in boom-bust sectors...
That suggests the need for regulators to take a broader view of the risks in the financial system. “All the financial regulators need to start taking charge and thinking through the structural consequences of what’s happening right now,” she [Saule Omarova] said
Absolutely not! I think the problem may be that regulators are taking "big views," like climate stress tests. This is basic Finance 101 measure duration risk and hot money deposits. This needs a narrow view!
There is a larger implication. The Fed faces many headwinds in its interest rate raising effort. For example, each point of higher real interest rates raises interest costs on the debt by about $250 billion (1 percent x 100% debt/GDP ratio). A rate rise that leads to recession will lead to more stimulus and bailout, which is what fed inflation in the first place.
But now we have another. If the Fed has allowed duration risk to seep in to the too-big to fail banking system, then interest rate rises will induce the hard choice between yet more bailout and a financial storm. Let us hope the problem is more limited - as Michael's graphs suggest.
Why did SVB do it? How could they be so blind to the idea that interest rates might rise? Why did Silicon Valley startups risk cash, that they now claim will force them to bankruptcy, in uninsured deposits? Well, they're already clamoring for a bailout. And given 2020, in which the Fed bailed out even money market funds, the idea that surely a bailout will rescue us should anything go wrong might have had something to do with it.
(On the startup bailout. It is claimed that the startups who put all their cash in SVB will now be forced to close, so get going with the bailout now. It is not startups who lose money, it is their venture capital investors, and it is they who benefit from the bailout.
Let us presume they don't suffer sunk cost fallacy. You have a great company, worth investing $10 million. The company loses $5 million of your cash before they had a chance to spend it. That loss obviously has nothing to do with the company's prospects. What do you do? Obviously, pony up another $5 million and get it going again. And tell them to put their cash in a real bank this time.)
How could this enormous regulatory architecture miss something so simple?
This is something we should be asking more generally. 8% inflation. Apparently simple bank failures. What went wrong? Everyone I know at the Fed are smart, hard working, honest and dedicated public servants. It's about the least political agency in Washington. Yet how can we be seeing such simple o-ring level failures?
I can only conclude that this overall architecture -- allow large leverage, assume regulators will spot risks -- is inherently broken. If such good people are working in a system that cannot spot something so simple, the project is hopeless. After all, a portfolio of long-term treasuries is about the safest thing on the planet -- unless it is financed by hot money deposits. Why do we have teams of regulators looking over the safest assets on the planet? And failing? Time to start over, as I argued in Towards a run free financial system
Or... back to my first question, am I missing something?
****
Updates:
A nice explainer thread (HT marginal revolution). VC invests in a new company. SVB offers an additional few million in debt, with one catch, the company must use SVB as the bank for deposits. SVB invests the deposits in long-term mortgage backed securities. SVB basically prints up money to use for its investment!
"SVB goes to founders right after they raise a very, very expensive venture round from top venture firms offering:
- 10-30% of the round in debt
- 12-24 month term
- interest only with a balloon payment
- at a rate just above prime
For investors, it also seems like a no-downside scenario for your portfolio: Give up 10-25 bps in dilution for a gigantic credit facility at functionally zero interest rate.
If your PortCo doesn't need it, the cash just sits. If they do, it might save them in a crunch. The deals typically have deposit covenants attached. Meaning: you borrow from us, you bank with us.
And everyone is broadly okay with that deal. It's a pretty easy sell! "You need somewhere to put your money. Why not put it with us and get cheap capital too?"
Update:
1) Old Eagle Eye's comment below is fascinating. I am getting the sense that the rules actually preclude putting 2+2=4 together here. Copied here in toto
SIVB did have a hedge put on during 2022, but it was limited to its available-for-sale securities ("AFS"). It was precluded from hedging its interest rate risk in held-to-maturity securities ("HTM") by U.S. GAAP rules. [My emphasis] Here is the explanation found at PwC:
[PWC Viewpoint Commentary: "The notion of hedging the interest rate risk in a security classified as held to maturity is inconsistent with the held-to-maturity classification under ASC 320, which requires the reporting entity to hold the security until maturity regardless of changes in market interest rates. For this reason, ASC 815-20-25-43(c)(2) indicates that interest rate risk may not be the hedged risk in a fair value hedge of held-to-maturity debt securities." "ASC 815-20-25-12(d) provides guidance on the eligibility of held-to-maturity debt securities for designation as a hedged item in a fair value hedge."]
[Extracted subsection:
"Chapter 6: Hedges of financial assets and liabilities.
