Hacker News Reader

Comment by daydream | Hacker News Reader

> The losses banks have taken on their assets are real. They already happened. They are survivable if banks remain liquid.

But… they aren’t real yet? They haven’t been realized. If held to maturity they will be paid back in full.

Which I know the author is fully aware of. So I don’t understand this point.

> I would suggest one has at least one backup financial institution. If one hypothetically does not, I would observe that opening bank accounts rounds to free. Thousands of perfectly good financial institutions exist.

Some people have investment accounts with brokerages like Fidelity or Schwab. Many brokerages (including the two mentioned) offer cash management accounts. They offer deposit insurance similar to FDIC and will give you tools similar to checking accounts. Debit cards, checks, bill pay, etc.

They can be excellent backup accounts that don’t add the additional overhead (however small) of yet another company to deal with.

View on HN

> > The losses banks have taken on their assets are real. They already happened. They are survivable if banks remain liquid.

> But… they aren’t real yet? [...] So I don’t understand this point.

If people withdraw their deposits, the bank will have to deliver the money somehow...by selling the assets that have lost money. So the point is that although, if nobody withdraws, the losses are survivable, if enough people withdraw, the losses are not survivable. As soon as depositors realise this situation, they will withdraw their money. So that's the problem.

Yes, if the bonds must be sold to cover withdrawals then the losses become realized (real) at that point. But not before.

From all the information I've gathered, there is an unstated aspect to this.

If any number of HTM bonds are sold to cover withdrawals, then all of them must be revalued and losses realised on the whole lot.

What. That’s insane. Essentially, if your deposit modeling is off just a little bit, you’re dead.

I'll turn that around: if as a bank your risk modeling does not account for outlier events, you're dead. Because the way banks and bank accounts are used these days (eg. payroll providers), the society is not willing to accept the second-order effects. As OP's example, a payroll provider was using SVB as their bank to hold funds in transit in order to make payments to third parties. Result?

    Every regulator sees the world through a lens that was painstakingly crafted over decades. The FDIC institutionally looks at this fact pattern and sees this as a single depositor over the insured deposit limit. It does not see 300,000 bounced paychecks.

But back to your point. Banks are over collateralised. Let's take an example from just this week. Credit Suisse is in trouble - their collateral has gone from ~170% to ~150% - they have the collateral to cover their total deposits, but they don't have nearly enough of them as liquid assets to cover the increased outflows.

Until last week I didn't know about HTM ("hold to maturity") and AFS ("available for sale") categories. I have since been educated.

> But… they aren’t real yet?

Barring something extremely abnormal happening, aren't low-yield bonds seeing real losses already due to inflation?

Like it doesn't have to be the spot price we're talking about, aren't many of them toxic already and others expected to track there?

> aren't low-yield bonds seeing real losses already due to inflation?

But bank deposits aren't in inflation adjusted dollars.

This is true, but there can be a lot of slight of hand when talking about dollars and future dollars.

Banks run on nominal dollars, and SVB would have remained capitalized if withdrawals hadn't overwhelmed their ability to get ready cash, which caused them to sell at a loss, which spooked everyone, causing a run.

Right, but if they're expected to be toxic you would mark them down nominally as well. It's not like the real losses aren't also nominal when realized.

And whether the metric you care about is the real losses or the expected nominal value at maturity depends on whether inflation continues to raise. As this also suggests interest rates increase, you get hammered on both sides.

No, because they still are with the same number of dollars, it it just that the dollar itself is worth less that it was.

If you say you have $100 worth of bonds, this is accurate at all points of time you hold it. What you can buy with $100 may be changing from year to year.

Reply on news.ycombinator.com