site banner
naraburns  nihil supernum  1yr ago (text post)   34500 thread views

Culture War Roundup for the week of March 6, 2023

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

16
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Silicon Valley Bank crashed just a day ago, and many folks in the VC/startup world are freaking out. I’ve seen predictions that 50-100 different startups will go bankrupt over the next month. This could cause a contagion effect and lead to worse effects, although I’m skeptical of a major economic collapse as some doomsday prophets have discussed.

Apparently the bank was mostly into mortgage backed securities, which lost a ton of value due to the Fed’s precedented* rate hikes. I don’t know enough about finance to confidently hop on my soapbox here - @BurdensomeCount may have a better idea of what’s going on.

As this collapse mainly affects very left coded super technical folks, I don’t expect many on the right to shed tears. That being said I do think this speaks to a larger issue of growth in the economy as a whole. Tyler Cowen has famously backed the stagnation hypothesis, or the idea that overall production has been slowing down.

Tech startups have recently been the major sector looked to for economic growth, especially with all the AI/LLM hype. This collapse not only will slow the industry but shows a marked incompetence from this growth sector which may cool investment there in the future.

How can we sustain economic growth without the recent massive gains from Silicon Valley technology?

We're heading into a period of elevated interest rates, an increased focus on cash flow for new companies (as opposed to quick growth), and likely recession. I'm not sure startups were going to be a major influence in the short to medium term anyway because of this, and I'm a believer that authors of mistakes should have to bear the consequences if we want to learn from those mistakes and gain the ability to hedge against them in the future.

In short, while things are still playing out, I'm not sure this is going to be catastrophic in any substantive way, and am inclined to believe that this will be good in the long term. It was stupid of a bank to cater solely to one industry segment in the first place, let alone one as volatile and vulnerable to insolvency as tech startups.

This will be good in the long term as long as the folks in charge don’t get tricked into the ””””depositor backstops”””” (read, bailing out VCs’ portfolio companies) that various VC and assorted tech people are telling us we need or else everyone will lose their jobs, innovation will grind to a halt, and China will win. I’m not super optimistic.

I agree that we're hitting the full maturation stage of, at least, legacy IT, marketing-tech, and eCommerce that drove the last ~15 years of growth (big caveat: along with the longest/lowest interest rates ever). Energy and physical technology also seems to be entering a new phase of innovation (something that's very interesting to watch is that the Department of Defense has explicitly said "we're going to buy most of our Space capabilities from the commercial space instead of building them in-house.")

I'd also add that the macro indications are great for the US and shitty for ... everyone else? Maybe not our allies in SE Asia but ...

China's demographics were already - long term - untenable. COVID accelerated that timeline and rising global rates are going to seriously ruin their state-fueled growth. Forget GDP, productivity, real estate bubbles ... in 2022, the Chinese population declined. That's 11/10 uh-oh-spaghetti time. Looking to Europe, their collective economics policies of the past 40+ years have led to an impending collapse of the Eurozone, persistently astronomical un-and-underemployment for young workers who are now in their prime working years, and utterly absent technology and innovation bases. IMHO, I think the RUS/UKR war may be a medium or long term blessing in disguise for Western Europe as it will force them to rethinking their energy dependencies and spur a little extra productive government spending (in the form of escalating defense budgets).

Emerging markets keep doing their thing. A friend once shared a good quote (not originally his) with me - "Brazil is the country of the future ... and always will be." Brazil, India, Indonesia, Malaysia, maybe some spots in MENA (though they are too small) theoretically have some demographic advantages, but all of these places do not have the governmental / law enforcement / contractual pieces in place to do business. Say you're an American oilfield services company and you want to help a Brazilian concern set up a drilling field. They're 90 days late on payment. Then 120. Then 180. What are your realistic recovery options in a Brazilian court? You get the impression.

Bringing it all back home ... there is no other market on the planet where anyone would want to park their money right now than the US - upcoming recession or no. It's not even a conversation of "oh, which markets are going to re-emerge first" etc. There is no other well developed market that has the technological, human capital, industrial, and governmental, and demographic resources in place to turn $1.00 today into more than $1.00 tomorrow.

I agree with all of this. Btw I am noticing a bunch of the “give us money” VC and founder crowd keeps menacingly referencing MBS in what feels like an effort to make us think this is some big systemic thing like 2008 all over again, and justify government handouts. Just something I’ve noticed.

I think you're right but my understanding is the direct cause was that startups were facing shaky circumstances so they started to pull their cash, so the bank needed to sell the treasuries they were holding, but because they purchased them in a lower rate environment and longer dated bonds are more vulnerable to interest rate risk, they simply weren't able to get enough money from the sale of those bonds to cover their liabilities (in this case the deposits).

I would argue that there is still immense and low hanging value to be had from IT, just not in the IT sector.

The state of IT in other sectors is more often than not atrocious and relatively simple things can improve efficiency a lot. Unfortunately the IT sector is consuming most of the IT talent so the rest of the economy is left with scraps.

But perhaps this is what you meant.

We’ve always let the dumbest idiot in any crisis fail. Big depositors should have had their cfo do credit checks. The bank didn’t interest rate hedge which every other bank does.

