I'm not saying that the system is perfect - far from it. But the whole banking system operates on the principle that the bank can invest their deposits (i.e. make loans) to earn interest and make a profit. If too many people (near 100% in SVB's case) withdraw their deposits at once, the bank will not have the liquidity to give them their money back until they liquidate their investments - hence the "run on the bank". What's the alternative? No banks? That means no mortgages, car loans, etc. - money in mattress time. The purpose of the Fed is to backstop a "run on banks" to keep the system functioning which is precisely what they did. And no, crypto is not an answer. Crypto is an example of no "lender of last resort" which is why it's so volatile. Imagine if the value of the dollar fluctuated by 10% or more per day - not good.
Since we don't experience bank runs very often, I conclude the banking system as a whole works pretty well. SVB, on the other hand, not so much at the end.
i can’t help but feel a sense of anticipation that all of your writing over the past few years has been training ground for intimately covering what’s unfolding right now in the economy, banking and tech (no pressure).
I’ve been on an info binge for the past week, and your writing stands out with a few others, that you’ve mostly linked!
… tldr; the banking crisis root cause is flawed data modeling. the entities represented in the system (for the purpose of both administration and legislation) do not accurately reflect reality.
… which, after my last 13 years of work in tech, does not shock me, but does leave me feeling deeply unsettled.
If you continued to analyze/cover this space, i’d pay to subscribe.
Thanks! This was the world I first worked in (back when I wrote a different newsletter: https://carnegieendowment.org/experts/analysis/458), so it's a rabbit hole I pretty quickly fell down. I have no idea if I'll keep at it or not, but either way, I won't make you pay for it (if people pay, they'll expect it to be good.)
And yeah, that was a good post. I think I see the root of all of this as much more economic and politic issue than a technical one. The infrastructure underneath it matters, but my view is that'll get shaped by economic incentives on top.
“the banking crisis root cause is flawed data modeling. the entities represented in the system (for the purpose of both administration and legislation) do not accurately reflect reality.”
Your criticism of economic/financial modeling need not be limited to these sectors, they extend to modeling period. Is there such a thing as a realistic data model then?
" People and companies losing their deposits wasn’t fair. Startups who banked at SVB didn’t do anything wrong, unethical, or even (we all thought) risky; all they did was open a checking account. It wasn’t their fault; they were victims of mismanagement and an economic shock; they deserve to be made whole. "
If you deposit more than 250k in a bank, losing money over 250k is *at least* partially on you.
If I'm a founder and I just got a 50 million dollar round, I'd be very careful about where that money goes because 49.75M of that is exposed. This is not opaque. As the founder I have a duty to protect that money. No one should assume that putting money in any bank is a riskless exercise. Sounds like the founder/VC/SVB bankers were all a little too cozy.
But I think the general view of the modern banking system is (was?) that the 250k isn't there to say that 250k is safe and the rest isn't; it's there to short circuit bank runs, and make deposits safe in general. Ie, if most people don't have 250k, most people won't run on the bank, so there's no critical mass to create a run, and everyone's deposits are fully safe. I get that's not the letter of the law, but I don't think it's unreasonable for most depositors to basically view it that way. And in that way, SVB's failure is more on regulators than depositors, since regulators are the ones who would've been in a position to realize that the 250k limit didn't protect SVB from a run. (Obligatory matt levine citation that makes this point: https://newsletterhunt.com/emails/25336).
On the cozy thing, I think it's more laziness than coziness. I saw reports that VC would tell people that they should use SVB, or that using SVB had some sort of signal value. In my experience, none of that was really true. It was just that SVB was there, they marketed to startups, and most companies didn't give much thought as to who they banked with. And if they did, it was about bells and whistles, not because people were assessing a bank's creditworthiness.
Mar 17, 2023·edited Mar 17, 2023Liked by Benn Stancil
"I don't think it's unreasonable for most depositors to basically view it that way."