"6.4 Hedging fixed-rate instruments
"6.4.3.4 Hedging held-to-maturity debt securities
"ASC 815-20-25-12(d)
"If the hedged item is all or a portion of a debt security (or a portfolio of similar debt securities) that is classified as held to maturity in accordance with Topic 320, the designated risk being hedged is the risk of changes in its fair value attributable to credit risk, foreign exchange risk, or both. If the hedged item is an option component of a held-to-maturity security that permits its prepayment, the designated risk being hedged is the risk of changes in the entire fair value of that option component. If the hedged item is other than an option component of a held-to-maturity security that permits its prepayment, the designated hedged risk also shall not be the risk of changes in its overall fair value."]
Source: PWC Viewpoint (viewpoint.pwc.com) Publication date: 31 Jul 2022
https://viewpoint.pwc.com/dt/us/en/pwc/accounting_guides/derivatives_and_hedg/derivatives_and_hedg_US/chapter_6_hedges_of__US/64_hedging_fixedrate_US.html
FDIC usually does resolution of smaller banks on a weekend. They come in on a Friday, close the doors, and the bank employees become FDIC employees. Assets and liabilities are sorted out and, in the best case scenario, a buyer is found and the bank reopens on Monday.../ No hawkish talk anymore by FED.. Brgds
ReplyDeleteSome claim that failure to make depositors whole will hurt innovation by depriving companies of funding. But how's this possible when VC's have hundreds of billions of dollars in dry powder (committed but not called capital). Can't solid portfolio companies just line up funding elsewhere? What am I missing? -Marc F.
ReplyDeleteEgads, and John Cochrane is right.
ReplyDeleteMy usual two cents is that if commercial banks want federal insurance, they must also have a fat layer of convertible bonds.
Let the market price the convertible bonds.
If the bank is not viable, then the convertible bond holders take over the bank.
This is an excellent idea! And it already applies to the nation's largest banks. FDIC and others have been clamoring to bring this requirement to regional banks. With all this we may yet see that happen!
DeleteIn terms of the regulators, it was everybody and nobody. This was a California state-chartered bank, so technically it was regulated by the California Department of Financial Protection and Innovation, which was actually the institution which declared it closed on Friday before turning it over to the FDIC. (It is worth noting that four of the biggest banks to come under stress on Friday (SVB, PacWest, First Republic, and Silvergate) were California state-chartered banks technically overseen by the CDFPI.) But as a FDIC member bank SVB also had to meet FDIC requirements and regulations, and as a Fed member bank it also had to adhere to Fed requirements and regulations. The Fed generally takes the supervisory lead for state-chartered banks if they are Fed members. Of course all of these banks also have to meet separate consumer regulatory standards from the Consumer Financial Protection Bureau. SVB's parent company, SVB Financial Group, was also a bank holding company, which actually subject it to a separate set of regulations. As usual, the U.S.'s obscene bank regulatory system with multiple overlapping regulators makes it hard to say who was supposed to be watching the shop.
ReplyDeleteAnd Silvergate obviously came under stress earlier and said it would wind-down on Wednesday, but it was indeed a CA state-chartered bank like the others.
DeleteAs a former director of 3 FDIC insured banks I will say this is a very good summary by Judge Glock. Small community banks actually have better risk management than large banks and the regulators are less lenient.
DeleteSVB was aFed member. As I wrote on March 17, “ But it is frankly unbelievable that the Fed and/or FDIC didn’t know what was going on. They looked at the situation and realized that the bank needed to be put out of its misery before causing any more damage. The best solution was exactly what happened. Close the bank, call in FDIC and let the professionals sort everything out.” The NY Times has now confirmed this. https://www.nytimes.com/2023/03/19/business/economy/fed-silicon-valley-bank.html?smid=nytcore-ios-share&referringSource=articleShare
Deletethis might help:
ReplyDeletehttps://www.levernews.com/svb-chief-pressed-lawmakers-to-weaken-bank-risk-regs/
This is the key point. The issue wasn’t incompetent regulators, it was defanged regulators. And Lael Brainard was Nostradamus… https://braddelong.substack.com/p/reading-now-why-did-joe-biden-pick
DeleteKey point of course! Maybe, this forum should focus more on regulation instead of trying to ban it in favor of growth? Regulation is a necessary evil when there is lack of honesty/transparency/information.
DeleteThere is another angle to this. SVB did not have a Chief Risk Officer (CRO) between April 2022 and January 2023 (the CRO left for reasons which were not clear based on public communications). I know from my own experience that regulators don't want to see the banks they regulate without a CRO. They also want to understand the exact reasons for a CRO departure and want to have this position filled asap, ie., which could mean an internal interim CRO. I am not aware they had one (please, correct me if they had an interim CRO. The new one joined in January). I would be interested in the composition of the Asset-Liability Committee (ALCO) during this period.
