And now that the fed said they will only "fully" back depositors in large banks that go under, there will be even more consolidation, so the next crash can consist of just one huge circle that acted gregariously and irresponsibly before its collapse, but no one could do anything because of its size.
What makes this even worse, is that Yellen would have lost nothing had she said they would back every bank and not just large banks - small bank failures are just not that common, and the risk is low, (and can be walked back in two years when the tension lowers).
Even worse, well over half of the SVB bailout went to <20 accounts.
IMO, they should have raised the amount that FDIC is insuring, which is unquestionably too low. [They still should do that.] They then should have insured everyone equally up to that point. And if Circle and Roku would lose 90% of their value from such a move, that beats the whole banking system consolidating into one or two massive players that can act with impunity.
OK, I am not an economist, and admit that I probably don't know what I am talking about, but HN is full of smart people with domain knowledge. Please (gently) broaden my perspective.
> What makes this even worse, is that Yellen would have lost nothing had she said they would back every bank and not just large banks - small bank failures are just not that common, and the risk is low, (and can be walked back in two years when the tension lowers).
FDIC is supposed to protect smaller individual depositors, not gigantic companies with hundreds of millions of dollars in deposits. Gigantic companies have financial teams that should be tasked with protecting their operating capital.
FDIC is also a last-resort. The idea is that bank assets are sold off and all depositors are made whole. It's very difficult and time-consuming to do this with a mega bank though.
Maybe FDIC limits should be raised, but the problem with that is someone needs to pay for it. Perhaps companies with $100MM deposits should be paying for private insurance.
I don't have time to research now, but there were some well detailed breakdowns of the large payments when the bailout happened.
The first relevant link in DDG is a statement by the Chairman of the FDIC. He claims that the top ten accounts held more than 13.3B between them, which is more than 10% of the total of all deposits in the bank, and more than half of the $20B that the government is expected to payout in total. [1][2]
The FDIC would otherwise have had to pay a scant 3M of that, so there's 2/3 of the bailout just in those 10 accounts. If we include the next ten accounts, it will be much, much more. (One of the articles at the time claimed more than 85% went to 15 accounts, but I didn't want to be extreme without time to research sources.)
I think that's a misinterpretation of the data. Based on reports, SVB had around 130B of deposits when it was taken into receivership. Something like 90% of that was uninsured, so around 115B of uninsured deposits. The 20B hole means that without FDIC backing the uninsured deposits, they still could have paid out over 80 cents on the dollar. So if the top ten accounts had about 13.3B of deposits, they only received less than 2.7B of the bailout - less than 15%, not "well over half".
I strongly suspect that any sources that claimed that simply made the same error that you did.
Looking at the source you provided from the Chairman of the FDIC, it actually says that the top ten accounts held $13.3B - not _more than_ $13.3B, as you claimed. The next ten accounts necessarily held less than that, so say a generous upper bound at $26.6B for the top twenty accounts. In order for them to get at least half of the bailout, they would have needed to hold north of $55B in aggregate. The FDIC itself would have to have gotten the size of SVB's largest depositors wrong by over double in order for your original claim to be true, in the absolute most generous case.
The large banks are more heavily regulated. Trump raised the limit at which a bank is called large from 50bil to 250bil. The small banks value their freedom from regulation more than they would value full FDIC coverage. The fed would love to make that trade:more regulation of small banks in exchange for full FDIC coverage for small banks.
Canada is dominated by five enormous banks, they are closely regulated, and Canada has had a better experience than USA with both fewer bank failures and higher real interest rates paid to customers.
Sure the Canadian banks are more stable but they are EXTREMELY conservative when it comes to lending out money. That's why the USA has a robust venture capital milieu while Canada does not.
In fact a Canadian bank would rather loan money to an American business to buy out a Canadian business than to lend the same Canadian business to expand.
Canada has higher real interest rates paid to customers? You sure? My Canadian bank account pays pretty much what my US one does, 0.00001% or something similar.
And if the circles’ areas, not their radii, were the failed assets. As it is, a circle’s apparent size (we judge 2D shapes by their area, not any one linear dimension — although this is difficult to do in practice, and linear marks are generally superior) is the failure size squared, which distorts the data a boatload.
Here is a, er, more faithful representation of the data. The recent failures don't look quite so crazy anymore!
The circles' areas are showing the failed assets. Observable Plot defaults to using a square root scale when encoding a quantity with the radius of a circle.
So weird, should it not be the area of the circle being the value? The radius should be √(Area/π) not the square root of the value or am I misunderstanding?
In this case, the x-axis is a different unit from the unit of data measured by the radius, so the fact that it's off by a constant multiple doesn't matter; it's just an arbitrary scale.
If both the x- and y-axes were encoding data in the same unit as the area encoded, then it might make sense to correct for the area, but I don't think I've ever seen a plot like that.
The volume of a circle would presumably be zero, but yes I've used the wrong word. I'll change it above. I sometimes wish there was a forum that allowed editing of spelling in other peoples posts and they could just accept the changes (same with Twitter). You could spend all your days correcting obvious mistakes...
> The volume of a circle would presumably be zero, but yes I've used the wrong word.
i guess?
> I sometimes wish there was a forum that allowed editing of spelling in other peoples posts and they could just accept the changes (same with Twitter). You could spend all your days correcting obvious mistakes...
nice and interesting idea except some people are awful and would abuse that feature :(
I've seen this claim a few times in discussions about such charts, but it's far from obvious to me that this is really true. My best guess based on my own perception is that we estimate quantities represented by circles as proportional to something between the diameter and the area.
Human beings are spectacularly shit at comparing areas. If you put two circles next to each other, one with double area of the other, most humans will say their relation is 3:2.
FWIW, I think the size of the insolvency is more telling of the failure than the AUM. E.g. First Republic was about ~$13B in the hole when it was taken over. Otherwise a large bank with a large hole looks way worse than a small bank with a huge hole.
I would also have a 2nd plot that has all banks / banking / bank assets with the failures in a different color, so we could identify the relative size of the failed banks, and also see the actual size of the big players in banking.
But they held "value" that was "gone" from one day to the other which triggered the financial crisis.
In reality they weren't worth anything but the realization that those values were worthless was the trigger. But those values were on the balance sheets and should therefore be visualized as well.
Only where "deposits" are defined as FDIC-insured consumer accounts. Clearly they held other people's money for them, which is pretty close to the economic definition of a bank.
But inconveniently long after savings accounts started being demand deposits due to online bill pay, atms and easy access to electronic funds transfers.
If you are going to complain about changing the definitions you should complain that they did it too late.
That needs to go back to the 80s to capture the SnL crisis. It dwarfs 08 in bank failures. It better indicates the conglomeration of the many banks into the few we have today.
I have faith that we can top S&L. We have the technology. We have the talent. There are six banks with over a trillion in assets in the US. I have faith that one of them has been doing some wild book cooking. I'd place a bet on Citibank, followed by Wells Fargo. There's an old saying in Tennessee — I know it's in Texas, probably in Tennessee — that says, fool me once, shame on… shame on you. Fool me… you can't get fooled again.
I'm just hoping this time it's something absolutely outrageous just for the lulz of it all. Like, let's get some FTX-style absurdity. All the absurdity happens in crypto right now but I still have faith in regular banking. Some people still like the challenge of regulated markets.
JP Morgan is the biggest one. It can do ANYTHING it wants and get away with it.
It can make 10 billion USD spoofing gold prices for a decade and get away with a 1 billion USD fine (and keep doing it) for example.