I do. It's unreasonable and irresponsible. It's not like banks have never failed. And cash is the lifeblood of a startup, so I'm more than a little surprised that apparently up to 50% of the startup community were at risk from a single bank failure.
I'm not arguing that this is *all* the fault of the depositors, btw. And sure, I get it that founders are focused on other things and most startups don't have a finance pro on the team. But still! Depositors have a duty to be the best stewards possible of their money and should understand the risk. And VCs should, too. Little surprised they weren't more on top of this, but I'm learning that VCs are not nearly as sophisticated as they would like us to believe.
Of course, there is no riskless place to put your money. But there are well understood ways to reduce risk - all your money in one bank is not the way. I don't think this requires a degree in finance to understand.
Hopefully founders learn from this and get less lazy.
Also explicitly *not* arguing that depositors losing money is 'fair'.
Also not arguing that the government doesn't have some responsibility for this. We all know banking regs exist and few people have time to understand what they will and will not do to minimize risk.
But having no control over the bad decisions of bankers and government is all the more reason to be careful with one's money.
Yeah, and to be clear, for a depositor with $50 million, I'd expect them to treat that differently than someone with $500k. My point is mostly that I believe it's the intention of the FDIC to prevent runs in their totality, not to encourage people to split up money into 250k increments across a bunch of banks. (So, in intention, it's like to a quality control inspector, who checks a few products with the intention of being able to say they're *all* safe, not just the ones they randomly test. And it's not - again, in intention - like heath insurance, where your policy has a cap and the insurer couldn't care less about your losses over that cap.)
Obviously, it's prudent and responsible to treat the $250k limit as though it were that sort of cap. But that intention seems pretty clear, given that FDIC more or less confirmed it with this bailout.
And on fault, I put a bit on startups for being lazy, a bit on the regulators for not scrutinizing things a bit more - though SVB lobbied them out of the building, so - a good bit on VCs for herding so much money in and out of SVB, a lot on SVB for not realizing all their deposits were from companies that were over the limit and were easily herded, and a lot on - and this is my old man moment - social media, for making it possible that these panics can unfold in a day.
> My point is mostly that I believe it's the intention of the FDIC to prevent runs in their totality, not to encourage people to split up money into 250k increments across a bunch of banks.
Yes! I think this point is being missed by a lot of commentators. The FDIC was created to prevent bank runs, after the US government saw how often they had occurred during the Great Depression era. And tbh the 250k is, at this point, an arbitrary amount.
I don't have much confidence that the bailout is evidence that the FDIC's intention is to prevent bank runs. People with lots of money at stake and politicians must have had a ton of influence. The FDIC originally was not going to save the depositors as of Saturday.
Eh, I'm less cynical than that. I could really see that going either way. The twitter VCs who were freaking out about the FDIC not doing anything pushed the idea that the FDIC wasn't doing anything, but reporting like that Wapo article show that they were doing a lot over the weekend. Obviously who knows what would've happened had there not been public pressure for connected people, but it wouldn't surprise me if we would've landed in a similar place.
(Or, well, I'm not sure the pressure made the difference. I think the fact that SVB's client base was what it was made a difference)
I thought this one was special since startups create jobs and the government has to protect those jobs, so that's why FDIC saved them. If a startup succeed and become the next Google, then it's gonna create more jobs and boost the US economy which is what the government would want, so that's why they get special benefits.
I think it has something to do with big companies already being billion dollar companies while startups are like little kids. I think the government views big companies as 'adults' while startups are like 'kids'.
Startups need more support just like how kids need support from their parents (government acting like a parent to startups), but once startups fully matured into a big company the government support wears off and the startup bears responsibility now.
I've also think that the public views it the same way as the government. From my observation, people tend to feel bad for struggling small businesses/startups while when a big company struggle, people don't feel bad. Probably has something to do with human psychology.
I get that argument, and - as there was for Garry Tan's - it's got some emotional appeal. But practically, I think it ends up being an additional form of regulatory capture for what's already a very well-off, exclusive, and influential industry.
hmmm by default there are no economies in the world that is free of government distortion: every industry/sector is impacted by government policies and laws in some form. At the end of the day the government will do what it deems to be fair at the moment when an economic crisis occurs.