ReplyDeleteI would like to provide a brief comment on the statements that have been made by some journalists, economists etc that this is not a "Lehman moment" and an isolated event of a tech bank. Yes, I assume and hope it is much more contained. However, some of the statements (eg., Krugman) created a false equivalency between a tech company and a tech bank going bust. We have bank-run problem, hopefully somewhat contained. I walked into my little local JP Morgan Chase branch this morning. I was told that they had never experienced such a huge sudden demand for new accounts and money transfers into their accounts before over the last 48 hours and had to ask people who had time off to come in for support. The Fed has arranged an emergency meeting for Monday. Let's see.
ReplyDeleteI am commenting from Australia so not an expert on US regulation. My understanding however is that SVB was exempt from stress testing because it was under a size threshold. I believe this exemption is driven by the US community bank association which has argued successfully that their member banks not be subject to the full force of the international banking standards that Dodd Frank translates into U.S. banking regulation.
ReplyDeleteYes, please see article on SVB's successful lobbying efforts
Deletehttps://www.levernews.com/svb-chief-pressed-lawmakers-to-weaken-bank-risk-regs/
How can anyone believe this duration mismatch story?! It’s so easy to regulate and check. This can’t just be a duration mismatch. This must be a sorry of corruption gone bad. Gave money to startups, received kickbacks…. Depositors’ $250k is guaranteed. How can a regulated bank be allowed to give money to startups?!
ReplyDeleteTo your point, Greg Becker, the CEO of SVB, was on the Board of Directors of the San Francisco Fed until March 10. Could that have affected the regulators?
DeleteThere is nothing worse than corruption going bad.
DeleteAside from the long bonds held at Fed it is unclear where a lot of the long term low coupon bonds are held, as helpfully shepherded by QE. Insurance companies and annuity providers are supposed to be sophisticated, but how many reached for yield down the curve too? Or is it just pensioners in their 401k who followed their dime-store financial advisor's advice to shift into bonds after age 50 who are taking the hit?
ReplyDeleteAnother question: I would love a Cochrane post on is the mechanics of the 7.8 trillion dollar short term debt market which rolls over this year (< 1 year maturity) -- should congress stupidly fail around the debt ceiling. Seems like a > 31B/day failure to redeem/roll this debt will be catastrophic to cash and money markets, especially again those which took deposits from the most risk averse investors. It is not clear fund managers have the tools to prevent runs and chaotic redemptions, or losses would be reversable, even should congress wish to later make investors whole. Does the Fed step in and buy defaulted bills at par? Can it? Is there a way to make these markets robust?
Yellen has been at both TY and Fed, does she have a plan, or is the strategy to throw the steering wheel out of the window in a game of chicken? Better make sure the opponent sees it. Everyone talks about armageddon as if that will make congress act, but betting that there are limits on congressional nuttiness seems to have turned into a questionable bet.
From an anti-corruption perspective I understand why you like narrow banking, but it's unclear to me why you like them from a financial stability perspective. Doesn't it just put a bunch of eggs in the fed basket and ignore the correlated risk?
ReplyDeleteWhat's extraordinary to me is that banks don't have mechanisms in place to stop a bank run in progress, e.g. they don't precommit to a three day individual bank holiday in the event of a run to give themselves time to shore up capital Or they could reserve the right to place a hefty fee on withdrawals in the midst of a run. Any number of mechanisms could help, but they don't seem to have any in place.
My understanding is that the SVB stock was still trading at 40% of its prior value before the regulators took it over. So while they were underwater from the perspective of assets minus liabiliites, they were not bankrupt from the perspective of the value of the company minus liabilities and should have been able to find a buyer. There are rumors now that the regulators will try to block any systemically important bank from taking over SVB, consumer welfare be damned.
Two possible market mechanisms that could help uninsured depositors: the first would be a central exchange for receivership certificates, i.e. they should be easily tradable rather than require wheeling and dealing without a clear market price. The second would be to require the government to immediately auction off SVB to any bank willing to pay more than $1 to take over all of SVB's liabilities. Hopefully the latter in some form will happen quickly, but it ought to be more automatic and less subject to regulatory whim. While the level of risk SVB took was obviously deadly, all depositors may yet be made whole and the stock price would seem to indicate that in a free market SVB would have been quickly taken over by a larger bank at no cost to the depositors. They had incredibly valuable contractual relationships with VCs and start up founders that should be of value to a bank like Goldman Sachs if only they are allowed to buy SVB.