The CEO can go on trips with Jeffrey Epstein, be friends with him and do business with him and get away with it.
It made tons of money off of the Madoff ponzi by providing Madoff with a bank account and not reporting the (from their perspective) extemely obvious ponzi that was going on for 15 years. Nobody went to jail and JP Morgan's fine was probably lower than what they made from the ponzi.
There are 100's of other examples of quite outrageous FTX-style crime. This is just what I happened to read about and remember. And that's only the publicly known stuff.
Let's turn it around: why would JP Morgan (and other big banks) NOT be engaged in extreme levels of crime that could be described as "financial terrorism"? If JP Morgan blows up it would be the end of the US and they know it and the US govt knows it. I repeat: they can get away with ANYTHING.
I want something GOOD like FTX running on Quickbooks. That's what people need to aspire to these days. That's why I like Wells Fargo. Create fake accounts… so absurdly brilliant.
JP Morgan might be the biggest, but I just don't have faith in them like I do Wells and Citibank or even HSBC. Some lame overly complex scheme isn't want I want. I want a decimal in the wrong place that everyone just ignores despite nothing ever adding up. I want vaults full of gold on the books that don't even exist… said to be held in countries that don't exist anymore. I want Superman 3 salami slicing, but maybe one that's been running perpetually since 1980… and it turns out that's actually the inspiration for the scam in the movie. I want Snopes to have to change something from "Legend" to "TRUE".
The world needs to be reminded that the USA is #1 and always will be.
There was a hedge fund manager / college professor from Irvine CA who figured out Madoff's scam in the 90s, and nobody paid attention to him whatsoever.
You might find Kenneth C. Griffin is more your cup of tea in that case.
Bonus: he hasn't been caught yet. His crimes are still on-going.
Same principal crime model as Madoff though: own a market making business + a hedge fund, sell stocks (naked), take the money, give nothing in return.
There's some other petty stuff like front running household investors through PFOF and instructing the broker you buy order flow from to shut down the buy button for retail investors when the price moves against you. But that's just the petty crime.
Add to the mix that despite all its flaws (which are many) crypto is way more transparent than a classical financial institution due to the fact that one can track all movements. Even for entities like FTX, we can guess what their wallets at and how many funds they hold. Part of the FTX debacle was due to depositors figuring out that it didn’t have enough money to cover their debts.
Unfortunately for USD backed currencies we still don’t have any idea where or who holds the backing assets. They might as well be non existent…
Alas, the biggest scams in crypto are more opaque than that. How many dollars does Tether hold, and in what forms? Where is all of this "commercial paper"? The biggest crypto falls are still to come.
If you like absurdity, FTX has recovered 7.3 billion out of the 8.6 billion hole and plans on relaunching the exchange to make the last billions back in fees
Most noteworthy is that this quick 8 month turnaround is partially thanks to the blockchain, and under no new laws being passed
It's not about the money, it's about the message. It's about running a global financial institution on Quickbooks. It's about not having a bank account. It's about not having stop losses. It's about wiping out losses by making your own money. It's about TOM BRADY. It's about the Larry David ad that ends, "Ehhhhh, I don't think so. And I'm never wrong about this stuff. Never."
FTX really elevated fraud to an art. I'm not even joking. It's beautiful. It's so insane when I think about it that I don't even think it should be illegal. He should through his entire defense behind the 1st Amendment, say it was all part of an elaborate roleplaying game, and somehow walk.
for me its more about how much this mismanaged business shook confidence in "crypto", instead of just this mismanaged business - the way we would judge any other sector. while the crypto aspect is helping resolve this far faster than other insolvent schemes of similar size and magnitude.
and Sam Bankman Fried is not involved in that.
yes, Sam did that elaborate thing, the people recovering and the bankruptcy court are not Sam.
It really didn’t though. Crypto is the never ending, infinite ponzi. It’s unshakable, unsinkable. The hype is real.
What Sam did was elevate things. Anyone can run a crypto scam. Literal kids do it. But to create art is something else. Something more human. Something timeless. SBF is perhaps the ultimate use case for crypto.
I had about $100 in FTX. Worth it. Totally worth it.
I’m stoked for the Coinbase collapse. My body is ready. Jesus, take the wheel!
I'm not sure we can treat crypto like any other sector - there are no others which are afflicted by a rapid succession of high profile scams, scandals and collapses. Confidence is shaken because there's no other rational response to this situation.
its really a choice to consider mismanaged companies as the sector itself, at least the construction industry started putting X days since incident as an effort to differentiate each site since nobody was hearing about sites that were operating fine. confidence isn't shaken for everyone in the crypto space, and there might be a need for services to point out how many days since incident they've gone, since nobody currently indexes that or reports on that while the majority of activity occurs within services that operate smoothly and as expected
I dunno, you're asking crypto to be compared with sectors that actually provide services people need who between them have had a few high-profile flameouts over the course of history. Whereas crypto has had a fairly quick boom-and-bust and the biggest players frequently either go bust or are heavily exposed to those who already have (and are desperate to convince everyone they're not and are actually fine).
I'm not sold on crypto and you'd have a hard time persuading me to change my mind, I'm afraid.
You only have to look at what you’re ignoring in order to hold that view
> actually provide services… need
You’re valuing entertainment at zero (nobody needs that), vice at zero (nobody needs that), financial services at zero, and a perpetual bug bounty at zero, those are the major components of the crypto space
and you simultaneously hold every participant in any of those sectors as both representative of the whole thing, and equally as relevant as the next participant
persuasion is not the word I would go for, the disingenuous nature of that perception is the main observation
The top token in that list, Serum, listed as worth $1.9 billion, had a trading volume of only $2 million yesterday on Binance. -2% depth is $60,000 lol.
we can debate liquidity and depth of the market for those assets, but they're also just using the same standard as what was lost as well, so does it really matter?
unless we’re going to start with “they didn't lose $8.6bn and an independent valuation put all lost assets at $2bn so now everyone’s solvent what an amazing turnaround”
One huge difference is that the $8.6 billion number probably included all the BTC and ETH that they were supposed to have and that was deposited with them. Those were actually liquid and worth a lot. When FTX failed they didn’t have the BTC and ETH people had deposited with them. The theory is that FTX/Alameda used it all to pump tokens they owned, buy real estate and make deals, and of course losing trades. No one in crypto values this new list at 7.3 billion. It’s actually a source of much ridicule and hilarity if it is similar to that list I posted.
it doesn’t and that wasn't the premise of why we can acknowledge that using that payment network saved everyone time in the clawbacks, despite the shaken confidence that the exact same event caused into that payment network
the main distinction involved here is that not knowing who to subpeona for records slows down everything, whereas with the blockchains used most of the participants consolidate funds into KYC’d exchanges and we know which ones they went to, speeding up requests for records and subsequent action
In other words, blockchains make it easier to track what we are doing with our money, without the justice system being involved, and that’s somehow good.
I actually think Madoff is a great example of comparison, and didn't mention that because I felt someone else would call that a strawman, ironically, or at least choose to say something about the difference in the size of those frauds.
The main difference is the time, you're choosing to ignore that. 8 months versus .... how many years for Madoff? A decade?
>But the biggest sum has been in “category A crypto” tokens with large and liquid markets. FTX now has more than $4bn of crypto assets under its control, a total that has been bolstered by a sharp recovery in cryptocurrency prices.
>Bitcoin, which had dropped below $20,000 after FTX’s collapse, this week broke $30,000 for the first time since June 2022 , with other cryptocurrencies including ethereum charting a similar course.