Let them fail. Depositors take losses. That’s normal capitalism. Don’t cry to me that some dude making 300k a year in tech doesn’t get his paycheck on Monday.

I’ve had financial difficulties. You made a bad decision take your loss. Fuck off.

Do you know why they all flocked to this particular bank in the first place? I can only assume that they we’re getting more favorable terms than they would receive at Wells Fargo or somewhere else.

Edit: nvm someone else answered this elsewhere

This tweet thread explains it well: https://twitter.com/jonwu_/status/1634250754486857729

Highlights:

"Say you're a founder (or VC) with a bunch of illiquid but sure-to-be-worth-something-someday startup equity or fund carry, and you need a house.

Typical mortgage lenders won't help you.

But SVB will, at the same insanely good terms they're known for in the venture debt world."

Because SVB understood startups, understood their business, their requirements, the realities of funding, of cash churn/spend, knew all entities involved in SV startup scene, from VCs, through insurance brokers, to startup lawyers etc, and do their business operations were very explicitly tailored for needs of SV startups. Their advice was extremely useful, often nearly as much as the actual financial services they offered, and as a result many if not most VCs explicitly encouraged their startups to bank there.

Apparantly they offered better terms as well as tolerated the flexibility that startup payrolls and accounts receiving typically end up needing much more.

For H1B the risks are way greater. Especially when hiring is slowed down and there is great competition with nationals. And VCs are a bit depleted financially. Not that the anti immigrant in me cares, but there is more at stake.

You have to be special kind of reckless - when your depositors are incredibly volatile to go very long with your portfolio. Especially in a situation in which it was clear that the covid measures will bring hard economic times ahead.

Sad thing is this was a top 20 bank with all the regulator stamps of approval. Will they never learn?

According to Matt Levine, the connection is that the bank made the same kinds of bets on perpetually low interest rates as its primary clients (tech companies) were making.

When interest rates went up, not only did the bank lose money on its low-interest rate mortgage-backed securities and treasuries, but its deposits decreased as the flow of money into dubious tech companies dried up.

The final straw was when the tech companies, who all talk to each other, panicked and withdrew $42 billion in a matter of hours, a hyperspeed bank run.

Silicon Valley Bank, in a moment of endearing innocence, made a public plea for the tech companies and VC funds to support them in their moment of need, the way that they had supported the industry for 40 years. This obviously fell on deaf ears.

The order taking possession of property and business states that, as of March 10th:

  • The Bank's liquidity position is inadequate, and it cannot reasonably be expected to pay its obligations as they come due.
  • The Bank is insolvent.
  • The Bank is conducting its business in an unsafe manner due to its present financial condition.

It also says that the bank was "in sound financial condition prior to March 9, 2023," which seems hard to square with it being insolvent (not just illiquid) less than 48 hours later. Is the bank large enough for its panic-selling of mortgage backed securities to temporarily crash the price?

Also, long-term interest rates (the kind that effect mortgage prices) are much lower than short-term interest rates right now. If the Fed has to keep raising rates, or keep rates high longer than currently expected, long-term rates could rise substantially too, in which case the holders of the mortgage backed securities have to wait until the loans mature to get their full payout.

If the loans are sold immediately at a loss, there won't be enough money to cover all depositors (because the bank is insolvent). How bad the damage is depends on how much of a loss they are sold for, and most companies don't plan for losses on bank deposits.

"Insolvent" is used as a technically term in different contexts with different meanings. As a legal term, and the CA DFPI order you link to is a legal document, it means the same thing as "unable to pay its obligations as they come due". As an accounting term, it means "book liabilities exceed book assets". As an economic term, it means "real liabilities exceed real assets". As a practical business term, it means "all the equity is wiped out and then some".

It is rare for an entity which is economically solvent to end up legally insolvent, because you can almost always do something to raise funds (if nothing else, selling the company) within the de facto grace period on your obligations. A bank run is a rare exception to this - partly because there is no grace period on banks' obligation to pay withdrawals promptly (there are various Twitter threads pointing out that operational improvements to banking have reduced the implied grace period caused by queue at the teller window, and how this made the SVB situation play out faster than It's a Wonderful Life) and partly because of the self-fulfilling nature of bank runs.

SVB's long-dated bonds are liquid, and the FDIC knows what they are worth. If the FDIC is promising to pay uninsured depositors in full at no cost to the taxpayer (which they were as of Sunday night), then they probably have a pretty good idea that the bond losses are limited to the point where they can be covered by zeroing out the common equity ($16 billion at book value) and the holding company debt ($5 billion). The bank would be legally, but not economically, insolvent. The other possibility is that the FDIC has multiple parties who are interested in buying SVB as a near-going concern for $1, and is confident that at least one will pan out. We know that just that has already happened for the UK subsidiary (sold to HSBC on Sunday night).

SVB’s depositors will almost certainly get almost all (or all) of their money back, likely very soon.

From what I recall, the last time this happened in 2008 to Washington Mutual and Wachovia, both banks were acquired within a few weeks by JPMorgan Chase and Wells Fargo, respectively. As far as I know the depositors were made whole as part of the deal. I obviously can't speak to the details here, but the Treasury has ways to incentivize such a merger, and if the shareholders and bondholders lose out, the assets might even cover the deposits.