One thing that doesn’t make sense though is that the depositors only came asking for their money on Thursday and only as a result of the loss realization in the bond portfolio and the inability to raise equity to cover the hole. So why did they realize the 20bn bond sale in the first place. And why did Moody’s call svb on Wednesday to tell them they will take the marks ina held to maturity portfolio and downgrade them (when BofA for example didn’t get that call and has a much larger portfolio). The facts as reported suggest the opposite of what is implied in the article: the loss triggered the run and not the run the realization of the loss. Any clarity would be much appreciated
ReplyDeleteI am not buying the published story. I am not a Econ PhD, but I knew better than to buy bonds having than two years maturity. There are a lot of investments easily available to banks that do not carry significant interest rate risks. Further there are lots of hedging options available.
ReplyDeleteMy hunch is that there is a deeper story that involves fraud and theft that really did this bank in. But, it will be a while before the indictments reveal it.
They were trying to raise funds when they were shut down. Too bad they didn't do that before. Presumably the firm had plenty of valuable non-financial assets: the organization, the clients, the tech, the brand, etc. So raising capital should have been possible, but you can't wait until a crisis manifests.
ReplyDeleteYou might be unaware that SVB President Becker back in 2015 successfully lobbied Congress to raise the asset level for banks to be covered by enhanced Dood-Frank regulations. https://amp.theguardian.com/business/2023/mar/11/silicon-valley-bank-weaken-risk-regulations-svb
ReplyDeleteLosses for startups/VCs are actually probably not very high so no need to bail them out at all in any case. If 90% of assets have an average residual maturity of 5 years and yield 1.5%, and equity is 4%, how much do the loss for uninsured depositors (90% of all deposits) represent? Back of the envelope says around 10% or so - bad but not catastrophic for high potential firms with lots of inherent volatility. It’s more a matter of managing panic right now.
ReplyDeleteWhere were the short sellers? If this story is really about corruption, why was everyone in on it?
ReplyDeleteTwo points:
ReplyDeleteAccording to the data on Yahoo, sometimes incorrect, as of March 9 -- the last day data is available, SVB had according to the stock market an equity value of 16.7 billion and traded at a 36% premium to book so it seems unlikely that the problems were so easy to spot.
Anyone who has worked in the US government knows that the worker bees are not there in order to work hard and produce a lot. That is just not the nature of most, though not all, people who end up as government employees. That reality needs to be observed in setting up a regulatory system. Some countries, perhaps Singapore, get some of the best and brightest in government employment. That is not the reality of the US system.
You people have failed to report on the REAL story.
ReplyDeleteDid the employees of SVB address their customers by their proper pronouns?
Maybe St. Greta can investigate this.
Why didn't they hedge their portfolio? Even with the bad management of the mismatch, if they'd at least gone into the futures market and sold STIRs, or laddered the futures position to match their cash bond position. They still would have lost a little money, but they wouldn't be out of business.
ReplyDeleteThere is no point of buying long bonds and then hedging them - the resulting yield will be little different from the yield on treasury bills, so it is simpler just to hold treasury bills.
DeleteSIVB did have a hedge put on during 2022, but it was limited to its available-for-sale securities ("AFS"). It was precluded from hedging its interest rate risk in held-to-maturity securities ("HTM") by U.S. GAAP rules. Here is the explanation found at PwC:
Delete[PWC Viewpoint Commentary: "The notion of hedging the interest rate risk in a security classified as held to maturity is inconsistent with the held-to-maturity classification under ASC 320, which requires the reporting entity to hold the security until maturity regardless of changes in market interest rates. For this reason, ASC 815-20-25-43(c)(2) indicates that interest rate risk may not be the hedged risk in a fair value hedge of held-to-maturity debt securities." "ASC 815-20-25-12(d) provides guidance on the eligibility of held-to-maturity debt securities for designation as a hedged item in a fair value hedge."]
[Extracted subsection:
"Chapter 6: Hedges of financial assets and liabilities.
"6.4 Hedging fixed-rate instruments
"6.4.3.4 Hedging held-to-maturity debt securities
"ASC 815-20-25-12(d)
"If the hedged item is all or a portion of a debt security (or a portfolio of similar debt securities) that is classified as held to maturity in accordance with Topic 320, the designated risk being hedged is the risk of changes in its fair value attributable to credit risk, foreign exchange risk, or both. If the hedged item is an option component of a held-to-maturity security that permits its prepayment, the designated risk being hedged is the risk of changes in the entire fair value of that option component. If the hedged item is other than an option component of a held-to-maturity security that permits its prepayment, the designated hedged risk also shall not be the risk of changes in its overall fair value."]