This seems to go against your claims. - the existing assets just became more valuable in USD terms. Actual recoveries:
>Recovery efforts have more than doubled that figure so far, court filings show, including $800m in recovered cash and a further $600m in “settlements and investments receivable”.
Sure but a court can order you to do a future transaction that effectively reverses the original. (Akin to how almost every single reversed transaction _actually_ works).
And if you refuse, they can order your local (or not so local) PD to jail you until you comply.
Blockchain still exists in the real world with its very real rules.
Because bitcoin's is on a very public blockchain so one can easily lookup what wallets received the transfers, notice that they belong to an exchange, court order that exchange to freeze/transfer the coins.
If it was done with wire transfers and etc they'd have to do many requests to learn what transfers existed while with bitcoin all transactions are public so its just faster. Imagine having to walk down to the local PD every time to approve your web search vs just doing them without that.
I think the problem is that knowing the wallets that have received transfers isn't enough. They also need to identify the parties that control the wallets. As far as I know blockchains don't do KYC, so I'm having a hard time believing that blockchains make the process of recovering stolen funds any easier or indeed faster, quite the contrary.
The blockchain itself doesn't do KYC; many individuals/companies interacting with the blockchain do.
Sure, for some transfers just knowing the wallets won't tell you who it is. The problem (for criminals) is that often they ultimately transfer the money to an exchange and that wallet is known. This is where the blockchain ends up coming in clutch. If you did a wire transfer of money to say Wells Fargo; the USG is going to have no idea it happened so they can't even think of freezing the money. However, each of yours (and everybody else's) bitcoin transfers are recorded to a public ledger so the second they notice a series of transactions starting from the theft to a known wallet they can immediately request the money to be frozen.
At this point USG can require the exchange to return the stolen money as well as whatever information they have on the account holder.
---
Can one steal bitcoins and get away with it? Yes!
Its just that the public ledger is not your friend in this instances and you'll need to account for it while this fact is not the case with many non-bitcoin thefts. If you commit wire fraud and then cash it out into physically currency and then re-deposit that currency into several banks there is no public way to link those two actions. (Privately one could track the serial numbers; but again private not public).
Honestly, I've long since made peace with not ever getting back those assets. In that way I accept the risk I took when making the trades in the first place. But what really irks me with the FTX bankruptcy is how my assets were suddenly impounded, effectively stolen from me, and I wasn't allowed to trade it anymore even to avoid further losses. Yes, contracts, articles, and so on, but I'm a simple man: I pay. I own. So, if I am to ask compensation for anything, it must be 1. to get my crypto back, and 2. to be paid damages for the inability to trade during a period of free fall. As it's now, however I guess I can count myself lucky if I even get back parts of my own crypto, if anything. So, I've decided to not spend energy on it.
we can debate liquidity and depth of the market for those assets, but they're also just using the same standard as what was lost as well as reporting where thats just asset price appreciation
08 was artificially low because many banks got merged at a fire sale. Wachovia, Merrill Lynch, Bear Stearns, and National City stick out. Other financial institutions got essentially nationalized and stock became mostly worthless like Citi and AIG, although the government sold most of their stock in 2011
Credit Suisse is about the same size as SVB, Signature Bank, and First Republic combined but it got “acquired” by UBS at a price 60% below its last trading price in a deal where $17 billion of debt was wiped out so it doesn’t count here
Credit Suisse is not included because it is a Swiss bank, not an FDIC insured US bank.
Lehman Brothers is also not included because, even though it was a US bank, it was an investment bank with no FDIC insured deposits. It was around the size of all of this year's failures, combined.
As you note, bank bailouts that were not FDIC bankruptcies are also not included.
I think GP knows this, but I also think you know that the graph is trying to paint a particular picture, and that picture is misleading because a lot of information is missing. We are not in the midst of a financial crisis that approaches 2008, and the graph is trying to make us think something different.
Not in an ‘08 sized crisis - yet. Wait until commercial property debt finally ‘looks down’. It’s been running off the cliff for a long time already, and is in exactly the same boat as the securities that took out SVB, etc.
I think someone found an interesting dataset, tried to visualize it, and thought it looked interesting. I doubt that there was any motive to the dataset other than, "Here's what I get from the FDIC, what does it look like?" Then shared code and source so that anyone else could reproduce it.
If you can find another data source that gives a fuller picture, you should. But compiling these data sources takes work. And the ones you get are all going to be a particular slice that represents some things but not others.
It depends on what we're trying to visualize. From an investor's perspective, a bank whose assets get sold for pennies on the dollar in a fire sale is essentially a failure. Lehmann Brothers was also a massive (investment) bank failure with huge second order effects on the economy.
This graphic seems to be modeling things from a taxpayer perspective. These banks failed and the government needed to step in to do something to ensure people could get their deposits.
The government stepped in in the missing cases too, just not the FDIC. Many of the missing cases had large securities trading and investment banking activities (e.g. Bear Stearns - ~400B), and so it was the Fed and SEC that were most involved in their forced sales.
WaMu was bought by JPM and is on the chart, presumably due to the FDIC involvement, whereas Bear, which was a similar size and was also bought by JPM is not.
FDIC premiums are not payed by taxpayers. What we're visualizing here are bank failures assumed by FDIC. The too big to fail banks didn't technically fail, but to give an accurate picture of a financial crisis they should be on the graph.
Agree, this representation also makes WaMu’s failure look like the worst in recent history but it felt like one of the smaller problems at the time with what was going on with the investment banks, Fannie/Freddie and AIG.
Tax payers are legally required to pay taxes in USD. I'm not actually sure if the IRS technically accepts cash but if so it would be extremely rare. Meaning all tax payers have a bank account and ultimately foot the bill even though it is technically funneled through the banks' books first.
"Footing the bill by having a bank account" is one of those very-hard-to-picture-or-feel things in days when most bank accounts are "free" and these banks have so many lines of business. E.g. am I paying for FRBs bailout by increased loan application fees if I buy a house or car or such? That's what I'd imagine, or maybe it's just that maybe otherwise savings accounts would pay a bit more interest or something?
Yeah it's tricky, if not impossible, to pin down exactly how any one person's money flows through the books but at the end of the day all bank profits come from their customers. We may pay account fees, overdraft fees, document and origination fees when they open a loan, interest on a loan if they don't sell it, etc.
Banks are in an interesting place because effectively any tax payer is going to have to have a bank account. In my opinion, that means tax payers are directly funding banks and the FDIC.
There are other types of customers for banks so I wouldn't argue that tax payers are exclusively paying those feels but it feel disingenuous to see politicians claim tax payers aren't footing the bill at all.
Aren't there some relevant details missing from this kind of analysis? Banks failing just means that the value of the banks assets fall below the value of their deposits, right? In which case the degree to which that happens seems to be highly relevant to this kind of comparison. E.g. the value of assets falling to 50% of deposits in bank failures in financial crisis A vs 90% in financial crisis B
At the very least that’s missing Fannie, Freddie, Bear, Merrill, Lehman, TARP and arguably AIG for another 1.2T+, granted a lot of this was eventually repaid as the FDIC will be as well.
You are correct. Additionally, the size of the bank(s) are not really what matters.
I want to see the scale (sum) of what was actually lost when they went bankrupt, and how much we (the public) have to put up to keep the system from collapsing. Does anyone have an actual visualization of how much we ponied up to keep our banking system from collapsing?
Did the public just provide a reasonable interest rate loan for a few months to a year? Or was it a sweet 0% loan for...ever? The important details are lost in the media reports and it would be nice to get a sense for what really happened.