On the other hand, those actions didn't prevent the rest of the excitement that happened in 2008.

I think the relevant “Silicon Valley” aspect of this is that since so many of their startup clients have been burning cash for the last several months, there have been lots of correlated “fundamental” (non-bank run) deposit withdrawals which have forced SVB to liquidate long-term fixed income investments at a time when they’ve lost a lot of value. It’s sort of fuzzy, imo, whether this is a liquidity issue or a solvency issue. Should you be marking to market on assets that are “matched up” against liabilities with a very long expected duration when whether or not you mark to market endogenously changes the actual duration of your liabilities.

Maturity transformation works when deposit withdrawals are not too correlated. A bank run is obviously an example of correlated deposit withdrawals but so is the slow bleed affecting many startups right now.

USDC has already depegged, they held around 10% of their total funds and a sizable amount of total liquid funds at SVB and currently trading at around 0.92 per coin (as a stable coin it's always meant to be worth $1). I have a small exposure to it personally (4 figures, nothing major) but the amount of drama being caused is well worth it at the moment.

What's the connection to usdc? I'll admit that isn't the peg I'd trust the most although better than tether.

Circle which is the organisation behind USDC kept a significant amount of their money used to back the coin with SVB. They sent orders to take out their money before the collapse but a day later confirmed that when the collapse happened a substantial order hadn't yet been processed and that money is now likely held up until the whole mess is cleared up.

What's the peg you trust the most?

https://twitter.com/circle/status/1634391505988206592 and https://www.circle.com/blog/an-update-on-usdc-and-silicon-valley-bank

USDC spent a few hours at .9, which was odd since only 8% of their reserves are in SVB and at least >50% recovery is very likely. It's climbed back up to .99 now though.

Circle banked at SVB and reportedly had billions on deposit there.

What I haven't been able to see in reports on this is what started the bank run. What or who started the ball rolling with "get your money out now, before this place crashes"? Or did nobody do it, it was just that the companies using it were drawing out money but the bank didn't have the reserves to cover it?

The bank had tons of deposits from venture investees storing their investments there, with far less demand for loans from their core clients, so they bought a ton of bonds (safe bonds bit relatively long dated bonds). As rates rose last year those bonds were dropping in value and with VC investments slowing there was drawdown of those deposits which meant SVB was selling bonds at a loss. Eventually someone saw the writing on the wall (the bank was particularly vulnerable due to needing to sell bonds to liquidate deposits and triggered the run). Most of their deposits are way over the FDIC insurance limit so I'm sure some of the deposit firm's have someone reviewing financials of their credit counterparties and would have triggered the run. I saw a VC saying they were suggesting to their portfolio firms to pull funds on Wed or Thurs but doubt they were first.

Thanks for the explanation, that makes the sequence of events much clearer to me.

The bank announced they were liquidating some assets or something.

That’s right. SVB announced that it lost $1.8 billion on sales of low-yielding securities to meet the bank’s liquidity needs. The bank’s deposits had declined each of the latter three quarters in 2022.

Also, Gary Tan of Y Combinator as well as Peter Thiel were advising their portfolio companies to pull out funds from SVB. https://www.livemint.com/news/world/peter-thiel-fund-advises-companies-to-exit-silicon-valley-bank/amp-11678456061549.html

If the FDIC or other banking entity does not cover deposits, any business that depends on SVB and has a > $125K bimonthly payroll will have to do furloughs or layoffs. That's basically any business above ~15-20 people.

Directors of a company are criminally liable if they ask people to work knowing they have no means to pay them. Wednesday is March 15 (payday, for work done March 1 - 15). That means companies unable to make payroll #2 in March need to furlough or have layoffs before start of work Thursday.

There's something on the order of 1,000 series A or higher deals per year (even in 2022, decreased from 2021). The average time between raises is about 2 years. Thus, conservatively there's something like 2,000 venture-companies that have > $125K bimonthly payroll, and many small businesses that use SVB but are not venture-backed are not counted in this.

SVB purportedly services 50% of all startups per their advertising. From a survey of my VC and startup friends, it seems reasonable to assume that 25% of that are extremely dependent on SVB (e.g. payroll, no cash sitting elsewhere, and incoming customer payments aren't going to cover anything).

If these assumptions hold, we're looking at around 10% of the entire startup ecosystem laying off effectively everyone in March (e.g. either by going under, or reducing headcount so drastically that they're cashflow positive... which for most startups would be extremely painful). Another large batch will effectively go under in April (e.g. they have one months' payroll at another bank but that's it).

So in the short term we're talking about somewhere on the order of tens of thousands of jobs. A lot of future value creation is lost. Sure, some of these startups are the Juiceros or latest crypto scam, but others are meaningful companies that provide meaningful services. The latter group typically doesn't get as much press because they're optimizing for value rather than hype.

In the medium term, if you're a business that requires having an account with a >$250K balance, why would you now use any bank other than JPM? Sure you can do "diligence" on your bank, but SVB had an A rating from Moody's and a "buy" rating from JPM. Now obviously those are bullshit but for anyone claiming that this collapse was obvious -- please share a screenshot of your brokerage account where you made tons of $ shorting SVB.

