Michael Barr, the Fed’s vice chair for supervision, testified before Congress this morning about the failure of Silicon Valley Bank:
SVB was flagged in a presentation to the Federal Reserve Board on the risks created by rising interest rates weeks before its stunning March 10 collapse....“Staff relayed that they were actively engaged with SVB but, as it turned out, the full extent of the bank’s vulnerability was not apparent until the unexpected bank run on March 9,” Barr said. “SVB’s failure is a textbook case of mismanagement.”
This is ridiculous. SVB was "flagged" in a presentation "weeks before" its collapse? That's infinitely too late to have any impact. What's more, Barr says that SVB's vulnerability "was not apparent" until the bank run actually began.
In what way is this a textbook case of mismanagement? Regulators, who were fully aware of the basics all along, issued a tepid warning a few weeks ahead of time and never raised any serious red flags. If that's the Fed's response to a "textbook" case of bank mismanagement, they might as well hang up their spurs.
I'm wide open to being proved wrong about this, but today's testimony does nothing but strengthen my belief that, in fact, there was nothing all that seriously wrong with SVB. They were plenty solvent and exceeded all the normal capital and leverage requirements—something that would have been true even if they were governed by pre-2018 rules. Basically, the regulations were fine and the regulators were fine.
In the end, SVB was hurt by a slowdown in the tech startup market, and then demolished by a sudden panic within the tech community. An investigation of what caused the panic ought to be everyone's top priority.
Hindsight is wonderful.
I have a story that fits the facts, that is a bit more specific. It is speculation.
First, FT has reported that the VCs were trying to put together a recapitalization of SIVB. I'm not sure when that started. But it's clear SIVB needed something, since I understand that several CFOs started worrying about their Friday payroll on Thursday.
The other thing I've been hearing is that SIVB's president was not well-liked or easy to get along with. I can easily imagine he pressed "too hard" on terms during the negotiation for recap, and Peter Thiel, who is a person who is not known to roll with the punches or take one for the team, decided he had had enough and pulled the plug. This was not appreciated by his fellow VCs.
This is a story, a speculation. Not an accusation or a representation of fact.
So, a vulnerability plus a personality clash at very high stakes. That's what it looks like to me, not a regulatory failure.
AKA "Silicon Valley"
Yes, I think this is as good a tale as any we're likely to get.
The bank failed to manage risk, the regulators failed to regulate.
The takeaway.....blame depositors who recognized these facts.
lol /facepalm
Yeah. It was a badly managed bank with an underwater portfolio of bonds, the regulators noticed this but failed to take decisive action, then the depositors noticed and withdrew their funds.
How could this unlikely sequence of events have unfolded? Such a mystery. We need a "top priority"investigation.
The regulators failed to regulate due to a long term Defang the Police effort that gained traction in the 1980s. Defund the Police is a pretty stupid slogan, but it seems to have worked for some people.
"Within the tech community."
This is another Peter Thiel special like Gawker. We all know it.
I wouldn't be at all surprised to find out that Thiel took a large short position on SVB before telling all his ventures to pull their money out.
Good heavens, the fed doesn't want to be proved "wrong" nor do the management. No one wants to be the fall guy/gal. I totally agree that the timing of the market, the race to the vault, and very likely Thiel were the culprits.
"Basically, the regulations were fine and the regulators were fine."
Not if a failure could endanger the whole system. Or did the FDIC make a big mistake by taking over when SVB really had the money to pay depositors?
The bank, regulators/regulations and depositors all deserve some culpability. Depositors because they shouldn't have had that much money in uninsured deposits in a bank. They probably should have put the money in money-market funds, which are not insured but do not loan at long term. But ultimately there should be regulations to prevent the situation from arising. Or if not the Fed should get a clue about the harmful effects of jacking up interest rates.
If over 90% of deposits are not insured and there is any possibility of depositors not getting their money back, it is a bank run waiting to happen. Pinning the blame on those who started the panic is not constructive. The system has to be made impervious to runs on large banks.
Agree.
Jerome Powell: "There's a need for possible regulatory and supervisory changes, just because supervision and regulation need to keep up with what’s happening in the world."
You can blame lots of things for the bank runs/bank closures -- rising rates, bank mismanagement, Peter Thiel, social media -- but none of that lets regulators off the hook. And if you claim SVB would have passed muster with regulatory scrutiny, that just points to a problem with the regulatory authorities. Either the design or operations of oversight didn't work.