There were winners and losers, but the government made money on TARP. That doesn't fit anyone's narrative very well, so you don't hear that much about it, but its a fact. So far not a public dollar has been lost in the current crisis. FDIC, like other insurance, is paid for by the insured. Every FDIC bank in the country is paying the cost of this.
The problem with those numbers is that TARP lost money if you account for inflation over that period. That said the point wasn't to make the government money anyway, so whether that matters is up to you.
Generally speaking, it turns into free assets. The people involved usually don't acquire the assets/associated debt without some kind of guarantee.
OneWest Bank for example after 2008 had a guarantee where if the assets defaulted above a certain amount they would receive full value of the loans in a payout from the government. They were actively foreclosing on people to justify catastrophic losses to get the bailout. Not sure how that ended up since I was only marginally aware of the start of that and everything went silent once the news got wind of the perverse incentives.
In terms of trends, the bailout game has been played consistently since the the dollar went off the gold standard (1971 iirc).
The ponzi is starting to unwind now that inflationary pressures are out of control. I expect concentration to eventually lead to nationalization followed by a new currency which will fail because they lost all credibility from their mismanagement as a private entity.
That's what's happened historically with every country that debases its store of value above the point macro effects become noticable which are around 3:1 ratio).
There's argument to be made that approximately any cost to keep the system from collapsing is a trade-off worth making if the alternative is the system collapsing.
Round 2 of the 2023 bank crises might be commercial real estate loans like the 1980s crises. Some downtowns are still 40% vacant. This more a consequence of covid than bad bank behavior.
Yeah these sorts of figures always frustrate me because the historical context for these things is so much broader, and it seems obvious to me to go back to the 80s, if not earlier.
> conglomeration of the many banks into the few we have today
There are still thousands of banks in the USA. I dont really see why there should be more than 100. Canada has 5 big ones and a few dozen tiny ones. Same in UK and Australia.
It’s not all roses - in Australia and Canada the few major banks operate in pseudo-cartel fashion. There’s few enough that they can effectively collude without doing it illegally.
Deposit banks were banned from crossing state lines and investing in equities markets after the Great Depression. These limitations were removed in the 1990s.
Mike Bostock knows far more about data visualization than I do so I assume there are good reasons for choosing this presentation. But it strikes me as hard to think about, so I must ask: when is stacked+packed circles a good choice?
The drawbacks to me seem to be:
- At any given point on the x-axis, the associated height of the pile is partly taken up by circles but is partly space lost to circle packing. This proportion is hard to estimate visually.
- Because circles extend in the x-direction, they contribute to height over a _variable_ time range. E.g. WaMu is still the tallest circle at the 2010 line. It's a bit like a kernel density plot where the _kernel_ is data dependent.
Because of these effects taken together, though it's tempting to think of the profile of the pile as a smoothed average of the rate of bank-failures, but this is pretty misleading, which I suppose is why there's no labeled y-axis.
The way I understand the graph is that is not the "number" of circles, but the area (money) that they consume. So those three bank failures are taking up the same space as all those other failures from 2008.
It is the total volume of money involved vs the count. The height of both stacks show just how much damage three did vs all the others in the left hand stack.
Yeah I get the count vs area distinction, but I think the point would have been more cleanly communicated with e.g. a stacked barchart with failures by quarter. With circles, the filled area over any span of the x-axis is not solely due to failures which happened during the corresponding time period. So the area above a short span in early 2010 includes area contributed by WaMu, which happened in Sept 2008.
My read of it was that the circles are centered on the date of the failure, and the stacking was some flavor of minimum height. The result is you get an approximation of "area under the curve" of bank value that has failed, while retaining a sense of whether a particular peak was a few large actors, or a mound of tiny. I find that clear distinction valuable.
The nice thing about Mike having posted the code is that, if you think there's a better way to portray it, the canvas is spread out before you. :)
For the life of me, I'm not able to understand why none of these lists include Lehman Brothers?
A commenter below posted:
>Lehman Brothers is also not included because, even though it was a US bank, it was an investment bank with no FDIC insured deposits. It was around the size of all of this year's failures, combined.
So does this mean, that since Lehman had no customer deposits, it doesn't count?
Lehman had no FDIC-insured accounts. So it shouldn't be included in a list of failures managed by the FDIC, because it wasn't.
The liquidation of Lehman was handled by the Treasury department (and a bankruptcy court), not the FDIC, and Congressional involvement was necessary to pass a bill to finance the resulting "bailout". The FDIC maintains its own fund to pay back depositors.
It might sound like a pointless technicality, but it's really not. It would be like asking why an NHSTA report doesn't include plane crashes.
Well, they're visualizing a dataset from the FDIC. So, it's more like it doesn't count due to a technicality. It's not clear to me if there's an "official" dataset that includes all non-FDIC institutions.
Both SVB and Signature Bank were atypical banks with atypical depositors.
SVB catered to risky tech startup and Signature Bank was involved in risky crypto shenanigans.
On SVB:
>As a regional bank in the Bay Area, SVB offered services specifically designed to meet the needs of the tech industry, and soon became the largest bank by deposits in Silicon Valley and the preferred bank of almost half of all venture-backed tech startups.
On Signature Bank:
>By 2021, cryptocurrency businesses had represented 30 percent of its deposits.
>Banking officials in the state of New York closed the bank on March 12, 2023, two days after the failure of Silicon Valley Bank (SVB). After SVB failed and in light of the closure of the cryptocurrency-friendly Silvergate Bank earlier in the week, nervous customers withdrew more than $10 billion in deposits.
First Republic Bank was also an atypical bank that offered high interest rates for high-net-worth depositors by not insuring deposits and having a very high loan-to-deposit ratio:
>Fitch Ratings and S&P Global Ratings downgraded First Republic's credit rating, citing "a high proportion of uninsured deposits" from wealthy customers who are more likely to move their money elsewhere and a loan-to-deposit ratio of 111%, meaning that it had lent out more money than it had in deposits from customers.
It makes perfect sense to focus on events defined in a uniform way across the dataset (in this instance US FDIC insured bank failures). Otherwise what you are depicting is an arbitrary collation of events that may have very different causes.
What I derive from the visualization (btw Bostock is a genious) are a couple of simple yet still tentative observations:
* A new cluster of failures might be forming. The pattern of correlated failure is not new and it may point to similar business models and/or exposure to the same factors. Measurable correlation may also imply some level of contagion. This is defined as one event increasing the chances of another (after conditioning on common factors). But it is anybody's case what the fully developed phenomenon will look like. Three events is still small number statistics. Dont run an AI model on this.
* There is at this point a remarkable absence of small bank failures. If this persist it hints that it is actually not the vanilla banking model (and poor risk management) that underlies what is happening, but more idiosyncratic factors that are specific to a few actors. In some sense that would be good news.
One thing is for sure. The crisis will be properly visualized.
> There is at this point a remarkable absence of small bank failures.
Not sure it's that remarkable? 1. not adjusted for inflation as others have commented. 2. There are less banks in the US, maybe up to 1/2 as many as in 2008 (https://www.statista.com/statistics/184536/number-of-fdic-in...). 3. You may only be seeing the leading edge of something similar to 2008-2010 range - too early to tell.
For a qualitative discussion of what might be happening inflation is a just a nuisance factor. There are other similar factors (e.g the size of the economy is not constant either) that we can ignore for this purpose.
The bank size distribution is more relevant but 50% of a large number is still a large number. You'd expect this to somehow show up in the statistics.