Right, all economies have some government distortion. I'm arguing that more is usually worse. And I specifically don't like special treatment for an industry, though I think exceptions are logical when an entire economy is facing existential risk...which is not what I think was happening with SVB.
Yeah, fwiw, I didn't really say this at the end on the bit about fairness, but the point of that was to ask "are we so sure that it's *fair* for Silicon Valley to get bailed out?" and not "why don't we help everyone like we did Silicon Valley?"
Yeah, I don't think it's fair to single out Silicon Valley for special treatment - and I've been a direct beneficiary of that industry. But it's an industry with lots of money and influence.
I'm watching how this shifts into a political narrative later this year, the 2024 Presidential race develops.
Like it or not, the two biggest sources of VC money are pension funds and endowments tied to universities and health systems - things that are ostensibly "socially good" for the poor and middle class, that help retirement, access to higher ed, and healthcare access.
Populists on both sides (Warren, Sanders on left; some MAGA remnants on right) have incredible leverage in calling out what fuels all of this.
I'm pretty sure most tech employees don't know that the fancy dinners, cushy office spaces, pet insurance they have is all essentially funded by retirement funds from people who make a lot less than they do - essentially a large, massive scale wealth transfer.
But that seems to mischaracterize why pension funds invest. It's still for the returns. It's not a wealth transfer because if the system works as it's supposed to, it's just an increase in wealth. (Eg, if a pension fund put their money into a hedge fund that returned 100%, sure, the hedge fund might get some nice dinners that are paid for with the pension fund money, but that's a trade that's good for the hedge fund and good for the pension fund.)
If there are no returns, and the whole asset class is a scam, that's obviously different, but that's a very different problem.
I tend to comment when I'm disagreeable about something, but I wouldn't come here week after week if I didn't appreciate your thinking and writing. I also hope Mode didn't suffer any adverse effects from SVB.
Thanks. And the whole brand of this thing is mostly me being grumpy about things I find disagreeable, so by all means, disagree away.
(And I appreciate that, we weren't directly affected by it. Obviously, if this torpedos the entire economy, that's not great for us, but it'd be weird if that was good for anyone.)
I don't think the question is about fairness at all. Is the Fed raising interest rates "fair"? Without a trusted entity (the U.S. Treasury in this case) providing liquidity, there can be no commerce. We would be back to bartering. We have a very recent example of applying the "moral hazard" rule in the case of Lehman Brothers in 2008. I don't think people realize just how close we came to complete collapse as a result (there's a great Frontline about this that's a must see). SVB made a bonehead move of funding low yield, long term investments with increasingly costly, short term deposits. Lehman demonstrated even more malfeasance and greed. Contrary to what people think, backstopping SVB didn't "bail out" SVB - they're history. Rather, bailing out SVB bailed out everyone in the U.S. (and beyond) who need to buy things to live.
Yeah, that was what I meant with the first argument for the bailout - you can make the case it was necessary for the world not to implode, and nothing else matters. But, to support that, a lot of people also said it was a fair thing to do too. Which, I can see, but feels... complicated, given the other things that would also be fair by the same logic.
(And on the term bailout, sure, SVB's execs, lenders, etc didn't get bailed out. Depositors definitely did, though that doesn't necessarily imply that they engaged in something reckless or didn't "deserve" it, or whatever. And yeah, the rest of the financial system was also probably bailed out from a bunch of losses. But again, that doesn't imply some wrongdoing to me; it's just that they changed the rules to protect people.)
I'm not saying that the system is perfect - far from it. But the whole banking system operates on the principle that the bank can invest their deposits (i.e. make loans) to earn interest and make a profit. If too many people (near 100% in SVB's case) withdraw their deposits at once, the bank will not have the liquidity to give them their money back until they liquidate their investments - hence the "run on the bank". What's the alternative? No banks? That means no mortgages, car loans, etc. - money in mattress time. The purpose of the Fed is to backstop a "run on banks" to keep the system functioning which is precisely what they did. And no, crypto is not an answer. Crypto is an example of no "lender of last resort" which is why it's so volatile. Imagine if the value of the dollar fluctuated by 10% or more per day - not good.