Source: PWC Viewpoint (viewpoint.pwc.com) Publication date: 31 Jul 2022
https://viewpoint.pwc.com/dt/us/en/pwc/accounting_guides/derivatives_and_hedg/derivatives_and_hedg_US/chapter_6_hedges_of__US/64_hedging_fixedrate_US.html
Greg Becker hasa BS in business from U. of Indiana. His LinkedIn profile shows absolutely no qualifications to be CEO of any bank. This was caused by rank incompetence coupled with arrogance. See https://gonzoecon.com/2023/03/silicon-valley-bank/ for more details.
Delete"Why did SVB do it? How could they be so blind to the idea that interest rates might rise?"
ReplyDeleteAs I understand it, a large part of the problem may be that SVB did not have a risk officer during this period, as reported by Forbes (https://fortune.com/2023/03/10/silicon-valley-bank-chief-risk-officer/). Laura Izurieta left in April 2022 and Kim Olson didn't start until January 2023.
One question that I believe is left unanswered by the regulation is the personal incentives and career risks of the risk officers involved. When they get early warnings that the train is starting to go off the rails, it may be easier to leave early than to stay and try to get it back on track. Especially, if they have significant personal wealth tied up in stocks in the company / bank (as it appears to have been the case)
If anything, it may be useful to revisit the regulation to require that there is always a responsible / liable risk officer on board with appropriate incentives.
What a shitshow…
ReplyDeleteI love you guys. The free market is great until careless people loose money and then it is “where are the regulators?” In this case they were exactly where the smaller banks have lobbied for them to be - not regulating. If you are a depositor putting millions in deposits in a bank, you are supposed to know what is insured and what isn’t and how to protect your funds. The best and the brightest apparently can’t read a balance sheet or don’t care.
ReplyDeleteThere was no "collapse of pension funds" in the UK - I do not think you can find a single example of a collapsed pension fund since last autumn. In fact, the pension funds as a whole as in a much better funding position than they were a year ago and the same is true for most funds individually. There were some disorderly hedge liquidations in autumn and a few funds lost money in these, but all pension funds are structurally short bonds. These positions may be relatively small for fully hedged funds and large for partially hedged ones, but there are no overhedged funds, so the rise in rates has been a boon and this far outweighs the hedge unwind costs for the few funds caught up in that.
ReplyDeleteThe chart by and link to Mr Cembalist provided great context. Was comforting to see that I was not the only one wondering how Treasuries could present a "liquidity" problem. In case anyone is optimistic about a government response to this simple/obvious problem, Senator Warner said on This Week today that there is not much regulation can do if you don't get Banking 101 (interest-rate risk) right. Hey! Maybe regulators could look at interest-rate risk!!
ReplyDeleteAs I understand it, the 2018 rollback of regulations all stress testing of banks with less than $250B at the discretion of the regulators. It also seems there should have been enough red flags to warrant this.
ReplyDeleteThe government should get out of the deposit insurance business. If there were private deposit insurance with risk weighted premiums: anyone who wanted their deposits covered could do so rather than dealing with this $250k limit. The private insurance entities would have incentive to keep close tabs on the actual risk and spot concerns ahead of time. If there is some flaw in this suggestion: it still seems likely that some private approach could be designed that provides incentives for better watchdogs than the federal bureaucracy.
ReplyDeleteThe $250k limit seems to be ignored because too many people just assume its academic and the government will keep their deposits safe above that (or mostly). There are actually tools like an ICS to let companies spread deposits among multiple banks to avoid risk: but people likely don't take the risk seriously enough to do so.
re: "Obviously, pony up another $5 million and get it going again. And tell them to put their cash in a real bank this time"
ReplyDeleteWhile on the surface that sounds like its only costing the VC money: actually its the entrepreneur who will need to give up more equity in their company. Yup: the entrepreneur's cash management was flawed though and they do deserve a hit for that, but the point is merely that its not only the VC that'll suffer. Also: that does use up cash that other future startups will not get because it was used to bail out existing companies so there are "unseen" people harmed by this. The next Apple or Google might not be funded because they needed to bail out their existing companies, which is an "unseen" and impossible to know potential cost.
Maybe those worried much about banks and their deposits might be better off buying gift cards from the likes of Amazon, Starbucks, etc! Just kidding !
ReplyDeleteDr Cochrane, I believe you are near-slandering the regulators. The thousands of pages of new bank regulation came down to a few key rules and tests. These included a Liquidity Coverage Ratio rule. In 2019, the threshold size of banks that fell under those rules was raised by Congress, per the lobbying of small-midsize banks like SVB. That change allowed them to speculate recklessly without so much accountability— Exactly as Lael Brainard publicly predicted it would, at the time. Please see https://braddelong.substack.com/p/reading-now-why-did-joe-biden-pick
ReplyDeleteIf there is no rule to watch duration mismatch 101, then yes we need one. And we need to burn the other 100,000 pages of rules if they can't enforce something so basic. Were there regulators saying "there is a huge duration mismatch but we don't have enough rules to stop it?" I don't think so, but would welcome a link.
