Hacker News Reader

Comment by fosk | Hacker News Reader

The Fed first said it would not increase rates, then it said inflation was transitory, then all of a sudden they did a 360 and increased the rates the highest in more than a decade, then they are saying they will keep increasing, then something will inevitably break, and then they will start cutting again despite their original plans.

Sure the banks should have managed risks better and this whole fiasco falls on SVB, but to be fair the Fed is doing a horrible job at setting expectations. Whoever trusted them in the past, got screwed. They are fundamentally a reactive organism that for some unknown reason talks as if they are the ones in charge.

Obviously they are not.

View on HN

When you're paid millions of dollars its your job to hedge risks.

When you're just a small renter, it's fine for when the handyman doesn't show up the apartment keeps flooding. When you're a 200 unit apartment owner, you better have a list of a dozen handymen that you can go down when the first one doesn't show up.

These banks are the equivalent of the apartment owner. They are expected to be able to survive any decision the FED makes short of the FED shutting them down and even in that case they should have known ahead of time and been working on an appeal.

> When you're just a small renter, it's fine for when the handyman doesn't show up the apartment keeps flooding

Not to overly quibble but it really isn’t. If we let the small renter get away with such egregious behavior the bigger renters will take note and do the same

The analogy is stretched, for sure. I think the analogy works a bit better if you say that one apartment building houses enough people that there aren't enough hotel rooms to house everyone if the building becomes uninhabitable, thus making the apartment building structurally significant to the local housing supply in some sense.

But again, it's just an analogy.

Still feels like it breaks. Given the new analogy, proper planning would be to set policy to steadily increase housing supply relative to renters based on likelihood of them being able to purchase a home, adjusted quarterly (or maybe yearly) and/or increase the number of available renter units relative to the amount of new renters coming online (IE, graduating college or moving into a new job that requires them to live in the area).

Don't be overly dependent on any one local thing in another words. You can't 100% plan for total failure of course, but this would blunt it significantly.

In another words, SVB did a bad job at hedging risk, and there's really no excuse for it.

> Still feels like it breaks

I'm not in love with the analogy, but agree with the underlying argument, fwiw.

> Given the new analogy, proper planning would be to set policy to steadily increase

Ironically, on this count, the analogy actually works!!! Bank runs sort themselves out on the sort of time frames that "steady" policies operate. Just like housing supply sorts itself out in the steady state (well, apparently not, but we need that assumption for your point to work) BUT steadily adding supply is NOT a good solution to "we literally don't have places for people to live because your apartments flooded and there isn't enough temporary housing to keep the city working".

> In another words, SVB did a bad job at hedging risk, and there's really no excuse for it.

Yeah I think we're violently agreeing. (Additionally -- FDIC/USFG bailed out SVB depositors above and beyond what they should've done.)

OK but the banks were punished. Their shareholders got their equity zero'd out. I'm not sure what more you want. The question was whether to punish the account holders at the bank.

By your analogy, that's saying the apartment renters needed to have handymen oncall in case the landlord couldn't manage it!

> I'm not sure what more you want.

Cover $250K then stop. You know, the actual contractually obligated about of insurance and not a penny more. I don't know why this is controversial or difficult to understand!

> OK but the banks were punished.

The banks weren't the only ones who benefited from risk taking, though.

The account holders were getting attractive returns on their cash, made possible by the bank's risks. Clients got to enjoy unlimited upside of the bank's risks while having the rest of society bailout their downside.

> unlimited upside

Oh man those interest rates must have been amazing! What were they?

>=400 basis points over the risk-free rate.

250k is an extremely small amount for a company of any significant or insignificant size.

Furthermore, the returns at SVB were not that spectacular, I assure you. They were just a very startup-friendly bank.

> They were just a very startup-friendly bank.

Americans have a million euphemisms for corruption.

You’re welcome to be more specific, but I suspect your snide sarcasm has no basis.

Being willing to work with startups has nothing to do with corruption, at all. Period.

It's not a euphemism for corruption.

