US banks have $620B of unrealized losses on their books
bloomberg.com
bloomberg.com
The CRE bubble is going to burst in the next 18 months as the 5-year commercial office leases signed in 2017+ are all going to come to an end, and the banks will be realizing these losses starting now and the next few years. Its going to get ugly soon.
Most of the entities on the hook for paying back the bank are commercial real estate development firms. As such they often have a mixed portfolio of residential (think apartments), office, retail, and entertainment properties. Offices are further divided into small business, professional services (doctors, dentist), and general office space. Unless their portfolio is extremely focused and not diversified, then they'll be alright.
Fortune 500 companies also aren't letting go of their iconic office buildings, as the building itself is an icon for the company.
Given all these factors, I think the idea that commercial real estate is going to crash and burn and that the banks are going to be left holding the bag are being quite a bit overblown.
But the looming CRE crisis isn’t making it into the mainstream consciousness, even though its big driver (WFH and its ramified impact) is being well reported on.
Often the companies buy property themselves, to use (since rents have been so high in he last decade or so). I don’t think banks are over-represented as owners, at least where I live. But I might be wrong.
Why is it that US banks are owning so much commercial real estate?
If the holding company can't rent out the property, at some point they stop being able to service the debt. This leaves the banks with a lot of illiquid assets and bad debt...
The shift to remote work has enabled replacing most of those effectively virtualized staff with AI agents that never sleep. Their takeover won't put those people back in offices, it obsoletes them to unemployed couch-potatoes. Likely notified of their new employment status via AI agent on a Zoom call....
During that time period (beginning of time to now), things actually have fallen apart, many times.
So what? What you say would be true even of all the predictions were right, but just 1% less in magnitude. It's setting an unreasonably high bar.
A lot of people don't want to think about the harm that their enthusiasms could cause, and are eager to discredit anything that isn't a rose-tinted assessment because it's a threat to their fun or ambition.
AI will be great for those who already have money, and they'll be the ones writing the history of its effects.
I’m not being dismissive of your fears here. It’s not like what people are talking about is impossible. But there isn’t much you or I can do about it at the top and generally the economy doesn’t crash a la 2007 every time a big shake up happens. So if you’re really concerned about it, do the best you can to insulate yourself from it and inform yourself about the possibilities to a reasonable degree. But we can’t sit here and agonize over every potential doom and gloom situation that could happen on any given day, especially when tons of times what people talk about on HN, the news, wherever, never come to pass or are not nearly as bad as predicted. It’s just no way to live.
So when cap rates double the implied value of the building falls by 50%. Effectively the building goes from being valued at a P/E of 50 (cap rate 2%) to a P/E of 25 (cap rate of 4%).
So the halving of the building's value, as cap rates move in response to interest rates, may mean that a 50% LTV loan is now underwater. Especially if the building is unoccupied or may soon be unoccupied.
With the Fed now expected by the bond market to pause rate hikes and start cutting rates soon, long term bond yields (as well as shorter term ones) have been coming down, increasing the value of those holdings.
You may have noticed this decline if you were shopping for CDs over the last few weeks. On brokerage platforms, top rates for 1 year CDs have gone down from a high of around 5.4% to 5% now.
The logic behind this seems a bit circular to me. Market participants accept lower long-term bond rates because they expect the Fed to cut rates in the future. They expect the Fed to cut rates because if they don't, there will be a financial crisis. There will be a financial crisis because long-term bond rates are high and banks are underwater on their bonds, except that bond rates are declining because the Fed is going to cut rates, because...uh, why is there going to be a financial crisis again? What reason does the Fed have for dropping rates? It's very much an "appeal to consequences" fallacy - some bad outcome is not going to happen because that would be bad. The world doesn't work that way.
IMHO the market is wrong, and the Fed is not going to cut rates. They've gotten too accustomed to being bailed out, and the Fed is on to them, and so they're not going to do that again, at least not in the form that the market expects.
For the last several months (since Jackson Hole) markets haven't been sure how quickly inflation would come down and therefore how high the Fed might risk taking rates. But today we have much more certainty. We now know the Fed really can't raise rates much more near-term given the risks and longer-term the stimulus and supply bottlenecks which created the majority of the inflationary pressures will ease.
Another thing you need to remember is that inflation easing is actually deflationary because real yields are calculated by nominal yields minus inflation. So as inflation comes down you're actually getting paid more to lock your cash up in bonds which continues to tighten financial conditions.