So the default will be to go with JPM, rendering most small and medium-sized banks uncompetitive.

At the end of the day, SVB's shareholders will (and deserve to) get wiped out. Their bond creditors and such will mostly (and deserve to) get wiped out. I am not for bailouts of either of those parties. And maybe how we think about the banking system where depositors are creditors should be re-interrogated, because who the fuck is wanting to risk all their money for like a 0.5% interest rate? But I do NOT think startups and small businesses deserve to be randomly decimated.

any business that depends on SVB and has a > $125K bimonthly payroll will have to do furloughs or layoffs.

I don't think your math is right. That's $650,000 to $870,000 per person per year. EDIT: Never mind.

Also, you're assuming that SVB doesn't have enough to pay any of the deposits. They probably can pay most if not all of them.

You're also assuming these companies won't be able to borrow more money. There's also the possibility that the depositors will be bailed out. Failing that, why wouldn't their investors reinvest the same amount? If these companies were good investments last week, they mostly still will be next week.

I'm going to register a prediction now with 80% confidence that there will not be a large number of layoffs because of this. Let's say under 1,000.

I don't think your math is right. That's $650,000 to $870,000 per person per year.

250000 USD / 15 = 16,667 USD per person per bi-week. 52 weeks in a year, 26 bi-weeks, 26 * 16667 = 43,334 USD/year. Which is... pretty low, these days, even by startup standards. Corrected: 433k USD/person/year, nevermind.

((In practice, any business worth mentioning has non-payroll expenses, on top of this.))

Also, you're assuming that SVB doesn't have enough to pay any of the deposits. They probably can pay most if not all of them.

The issue is less how much SVB can pay out eventually, but also how long it takes for that to be resolved. If you don't have a functional bank account Monday, it's probably illegal to let anyone work for you even if you have a very-likely-to-be-valid IOU. And it might be the better part of a month before you get all of the assets you're going to get.

((This may even be a problem the other direction: if you do payroll monthly and have until the 31st, that's great, but you might still be trying to spin up a whole new bank account and connect your payment processing system to it because some processors won't hold money for you (or in Paypal's case, shouldn't be trusted to).))

Now, this ultimately boils down to a Weird Cashflow Problem, and Weird Cashflow Problems are absolutely the sort of thing that your typical startup (and even a lot of small businesses) have a lot of experiences dealing with, and the ones new enough to not have that experience can also have a ton of vendors quite happy to be paid for the privilege of giving that experience. Some will have open lines of credit; some very few will actually have available cash on it. Most businesses will get loans, or additional venture funding, or pull cash out of a funder's 401k, or sell a kidney, whatever, if nothing else comes up; some few others will be able to pause for a week or two, or have an executive willing to bet his or her bacon that his employee #2 means it when promising not to sue for a late paycheck.

On the flip side, that Weird Cashflow Problems are a thing in startups also makes employees incredibly risk-averse about them. A car mechanic may -- though I wouldn't be the business on it -- look askance at an employee who fucks up direct deposit once, but he's got Options. Startup employee, less so.

Failing that, why wouldn't their investors reinvest the same amount? If these companies were good investments last week, they mostly still will be next week.

Investors will usually evaluate every opportunity individually; there's a specific name for anyone not doing so, but the general one is 'fool'.

But more specifically, this event itself is a reason to reevaluate a lot of investing. Even outside of the question of more widespread bank runs, since other banks have a lot of exposure to interest rate risks, this particular institution was very well-regarded mere months or even weeks ago. It's normally great business to be selling shovels, but this is the sorta time that approach dies.

I've read elsewhere that the FDIC will make sure they have access to their deposits by Monday. But failing that, I don't understand why a company couldn't open a bank account at another bank on Monday and take out a loan.

I've read elsewhere that the FDIC will make sure they have access to their deposits by Monday.

The current FDIC page says "All depositors will have full access to their insured deposits no later than Monday morning, March 13, 2023. The FDIC will pay uninsured depositors an advance dividend within the next week. Uninsured depositors will receive a receivership certificate for the remaining amount of their uninsured funds." (emphasis added).

((And a lot of the assets for that might be currently worth only ~60-70 cents on the dollar. Which won't be that way forever! But that risk is a discount factor.))

But failing that, I don't understand why a company couldn't open a bank account at another bank on Monday and take out a loan.

The median company affected probably will over time, although that comes with its own costs and overhead. Monday specifically is a harder sell; business accounts and business account transfers can be a lot of paperwork and checking, and even if the startup has crossed all their is and dotted all their ts today, it doesn't necessarily mean that the banks will (or may even be able) to finalize an account the same day, especially for anyone without an SSN or with a more complex situation. There's a lot of frictions to KYC that aren't obvious to normal people.

This is further complicated if most uninsured depositors are taking 10-20% losses on their accounts, because then most mid-sized businesses who aren't already banking with someone too-big-to-not-bailout will be trying to make a new account with them, startup or otherwise; anyone who's read and understood what this chart means knows that they can't be sure their small bank isn't vulnerable to the same problems.