Quite literally a bank is not required to make you a business account. A bank is not required to give your business a loan. The list goes on.

SVB was much more willing to do things other banks wouldn't for startups; this is what people mean.

Now, there are a bunch of claims that SVB gave startup founders discounts on personal mortgages for having their business accounts there. This is just gravy on top of the fact that SVB may have been the only bank to accept the business account anyways.

Precisely this.

>Now, there are a bunch of claims that SVB gave startup founders discounts on personal mortgages for having their business accounts there.

Exhibit A in why it’s not even worth replying.

> Quite literally a bank is not required to make you a business account.

Banks actually want deposits, so unless you are involved in a shady enterprise, are penniless or insolvent, or are the type of business that presents an outsize risk of demanding your deposits all at once, you are highly likely to be granted the privilege of lending the bank your money. Not to mention being paid relatively little for the risk you are taking.

> 250k is an extremely small amount for a company of any significant or insignificant size.

How does this justify handing out free insurance post facto on everyone else's dime? Running a business involves taking and managing risk. Or at least that's what we are told when founders and VCs are the only ones who get rich during exits. But as soon as they get burned the masses are there to bail them out. Tails you win, Heads I lose.

> Furthermore, the returns at SVB were not that spectacular, I assure you.

Free money/services are free money/services. It's nice that the additional interest and in-kind services are pennies and nickles to you multi-millionaires and billionaires, but lots of Americans could've happily retired on the total actual value of those "not spectacular" returns, so the argument that "it's not that much money" is going to fall on mostly deaf ears.

> How does this justify handing out free insurance post facto on everyone else's dime?

Wasn’t trying to; was stating a fact. I believe the insurance limits should be higher in general. Also, this was paid out of bank-funded insurance, not any direct taxpayer funds.

> It's nice that the additional interest and in-kind services are pennies and nickles to you multi-millionaires and billionaires

You’re making a lot of assumptions here, so perhaps consider checking yourself.

Furthermore, the SVB returns, for a consumer or bank customer, were not out of line with other banks. The unique part about SVB was their willingness to work with startups, where other banks simply wouldn’t consider it.

Nobody said “it’s not that much money.” It being a lot of money doesn’t somehow instantly mean it isn’t worth backstopping a ton of private citizens, most of whom (no matter how loudly you protest otherwise) are not particularly wealthy, and startups that relied on SVB having been the 16th largest bank in the country, and thus “generally regarded as safe.”

> Also, this was paid out of bank-funded insurance, not any direct taxpayer funds.

It was paid out of unrelated bank-funded insurance. If I insure my cat, and I get a payout when my dog dies, it's not from "bank-funded insurance." Not only was my dog never insured, so I never funded its insurance, but the cat insurance also wasn't somehow transformed by proximity or similarity into dog insurance, although they both lived in the same house and are both pets.

The reason this is clearly true is because if hypothetically a week from now a few banks failed that held enough accounts under the federal insurance limit to deplete the remainder of the insurance fund, the failure of the insurance would no doubt also be backstopped. It's like stealing from the cancer fund to pay for plastic surgery, because you know that if the cancer fund runs out, people are going to be willing to shore it up.

> It was paid out of unrelated bank-funded insurance.

Every insurance system on the planet works like that, or is intended to, and FDIC insurance is no exception. Insurable risks covered by a large customer base. Institutions that present a higher risk pay higher rates.

The FDIC cannot actually go bankrupt under any conceivable conditions because by law FDIC member banks are required to cover the costs, even extraordinary costs as might be presented by a major financial crisis. And fortunately the sector as a whole has ample resources to do so even under conditions as severe as the Great Depression. The $128 billion that the FDIC has on hand is nothing compared to the ~$7 trillion in bank assets in excess of liabilities that are backing it up.

>Running a business involves taking and managing risk.

There's really no decent option for most businesses to mitigate this risk. Most services you find can only do up to a couple million.

> Furthermore, the returns at SVB were not that spectacular, I assure you. They were just a very startup-friendly bank.