DeleteThere is certainly room to wonder where the regulators were, but it’s odd to blame regulations that have in fact been quite effective for the banks subject to them.
DeleteSince 2000...the system has been RUN TO RESCUE THE RICHEST...while rural America Main street...can't afford to paint their houses. Take a trip to upstate NY, then the Hamptons! The First lesson of today's college...how to DEFAULT on your Student loan for your failed education! All non-profits...including colleges should be taxed if give anyone $100k+! End all federal aid for colleges and cities...make these entities SELF FUNDING!
ReplyDeleteWe have a centralization of power...where the richest now are rescued...as they GET RICHER!
The problem is you think these government entities want to do good! The IRS, FED, DOJ, FBI, etc....now are just arms of the Democrat Woke Fascists!
ReplyDeleteI smell a congressional testimony in the near future ;)
ReplyDeleteBailout of the unsecured depositors already started. Almost took the whole weekend...
ReplyDeletehttps://edition.cnn.com/2023/03/12/investing/svb-customer-bailout/index.html
SVB was the 'canary in the coal mine'. On Sunday, it was announced that Signature Bank has been closed. Furthermore, depositors in SVB and Signature will be made whole--all of them and every dollar on deposit with either bank. The Fed has opened a systemic risk facility offering to back any bank that is likely to or will be subject to a run on its deposits.
ReplyDeleteWhatever you might think of this development, in terms of moral risk, the Fed, the Dept. of Treasury and the FDIC are not willing to take a chance on the resilience of the U. S. banking system.
Will the FOMC get the message and pause the Fed-funds rate increase this week? If it doesn't, it will make the task of stemming the mid-sized bank failures that much more challenging given the pro-cyclical effect of the Fed-funds rate on the market value of T-bills, T-notes and MBS.
Re: "Update". Banking with just one bank is not atypical for small and medium-size commercial firms that lack a sophisticated treasury function within the firm. The V.P. finance is usually a CPA that has limited experience in money-management. He/she is often comfortable off-loading that function to the firm's bankers. The owner/founder/entrepreneur often doesn't have a background in finance, and is seldom capable of discerning one bank from another in terms of stability and probity. His venture capital backers may or may not have that capability or insight into the bank that they recommended. This much is clear from the panic on Thursday of last week. In a perfect world, this event (or, 'shock') would have been long anticipated and guarded against. As events have shown us, to err is human; yet, to forgive is other-worldly.
ReplyDeleteRight on, Old Eagle Eye. Innovators need to spend their time, and limited financial resources, innovating. Trusting a bank with administration of money, payroll etc. is a vital enabler. Having the confidence to trust your bank is what makes this system work. When that trust is violated we are at risk of bringing innovation to a standstill.
Deletewanted to post a quarterly plot of interest bearing and non-interest bearing deposits but no graphs here
ReplyDeleteNon-I peaked 4Q21 and held til 1Q22 then slid from 120+ billion to under 80 billion 1Q23 est
The notional value of their derivatives portfolio is a ? too
This is astonishing to me
In 2021 who could not see interest rate increases coming?
Also blythe talk of protecting "the poor depositors" is an 80 billion ticket
Well Signature bank is now closed by regulators as well. The Barney Frank bank as some call it, and a board member to boot. This may be the most ironic story in all of banking.
ReplyDeleteWhat about the fact that Kim Olsen, a former fed official, joined SVB as CFO in January 2023? Obviously she knew what was going on.
ReplyDeleteKim Olsen joined too late. After you have obtained access to the systems, reports etc, understood how this place works, etc it was too late. I am more puzzled that they had no CRO for most of 2022
DeleteI smell a congressional testimony coming up ;)
ReplyDeleteBy the way, here is a bank that seems to get it: https://mercury.com/vault
Also interesting - FDIC changed the format and procedure for Quarterly Financial Data SDI Data downloads on the 3rd quarter of 2022. Instead of one download file for ALL 64 categories of CALL reports, it was changed where it required a specific URL for each CALL category PLUS that URL required specific fields for the request.
ReplyDeleteNot sure why this 'streamlined' download change happened mid-year and 'why' adding more complexity to fetch and parse the quarterly filed data?
Mea Culpa! I posted this on the journal bias post. I meant to post it here.
ReplyDeleteJohn, as you stated, undergrad finance majors and in your province, MBA and PHD candidates are grounded in bond convexity and duration risk. All of the investment banks have intensive training training programs that teach financial engineering in terms of risk management. Forget the regulators! WTF were the risk managers? I explained this collapse to my 18 year old nephew. I demonstrated bond convexity arithmetic and the change in bond price formula with a simple excel spread sheet. He was astonished...forgive the hyperbole...at what a 100 bps increase in interest rates do to long dated bonds. When I told him the CME had been trading the 10 year t-note treasury for years, the most actively traded BTW, he began to understand how this debacle might have been avoided.