Timing is indeed a key aspect. The actors involved have quite a bit of discretion and options and it takes time for all these to play out. But if it does transpire that small banks are not as seriously affected as in the last crisis and its more of a "middle class" bank disease this would be a strange new thing.
Indeed. ZIRP was a new thing [1]. My laymen understanding is these big banks got caught out making really stupid (in hindsight anyway) bets interest rate would stay near zero and ran into liquidity issues.
Is there a reason small banks wouldn't also make stupid bets like this? Maybe they have less money slushing around that isn't in loans or something.
Its still too early to connect the dots. That all sorts of low-quality bets would be unwound with rising interest rates was indeed a given. But in such an opaque financial system, the surprises can come from any corner.
A vanilla bank business model actually benefits from rising rates as this generally widens the spread between their lending and borrowing. So then you have to look at second order effects. According to reports one causal factor is traced to their "low-credit risk" investment portfolios and various combinations of real hedging vs creative accounting for their interest rate sensitive positions. It may be that the really small banks do less of that.
In any case the long period or low rates is really not an excuse. Bankers are not extracting rents from society to play being idiots. Managing risks is what they are supposed to be paid for. You don't manage risks by assuming tomorrow will the same as yesterday.
Highly suggest sorting by date rather than size. Putting the smallest failures at the top causes a little bit of an optical illusion making the first tower appear smaller than it really is. It also highlights that whatever is happening here is drastically different than what happened in 2008 - here we have three “big bank” failures while 2008 had only one, but we have zero “little bank” failures while 2008 had hundreds.
Community-scale banks (aka credit unions) are a better idea for local-regional communities (aka cities and towns and agricultural regions) because their managers have to live with their clients.
Take a community of 100,000 families, in an economic system where they're all collecting income and paying bills and so on. The idea behind a bank is that they hold the community's money securely while making their own money by providing loans at a greater interest rate than the interest rate they pay out to their depositors. Fundamentally, this means the bank is dependent on the fact that customers don't try to withdraw all their money at once (as the above dynamic means the bank is never able to meet this demand).
FDIC guarantees by governments are intended to bolster this customer trust, as if everything goes sour the deposits (to a certain limit) are guaranteed. The important point is that the regional bank has a profit model based on giving successful loans to individuals and small businesses for things like houses, cars, business expansion, etc. The bank will of course want to know a fair amount about the people they are loaning to (drug addicts are not good bets, etc.) in order to ensure they get paid back.
I hear people laughing in the background about the naivete of this picture in the modern American financial system. Since this post is already a bit long, go watch the movies "Margin Call" and "The Big Short" to get an idea of what's actually been going on.
The model you describe is exactly why S&L happened. Small banks with highly correlated deposits holding their own loan books is a recipe for maximizing vulnerability to economic shocks. Interest rates go up and the loan book loses value; the local housing market drops and the loans get foreclosed and lose value; a major local employer goes out of business and depositors all start pulling their money out instead of reliably depositing paychecks.
Banks don't work that way any more because it's a really bad way to run a bank.
This is a rather skewed perspective that ignores the fact that if you have FDIC insurance for depositors then that can be gamed unless you have strict regulation of the banks:
> "The roots of the S&L crisis lay in excessive lending, speculation, and risk-taking driven by the moral hazard created by deregulation and taxpayer bailout guarantees."
This is why a lot of people worry that Silicon Valley Bank, First Republic, etc. might be the tip of an iceberg. If they've all leveraged themselves on risky speculation bets in the hopes that they'll get bailouts if it all goes sour (even though depositors greatly exceeded FDIC insurance limits) then you could have a domino situation.
Note also that it's a perfectly good way to run a bank as long as you don't get greedy and go for big risky bets, but the only way to ensure bank managers don't get the Las Vegas bug is to enforce the banking regulations in a fairly strict manner.
SVB and FRC did not “leverage themselves on risky speculation bets.” They made lots of very safe investments but did not adequately hedge against large and rapid rate increases. Also not good but it’s a big difference, IMO.
Not even most hedged investments at large scales. Hedging doesn't make risk disappear: It just shifts it somewhere else, either to someone even larger, or dispersed among many parties. When the risk is correlated with the risk from others, like the trillions in underperforming treasuries that the bing banks have in their balance sheets, who is the safe counterparty? Not the next bank, which also has a similar position. We saw this with the housing crisis and synthetic CDOs, and we see this on insurance against sufficiently large natural disasters.
So sure, small banks failing to hedge is irresponsible, but at a large enough scale, someone holds the bag.
I'm not an expert on interest rate securities, but with e.g., physical goods the bag holder has physical goods or contracts for supply that allows them to meet their obligations.
Is something similar not possible with interest rates? I imagine for instance someone with lots of cash and little desire for risk could lend their money to the banks at overnight rates, collect the interest, and then offer swaps against that steam, no? Of course, no one's going to get their 10th mansion off of this. A little riskier, someone holding variable payment debts could do the same, as long as the loans are diverse the risk could stay low. In this way, the risk at least shifts from "we're screwed if interest rates change" to "we're screwed if interest rates change and many diverse loans start to fail to make payments" in which case you're probably screwed regardless.
Oh absolutely. My point was just that if you truly wanted to make a big risky bet, you would definitely not be going out and buying a bunch of agency bonds.
I read one of their customers on here raving about the mortgage rate they got through SVB because it was so much better than anything offered by anyone else.
I don't see any evidence of any overall prudent investing on their part considering the entire bank had to be bailed out and FDIC limits relaxed.
I think an overlooked issue here is that, beginning in 2020, SVB suddenly had tons of deposits. Yields all around were zero-ish and they had money burning holes in their pockets. I can sorta see how they could be incentivized to close some loans.
Also, I think that as with FRC, their loans tended to have fairly low credit risk, so again not super sketchy, just poorly hedged.
> "SVB's focus on the innovation economy was a big winner in the past but may not remain so in the future, according to Dick Bove, the prominent banking analyst at Odeon Capital Group. The U.S. economy, in Bove's view, is shifting from a consumer-oriented economy driven by plenty of low-cost capital to a manufacturing economy marked by limited access to capital that's relatively costly."
> "And despite his downcast report, Bove maintained his hold rating on SVB's stock. Investors seem to be a bit more optimistic. After SVB's shares lost two-thirds of their value last year, they're up nearly 11% so far this year, closing Wednesday at $254.99 a piece."
Reading the tea leaves is an imprecise art, I guess.
The S&L crisis started before the deregulation. In fact, the deregulation of S&Ls was an attempt to address the fact that many of them were already insolvent due to holding long-term fixed-rate mortgages in a rising-interest-rate environment.
One thing I don't understand, and perhaps you could explain, is why anyone in the US would ever keep more cash in any one bank account than what was covered by FDIC insurance. It's precisely the reason I don't e.g. take my savings to an offshore bank that offers much higher interest rates. Is this just a matter of people taking trust in a bank's solvency for granted?
Banks can offer incentives for certain levels of deposits. Better interest rates, better cards, better loans. SVB offered perks to bank exclusively with them.
I’m not sure about personal banking (since I’ve never had enough in a bank to worry about it!), but in commercial banking I imagine it’s routine. I work at a medium sized firm and I authorize multiple payments a week that are over the FDIC limit. I’m not sure how much juggling it would take to transact at volume and never have an account go over the limit, or for long.
FDIC limit is 250k per bank, so if you have 1 million you'd need 4 banks. If you have 10 million you'd need 40 banks. Having money spread out like that doesn't seem easy to manage. Also having 10 million in one bank gives you better interest rates and service at that bank than if you only had 250k.