4.91% when most other banks were offering in the region of 0.1-0.2% is fairly spectacular.

No, but they should have had renter's insurance to cover their lost property.

> These banks are the equivalent of the apartment owner.

The issue is that they ARE important enough to the economy to be considered an "large apartment building owner" (see: de facto bailouts) but were NOT regulated as such. Pardon my continued use of a somewhat tortured analogy.

You and me and everyone else without >$250K in an SVB account should be angry about this. Why? Because the customers of SVB were getting attractive returns on their deposits. So they got exposure to the upside of the bank's risk, but were bailed out of the downside.

> So they got exposure to the upside of the bank's risk, but were bailed out of the downside.

This is a huge flaw in the current way things are operating, but it benefits those people making the decisions and their friends so I don't see it changing.

Facing consequences for bad investments is apparently something only poor people have to worry about. Which is insane.

> customers of SVB were getting attractive returns on their deposits

Can you quantify that? If a startup had $1M in a business checking account, what interest rate were they getting paid?

I was under the impression that depositors at SVB were getting 0% like everyone else in the 0-interest-rate era.

Around 4.91% interest up to $4M.

> Around 4.91% interest up to $4M.

Anyone being paid interest like that on demand deposits should expect their bank to fail, unfortunately, and consider themselves lucky to get cents on the dollar when it does.

> So they got exposure to the upside of the bank's risk, but were bailed out of the downside.

The fastest way to cure the risk of banks like SVB would be to make it illegal to pay interest on demand deposits and have those that want to earn interest to either purchase certificates of deposit or invest in some kind of mutual fund. There are hundreds of money market mutual funds that are safer than making an uninsured deposit at (read: loan to) a mid size commercial bank.

It would also be necessary to require all deposits to be insured and have banks pay appropriately risk adjusted deposit insurance premiums of course.

The Fed just spent a decade making fools of anyone who tried to manage risk sensibly; anyone who believes the stock market is currently measuring some sort of value would have to rate COVID as one of the best things to ever happen to the US. It is in fact just measuring the money printing activities. It is a bit rich to tut-tut at the money managers - they are probably idiots, but the reason idiots are managing all that money is ... the Fed incentivised it for more than 10 years.

Furthermore, I still think it is unclear that SVB was a true failure for the decision makers - in the sense that it seems possible that the people making active decisions might have made money. There has been a lack of clarity around who is losing out in this crisis and why ("shareholders" doesn't count - who are they? how much a head?). It seems possible the SVB owners who lost money were either index funds who pass the costs back to small players who were forced out of bonds by the fed, or entities caught up in some web that meant they simultaneously benefited from the bail outs. The people getting bailed out aren't normal depositors.

There is absolutely no hedge that can stop a bank run, there is literally not a single financial invention that allows both banking and enough liquidity to stop the run.

Except for insurance on that sort of bank run.

Really it comes down to irrational VCs starting this bank run, which inevitably led to the destruction of the only bank that really understands tech startups. And now the same folks are making fools of themselves doubting all economic data (public or private, they all agree) that shows that most sectors are doing fantastically, just not tech. They are also backing a presidential candidate that thinks WiFi causes cancer, and has some of the worst anti-science views out there; in addition these same VCs are pretty much solely responsible for the community backlash against tech, the change from appreciation of tech to hate of tech, as they are outliers with their weird political obsessions that don't even match the rest of the industry.

The result of the interest rate hikes was that there were going to be some bank failures, that's the whole point of trying to cool things off. That it happened in tech and not elsewhere is due to a few bad actors that have more economic power than wisdom to wield it well.

1) What's irrational about taking your money out of an insolvent bank?

Sure SVB collapsed immediately from the bank run; but whose to say it wouldn't've gone bankrupt 5 months from now (it's not like 2% mortgages are worth any more now than then!)?

2) It's not like the Fed didn't warn them quarters ahead of time that SVB wasn't hedged against rising interest rates. The Fed does not need to give banks advanced notice of what they're doing; the bank should be hedged for w/e the Fed does so that they don't become insolvent triggering a bank run.