Dr. Cochrane, do you favor larger / more stringent reserve requirements? Do you think they would have helped prevent the SVB failure? Or were the requirements adequate, but the regulators asleep at the switch?
ReplyDeleteAndy Kessler pointed out in WSJ ("Who Killed Silicon Valley Bank?", March 12) that SVB has an unusual cash-flow pattern. Most bank deposit accounts are, on average, pretty stable. But venture-funded startups get a big chunk of cash which they steadily draw down. If the deal flow shrinks, then SVB's cash flow will rapidly turn very negative. This happens exactly when rates climb...
ReplyDeleteThe proximate cause of the failure of SIVB -- SVB Financial Group -- can be summarized as follows:
ReplyDeleteThe 2022 audited financials show a negative $15,152 million difference between fair market value of 'hold to maturity' ("HTM") securities and the amortized book value of those same securities. The consolidated group's book equity as at 12/31/22 comprised preferred shares ($3,646 million), common par value ($59 thousand), common paid-in ($5,318 million), retained earnings ($8,951), and accumulated other comprehensive loss (neg. $1,911 million). Total book value of common shares (incl. minority shareholding) was $12,358 million. Total book value of shareholders' equity was $16,004 million (down $232 million from the same period of 2021).
What was the reason that SVB Financial Group held a disproportionate amount of financial assets in HTM securities as at 12/31/2022?
The answer is obvious: The write down of HTM securities that reclassification of the securities would have severely penalized the book value of common equity. Subtracting the $15,152 million mark to market losses on conversion to AFS from the $12,358 million recorded book value of common equity would have resulted in negative $2,794 million common equity and left but positive $852 million in total equity on SVBFG's balance sheet as at 12/31/2022.
KPMG, the group's auditor gave the SVBFG an clean opinion. Yet, the management's analysis section (p. 49) shows that SVBFG's common equity as a percentage of average assets declined progressively from 8.25% (12/31/20) to 6.24% (12/31/21) to 5.75% (12/31/22). This statistic alone should have prompted the auditor to delve more deeply into SVBFG's status as a "Going Concern", esp. in light of the FMV of the HTM securities held on the books of the consolidated corporation.
SVBFG was under-capitalized at the end of 2022. It had an apparently strong franchise in the product-market sectors that it chose to compete in. It was undone by decisions to hold the majority of its liquid assets in 1-5 year and 5-10 year maturity government and agency debentures and notes, presumably made to reach for yield, when it ought to have stayed in the short-term end of the yield curve during 2022.
There is no indication at all that the extension of loans to VC partnerships played any part in the failure of SVBFG. The real estate loans are largely collateralized and liquid (insofar as real estate is a liquid asset). Had SVBFG been successful in recapitalizing equity, the failure on the 10th might never have occurred.
Was this a failure of management? Very likely so, esp. of the risk management committee members.
Source: SVB FINANCIAL GROUP, Form 10-K, For the fiscal year ended December 31, 2022
URL: https://d18rn0p25nwr6d.cloudfront.net/CIK-0000719739/f36fc4d7-9459-41d7-9e3d-2c468971b386.pdf
hey these people can't even TELL a MAN from WOMAN.
ReplyDeleteThey are religion is Democrat Fascism! WOKETOPEIA. Main Street has to stop allowing these OVERPAID criminal being rescued! Make the OTHER BANKS rescue them...from THEIR BONUS MONEY! Goldman FIRST!
A failure of bank regulation as an idea, or as practised? I wonder whether it _can_ work? It involves second-guessing millions of the bank's own decisions: isn't that what Mises and Hayek pointed out had to be done 'locally'?
ReplyDeleteWe think the regulators keep getting it wrong, or that there is too much "moral hazard", but I think the regulators are getting exactly what they want, they just keep calling it something else. In their statement announcing the SVB bailout, the regulators (Fed, Treasury, FDIC) say, "This step will ensure that the U.S. banking system continues to perform its vital roles of protecting deposits and providing access to credit to households and businesses." Providing "access to credit" means making illiquid risky investments --- going long credit is just selling puts on the underlying assets of the borrower after all (Merton Model) --- and "protecting deposits" means backstopping depositors. So, regulators (and many policymakers) define the "vital role" of the banking system to be a system that takes taxpayer-backstopped short-term deposits to make illiquid risky investments and, low and behold, that's exactly what we end up with! Btw, since taxpayer-backstopped short-term deposits are also known as "cash equivalents" and going short cash is also known as "using leverage", regulators actually consider the *ideal* bank to be one that makes levered risky illiquid bets. No wonder that every attempt at regulatory reform results in...banks making levered risky illiquid bets.