Since we don't experience bank runs very often, I conclude the banking system as a whole works pretty well. SVB, on the other hand, not so much at the end.
i can’t help but feel a sense of anticipation that all of your writing over the past few years has been training ground for intimately covering what’s unfolding right now in the economy, banking and tech (no pressure).
I’ve been on an info binge for the past week, and your writing stands out with a few others, that you’ve mostly linked!
Here’s my favorite (i saw you reference patrick so maybe you saw) https://www.bitsaboutmoney.com/archive/banking-in-very-uncertain-times/
… tldr; the banking crisis root cause is flawed data modeling. the entities represented in the system (for the purpose of both administration and legislation) do not accurately reflect reality.
… which, after my last 13 years of work in tech, does not shock me, but does leave me feeling deeply unsettled.
If you continued to analyze/cover this space, i’d pay to subscribe.
Thanks! This was the world I first worked in (back when I wrote a different newsletter: https://carnegieendowment.org/experts/analysis/458), so it's a rabbit hole I pretty quickly fell down. I have no idea if I'll keep at it or not, but either way, I won't make you pay for it (if people pay, they'll expect it to be good.)
And yeah, that was a good post. I think I see the root of all of this as much more economic and politic issue than a technical one. The infrastructure underneath it matters, but my view is that'll get shaped by economic incentives on top.
“the banking crisis root cause is flawed data modeling. the entities represented in the system (for the purpose of both administration and legislation) do not accurately reflect reality.”
Your criticism of economic/financial modeling need not be limited to these sectors, they extend to modeling period. Is there such a thing as a realistic data model then?
Preach it!!🙌🏽
" People and companies losing their deposits wasn’t fair. Startups who banked at SVB didn’t do anything wrong, unethical, or even (we all thought) risky; all they did was open a checking account. It wasn’t their fault; they were victims of mismanagement and an economic shock; they deserve to be made whole. "
If you deposit more than 250k in a bank, losing money over 250k is *at least* partially on you.
If I'm a founder and I just got a 50 million dollar round, I'd be very careful about where that money goes because 49.75M of that is exposed. This is not opaque. As the founder I have a duty to protect that money. No one should assume that putting money in any bank is a riskless exercise. Sounds like the founder/VC/SVB bankers were all a little too cozy.
Kinda? I get that, and certainly think things like the Roku case are pretty irresponsible. https://www.cnn.com/2023/03/10/business/roku-svb-cash/index.html And sure, of course it'd be prudent to put most money somewhere else.
But I think the general view of the modern banking system is (was?) that the 250k isn't there to say that 250k is safe and the rest isn't; it's there to short circuit bank runs, and make deposits safe in general. Ie, if most people don't have 250k, most people won't run on the bank, so there's no critical mass to create a run, and everyone's deposits are fully safe. I get that's not the letter of the law, but I don't think it's unreasonable for most depositors to basically view it that way. And in that way, SVB's failure is more on regulators than depositors, since regulators are the ones who would've been in a position to realize that the 250k limit didn't protect SVB from a run. (Obligatory matt levine citation that makes this point: https://newsletterhunt.com/emails/25336).
On the cozy thing, I think it's more laziness than coziness. I saw reports that VC would tell people that they should use SVB, or that using SVB had some sort of signal value. In my experience, none of that was really true. It was just that SVB was there, they marketed to startups, and most companies didn't give much thought as to who they banked with. And if they did, it was about bells and whistles, not because people were assessing a bank's creditworthiness.
"I don't think it's unreasonable for most depositors to basically view it that way."
I do. It's unreasonable and irresponsible. It's not like banks have never failed. And cash is the lifeblood of a startup, so I'm more than a little surprised that apparently up to 50% of the startup community were at risk from a single bank failure.