Sure it would be much better if the VCs had gotten together and agreed to either loan SVB money or etc so it would be paper solvent. But my original point is, don't be blaming the Fed for SVB's collapse. SVB had a duty to manage their risks and they did not.

If it were not for the bank run, the bank wouldn't have gone under.

There was no need to loan SVB any money! The only thing that causes this run was a coordinated withdrawal of lots of accounts all at once, and without that, with norma behavior everything would have been fine with SVB.

The biggest critique of SVB isn't that they needed money (they didn't) or that they didn't hedge properly (it was pretty standard stuff) it was that they had correlated clients that were far less diverse than one would guess by looking at the number of accounts.

That was the risk that SVB "did not manage" in that a lot of the money was actually controlled by a very tiny number of people, even if it was in the account of a startup, that startup was going to follow the advice of they major financial backers.

The bank wouldn't've gone under when it did without a bank run. This doesn't mean that the bank would've never failed.

The bank was literally (pre-run) trying to raise cash; they already didn't have enough cash before people tried to withdrawal 40B! If your bank is insolvent and trying to raise money; it's incredible rational to withdrawal your money before it collapses.

> If your bank is insolvent and trying to raise money; it's incredible rational to withdrawal your money before it collapses.

To me one of the concerns there is it certainly seemed like Peter Thiel had a lot more specific insights on how real this solvency concern was, given that late Wednesday night he told his companies to have their money out of SVB by COB the next day, and that a loan application app needed to be up and running by noon Friday.

Something's wrong when a business's excuse for failure is that each of their customers acted rationally.

Or something old and understood is happening. Tragedy of the Commons? The same behavior that causes traffic jams? (If everyone slowed down and evened traffic out, there wouldn't be waves leading to a cascading stoppage.)

You are right if it's only a single business failing. However, even JPM or BofA won't survive if >25% of their customer deposits are withdrawn in a matter of days, let alone other smaller banks.

If 100% of businesses in a certain sector are a "failure" according to your definition, then I suggest your definition is wrong.

> What's irrational about taking your money out of an insolvent bank?

Nothing. But what happened here was that multiple people decided to take money all at once out of a solvent bank, thus making it insolvent. Banks are in fractional reserve banking business, not custodian business. So no bank in the US can survive >25% withdrawals.

It wasn't "irrational VCs" starting a bank run that caused the SVB problems.

It was the very clear fact that the bank was insolvent, except under "we can pretend all their customers will leave their money at below-market interest rates for a decade" accounting.

Perhaps if SVB had less-informed customers, the bank failure would have happened at the speed of First Republic. But the writing was on the wall: SVB was certain to require some form of bail-out or acquisition.

They are only "insolvent" if they mark to market rather than holding the treasuries to maturity; which as I said above is true to pretty much every bank out there. This is inherent to banking, if there's a run then there is not enough liquidity.

In all my reading of SVB, I have never once seen an anaylsis that says that a bailout would be necessary. For that matter, neither have I seen an analysis claim that acquisition was necessary. An acquisition would have been far preferable, and causing a bank run was far more irrational than allowing an acquisition to happen.

This is inherent to fractional reserve banking - not all banking. Demand deposits should not be lent out, if banks want to make loans then the capital should come from longer term assets not demand deposits. This isn't hard, it's just far less profitable.

> There is absolutely no hedge that can stop a bank run

In this case there were a significant number of hedges available to prevent a bank run.

1. Diversify Depositors (Sequoia's terms to keep balances at SVB is a risk to the bank, you hedge this by marketing to mom and pops);

2. Offer high rate CDs until your long term depositors are > X% of deposits. Yes this results in negative marginal returns until the bank rebalances its investments.

3. Private insurance (charged as fees to depositors) to accounts over FDIC limit (encourages #1 and #2)

4. Shorter term financial instruments

> Really it comes down to irrational VCs starting this bank run

I really wish there was more investigation of Peter Thiel on this. He issued very specific edicts to startups he was invested in on Wednesday night saying that they were all to have all funds out by COB Thursday.