Cash sweep products are a thing. For example, Wealthfront will sweep your cash into a bunch of smaller banks, putting 250k in each, so on the off chance they fail, your money's not been disappeared.
Doral Bank failed in 2015, so 8 years? Not quite 10. The GFC was 2008 which was 15 years ago though but people still remember that one.
Even though the FDIC chose to make depositors whole for SVB, Signature, and FRC, there's no written legal guarantee that they'll keep doing this, so in the face of that, I don't think people are forgetting the $250k FDIC limit.
Anyway, my point is no one's walking up and down Main St with their $10 million and opening 40 different bank accounts by hand because the finance industry invented a product (prior to SVB, even) so no one has to do that.
oh, I don't know. Lots of people walk up and down main street. No one I know keeps more than $250k in any given bank under the same name. Obviously having LLCs, wives and children etc let you spread things out in the same bank. I had to wait 10 months in 2008 for FDIC to make me whole when a local bank I had most of my savings in at the time went under. Most people have a living memory of that.
Also, I have 8 accounts at 6 different banks and I'm not even worth $1m so I can't imagine it's too hard for someone worth $10m to figure this out.
Figure what out? I'm saying people with that kind of money have private bankers*, and don't need to spend the time making 40 different accounts and managing that because their money is already protected through a sleight of financial trickery called cash sweeping.
If you like seeing the inside of bank branches, and having unnecessary zoom meetings where the background is a picture of the inside of a bank branch, that's entirely up to you.
Well, there are private banks/banks that specialize in private banking (like First Republic), and private bankers, which specialize in navigating the Byzantine bullshit endemic at a normal bank. I believe you pointed to the private banker division at Chase, and most large banks have them somewhere, for large net worth individuals like you’re saying.
It is definitely an entirely different experience (either way), no doubt.
If you’ve ever tried to practically use multiple banking institutions in the US, especially through the 80s-90s, you’d immediately relate to only using a single account regardless of what the statistical hazards are. It’s 2023 and my institution limits Zelle transfers to $2,500/mo. Want more, just as fast? Back to human wires and fax machines…
Really? You could always just go get a cashiers check and walk it over to another bank. Now I've got accounts at several banks with free online transfers between them (2-3 business days)... it was more of a process in the 90s, sure, but now you can open a checking account at any major bank, link it to another bank and fuel it in 5-10 minutes. Not with Zelle, just a normal domestic wire which is usually paid for by the receiving bank.
My credit union (it’s actually very worth it, so I’m not switching) has no branch within multiple states from me. A domestic wire requires phone calls and fax, can’t be done entirely online. It’s certainly more doable with a behemoth like Chase, but then you’re dealing with all the negatives of using Chase and employees that have no leeway to use common sense when solving problems.
They're just the tip of the iceberg. Pretty much all medium size banks have lots of underperforming assets, and they're just one minor mistake from going under, like these 3 banks. Also notice that there are still many shoes to drop: comercial real estate (a disaster waiting to happen), car loan defaults, etc.
Keep in mind that car loans get paid down pretty fast. For example, Capital One's <620 FICO customer segment has already paid off about half the auto loan principal from 2021. Their delinquency rate has flattened, quarter over quarter, as well.
New car loans are not only longer (> 6 years) and also more expensive than previous loans. Also, people stop paying when they lose their job, so the trigger may be the start of a recession.
Some of them. A bit less than half of loans at Capital One are more than 5 years, for instance. (I'm just overly familiar with that company.) But a 6 year loan from June 2021 is still almost 1/3 paid off.
The other question is if there's a recession, how much will used car prices, for cars that were new in 2021, drop.
The bank manager can take more risks, knowing that the depositor's won't lose their money (over $250k if the bank they manage fails. Which is important because people tend to get mad when they lose large amounts of money.
Thus, buy bailing out depositors to an unlimited amount, bank managers are then incentivized to take more risks investing the depositor's money because the more risks they take, the more likely it is that one will pay out, raising the bank manager's bonus, and what their stocks are worth. Of course, by taking more risks, they also increase the chances that one will fail catastrophically, but since the depositors are all covered, up to an unlimited amount, eh.
Evidently, the best way to run a bank is a 0% cash reserve minimum (as was granted to US banks starting during COVID), so you can make money off of literally every last penny, and then discourage and prevent customers from taking their money out, then [externalizing all costs]/[socialize losses] onto government (taxpayer) via bailout or customer via bail-in, while privatizing all profits in the meantime. And charge poor people $30 every time they dip under $0 even by a penny.
Doesn't mean it's a good idea for the rest of us non-banks.
> Banks don't work that way any more because it's a really bad way to run a bank.
As compared to what we have now? With dubious financial instruments so opaque I'd have to spend 30 years lurking underneath desks on Wall Street eavesdropping on conversations just to have any clue at all how the fuck those work?
I'm almost comforted when there's a Bernie Madoff, because at least I can wrap my head around how a Ponzi scheme works (ignoring the ethics, obviously). I'm (irrationally?) worried that some of what's been going on the last few years actually makes plain Ponzi schemes look legitimate by comparison.
Not 25 years ago, we used to laugh about the stupid books with titles like "Dow Jones 100,000!" and whatnot, I think it was a Slashdot post way back when. And while we haven't quite reached that pinnacle of absurdity, it did hit 37,000 not so long ago.
None of us may be able to pick the queen of hearts, but some of us have caught on to the fact that it's three card monte. When it all falls down, should we console ourselves with "well at least the banks weren't small and local and vulnerable to economic shocks"?
The difference between a ponzi and real finance is mostly lying.
In a Ponzi you lie about asset growth. In real investments you report real asset growth. If the system fails it won't be because people lied about what their assets were. It might be because they mis-estimated what those assets were worth. But that isn't a Ponzi.
Crucially though, it is actually possible for assets to grow. As long as there is new good business to finance, the whole thing doesn't need to be zero sum. And annoyingly, growing business is so much faster with a conplex financial system, that countries with such financial systems stand no chance. So even if you don't like the risk and excess in a complex financial system, it's really hard to go without.
It’s more complex than that. The ideal situation for a bank is when all its depositors are customers of each other. The deposits never leave the bank so the bank has very stable liabilities.
So a local bank will be great when a community transacts mostly with itself. SL happens right around when multinationals and corporate centralization started becoming more of a thing..
We live in the era of trillion dollar corporations, fintech, and the internet so a community bank ends up more like a small big bank. Its depositors are getting paid by corporations from way outside the community, and they’re spending money on the internet or at giant multinational retailers and thus shifting deposits outside the community. In this paradigm, it’s actually better to be huge and have a diversified customer base because the best way to keep your liabilities stable is to have the largest market share possible.
You’ll notice, banks usually have generous incentives for setting up direct deposit. Direct deposit is the best way for them to have a predictable measure of incoming deposits when there isn’t a high likelihood of one particular account transaction with another.
Might be a bad way to run a bank, might not, but it's definitely not the cause of the S&L crisis. That was a story of deregulation leading to massive moral hazard, rampant speculation, and multiple high profile cases of outright fraud.
Economic shocks and rising interest rates aren't usually the reason that financial institutions engaging in risky or fraudulent behavior get caught, they usually just make it so the clock runs out on their scheme to roll over the losses or otherwise avoid getting caught.
The liquidity crisis of 2008 didn't cause Bernie Madoff's fund to collapse, the fact that it was a complete fraud from inception did. It just made it impossible to ignore.