I'm not arguing that this is *all* the fault of the depositors, btw. And sure, I get it that founders are focused on other things and most startups don't have a finance pro on the team. But still! Depositors have a duty to be the best stewards possible of their money and should understand the risk. And VCs should, too. Little surprised they weren't more on top of this, but I'm learning that VCs are not nearly as sophisticated as they would like us to believe.
Of course, there is no riskless place to put your money. But there are well understood ways to reduce risk - all your money in one bank is not the way. I don't think this requires a degree in finance to understand.
Hopefully founders learn from this and get less lazy.
Also explicitly *not* arguing that depositors losing money is 'fair'.
Also not arguing that the government doesn't have some responsibility for this. We all know banking regs exist and few people have time to understand what they will and will not do to minimize risk.
But having no control over the bad decisions of bankers and government is all the more reason to be careful with one's money.
Yeah, and to be clear, for a depositor with $50 million, I'd expect them to treat that differently than someone with $500k. My point is mostly that I believe it's the intention of the FDIC to prevent runs in their totality, not to encourage people to split up money into 250k increments across a bunch of banks. (So, in intention, it's like to a quality control inspector, who checks a few products with the intention of being able to say they're *all* safe, not just the ones they randomly test. And it's not - again, in intention - like heath insurance, where your policy has a cap and the insurer couldn't care less about your losses over that cap.)
Obviously, it's prudent and responsible to treat the $250k limit as though it were that sort of cap. But that intention seems pretty clear, given that FDIC more or less confirmed it with this bailout.
And on fault, I put a bit on startups for being lazy, a bit on the regulators for not scrutinizing things a bit more - though SVB lobbied them out of the building, so - a good bit on VCs for herding so much money in and out of SVB, a lot on SVB for not realizing all their deposits were from companies that were over the limit and were easily herded, and a lot on - and this is my old man moment - social media, for making it possible that these panics can unfold in a day.
> My point is mostly that I believe it's the intention of the FDIC to prevent runs in their totality, not to encourage people to split up money into 250k increments across a bunch of banks.
Yes! I think this point is being missed by a lot of commentators. The FDIC was created to prevent bank runs, after the US government saw how often they had occurred during the Great Depression era. And tbh the 250k is, at this point, an arbitrary amount.
I don't have much confidence that the bailout is evidence that the FDIC's intention is to prevent bank runs. People with lots of money at stake and politicians must have had a ton of influence. The FDIC originally was not going to save the depositors as of Saturday.
Eh, I'm less cynical than that. I could really see that going either way. The twitter VCs who were freaking out about the FDIC not doing anything pushed the idea that the FDIC wasn't doing anything, but reporting like that Wapo article show that they were doing a lot over the weekend. Obviously who knows what would've happened had there not been public pressure for connected people, but it wouldn't surprise me if we would've landed in a similar place.
(Or, well, I'm not sure the pressure made the difference. I think the fact that SVB's client base was what it was made a difference)
I thought this one was special since startups create jobs and the government has to protect those jobs, so that's why FDIC saved them. If a startup succeed and become the next Google, then it's gonna create more jobs and boost the US economy which is what the government would want, so that's why they get special benefits.
But why would big companies that are already employ a bunch of people not get those privileges?
I think it has something to do with big companies already being billion dollar companies while startups are like little kids. I think the government views big companies as 'adults' while startups are like 'kids'.
Startups need more support just like how kids need support from their parents (government acting like a parent to startups), but once startups fully matured into a big company the government support wears off and the startup bears responsibility now.
I've also think that the public views it the same way as the government. From my observation, people tend to feel bad for struggling small businesses/startups while when a big company struggle, people don't feel bad. Probably has something to do with human psychology.
I get that argument, and - as there was for Garry Tan's - it's got some emotional appeal. But practically, I think it ends up being an additional form of regulatory capture for what's already a very well-off, exclusive, and influential industry.