ReplyDeleteNow, of course, no respectable regulator or policymaker would say that the goal of regulation was to produce banks making levered risky illiquid bets, all backstopped by taxpayers. But, when they say they want a banking system that protects deposits and provides access to credit, they are saying the same thing, just using different semantics. They are suffering from some sort of semantic illusion.
We wont stop the bank runs and bailouts until regulators and policymakers stop saying that the "vital role" of the banking system is to provide access to credit, aka make risky investments. We have plenty of investment funds, not backstopped by taxpayers, to make those investments, e.g., funds that buy securitized loans and mortgages and certainly long-term treasuries. The most damning evidence that regulators and policymakers are pursuing the wrong target was when the Fed quashed narrow banks, which John correctly pounds the table on constantly. When regulators had the ideal bank (Narrow Bank) dropped in their laps, they quashed it! That bank was not going to "provide access to credit", aka make levered risky illiquid bets backed by taxpayers.
Should not the Fed have been aware of situations like this going forward when they started jacking up interest rates so quickly after such a long period of very low interest rates? Every bank may not go under but all banks have huge exposures because has mentioned treasuries are the safest investment product. How big is this bond problem anyway?
ReplyDeleteAll banks are illiquid. The duration of their assets always exceeds the duration of their liabilities. But SVB deliberately increased duration risk chasing yield. Big mistake.
DeleteThey were warned by their own risk management as early as 2020, but rather than reduce profit by $50 million to fix it, this impairing C-suite stock options and Q bonuses, told risk team to pound salt. Criminal malfeasance and doc proving it already made public.
ReplyDeleteDon’t forget that SVB let their risk manager go and didn’t bother to fill the position for at least six months in 2022.
DeleteGovernor Michelle W. Bowman, a member of the Board of Governors of the Federal Reserve System, made and address today at the At the Independent Community Bankers of America ICBA Live 2023 Conference, Honolulu, Hawaii. The essence of the speech was, “How regulation fosters innovation.”
ReplyDeleteOne point of interest is the complication involved in understanding how to regulate crypto-currency which, for the Fed, raises questions of “legal permissibility.” Also, the regulation of climate risk management for “the largest banks” is a scary creep of government oversight.
Look at this excerpt from Governor Michelle W. Bowman “2022 Year-End Message from Governor Bowman” (found under this heading):
“Over the past two years, many banks purchased securities with extended maturities to enhance earnings. This strategy supported net interest income while interest rates and loan demand remained low. As interest rates have risen, the book value of these investment securities has declined. In a small but growing number of banks, these losses are eroding tangible common equity, even though for most banks the losses do not directly impact regulatory capital levels.
“Nevertheless, banks with unrealized securities losses need to carefully consider the potential impact of holding securities with below-market interest rates, including, among other things, the impact on their liquidity, capital, and earnings. Unrealized losses can strain banks' liquidity, as the lower market value of their securities may require banks to pledge a greater number of securities as collateral, which in the aggregate can reduce their borrowing capacity. Likewise, banks' sale of securities with unrealized losses to meet funding needs would result in the realization of those losses and would negatively impact their regulatory capital. In addition, these depreciated securities could result in lower earnings due to yields below current market rates on some longer-dated securities. This environment presents potential challenges including how to best manage the unrealized losses and the associated risks to liquidity, earnings, and capital.”
She knew for two years that increasing the interest rates would cause these banks to fail. This was no surprise by anyone, including the banks. How and why, then, were Silicon Valley and others able to fail? And…get bailed out? Could it have anything to do with the fact that Silicon Valley bank was a main source of green energy start-ups?
You may want to revise a few of your assessments re oversight and the Fed.
ReplyDeletehttps://www.nytimes.com/2023/03/19/business/economy/fed-silicon-valley-bank.html
To everyone who thinks this was a failure of regulation. Please read the NY Times article linked above. Then tell us what additional actions the regulators should have taken. They are regulators. They do not run the bank. If Greg Becker and his pals ignored their advice, the only recourse is … exactly what they did.
DeleteA factor not yet mentioned is the multiplicity of regulators, in this case the California Banking Dept, the Fed and the FDIC.
ReplyDeleteIn my bank we were examined in alternating years by either the Fed or the New York Banking dept. The caliber of people is very different but there is also a lot of jealousy and politicking among the various agencies. In the case of SVB, it could be that the agencies weren't adequately communicating with each other or were each focused on various financial elements with others falling through the cracks.