I can easily get a real person in front of me, there's not a million layers of bureaucracy, and most importantly - no bullshit fees. I don't constantly get my bank drained just for having an account in the first place.
To join mine, I bought a $5 share (as a minor) and continue to get a $5 dividend many many years later (but I gotta close the account to get the accrued dividends...).
I gotta go to the AGM one day. I hear it's catered.
This doesn’t work. The very failures we are seeing today are why this doesn’t work.
Why would someone leave their money in the regional bank offering 0.3% that all of their issued loans average barely above when people can take it to a bank offering 3% or get CDs offering that?
When Chase, BofA, and Citigroup are all offering approximately 0.0000% APY on their CDs, and it's credit unions that are offering 3+% (you should be shooting for at least 4% right now), it's not clear that big banks are actually better than a smaller, community-focused bank or credit union.
I double checked, and it turns out they all have a couple of products offering closer to 4.5% APY, but the tables on those linked pages still have far too many numbers like 0.05% APY on them to be taken seriously.
Credit Unions are better also because they are literally libertarian socialism. Imagine Uber with no shareholder class - as a cooperative, and so on. No one to extract rents from the two sides of the marketplace, the people themselves own the network.
So are housing cooperatives. They have low rent because there is no landlord class. Look at the Mitchell-Lama program in NYC, 30 years later they the most desirable and awesome apartments for the price, all around the city. Meanwhile next door the capitalist landlord-owned buildings are 3x as much for worse amenities.
So yeah, I'm a libertarian socialist. People in the US hear socialism and they think of a big, oppressive government, but actually government is just as much if not more on the side of the capitalist, the industrialist... bailing out large banks and injecting trillions into corporate equity to prop up markets. In the 19th century they actually violently put down strikers, evicted people, operated debtor's prisons etc. which is why most libertarians back then were socialist. It was cool. Today that lefty libertarian stuff is mostly in Europe.
That article at therealdeal has me pretty stunned.
The headline story is that landlords are keeping rent controlled units vacant rather than renovating them which would be unprofitable, and daring the toothless regulators to do something about it. Ok, makes sense.
Then as a side note, the article mentions that they’re also deferring maintenance on whole buildings: roofs, boilers, cameras, etc. The fact that these landlords are incentivized not just to let a few units turn into roach hotels, but to let their whole buildings crumble, without the remaining tenants leaving to a better building - that’s a fundamentally broken market. Way beyond a few rent controlled units.
Sure, "Coffee is a drug" just as much as Caffeine and sugar are. OP likely meant that those who abuse scheduled drugs (outside of weed) are more likely to develop a habit of putting off work to feed that addiction, or to be fired for misconduct in connection with their use, resulting in more late payments and eventual repossessions/foreclosures on loans. It's all in terms of managing risk, but I'm sure chronic alcoholics are just as risky, if not more risky, as meth heads, since they blend in and have a good credit score but are more likely to lose their job for being intoxicated at work or go to jail for something like drunk driving.
Are there no small failures this time around because of ongoing consolidation? This seems to be a dominant trend in economics across the board. Does anyone have a good answer why?
Some midsize banks got a massive inflow of deposits during the pandemic and they didn't know what to do with the money, so they parked it in "safe" investments like mortgage backed securities.
Then when the pandemic subsided, people started burning down their savings accounts, and interest rates started skyrocketing. All the banks' money was tied up in all these 30-year mortgages that nobody wants anymore because they're paying like 2.5% and inflation is 8% now, but they're still forced to sell them for terrible prices just to keep their ATMs full. Eventually word gets out about what's happening, there's a bank run, and the bank starts death spiraling.
Massive banks have nothing to worry about because if the government allows them to fail they'll take the entire economy down with them. And (presumably) smaller banks weren't under the same pressure from investors to generate returns on idle capital, so they didn't commit the same sins as the midsize banks.
SVB's fail was because of radical speed in deposit inflow, which was invested in fixed rate bonds just at the moment that the Fed was on a rising rate ratchet. Because of an affluent clientele, they had unusually high % of uninsured deposits. Very unlikely that a small bank would have had the need to layoff so many deposits in such a short time span. Very unlikely that a smaller bank would have had the high % of uninsured deposits. These are the funds that jump ship immediately.
First Republic fail was because of large pool of fixed rate assets....(jumbo low-rate mortgages and Treasury Notes) in a rising rate context. They also had an affluent clientele and thus a high % of uninsured deposits. The fixed rate assets lost so much value while rates rose, that they had no tangible equity. Thus, the hot money deposits raced out of the bank.
Smaller banks typically don't have such a concentration of affluent customers, which means that more of their clientele would be under the threshold for FDIC insurance. Smaller banks probably have a good book of commercial loans which are commonly priced at a variable rate.
I am baffled that SVB and FRB did not hedge their fixed rate portfolios with interest rate swaps....Maybe their mind was on staying abreast of the white-hot tech sector instead of wringing their hands about the next Fed rate decision.
It's just a matter of visibility. The failed banks so far were all family of the tech industry. Since the tech industry is so far one of the most impacted by contracting conditions, its banks are being poked at first. Now that it's done, the other banks are going to get some love.
I generally love Mike Bostock's work, and this is a beautiful visualization, but I think this comment points out why circles are just a bad way to represent anything visual other than pizzas: it's a leaky abstraction (which is a real sin in visualization)
You have to know how the chart is generated in order to correctly interpret what you're seeing. The value can be the radius, diameter (how pizzas are measured) or the area itself! All of these choices lead to different interpretations of what you're seeing.
Then, on top of that, we're just not great and visually comparing areas. Pizzas are a great example of this. A 12" pie and a 16" pie don't look that different but you're talking about nearly twice the amount of pizza! A good example in this visualization is Signature vs SVB
Then on top of that we have the dimensions of the circle forcing the visual to overlap on the x-axis without this meaning anything concrete.
All that you can meaningfully take away from this visually is "2008 had one big and a ton of small failures, and currently we're seeing 3 big, but not quite as big failures", and unfortunately there's not a lot more you can get out of this. Because of the x-axis problem it's hard to even tell if WaMu preceded some of the others or not.
Or else replaced with a fixed-width rectangle and scale the height by assets. Would fit more easily on a timeline and humans are much better at comparing heights than we are areas.
I've seen a bit of discussion about concerns around office property mortgage and such (e.g. JP Morgan wanting to be insulated about losses on outstanding loans) for FRB which seem concerning as a post-Covid systemic risk.
I wonder if the consumer deposit side, though, has a lot to do with an unanticipated risk of the particular business model SVB and FRB were using: higher-income people. Sounds great - less risk of defaulting on loans - but then you are highly-concentrated in larger deposits for a smaller total population (larger meaning less insured, so easier to spook; smaller population means the panic doesn't have to spread as far)... so it works well for decades and is seen as something to aspire to... until it suddenly doesn't.
Weird part about this whole thing is, we have been repeatedly told that banks are good and they learned a lesson in 08. Now both tech and banking are in trouble again.
> we have been repeatedly told that banks are good and they learned a lesson in 08
This is a wholly different lesson.
In 2008, banks were making bad investments.
In 2023, the changing interest rate environment caused good investments to become worth less than their original value. If held to term, things would be fine, but liquidity issues put stress on the system.
These are not the same, and we have better means of dealing with this problem. SVB and First Republic were handled appropriately. Investors written down to zero, depositors made whole.
The future banking system will be even more robust after having learned this new lesson. I've no doubt that we'll have regulations that demand a better mix of investments that are regularly audited and stress tested.
We as a society also have to stop thinking of banks as a means to earn interest on deposits. Chasing the best rate has led to this problem. Deposits are liabilities.