This month saw the second and third largest bank shutdowns in history. Regulatory oversight needs to change to avoid reruns of what happened to SVB and Signature, which we now can see are not just vulnerabilities to individual banks but to the system.
One thing that has changed is that we're seeing rapid rate hikes to combat inflation for the first time in decades. That creates different vulnerabilities for banks than what happened with the GFC. Rate hikes generally do give banks more room to generate higher revenues. But they also depreciate asset values of bank holdings, and encourage depositors to pull money from banks (there's been record outflows of bank deposits recently, much of it going into money market funds). Regulators need to build better oversight for a higher rate environment. (The Fed needs to consider the unintended effects of its rate hikes too. Just saying banks have been warned is not enough.)
The problem was the excess deposits. Companies put their eggs in one basket which put the bank at an increased risk of a run. Banks with a higher percentage of insured deposits are at a much lower risk. The companies had to pull their deposits because they put their business at risk.
This. Small depositors have no reason to panic and start a run, because their risk has been taken on by FDIC and pooled with the risks of all other insured depositors.
Since we find ourselves in a situation in which the failure of a bank with a very unusual financial structure* caused widespread tremors, I think that expanding the scope of depositor insurance would be a much better policy than ad hoc bailouts. But the premiums charged to banks should be priced according to their financial resiliency, based on empirical and historical analysis.
*Commenter Lounsbury explained this in comments to earlier posts
In the first days of the SVB troubles, there were several posts by Kevin that were filled with lots of cogent comments from readers. That, in itself, was a terrific improvement over the usual reactions of lots of us who simply blame everything on the Republicans without any deeper analysis.
One commenter who made a lot of sense was Lounsbury. He seems to be knowledgeable and qualified by experience in this particular field. The main point he kept emphasizing was that SVB was in an almost uniquely bad position to deal with what should have been a pothole in the road, rather than huge sinkhole that could swallow it whole. What was so bad about SVB's situation was synopsized in three related facts:
(1) SVB had a huge amount of deposits in excess of their normal lending demand, so they put abnormally large amounts of their capital in long-term federal debt. That debt is the best possible guarantee of preservation of capital, but it becomes extremely non-liquid when the interest marketplace makes a big upward correction. So they were solvent, but not liquid. (Actually, of course, they were "liquid" in the sense that they could have sold their federal bond assets at the going rate, but that would have been a fiscal disaster that would surely have driven them into bankruptcy.)
(2) The reason SVB had such an excess of deposits was due to the nature of their big clients--Silicon Valley venture capitalists. Unfortunately, those are the exact people who, unlike the normal Mr. and Mrs. Smith retail clients of most banks. are most likely to make sudden large demands for their deposits, when they discover something they want to use their money for. This is the double-whammy facing SVB--excess capital invested in long-term debt instruments, mirrored by deposits that are most frequently characterized by sudden large withdrawal demands.
(3) SVB was supposed to have a Risk Officer, a specialist in the banking industry who is expert in recognizing the sorts of things that can go wrong and ensuring the bank is well protected against the unlikely, but potentially disastrous, situations that occasionally cause terrible disruption in normal financial operations. SVB lost their Risk Officer a year ago, and never filled the position after that.
Lounsbury's point was that SVB management should have been aware of all three of these problem areas. If they didn't have an easy fix, they should have been looking for solutions while the problems were still just hypotheticals. Had they recognized the risks, there were actions that they could have taken that could have defused the financial ordnance before it exploded in their vault.
I think, in fairness, that nobody expected a $42 billion bank run. It really didn't matter what SVB did or hadn't done. There's probably no bank in the world that could've survived that size bank run and gotten its hands on that much cash quickly enough. The problem is Peter Thiel and the question is simply whether he panicked or whether he was short bank stocks when he started the bank run.
"...nobody expected a $42 billion bank run>"
That may be true. But in the aftermath of SVB's failure, we can all see pretty clearly how that $42 billion bank run happened.
The "nobody expected ___ to happen" excuse doesn't let bank management or regulators off the hook because their job is to anticipate what could go wrong and to protect against it.