The risk of a larger financial crisis is very low given this isn't a credit event, but a duration event. As long as the Fed doesn't raise rates further we can rule out a financial crisis and so long as you don't believe there is a structural inflationary problem in the economy rates should trend downwards over the coming months.
Inflation was kept low through the 2000s despite massive quantitative easing because you had the entry of 3 billion workers into the global economy, keeping wages low even though the money supply was high. That trend has reversed. China hit peak workforce employment in 2015 and their workforce is now shrinking fast, the result of their One Child Only policy in the 70s. The same demographic headwinds affect all major developed economies. Fertility rates fell below replacement rate in most of the developed world in the 1990s - that baby bust is now entering the workforce.
Similarly, technological progress has slowed down over the last 40 years. Much of the low-hanging fruit has already been picked; only a select few industries still show significant technological development. In particular, the major need-to-live industries like housing, health care, food, and energy have seen few yield-increasing technological developments over this period, and their price has risen accordingly.
We don't understand the drivers of inflation very well and people have different theories on this.
You may be correct though. Fundamentally inflation is driven by the change in the supply and demand for goods and services, and in a globalised world with rapidly changing geopolitical landscapes and technological advancements it's hard to predict actually how supply and demand pressures will balance out in the coming years.
However, what we do know is that demographics have historically correlated well with inflationary pressures. The inflation in the 70s was likely in part driven by the huge growth in demand which came from the baby boomers entering the economy. Then more recently you have examples like Japan where demographics have clearly played a role in their sluggish economic growth and low inflation.
But to be clear, I'm not saying the market is necessarily right in its pricing, I'm just suggesting that there's a lot of good reasons to believe the Fed funds rate is currently too high and will need to come down. Even if you're correct you would still have to explain why inflation rocketed up in 2021 right after trillions of dollars of stimulus entered the economy. The trends you're describing have been in place for years. The inflation we've seen over the last couple of years in my opinion is just far easier to explain via the combination of reckless monetary policy, fiscal stimulus and lockdown policies.
It's not really supply chain bottlenecks. It's greed. Same thing for "14 year olds can work in meat packing plants".
Greed is having a moment and must be countered by policy (top marginal tax rate of 90%, let's go!)
That's the thing, you can't know that. Rates could go up, down, sideways, or in circles over the next 10 years. A 10-year bond locks it in to a predictable rate and payment schedule.
The world does work that way! Most people don't want bad stuff to happen, and will take action, coordinated or otherwise, to prevent bad stuff from happening.
To me, sounds like the Fed is trying to slow this thing down without spooking people. So, promise to cut rates soon, then see what people can actually tolerate, and maybe leave them high enough to chafe a bit.
High interest is killing stock market returns.
No administration wants to enter election season with decimated stock market.
Hence, if nothing else, there will be political pressure on FED to lower interest rates to juice up stock market.
Second, some kind of recession is expected by many so if it happens, that will a reason to cut rates.
Inflation, however, will reliably kill a presidency. Note that we went through 4 presidents in the 8 years of the 1970s inflation, while the average president in the 40 years afterward served for 8 years. This holds universally across cultures as well - globally, countries like Venezuela, Zimbabwe, Weimar Germany, post-war Eastern Europe, etc. go through rapid regime change during periods of high inflation.
[1] https://www.pewresearch.org/fact-tank/2020/09/25/few-in-u-s-...
The fed historically does take actions adverse to employment in election years.
https://www.advisorperspectives.com/commentaries/2016/08/25/...
The stock market (as measured by SP500) is down only 15% from its all time high, and is currently higher than any point before 2021.
I'm not saying this is proof that they won't hike rates, just that what they say should be heavily discounted.
They might just pivot on a dime.
What I find interesting is that literally everyone is saying that FED is wrong.
Like find me anyone outside FED that says: yeah, rising rates is good.
Granted, many people who comment (wall street) have self-interest in rates being low but I don't think there's any notable economist who isn't critical of FED.
Hard to say the Fed won’t cut rates when their own projections have them cutting rates.
SVB didn't do anything to hedge its duration even well after the "inflation is transitory" period was over.