But at a deeper level there's a small number of startups that probably can't get an account in today's environment, and lot of people in the startup realm that any sane loan provider's just going to turn up their nose against for the terms they need. Some people went to SVB because it was a lot more willing to accept situations like "oh, I get one lump payment of ten million dollars, and don't have any other employment history ever and probably won't have any more cash coming in for two years", or your business case requires a big-boy credit card and you're not getting one without a credit score you don't have, or because the smaller banks are not Wells Fargo and that can matter.

Some of that's not even wrong -- a Jucero-level business that has a perfectly good piece of paper claiming it will get a payout in mumblemumble months isn't actually guaranteed to have that payout, or whatever fraction of that payout makes it through the fire sale, or even that perfectly good piece of paper in six months. It doesn't even have to be intentionally-fraudy!

From that chart’s thread:

In 2008, the problem was that people lost their jobs, stopped paying, and couldn't sell their house because EVERYONE had lost their job.

So what we did was say that any foreclosed mortgage is instantly paid back by the US Government. No more interest, but no risk.

Is that what the zero interest regime this past decade really means? The idea shakes me.

I'm... not sure. The core definition of (near) zero interest is 'just' that the Federal Reserve will only offer (nearly) zero return for reserve funds or other effective loans to the government, and thus banks who are looking for investments should start looking at anything even slightly better than par once other risks are included. Which has a lot of complicated downstream effects, admittedly, many of them either marginal or just-not-good.

I think PoiThePoi is referencing the bigger emphasis on mortgage insurance and (effective) backstops from the federal government for mortgages, along with some other complexities. I think those remain even (maybe especially) at higher interest rates, they're just more boring.

anyone who's read and understood what this chart means knows that they can't be sure their small bank isn't vulnerable to the same problems.

I saw that tonight and could not figure out what it meant.

As financial products, rather than bank services, long-term securities like bonds have a sale value that can be (and often is) entirely different from the face value or estimated face value, and the transition from one format to the other can be very expensive. The best-known reasons for a discount over the face value is credit risk, that of the contract not being fully fulfilled (such as for bankruptcy, default, death, etc).

Traditionally, some assets were considered very-close-to-perfect for that sort of risk, for which you'd also get very little interest on them. Patio11 uses "marketable Treasury securities" as an example of something that you can almost always sell with very little slippage as recently as July of last year, with long-term known-good home mortgages just under that. And quite a lot of banks had a lot of this class of product: SVB was higher at ~50-60%, but a lot of very traditional banks have around 20-30%. While these are virtually always going to pay out, their sale value depends on other options.

Almost all of those options are downstream of target rates set by the Federal Reserve. In a variety of complicated ways, the Fed will offer or encourage people to offer loans at certain rates, or tell people that they can get a certain rate of interest for leaving money in their reserve accounts; most normal-people long-term loans face some further >2% (sometimes much more, if your credit sucks or the loan duration is weird) rate above that. If these change rapidly, the sales values of loans from before the change can get weird, because there are often going to be better options for your cash.

Today, someone buying 10-year treasury bonds (uh, 'note') is comparing a 2020 bond at 1.5% (or even 0.6%!), or at 2023 bond at 3.5%-4% (or even a seven-year 2023 bond at those rates). You might still make that first purchase, if only because there aren't infinite stock of the latter. But you're not going to get anywhere near the same offers as you did in 2020, even as the final value of the bond is still the same. And the same goes for non-t-bill loans.

In many cases, this is a pretty big discount now. Banks have to actually mark on their sheets the losses for available-for-sale securities (the brown lines), but they don't have to mark hold-to-maturity ones (the blue lines) unless they have to sell them or anything else in the same portfolio. But from a strict definition of liquidity, both count. And that chart -- from the Fed! -- says that they've lost about 600 billion USD. Not all of that's 'real', in the sense that a lot of the hold-to-maturity stuff (and even a small amount of the available-for-sale stuff) will actually get held to maturity and cashed out at its actual face value, or at least held until the "this cashes out soon" is worth more than the "the rate sucks" bit. But the more people have to sell in the short term, the more 'real' it gets.

Which wouldn't be that bad, except 600 billion is about 1/3rd of the 2 trillion dollars that makes up all equity in the US banking sector period. That's not quite as apples-to-apples comparison, and a lot of it's really localized (again, SVB had nearly twice the exposure as the average bank), and even if all the equity vanished that's the non-depositor money. And if nobody calls the pot, it doesn't real. But it's a big chunk of what's available for day-to-day operations.

And that chart -- from the Fed! -- says that they've lost about 600 billion USD.

Who has lost it? The fed or the entire bond market? That's what I'm confused about.

More comments

52 weeks in a year, 26 bi-weeks, 26 * 16667 = 43,334 USD/year. Which is... pretty low, these days, even by startup standards.

You're out by an order of magnitude. It should be 433k, which seems unrealistic even in Silicon Valley.

Corrected, and thanks. That does significantly reduce my concerns. Still a huge problem, but every 40- or 50-person employer is a smaller field.

SVB purportedly services 50% of all startups per their advertising.

Why? Do big banks not want to service startups? Did SVB offer better terms?

It's more a historical thing. 25 years ago, when startups were less of "a thing", a lot of traditional banks didn't approve a startup account because the below looks really weird if you're used to servicing traditional businesses.