Every time someone says "stress test", I will point out that the stress tests did not model for interest rates rising like they did, which triggered the current problems we're having (or, really, interest rates being so low for so long was the actual problem IMO, but that's a whole separate conversation). Stress tests as currently implemented would not resolve the issue, and even if they were so wise to reactively throw in more aggressive changes to interest rates to these stress tests, would that be enough to catch the next problem?
See Tables 2.A and 3.A in 2022 Stress Test Scenarios (PDF), covering a 3-year period starting in Q1 2022. Look at the 3-month Treasury rate. In 2.A, rates go up to 1.5 by Q1 2024 and sit there. In 3.A, rates sit at 0.1 for three years.
Rates are not the only variables, but the point remains that the interest rate changes that triggered these problems do not appear in the stress tests.
I strongly disagree, the current mess is largely caused because some banks didn’t anticipate interest rates going up from historic lows. That’s profoundly bad risk management. A child could have done better.
Well, yes. But it's also bad economic policy. The banking sector just happened to have adapted to a decade of low interest rates. Then during the Covid crisis of 2020 the FED pumped dollars like crazy into the markets [1] -- adding fuel to the fire. If you set bad incentives you'll harvest bad behavior. That is the tragedy of FED-style economic planning in a nutshell.
What lesson are they going to learn? Go big or go home? I don't foresee any kind of legislature becoming law over this, especially with our divided government.
However, this time it looks like the government failed, not the banks. If you look at the US government (bonds) as just another business the bank can invest in, you can't help but notice that the business had not been very well managed.
It promised that bonds will keep their value, but they did not (currently at market price and in the future, due to their yield suffering from inflation). We still don't have a clear message on the situation with government debt as well, but we are probably headed towards more inflation and/or some (hopefully soft) default(s). And as always, fewer people will have more assets/control and the rest of us will pay the bill.
The issuer of a bond only promises that they will pay the principal of the bond back with interest. They don't promise that the bond will hold its value in the market.
Yes, but here the issuer is strongly related to the levers of interest rates and money supply which directly impact the long term value of the issued bonds.
This honestly looks much less frightening that I would have thought. But then my experience of '08 was "that wasn't so bad?" which I understand is highly regional. Visited some colleagues in 2015 in the UK who kept referring to "the recession" when talking about struggling customers or abandoned buildings. I had to ask whether they had experienced another recession or whether they still meant the 2008 one...
Where are Lehman Brothers, Bear Sterns? And where are all the non US banks, such as e.g. Credit Suisse? They would make the other numbers in this graph dwindle ..
Way easier to diagnose & save 3 banks than the multitudes in 2008 from a time spent analysis perspective (wether or not we should is a different matter). The question is the ripple effect here - it seems to have stemmed 3 large breaches where as in 2009 it was death by a thousand cuts (with some large breaches).
I feel like it should include earlier time periods (80's? Depression?), and have some adjustment for different times. I'm not enough of an expert to guess whether to use inflation adjusted dollar amounts (using what inflation metrics?), or as a % of GDP, or a % of the banking system..
"investment banks" are also misleadingly named in my opinion. This dataset seems to stick to what the more common idea of a bank that accepts deposits and offers checking/savings accounts.
Is it me, or does this kind of illustrate that the banking industry doesn't know what it's doing and cannot be trusted in the future? I don't know what the answer is, but these "fat cats" running banks really aught not to.
The person who wrote this blog also wrote the very popular and famous D3 visualization library for Javascript. It also looks like this is javascript from the code.
If two regional banks failing with little realized loss is massive and unprecedented how would you describe 2007-2008 when the entire world economy was brought to its knees with napkin math ~1.2T of bailouts/failures in the US let alone Europe with several banks nationalized (e.g. UBS and RBS) and nation states defaulting on debt?
This graph only shows consumer banking failures which captures a tiny fraction of the financial institution failures in 2008.
Additionally, this dataset by definition doesn’t consider the several financial institutions (including much larger consumer banks) that were deemed “too big to fail” and were therefore not allowed to fail and didn’t make this list.
Finally, this graph reports deposits which are not “lost” in a bank run, the bonds still have their original value if they can be held to maturity. The actual cost to the government/FDIC will be no where near what the total deposits number is.
But for context here are some much larger numbers following the same criteria:
Fannie Mae and Freddie Mac had $5T in mortgage backed securities and debt when they went under.
AIG was backstopping some ~$600B in financial instruments for banks and needed a $180B bailout.
Wonder what year that massive spike is. Oh no, can't zoom in on mobile! Guess I'll have to check later on my desktop, if I remember. Until then, it is a mystery.
Excellent point. To Joe Consumer, bank vs credit union accounts are ready alternatives. Where might one find data on failures? A quick scan of this site offered no clues.
National governments should control their own money, scrap interest entirely. Let the market decide what each currency is worth for international trading.
The status quo international banking system is designed to enslave, and governments can't do anything about it because the banks are above them.
How anyone can defend the current system is beyond me. It's not remotely close to being the best we can do, it's a scam at every level, way more complicated that it needs to be, by design.
> National governments should control their own money, scrap interest entirely. Let the market decide what each currency is worth for international trading.
Isn't this how the current system already works? Except for the "scrap interest entirely" part, which I'm not sure know what it means.
I don't think that this makes any sense. Debt is inherent to the act of lending. A loan that doesn't have a debt attached to it isn't a loan, it's a donation.
Are you actually that dense, or are you being obtuse? Of course, the initial loan amount is owed. So when I say a debt is attached, it means it's added on top of the loan amount, that's called interest.
What I mean by scrapping interest, is that no extra debt is created when money is loaned after being printed. If 500 is loaned, 500 is owed. The way it currently is, the loan amount is owed plus interest, creating a debt that can never be paid because there isn't enough money in existence to pay it.
The Germans had a saying - "One Mark for one Mark's worth of work or goods produced". This is how they solved the inevitable hyperinflation of the international system.
They went from a situation where you couldn't buy a loaf of bread with a barrel of Marks, to being the richest nation in Europe within 3 years. That's why WW2 happened, it was a clash of economic ideologies.
I see... you mean banning interest. That would be a form of price control, and price controls don't tend to work very well, historically. The idea that there isn't enough money to pay interest is "interesting", but unfortunately it's also wrong. At any rate, when there isn't enough money what happens is deflation, not inflation.
This is not even remotely close to 2008… This figure conveniently excludes the largest 6 failures of 2008, TARP, and major international banks and insurers simultaneously affected.
As a reference the prospect of AIG failing without a bailout was so significant it became an international issue threatening both US and EU economic stability. SVB and FNB are borderline irrelevant in comparison.
The amounts reflected in SVB and FNB are also exaggerated as the money isn’t “lost” as if parent can obtain liquidity elsewhere (via government or merger) the asset will return principal + interest at maturity.
What makes this even worse, is that Yellen would have lost nothing had she said they would back every bank and not just large banks - small bank failures are just not that common, and the risk is low, (and can be walked back in two years when the tension lowers).
Even worse, well over half of the SVB bailout went to <20 accounts.
IMO, they should have raised the amount that FDIC is insuring, which is unquestionably too low. [They still should do that.] They then should have insured everyone equally up to that point. And if Circle and Roku would lose 90% of their value from such a move, that beats the whole banking system consolidating into one or two massive players that can act with impunity.
OK, I am not an economist, and admit that I probably don't know what I am talking about, but HN is full of smart people with domain knowledge. Please (gently) broaden my perspective.