He also then specifically directed one of his companies to have a loan application app up and running, and that it would need to be ready by noon on Friday (the Feds shut the doors Friday morning).

These were remarkably ... prescient ... predictions, and very precise and specific.

I cannot help but wonder if Thiel had some insider information on this going down. I'm sure he's probably quite friendly with SVB's board given how much money he's pushed their way.

It should also be noted that encouraging or pushing a bank run is in fact a felony.

I find these takes truly bizarre. Your basic assumption is that the future is predictable and there's a group of powerful, but incompetent people controlling it.

Central Banks' mandate is to control inflation while still ensuring employment is healthy. They can try and model and predict the future as much as they want, and they can tell you what they're thinking at any point in time, but this is fundamentally not something that anyone can do. So the line became "The fed promised us they wouldn't raise rates, but then they did, so my shitty risk management is their fault, not my fault". No, it's your fault, because the Fed's job isn't to make pronouncements and stick to them regardless of reality, it's to keep inflation and employment at healthy levels.

So the Fed did its job. Inflation is being brought under control without a recession or a hit to employment which so far is quite remarkable. If they continue this for another year or two and bring the situation completely under control it'll actually be a historic achievement.

I am not arguing that they failed at predicting the future: they failed at managing expectations.

I am suggesting that the Fed should have not be authoritative on these topics since obviously they have no clue. Only recently they tried to tone it down a little by asserting they are a "data-dependent" and even then, they are still authoritatively making assumptions every week on things they have no clue about.

Again, dig into your own premise. You can't set and manage expectations about a future you cannot predict. You can set expectations about your behaviour given a certain situation (ie, if kid misbehaves, then no video). You cannot set expectations about the economy, because the economy is a complex and unpredictable system.

This is the reason that while companies can issue guidance and forecasts in their quarterly reports, they come with pages of disclaimers. You've essentially taken the equivalent of those forecasts at face value, ignored the disclaimers and are angry your trades didn't work out. You can blame other people all you want, but your ideas are just weird.

> Central Banks' mandate is to control inflation while still ensuring employment is healthy.

Then the FED shouldn't promise they won't raise rate and instead should just SHUT THE FUCK UP.

It's great that it's doing its job. It's not great that it's ran by downright forked lying tongues.

It was relatively cheap to hedge the interest rate risk. Most banks did that. SVB didn’t.

Buying the 10 year and hedging all risks down to the two years is equivalent to just buying the two year.

I think most other banks tried to get interest revenue from something other than going turbo long long-term treasuries? Or just accepted lower profit?

Could also hedge with swaps or swaptions or mortgage pass throughs. Banks who are very long mortgages frequently short passthroughs. They take enough risk off the table to not have the risk exposure of a hedge fund.

https://en.m.wikipedia.org/wiki/Mortgage-backed_security

Many banks are still building deposit reserves by competing on CD rates.

The lesson really is that the market is in charge of the market. I don't think the Fed lied. I think they really thought they wouldn't have to raise rates much, if at all. They were just wrong because they're not in charge. They only think they are.

> > Whoever trusted them in the past, got screwed

Righfully so. It's econ101. Central Banks modulate the interest rate based on inflation.

Nobody knows how inflation originates or subsides

Of course this is in fact a total lie and reflects no truth in rality.

I don't have a view on the topic itself but post like yours - assertions of what is true or false with zero fact or argument - make me favor whatever the other side is.

Presumably if you had something that supported your point you'd share it.

The only thing we should take away is that there are no rules. The Fed will decide the winners and losers without the deference that a system claiming to be for the benefit of people should have. If you’re not too big to fail, then you’re too little to succeed.

"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks"

I don’t understand why there hasn’t been a push to replace Powell as fed chair. He’s been so wrong on everything. Everyone knew inflation wasn’t transitory to the point even Janet Yellen made a comment that it wasn’t.

Reply on news.ycombinator.com