  1. Someone with no commercial history or credit

  2. Who wants a credit card

  3. Then who one day deposits millions of dollars

  4. And the next months dollars get sent out and the bank balance goes down

  5. With minimal consistent income

These days it's more that you ask some random person in the startup world, VC, or lawyer, and they go "yeah, a plurality of the people I know use SVB" and that's not where you spend your precious hours as a founder trying to differentiate your company so you just go with the flow.

Though, next week every single founder is going to be taking money out of First Republic, Citizens, Fifth Third, Capital One, BNY Mellon, etc. and wiring it straight to JPM. I suspect there will be a broader bank run.

Though, next week every single founder is going to be taking money out of First Republic, Citizens, Fifth Third, Capital One, BNY Mellon, etc. and wiring it straight to JPM.

Isn't that a bad idea, though? Doing the exact same "all our eggs in one basket" that took down SVB? Being an ordinary idiot, I would have thought the lesson here was "Okay, for the love of God let's keep our payroll in this bank and make sure we have enough to cover six months' wages, then put the running costs money in this other bank" and so on. I realise that makes it a lot of bother to set up and maintain individual accounts, but if you've just had "oops, the bank where we kept all our money just imitated a dead goldfish", wouldn't you try and spread the load around?

It doesn't have to be JP Morgan in particular. Any of the Big 4 (JPM, Bank of America, Citigroup, and Wells Fargo) are both well-diversified and too-big-to-fail. I don't know why anyone with deposits over the FDIC insurance limit of $250,000 would have sensitive money anywhere else.

Apparently the traditional banks were right to be worried.

How so? I’d really like to understand the logic of this position.

I suppose if you're a big bank and startups are only a tiny fraction of your business it would be okay, but its hard to manage risk if you can't predict when customers will deposit and when customers will withdraw.

Cramer cursed JPM, though, so the question becomes- how superstitious are the people making decisions about this?

SVB offered higher deposit rates than most big banks. They were paying 4.5% on money market deposits. Chase is paying 0.01% on everything account short of 48 month CDs which get a generous 1.49%.

This is the key detail I was looking for and it makes me a lot less sympathetic to the depositors now demanding a bail out.

In fairness, it's not good if the safe (because the government can't afford to let them fail) banks use their position to pay peanuts on their deposit rates but attract deposits because no one can bear the risk of banking elsewhere.

I think it's perfectly fine if lower-risk deposits pay substantially lower interest rates

It's one thing if JP Morgan/BoA/Wells were lower risk because of better management but they're lower risk mainly because the US Government can't afford to let them fail so if this happened to them the Fed would have swapped their bonds at par or launched a new round of QE.

I'm not a huge fan of the government picking winners and then those winners taking monopsony rents as a result.

The big banks are regulated in a way which reflects their too-big-to-fail status. As part of this, they are required to prove that they are not doing the specific thing that SVB did - i.e. taking non-mark-to-market interest rate risk by investing floating rate customer deposits in long-dated fixed-rate securities.

Medium-sized banks like SVB were exempted from these rules in 2019 - to quote from the SVB annual report,

In October 2019, the federal banking agencies issued rules that tailor the application of enhanced prudential standards to large bank holding companies and the capital and liquidity rules to large bank holding companies and depository institutions (the “Tailoring Rules”) to implement amendments to the Dodd-Frank Wall Street Reform and Consumer Protection Act (the “Dodd-Frank Act”) under the Economic Growth, Regulatory Relief, and Consumer Protection Act (the “EGRRCPA”). Under the EGRRCPA, the threshold above which the Federal Reserve is required to apply enhanced prudential standards to bank holding companies

increased from $50 billion in average total consolidated assets to $250 billion. The Federal Reserve may also impose enhanced prudential standards on bank

holding companies with between $100 billion and $250 billion in average total consolidated assets.

(SVB had $211 billion in total assets)

SVB CEO Gregg Becker was personally involved in the campaign to relax the rules. He stood up in front of a Congressional Committee and said it was OK for banks like SVB to do this stuff because they were not too big to fail and wouldn't need a bailout.

Among other things, SVB pioneered extending loans to startups as part of their funding rounds.

It's not the lack of cash, it's the timing of it.

Directors of a company are criminally liable if they ask people to work knowing they have no means to pay them. Wednesday is March 15 (payday, for work done March 1 - 15). That means companies unable to make payroll #2 in March need to furlough or have layoffs before start of work Thursday.

How many will be able to secure funding from VCs (who may themselves have funds tied up in SVB) before Thursday?

Probably almost all of them. Why would an investor let their investment go to waste by not providing the cash letting the company fail?

I'd like to register the prediction that zero startups with a valuation over $100 million will fail to meet payroll due to this.

For one, there is a high likelihood that SVB will be purchased and all depositors made whole by Monday, or at latest mid-week.

At what odds?

85%. Am I willing to be actual money with an internet stranger about it? No.

Edit: Bill Ackman just tweeted the following which increases my confidence in this particular prediction but overall seems to indicate potential contagion. It also makes me wonder what trash Bill Ackman is currently holding which is poised to dump without government intervention:

From a source I trust: @SVB_Financial depositors will get ~50% on Mon/Tues and the balance based on realized value over the next 3-6 months. If this proves true, I expect there will be bank runs beginning Monday am at a large number of non-SIB banks. No company will take even a tiny chance of losing a dollar of deposits as there is no reward for this risk. Absent a systemwide @FDICgov deposit guarantee, more bank runs begin Monday am.