Really disagree that the government should be protecting these jobs. Government's general track record is to distort economies more than help them.
hmmm by default there are no economies in the world that is free of government distortion: every industry/sector is impacted by government policies and laws in some form. At the end of the day the government will do what it deems to be fair at the moment when an economic crisis occurs.
Right, all economies have some government distortion. I'm arguing that more is usually worse. And I specifically don't like special treatment for an industry, though I think exceptions are logical when an entire economy is facing existential risk...which is not what I think was happening with SVB.
Yeah, fwiw, I didn't really say this at the end on the bit about fairness, but the point of that was to ask "are we so sure that it's *fair* for Silicon Valley to get bailed out?" and not "why don't we help everyone like we did Silicon Valley?"
Yeah, I don't think it's fair to single out Silicon Valley for special treatment - and I've been a direct beneficiary of that industry. But it's an industry with lots of money and influence.
I'm watching how this shifts into a political narrative later this year, the 2024 Presidential race develops.
Like it or not, the two biggest sources of VC money are pension funds and endowments tied to universities and health systems - things that are ostensibly "socially good" for the poor and middle class, that help retirement, access to higher ed, and healthcare access.
Populists on both sides (Warren, Sanders on left; some MAGA remnants on right) have incredible leverage in calling out what fuels all of this.
I'm pretty sure most tech employees don't know that the fancy dinners, cushy office spaces, pet insurance they have is all essentially funded by retirement funds from people who make a lot less than they do - essentially a large, massive scale wealth transfer.
Will be interesting to see how this plays out!
But that seems to mischaracterize why pension funds invest. It's still for the returns. It's not a wealth transfer because if the system works as it's supposed to, it's just an increase in wealth. (Eg, if a pension fund put their money into a hedge fund that returned 100%, sure, the hedge fund might get some nice dinners that are paid for with the pension fund money, but that's a trade that's good for the hedge fund and good for the pension fund.)
If there are no returns, and the whole asset class is a scam, that's obviously different, but that's a very different problem.
Here for the comments
BTW, Benn -
I tend to comment when I'm disagreeable about something, but I wouldn't come here week after week if I didn't appreciate your thinking and writing. I also hope Mode didn't suffer any adverse effects from SVB.
Thanks. And the whole brand of this thing is mostly me being grumpy about things I find disagreeable, so by all means, disagree away.
(And I appreciate that, we weren't directly affected by it. Obviously, if this torpedos the entire economy, that's not great for us, but it'd be weird if that was good for anyone.)
I don't think the question is about fairness at all. Is the Fed raising interest rates "fair"? Without a trusted entity (the U.S. Treasury in this case) providing liquidity, there can be no commerce. We would be back to bartering. We have a very recent example of applying the "moral hazard" rule in the case of Lehman Brothers in 2008. I don't think people realize just how close we came to complete collapse as a result (there's a great Frontline about this that's a must see). SVB made a bonehead move of funding low yield, long term investments with increasingly costly, short term deposits. Lehman demonstrated even more malfeasance and greed. Contrary to what people think, backstopping SVB didn't "bail out" SVB - they're history. Rather, bailing out SVB bailed out everyone in the U.S. (and beyond) who need to buy things to live.
Yeah, that was what I meant with the first argument for the bailout - you can make the case it was necessary for the world not to implode, and nothing else matters. But, to support that, a lot of people also said it was a fair thing to do too. Which, I can see, but feels... complicated, given the other things that would also be fair by the same logic.
(And on the term bailout, sure, SVB's execs, lenders, etc didn't get bailed out. Depositors definitely did, though that doesn't necessarily imply that they engaged in something reckless or didn't "deserve" it, or whatever. And yeah, the rest of the financial system was also probably bailed out from a bunch of losses. But again, that doesn't imply some wrongdoing to me; it's just that they changed the rules to protect people.)
"Rather, bailing out SVB bailed out everyone in the U.S. (and beyond) who need to buy things to live."
That seems a bit extreme. If a single bank - 16th largest in the US - is that impactful on the US and global economy, something is wrong.