[dangjc]

They decided not to hedge on purpose, take the risk, hoping to earn the extra profits. Slate money did a good episode on this last week. SVB has high touch relationships with their clients, and they need money to pay for all that client service. This was a bet that went bad (in a very foreseeable way).

[kwillets]

If they wanted to increase risk/return, why did they buy T bonds in the first place?

[Jensson]

Regulations, they aren't allowed to buy that much of other risky assets.

[dragontamer]

And T-bonds are the riskiest that those regulations allow for.

One can say that maybe our regulations should be tighter. But I'm not sure if we can solve the "banks taking risks that are inappropriate for their own damn survival" problem. Its clear that there will be a bank that's too stupid to figure out and hedge its own risks here on out.

If we regulate with lists of "these are safe" and "those are not safe" piles, they'll still figure out how to buy the "riskest of the safe pile", and get into trouble.

--------

Case in point: if they bought T-Bills instead, they would have been safe. But no, they wanted the higher risk T-Bonds.

[jldugger]

> And T-bonds are the riskiest that those regulations allow for.

From finance podcasts I have learned that the most important bit may have been that the Basel capital requirement for T-bonds is zero. (bonus citation)[1]. From a repayment risk perspective, that makes sense. From an FDIC puts your bank in receivership perspective, it's kind of a moral hazard.

> Its clear that there will be a bank that's too stupid to figure out and hedge its own risks here on out.

Everyone says hedge risks but that costs money that will likely be equal to or in excess of the entire "go long on interest rate risk" strategy profit.

[1]: https://www.investopedia.com/terms/r/riskweightedassets.asp

[dragontamer]

> Everyone says hedge risks but that costs money that will likely be equal to or in excess of the entire "go long on interest rate risk" strategy profit.

Its SVB's responsibility to understand its own depositor base. It turned out that Silicon Valley was full of squeamish lemmings who'd jump off a fiscal cliff at the slightest provocation if their VC superiors told them to.

SVB needed to understand that its $100 Billion in deposits from 2020 through 2022 was based on hype, low-interest rates, and a sudden surge of VC capital. It also needed to understand that taking on more interest-rate risk in the form of buying long-term bonds was a bad idea, because you're now double-dipping in the risk pool.

SVB had something like 50%+ of those deposits in long-term held-to-maturity risky bonds (either 10Y+ Treasuries, or MBS). A more regular bank has only ~20%.

And SVB *ALSO* has a depositor-base who withdraws money as interest rates rise, because VCs lose their money / capital in rising interest rate environments. A more typical bank, like Ally Financial (risky but more typical) has a depositor base of regular ol' Joes who are well under the FDIC $250,000 limit and have no reason to withdraw their money in any economic condition.

-----------

So we're seeing SVB's executives trying to blame everyone else, as if it were the Regulator's job to identify these risks. No you dumb dumbs, its SVB's job to figure out their unique risks associated with Silicon Valley.

And once that risk is understood, maybe then (and only then) is it appropriate to apply strategies that are happening at other banks at SVB.

And "But everyone else was doing it" and "Regulators didn't regulate this issue" are crappy reasons for doing something. No one else was serving Silicon Valley like SVB was. No one else had the stupid amounts of interest-rate risk associated with having a Silicon Valley tech-startup as their #1 depositor / typical customer. No one else had 95%+ uninsured deposits.

-------------

Don't get me wrong. I understand that "doing things properly" would have made less money. But I also understand that "doing things properly" would have led to SVB's continued existence. This second bit is the bit that so many people seemingly forget in this discussion.

[lazide]

Eh, I think they’re pointing out that insurance companies (or counterparties to swaps and similar) try to not be so dumb as price insurance for something that will definitely happen at less than the cost of the thing that will definitely happen.

If the value of these bonds was almost certainly going to drop 10% (say), and that was going to be $x billion - then it’s not like SVB was going to find some way to hedge that and only spend $x million on it once that became obvious. Which by 2022 it was obvious.

[jldugger]

The flighty deposit base is also complicated. There are rules around depositors and capital ratios, and they include primitive models of depositor types. It turns out VC money is more like a "deposit broker" than a retail or merchant account, even when you have rules in place to make borrowers use your other services. We know this in retrospect, but I can't fault them too hard over deposit risk modelling. I would have thought "don't cause a bank run on your own bank" was something a dude who used to work for Credit Suisse would already know. Maybe thats why I shitpost on HN instead of running a high finance company myself.

> Its SVB's responsibility to understand its own depositor base.

Its SVB's responsibility, and they paid a price for it by getting zero'd out. But it's _also_ the regulator's job to look out for such risks, to protect deposit insurance among other things. One professor puts the blame on a shift from regulating risk to regulating process[1], promoted by Alan Greenspan. Who later said of the 2008 crises "those of us who have looked to the self-interest of lending institutions to protect shareholder's equity – myself especially – are in a state of shocked disbelief."

Said differently the regulators aren't there to protect shareholders from the risk of owning a bank. That's just a handy benefit. It seems likely though that this risk would have been easy to spot and the power prevent had regulators been motivated to do so.

For whatever reason, banking seems to attract a lot of "heads the exec suite wins, tails shareholder loses" agent-principal problems. Thats really my point: the cost of hedging risk is the profits -- and profit-sharing -- that you might have earned. Probably _all_ of it. "Not hedging your risk" then isn't just staff and execs being "dumb dumbs" who forgot about it, but a deliberate enterprise and strategy. In the fullness of time we'll probably learn how execs were incentivized to take the risks that took the company down and how they influenced the board to give them those incentives. IDK how relevant it is but Greg Becker was on the board of the SF Fed, while running a regulated institution.

> No one else had the stupid amounts of interest-rate risk associated with having a Silicon Valley tech-startup as their #1 depositor / typical customer. No one else had 95%+ uninsured deposits.

About that... if fed keeps raising rates, I expect more banks to float to the surface. Collateral damage, in both senses. SVB was hardly unique in the 'unrealized losses' category.

[1]: https://podcasts.apple.com/us/podcast/the-regulatory-blunder...

[hbarka]

Regulations weren’t tightened. In fact they were loosened at the behest of SVB. https://theintercept.com/2023/03/11/silicon-valley-bank-used...

[TMWNN]

The 2018 Dodd-Frank revisions had nothing to do with SVB's collapse. <https://news.ycombinator.com/context?id=35231174>