I see Peter Thiel getting blame, and my opinion of him in general is that he's a "scurvy little spider" just like Mr. Potter, but I don't know enough to say whether the bank run would have happened with or without him. He was not the only one to hear of the bank's vulnerability. One investment company head on CNBC was asked about his own role in spreading word about SVB. He said you can't blame him for talking about what everybody was already talking about. He needed to get word out for his firm and his clients to protect their assets. Does the fault go to the first person to spread word about a vulnerable bank, or to those responsible for the bank's vulnerability?
The "vulnerability" at SVB was more than just that they were vulnerable like every other bank (unable to withstand a $42 billion bank run). It didn't happen at my credit union or at Chase, but at SVB for a reason.
The problem was not Thiel or other depositors.
If it wasnt creepy Thiel, we wouldn't need to overlook the desperate financial situation of the bank in a desperate search for a boogeyman to blame.
The bank literally was insolvent and unprepared for the large number of uninsured depositors who had legitimate claims and concerns as to whether or not the bank could meet its obligations. Blaming depositors is easy, but it gets the problem backwards.
"There's probably no bank in the world that could've survived that size bank run"
Perhaps not, but a bank that gambled on long-term bonds just before a rise in interest rates and that has insane amounts of uninsured deposits is far more likely to experience a run than a well managed one.
::Peter Thiel crashes through the wall, wearing full Cardinal regalia::
*NObody* expects a $42 billion bank run!
Our weapon is surprise! Surprise…and fear! Fear and surpri-
Our TWO weapons are fear and surprise!
…and uninsured bank deposi-
Our THREE weapons are fear, surprise, and uninsured bank deposits!
…and an almost fanatical devotion to the tech industr-
Our FOUR weapons - AMONGST our weapons - amongst our WEAPONRY are such elements as: fear! Surprise! Unins- I’ll come in again.
::scene::
Perfecto!
it seems like mismanagement that they did not recognize the risk of fed interest rate hikes combined with a huge percentage of uninsured deposits by a relatively small number of depositors who were concentrated in a single industry (along with i'm sure many other things). this is more complex that just one metric, and that's why you have risk management departments at banks and regulators who should actually regulate and not just send memos.
I don't want to sound like a broken record but the problem is that GAAP is just stupidly wrong when it comes to changing interest rates and their effects on financial instruments.
GAAP and regulators are better than they were when I started with the FDIC but they are still too scared to open their eyes and see what is wrong. That is terrible. But what is worse is that Wall Street analysts who are way smarter than I am are too scared to do the right thing.
The problem is historical cost accounting.
When you buy a share of JP Morgan then nobody will seriously question how you should value it. Whether you paid $104 September or $143 in February, we all agree that it should be valued at today's price of $128. There is a deep market and it is easy to value.
However, if I buy a JPM bond with a 5% yield at par and you bought it two years ago at $120 then we both should value them at the same price since they are the EXACT SAME THING.
Held to Maturity bonds, or loans, are valued at their historic cost. So you can PRETEND your bond didn't go down in value.
SVB had a bunch of Mortgage Backed Securities that HAD, past tense, gone down in value before SVB sold them and RECOGNIZED the loss that everyone knew already existed.
Yes, they could have held them to maturity but Goldman, the company that bought them, could also hold them to maturity too. SVB's accountants AND REGULATORS valued the bonds $2 billion higher than Goldman's accountants are regulators valued the EXACT same bonds just a few nanoseconds later.
Management, Wall Street, the auditors, and the regulators all knew that the value of the assets was far less than the book value. The problem was that selling the bonds forced all of them to stop hiding the truth.
This was all know months ago. It was know years ago that IF interest rates went up so much that the bonds, and the fixed rate loans, would be worth less than book value.
On the other hand, the value of the liabilities also decline as rates go up. This often offsets the decline in value of the assets.
The big problem is that many of the liabilities usually come with the option to redeem at par, or a small penalty.
If I bought a 5 year CD at 1% a year ago then I could cash it in and pay a "substantial penalty for early withdrawal" of around 1%. So I could take it out, get $99 and invest it in a 4 year CD at 4% and come out way ahead. If I had a money market fund then I could transfer it out at par.
When a run starts, the liabilities go up in value because the lower interest rates paid immediately stop. So you end up with an insolvent bank.