The Fed has been very explicit that they need to tighten labor market conditions. "Tighten labor market conditions" = "cause layoffs". That's how it works, they just can't say it out loud because it'll spook people. They can't bring down inflation until they raise rates high enough that companies start going under, and they have no incentive to lower rates unless actual bad things are happening in the real economy. (Remember also that when rates are high, the Fed has more room to cut when actual bad things do happen, because there's a floor at zero.)
Both the bond and stock markets have been consistently wrong about this for the last year.
Disclaimer: I bought a 5.4% CD a few days after the collapse of SVB.
It seems to me that "inflation" is just another word for "worker power." It means there is low unemployment and lots of opportunities to hop to a higher paying job. It means you can work remotely and tell your boss to get stuffed if they want to RTO. Prices go up but if you play your cards right, you can increase your income by significantly more.
Not worthless, worth less. The Fed has always let banks borrow against Treasuries at the discount window. The discount is to their fair market value. They created a new mechanism, the BTFP, which lets banks borrow against the face value of their Treasuries at a higher rate [1].
[1] https://www.federalreserve.gov/newsevents/pressreleases/mone...
Your observation is likely over fitting to some of the CDs tied to risker banks.
even the CNBC talking heads go only so far as to talk about a pause in rate hikes
I’m attaching a call report from FFIEC on JP Morgan chase. You can see in the document JPM has over 1T in uninsured deposits. No bank run despite of JPM also holding paper losses in the portfolio.
https://www.dropbox.com/s/xahdu1092n5axfg/Call_Cert628_12312...
Search for the phrase “uninsured”.
The interesting aspect of the drama is that when the venture fund incurs a paper loss the same way as the bank does, it goes to extreme lengths to avoid marking it down for as long as possible.
On the other hand, 80 years is a long time. Maybe Hollywood needs to do a remake. With Marvel characters.
Clearly no bank supports withdrawing it in cash, fine. But the transfer should just be a decrement of a number in the first bank and an increment of it in the account on the second bank.
What other effects do I need to think about? (Serious question)
The transfer actually decrements two numbers at the first bank and increments two numbers at the second bank, and typically also mutates two further numbers at another institution, as so:
If you transfer your deposit from a bank, then that bank needs to decrement it's liability to you. If it's decrementing it's liabilities, it also needs to decrement it's assets by an equal amount (otherwise it'd be making a profit with this transaction, which it's clearly not).
We therefore know from this that the first bank needs to transfer an asset of some kind from itself to the second bank.
The second bank then needs to increment their liabilities to record the fact they now owe you money, and they need to increment their assets to record the fact that they now own this new asset they've just received.
The asset that changes hands is typically Central Bank reserves, which is essentially money that is held in an account with the Federal Reserve. In order for the money to change hands, the Federal Reserve has to decrement the amount of reserves that are owned by/sit in the account of the first bank, and increment the amount of reservers that are owned by the second bank.
The other effect you need to think about is what happens if the first bank doesn't have a sufficiency of reserves to settle for this transaction? In this case, the sending bank would need to sell some of it's other assets so that it receives reserves from other institutions, before sending those reserves to the second bank, or alternatively it can go to the Federal Reserve and ask the Fed to create new reserves and lend those reserves to the bank, securing the loan against eligible collateral (such as US treasuries, as in the new Bank Term Funding Program), which it agrees to repay later (either with reserves that it receives in the future from other inbound transfers, or by ultimately selling the assets).
If it was I could manufacture money out of thin air just by never decrementing my ledger.
What actually happens is that banks have accounts at other banks. If you transfer say $230 from your account at SVB to PG&E at First Signature then what really happens is SVB decrements 1k from you and increments 1k to themselves and tells First Signature to decrement 1k from SVB's account and increment 1k to PG&E's account. (Take a look at #4 on how a wire transfer works [1]).
So when you want to transfer large sums it causes a bit issue as SVB might not have an account that size at the given bank. So they'll need forewarning so they can do a bunch of other transfers so that they do.
There's a practically issue with SVB having accounts at literally every other bank so if SVB and First Signature have an account at a common bank that works as well with just an extra step of transfers.
https://www.bitsaboutmoney.com/archive/bank-transfers-as-a-p...