Banks typically don’t move that fast. Credit funds will make some money.

n=1, our exec team spent all of Friday working on this and report we will be able to make payroll and that our two big name VC funders are ready to help if needed.

Given our company size I’m guessing every payroll is about $600k.

I’m of the opinion that the overwhelming majority of tech startups are overvalued and only existed as a result of absurdly low interest rates that no longer exist. Many of them would have failed eventually as a result of their inability to raise money and if this pushes some of them out sooner that’s a good thing.

More fundamentally I hope that this pushes some of the intellectual capital currently wasted on basically pointless ventures into more productive parts of the economy.

China did this explicitly a couple of years ago by straight up banning tech startups and I believe it was good policy (https://www.wired.co.uk/article/china-tech-giants-policy). I’m glad that interest normalization is having the same effect.

I’m of the opinion that the overwhelming majority of tech startups are overvalued and only existed as a result of absurdly low interest rates that no longer exist.

Speaking as a software contractor who primarily works for startups, and lost a client yesterday due to SVB, I would say this is not just your opinion but the general understanding of how all this works. Nobody in my world is particularly surprised by any of this; it's just how the wheel turns.

I think most people have no experience with startups and thus no appreciation for how absurdly productive a small group of unbridled engineers can be.

I'm not generally worried about the engineers being productive, I'm usually worried that the leadership has no clue of what the engineers should build, and leadership is like throwing around marketing terms and buzzwords to get capital with little idea of if they'll be able to get customers.

And not only that, these insiders often use sky-high valuations to cash out for hundreds of millions of dollars. Index investors and pension funds are left holding the bag while executives in Carvana, Lemonade, and other "tech" garbage get rich. There are so many companies that never had a viable business model and should never have existed.

This is the tech bro version of the labor theory of value. I think nobody seriously disputes that a small group of unbridled engineers can be absurdly productive in creating X; the question is more like, “who the heck needs X?”

Or, in many cases, the far more fundamental question of ‘what is x and does it have potential applications?’.

I think you may give media too much credibility in portraying what startups work on.

A company like Juicero is exceptional, but it’s also not even an a priori bad idea. The rich buy all kinds of luxury goods that seem absurd to most people.

I have a $3,000 chair that provides massages that are maybe half as good as those provided by a person. The Osaki company that makes it, as far as I know, is still in good shape.

I have a robot that does a poor job of cleaning my floors and that frequently gets lost or stuck. iRobot is still around and I’m guessing profitable.

I guess the question is how often do you use your chair and how often would you get a chair massage in real life. One could imagine a pay back period in 2 years. Then provided the chair pats for say 10 years you get 8 years of half as good massages for free. Maybe now you only go once or twice a year for the real deal. Could on net be worth it.

Same with iRobot. It isn’t a replacement to cleaning generally; but can reduce the need to clean.

I live/work in the middle of all of this with what I think is a pretty good big picture view. Almost everything these guys do lately is some combination of absurd and/or hyper niche, and at the very least incredibly overvalued.

Overvaluations were at least partially corrected last summer. Lots of startups have been unable to raise new funding rounds as a result.

Though I'd argue about the definition of "overvaluation." Is something overvalued if the market has access to practically free money and just needs to park it somewhere?

That’s true, I have never worked at a start up and am basing my opinion off of some wasteful enterprises I have heard about second hand. Although I would ask at what level you think svbs depositors should be bailed out at above 250k if any?

From a moral hazard perspective, zero.

That’s harder to justify if the people bearing the cost of avoiding the hazard nearly all had nothing to do with it. There’s a risk of cascading failure that could cause failures for companies barely related to SVB, but I’m not knowledgeable enough to quantify that risk.

It’s not a question of the productivity of the engineers, it’s a question of the value of the products they’re building.

That some of what’s being built will end up not being worth anything is built into the startup funding ecosystem.

If VCs could scry a crystal ball they would, but judging what will turn out to be valuable is hard. Even so, with all the failures, the sector as a whole is still absurdly productive.

It’s almost criminal that so few people are able to participate in this part of the economy due to accredited investor rules.

What are those "more productive parts of the economy" in your opinion?

Arnold Kling on SVB points out mark to market vs held to maturity.

When interest rates go up, the value of a portfolio of fixed-rate bonds or mortgages goes down. Roughly speaking, if the bank paid $100 to buy a long-term bond with an interest rate of 2-1/2%, and now the interest rate on a comparable bond is 5%, the bank’s bond is worth about $50.

The regulators should make you mark down the value of your assets to their current market value and force you to shore up your capital. They should make you stop paying dividends and executive bonuses, for one thing. You should not be allowed to make any more risky loans, because of the moral hazard: if the risk pays off, you return to profitability; if it goes badly, then the taxpayers take a bigger hit via the deposit insurance fund. You have an incentive to take desperate gambles.

If the bank is genuinely solvent on a mark-to-market basis, then any gambles it takes are with the shareholders’ money. Once it is insolvent on a mark-to-market basis (or “semi-insolvent” as Tyler put it), it is taking gambles with the deposit insurance fund’s money.