If JP Morgan, or my bank, had to redeem 100% of deposits then we would find that our banks are insolvent. We couldn't sell the bonds without "recognizing" a significant loss, we couldn't borrow enough, our buildings can't be sold instantly, our receivables can't be collected immediately, etc.
Banks can easily run with negative equity because USUALLY we don't need to redeem our liabilities immediately.
If liabilities come due tomorrow then we, like virtually every business in the S&P 500 will be in default tomorrow.
Regulators, Auditors, Accountants, and management need to understand that just because Wall Street analysts are too stupid to understand the actual value of financial instruments doesn't mean you can always ignore reality.
I thought this was blindingly obvious. Peter Thiel pushed this, hard, and caused the panic in order to destroy this bank, probably to test how effective it would be as a technique.
All banks are always somewhat vulnerable to a run, because they all loan out far more than their deposits, and that vulnerability is multiplied greatly by having their depositors come from a small/coherent base, i.e. much more vulnerable to panics and the madness of crowds. Really, the Reserve Bank should have stepped in to stop the run and ensure the bank had sufficient cash to handle all demands, and been vocal about it as part of its job.
All banks are not insolvent. All banks do not have an extremely high proportion of uninsured deposits.
Blaming Thiel overlooks all the actual problems that created this situation.
Would this have happened if Thiel had not encouraged a run on the bank? That's the problem with the straw that broke the camel's back arguments; all of the parts had to be put in place before subsequent events unfolded as they did. All that's left is to argue which straw(s) were the most important. Which has already been done to death, here and elsewhere.
Once the bonds matured, no. Thirty years from now.
You're not arguing with me. As I think I rather clearly explained already.
I know, I just like to add color to the portrait of the bank on the gallows.
Why did the tech community panic? Because they had huge, uninsured deposits at SVB (Roku with almost $500B) and did what reasonably normal folks did when their money is at risk.
Where the regulators asleep? Don't know, but a better question (which I can't answer) is did they have the authority to do anything since on a GAAP basis, SVB looked fine.
Was SVB really solvent? SVB had a very unusual depositor profile, which made it extremely susceptible to a bank run, and a very unusual asset profile, with about 40% held in T bonds. With most finance companies, you look at the quality and diversity of their loan portfolio for risk assessment. Banks I worked with only lent floating rate to their C&I clients, and required borrowers, at times, to enter into interest rate swaps to mitigate the risk of rising cost of funds. Banks would also limit their hold positions of each credit based on risk assessment.
GAAP is wrong? Yes, in many cases. And it takes a long, long time to change it (just look how long it was for GAAP for leasing transactions to get amended). My experience is that rating agencies and finance company credit analysts would generate adjusted credit stats to handle off book exposures and would absolutely stress test for risk, including interest rate risk. Doesn't mean that their recommendations are always accepted by the higher ups in any organization.
Could/should the FDIC offer insurance (at a cost) to holders of accounts over the $500,000 limit? If enough people bought it, it would stabilize the banks a bit, as the insured would no longer have an incentive to get ahead of a bank run.
How much do banks pay for FDIC insurance now?
Once that stabilization was reached, wouldn't the banks then just adjust their risk taking to eventually match and we'd be back at square one?
A proper insurance program requires adherence to conditions set by the insurer. That’s what fire codes, fire marshalls, UL, and other laws, officers and agencies are for. Bank regulators should have access to banks’ information sufficient to set premiums according to the level of risk.
I've seen some discussion on nationalizing a bigger portion of the financial markets and banks. In a real sense, SVB is now nationalized anyway. Why should our money, our value, our labor be held by others when it can be held by ourselves.
I don't know if this is even a good idea, but I don't like paying bankers.
You realize that regulators these days are taken from the industry being regulated "because "experience") and fully intend to go back there once this administration is over. Or maybe before.
All else follows from regulatory capture.
"... They were plenty solvent .."
No they weren't. The FDIC estimates SVB's failure will cost the FIDC $20 billion. And that estimate is probably a lowball.
From a NYT article last week:
In reality, this was a slow-moving disaster of SVB's own making until it became a fast-moving one, apparently triggered specifically by Peter Thiel.
Was none of this public information? Do large depositors do any vetting of institutions they entrust with millions of dollars? Sounds like this was indeed a run waiting to happen.
The Big Short. Watch it.
Or read the book; it’s well and clearly written.