What happened here is that the VCs generated a run such that SVB had to liquidate a lot of assets. Moreover, these assets were long term investments. Meaning unlike your sudden and immediate demand for the money, they had to sell those assets on the open market. Unfortunately, since interest rates had gone up the loans the bank had given out were not as valuable on the market and the bank had to take a loss. For example, it had given out a million dollar loan at 3% over 30 years but now it was selling it at 750k on the open market because no one is buying 3% interest rate loans because currently the rate is ~5% (they couldn’t wait the remainder of the loan duration). That created a huge hole in their balance books. They tried to shore it up with a loan which would have plugged the hole but that fell through for some reason I forget (not sure why they announced they were looking for a loan before they had closed it).
So in practice you don’t have to think about anything. Moving $2 million is nothing. It’s when you move $2 million and convince 1000 other people to move it all at the same time that you can make the bank feel some pain (I don’t have a sense on the amount of money moved that cause a problem - all numbers here made up).
What I don’t understand is that a simple change to the regulation would close the ability of random people to create bank runs. If you want to transfer > X million ahead of time, you must give an N day notice. Additionally, the bank can choose to deny it if it’s coming from a regular (ie not money market) account (chequing / savings). You have a special carve out for regular periodic payments and you’re done. It’s kind of ridiculous that you can give a loan to someone and then arbitrarily ask for the entirety of the money immediately without warning.
We have those kind of fixed-term and notice accounts, but many businesses and consumers prefer to use more flexible demand-deposit accounts. You could theoretically regulate demand-deposit accounts out of existence and say everyone now has to use fixed-term and notice accounts, but that might have significant implications on the level of deposits that people choose to hold with banks (as opposed to just handling the cash themselves, where they have no such notice requirements), and this will have significant knock-on impacts to the availability of credit.
> Additionally, the bank can choose to deny it if it’s coming from a regular (ie not money market) account (chequing / savings).
Under what circumstances could they choose to deny it? For how long can they choose to deny it? I wouldn't personally be too fond of lending my own hard earned money to an entity that can arbitrarily choose not to give it back for an indefinite period of time in a way that gives me no legal recourse... Again such a regulatory change could have significant knock-on implications to the quantity of deposits and the availability of credit.
The tricky things with bank runs are there aren't any easy solutions - the best one we've come up with so far is deposit insurance, although this has it's limits (both metaphorically and literally).
If you’re converting currencies there are also FX mechanics where the bank will quote your transfer on the FX market and try to make a buck here and there before releasing your transfer which can add delays.
Banks are not piggy banks. They don’t just pile the deposits into a vault. They estimate how much money is likely going to be whitdrawn in a certain period and invest the rest. That means that the bank might not be able to pay out all the money they owe if a large percentage of the depositors all at once request their deposits returned.
It isn’t that large a percentage of deposits - less than 20% for most banks by my estimate. Notably in 2020 the Federal Reserve reduced the required reserve ratio for banks to zero, though they do have other requirements such as a minimum equity ratio and such.
The reality is that fractional reserve banks have a massive failure mode if even a modest percent of demand accounts were to withdraw in a short period.
My sincere hope is that these recent developments raise awareness for how fragile the current system is and there is enough public outcry to effect meaningful change.
As it stands a checking or savings account at an FDIC insured bank is an unsecured loan to a leveraged counterparty that uses your funds for speculation, backed by an insurance program with $142 billion protecting $17 Trillion in deposits. (0.7% coverage)
If the FDIC fund is insufficient in a crisis either the US Taxpayer or all dollar holders will be left holding the bag.
Now that the medium of exchange and store of value is fiat currency that is digitally printed or burned by the Fed, it's the opposite. Most transactions happen with nothing physical changing hands. The marginal cost of a bank to take on more deposits is nothing. It's no longer natural for banks to be paid by depositors directly as custodians, which is why they have to make money elsewhere; as you said, leveraged (but regulated) speculation. We've decided that this is overall a good thing for society.
> backed by an insurance program with $142 billion protecting $17 Trillion in deposits
This part's kind of misleading, since it's not like all $17 trillion (or even some significant part of it) will leave the US banking system entirely. Almost all of it just goes from one bank to another. When too much leaves one particular bank at once, you get a bank run. But the system as a whole remains balanced, and there's no conceivable scenario where enough people will take money out of the banking system for it all to crumble. Where would the money go, under people's mattresses or other currencies?
I do find this confusing in a way - need to think about it some more.
Isn’t this only for smaller banks? if I recall correctly something like banks with a market cap over $500B have to maintain a proper reserve ratio.
https://www.federalreserve.gov/monetarypolicy/reservereq.htm
In other words, you can practically bring down any deposit-taking bank by starting a large-enough bank run. (Except maybe if the central-bank intervenes.)
Standing on some sort of "customers shouldn't participate in bank runs" principle is a great way to end up in some sort of trouble or other.
It also isn't exactly clear to me how to stand "customers shouldn't participate in bank runs" on any sort of firm ethical basis that both stands up to theoretical scrutiny and also is an ethic that can be practically lived by. You put money in a bank. The bank counters with a promise that you can withdraw it whenever you like. Is it really somehow immoral for you to take it up on that promise when you become concerned they can't hold it up? Having already put your money in the bank, is it now somehow also your responsibility for them to be able to live up to their promise, let alone their promises to other people? Am I morally obligated to see that my bank has a reasonable chance to not fulfill their promise, but just live life as if it's not?
You might be able to work this into an ethic, but I would submit that at the very least it's going to be a long journey, not something that can be simply asserted as an unexamined premise. Taking a consequentialist point of view is superficially appealing, but it is unclear to me that across any time span other than a hyper-short one that a full accounting of the situation leads to an obligation to depositors to ignore threats to their deposit. You can easily simply end up destroying the bank system as a whole as you require people to incur risks they're not willing to incur in order to do some banking and other such things, so they choose not to bank at all. "You need to leave your money in even when it's at immediate risk and also you're not allowed to consider second and higher order consequences because that's immoral too and also you're not allowed to ask the banks to consider them either" gets to be rather untenable on a number of levels; even if you want to propose such an ethic theoretically it clearly fails the living-by-it-practically test.
But, again, I'm not saying you couldn't possibly work this into a coherent ethic, I'm just saying, it's pretty challenging. It's hard to avoid that you basically end up writing in that you should allow people to basically lie to your face about the promises they made to you and you're morally obligated to live by those lies.
Knowing that only $250k of those funds were FDIC insured, would you have left your funds in the bank and hoped for the best?
If you answered Yes, how would you explain your decision to your investors or employees?
There is zero benefit for them to take the risk of leaving their funds with the bank and considerable downside.
Could just package it as Priority Withdraw Guarantee and charge it as an add-on and part of the guarantee is that your cash will be kept in a reserve independent of investment activities.
there is no modern western bank that is capitalized 100% - if everyone goes to every single bank and withdraws their money, the entire western economy collapses.
it's up to the depositors (the customers) to know that if they withdraw 100% of their cash, the entire system collapses.
as a human living in the current era, you either buy into this, or you go and live in a compound with gold bricks and weapons stashed in the basement, or you become a cryptobro, or you become like Russia and start wars to gather resources and land.
If there's a reasonable chance you'll have an urgent need to sell that CD well before it matures, then yes, otherwise probably not.
Technically yes, but you only realize it if you sell. And if you hold until it matures, then there is no loss, since you get everything back (barring default).
But as other commenters pointed out, there are only very few specific cases where you would need to mark this to market and have it matter enough for you to become a forced seller.
Yes. If you sold that CD, you'd sell it for less than par. (EDIT: Check if you have the right to take your money back at par. If you do, it may make sense to call that option and re-invest.)
Which would be nonsense, because a CD* isn't a deposit.
*There are CDs and there are CDs. Many CDs have a put which lets you redeem your principal at any time. Those don't lose value when rates go up. But many CDs don't, particularly tradable ones, and that's part of the risk one takes when trying for a higher return. How a bank counts deposits for waiving fees is a separate question from fair market value.
What does CD stand for?
There aren't many assets like that. If the value of a property or an equity goes down, all you can do is hope that it goes back up. With bonds you just have to wait, and indeed every day that passes the value will converge back to par.
Inflation mostly caused by thr central bank (effect of Ukraine war on food/oil is much smaller).
Meanwhile economy books claim that central banks are good. When they run insane policies that screw the the people..
Most central banks are so tertible that they make 2-3% inflation just for using cash. This hidden tax is incredible. Yet most fail to see that, apart from maybe cryptobros.
I think this is a great buying opportunity for banking stocks and have put my money where my mouth is.
Put another way - if they lose 10%, they are finished.
You think the US government is gonna go "whoops, our central bank failed, and we aren't willing to pay a fraction of the cost of the F-35 program to bail them out"?
They sent $109B to the Treasury in 2021, and $58B in 2022. Some years being a down year isn't shocking.