But instead of requiring banks to mark long-term bonds to market value, the regulators give banks a loophole. They say that if you have your bond in a “held to maturity” account, which means you do not intend to sell it, you can pretend that it is still worth $100.

Indeed, it is true that when the bond matures, say, in twenty years, you will get your $100 in principal. But in the meantime, the interest rate that you pay to depositors will have gone up. If you have to pay 4-1/2 percent interest and you get 2-1/2 percent interest on your bond , then you lose $2 per year. For twenty years. You will run out of capital and be busted before the bond matures. When people can see that coming, there will be a run on your bank, and you will be busted right away.

The “held to maturity” loophole blinds regulators to the true condition of the bank. It allows the bank to keep taking in short-term funds, with which it can then try desperate gambles where the losses are borne by the deposit insurance fund. That is what the S&Ls did back in the 1980s. It did not end well.

First off, setting his bond calculation math aside… this is a FASB accounting rule, not a regulation per se. Also, most banks don’t designate their bonds as HTM. Regulatory capital rules allow most banks (except for the very largest) to exclude unrealized gains or losses from capital calculations. Marking them to market would produce significantly more volatility in capital levels. Shoring up capital is well and good, except that most banks don’t have easy access to capital markets. They can’t just raise more capital on a moment’s whim, even if these are paper changes in value. If a bank chooses to hold onto a low-yielding bond that drags against their net interest margin, then you’ll see that reflected in poor earnings performance. This information is reported to the public each quarter by all banks, so it’s not a question of transparency.

The accounting designation rules for securities is not the main problem here, it’s that interest rates stayed too low for too long, and the bond market had its worst year in 40+ years in 2022. The Fed’s recent announcement to keep interest rates elevated was a nail in the coffin.

the Fed’s unprecedented rate hikes

Unprecedented, or unexpected? Even the softer version seems tough to defend. Were we going ZIRP forever?

Well, where is r* and where has it been trending?

You don’t need to know a lot about finance. Basic principle is that bonds price is inversely related to yield. So if I own a bond with a yield of 5% and the general interest rates rise, then my bond if I sell it is worth less. If the yield decreases, then my bond is worth more. Both assume the creditworthiness of the borrower is unchanged.

So the bank had a lot of assets with low yield. Yield increased meaning the bank’s assets decreased. Which meant people started getting worried that the bank didn’t have sufficient assets to cover liabilities. This, among other things, led to a bank run.

To explain this relationship further to the layman audience, let's say the interest rate (which is directly related to yield) on Treasuries is 0% (and for simplification, that's the only other investment option, and the market doesn't price in any potential for that interest rate to change), and you buy a $100 bond that promises to pay 2% interest over the next 30 years. You're buying the bond and anticipating getting $182 at the end of 30 years!

The day after you buy the bond, the interest rate on Treasures gets raised to 5%. Now, in order to get $182 in 30 years, someone can just buy $42 worth of Treasuries today. So your "bond" is now worth less than $42 -- because why would they buy your bond when they could buy the Treasuries instead?

Tech startups have recently been the major sector looked to for economic growth, especially with all the AI/LLM hype. This collapse not only will slow the industry but shows a marked incompetence from this growth sector which may cool investment there in the future.

I mean, is it a bad thing if these companies go bankrupt? Do we really need more companies in the world like this?

Imagine you had a team of people that could scour the internet and start finding the clients that most needed your services. Now imagine being able to do that without hiring anyone. Wouldn’t that feel like cheating?

That's what you get with Leadsniffers. An AI trained with billions of data that knows exactly who would benefit most from your service according to your niche and industry. Simply fill out some information about your offer and let the AI do the work for you. [...]

Our AI analyses each of your leads by compiling hundreds of data points from millions of pages on the web, from websites to social media platforms.

This provides you with the ideal message to send with your Gmail account to maximise your efficiency without having to sacrifice quality and personalization. You can choose to send the message automatically or to wait until you approve it.

Plenty of conventional businesses are shit too. What about Helion Energy or OpenAI or Stripe, or something like Samsara or anduril?

The alternate to AI doing it is people doing that. I did that by hand at my job. When publishers make books, they need to get someone to buy them - preferably professors who then proscribe them for their class.

This is inevitable unfortunately. Spam communication is about to 100x my friend. LLMs are coming for a sales team near you.

I think I've already gotten a couple of these already. Very personalized emails that reference OSS work I haven't actually done but is vaguely related to other OSS work I have actually done. The same type of error LLMs always make when you ask for book recommendations where they make up non-existent books that sound plausible.

Fed’s unprecedented rate hikes

The rate hikes aren't unprecedented. Here's the history of the bank loan prime rates. Any bank that blew up because of that 2022 change was engaged in catastrophically bad decision making and pretending that the extremely low-rate environment was permanent.

Yep, there's a few of these firms every rate cycle last time it was FBR the DC investment bank that drank the koolaid on leveraged MBS.

Well that’s what I get for sourcing my information from hacker news.

Hacker News commenters are a very mixed bag. Sometimes you get an expert or insider sharing gems you won’t find anywhere else.

Most frequently you’ll get midwits sharing their slightly informed speculation as if it were fact.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats