Banks forced to realize losses in bonds

Banks Lose Billions in Value After Tech Lender SVB Stumbles

Index of banks posts biggest drop since pandemic roiled markets nearly three years ago

By Jonathan Weil and Ben Eisen, The Wall Street Journal, March 9, 2023

The four biggest U.S. banks lost $52 billion in market value Thursday…

hursday’s rout is another consequence of the Federal Reserve’s aggressive campaign to control inflation. Rising interest rates have caused the value of existing bonds with lower payouts to fall in value. Banks own a lot of those bonds, including Treasurys, and are now sitting on giant unrealized losses.

Large declines in value aren’t necessarily a problem for banks unless they are forced to sell the assets to cover deposit withdrawals. … Banks don’t incur losses on their bond portfolios if they are able to hold on to them until maturity. But if they suddenly have to sell the bonds at a loss to raise cash, that is when accounting rules require them to show the realized losses in their earnings…

The Federal Deposit Insurance Corp. in February reported that U.S. banks’ unrealized losses on available-for-sale and held-to-maturity securities totaled $620 billion as of Dec. 31, up from $8 billion a year earlier before the Fed’s rate push began

“This is the first sign there might be some kind of crack in the financial system…People are waking up to the gravity that this was one of the biggest financial euphoria episodes.” [end quote]

Bond values fall when interest rates rise. Treasury debt, which has no default risk, is often portrayed as a low-risk investment (compared with stocks). But the risk of loss is very real when a period of low interest rates is followed by high interest rates.

The 10 year Treasury yielded only 0.55% on August 3, 2020, when the Federal Reserve suppressed interest rates with QT during the Covid emergency. A large portion of the billions of dollars of Covid relief fiscal stimulus was also stashed in safe bank accounts, suppressing interest rates.

Due to the Fed’s anti-inflation campaign, the 10 year Treasury yield is now close to 4%. Bonds that were bought in 2020 and 2021 are worth only a fraction of their book values. This is true of the Federal Reserve’s giant asset book as well as many banks. This isn’t a problem as long as the banks can cover their cash flow needs without selling the bonds until they mature – which could be up to 30 years from now.

Big banks hold a range of assets and serve companies across the economy, minimizing the risk that a downturn in any one industry will cause them serious harm. Small and focused banks like SVB and crypto banks are higher risk since they have lower liquidity.

All METARs are probably aware that the FDIC only insures $250,000 per owner so it is unwise to keep more than that at any FDIC-insured bank. Meanwhile, money market accounts at discount brokers yield over 4% so many bank customers are pulling out their savings from low-yield deposit accounts.

The large scale of unrealized losses won’t lead to a banking crisis as long as banks have enough liquidity to handle their cash needs. But runs on banks do happen…and bank stocks are part of indices and mutual funds that are being punished by the market.



I am not expecting a full scale meltdown. That said this worries me. SIPC can become worthless. It is not the federal government’s obligation at all.

Ironically buying equities in a timely fashion can reduce your risks. The brokerage or bank can go under but you own the shares of your investment either way. If the brokerage has used them as margin etc…you still have the rights to your equities. Those are recognized by the corporations you invest in. I imagine you just resurface at another brokerage.

If someone or one of our lawyers knows more about margin and the use of shares let us know.

This has become worrisome to me now. SVB is not a huge bank, like Wells Fargo or Goldman. But it is a large, credible bank in the venture funding world of tech. This is a bit of a shock. I’m wondering now what will happen to some of these start-ups that have funding that is involved with SVB. And how the smaller banks are going to go now.

We have a weekend coming up. And I’m on vacation next week. I’m not kidding that I’m considering moving a large chunk of the IRA out of QQQ for the week (50% of my holdings).


@bjurasz I never give investing advice. But I had to gasp when I saw that you had 50% of your IRA in a single ETF in a volatile sector.



Its worse than you think.

QQQ - 50%
VOO - 35%
VYM - 12%

I’ve gotten out of all individual stocks and in only 3 ETFs now. Sure, that is a lot of individual companies and relatively diversified on the surface. But as Fidelity told me two weeks ago “you have little downside protection with this portfolio”.

@bjurasz it’s better to own stocks (ETFs, etc.) outside of an IRA because of taxation. If you have losses in a non-IRA account you can offset the loss against ordinary income. Also, capital gains and dividends get favorable tax treatment in non-IRA accounts.

It’s better to put bonds in an IRA because interest is taxed as ordinary income in non-IRA accounts. Interest can compound tax-deferred in an IRA until the money is withdrawn. Keep careful records of your IRA cash contributions so the money isn’t taxed twice.

All your ETFs are good funds but you are taking a high risk since the entire market has been in a bubble, especially tech stocks.

There is also a lot of overlap between VOO and VYM.

Many people say, “Don’t time the market. Hold through the ups and downs.” Personally, I pulled out of the stock market in 2021-22. I think the Fed will continue to raise interest rates until a recession results. I think the stock market doesn’t believe that yet.

Only you can decide what’s best for yourself. We each have our own risk tolerance and timeline.


This IRA is a roll-over IRA. I cannot contribute new money to it. It is roughly 90% of my retirement assets. It’s the summation of decades of work and old 401k plans. New money goes Into the 401k. Taxable money is currently in money markets because it is 1) emergency cash, and 2) cash for big purchases within next 3 years.

New money goes into the 529 plan. But I am getting disgusted with adding $400 a month and seeing a flat balance for a year now. And that is NOT aggressively invested.

I have considerable unvested RSU’s at Nvidia that I cannot do anything with yet, none of the shares are mine yet, and I treat them like forced early retirement insurance. Honestly, Nvidia is one of the few companies I trust at the moment, investing wise, but I refuse to put any additional money into that stock. Volatile? Yes. But we own the future.

I am likely to trim QQQ by a large chunk soon. Honestly, I have considered even doing the whole target-date-retirement fund route with the rollover-IRA. I need a double in 9-10 years. In a reasonable world this is not difficult.

1 Like

Double in 10 years! In a reasonable world! LOL!

In the 1930s, economist John Maynard Keynes said: “Markets can stay irrational longer than you can stay solvent."



Hate saying this in public…but I will…I know it is not something I should say it is not my business…hold into 2030 at least. Do not sell low. The QQQ will be strong. I get the market will be rough for another year in all likelihood. I do not know when you bought the QQQ. If you bought lower disregard.

Selling low is wrong. Holding becomes the winning solution.

Only the 1948 bottom was long. This is a long bottom as well. But the upside will be there for years to come.

1 Like

@bjurasz I forgot to mention…

In a reasonable world, the price growth of the stock market should parallel the growth in GDP. If the stock market grows faster than earnings it’s due to higher P/E ratios (sometimes called “expanding multiples”). That happened in the past 20 years due to the Federal Reserve suppressing interest rates by pumping monetary stimulus into the bond market which sloshed over into the stock market.

Real GDP growth fluctuates but it has almost never been over 5% since 1990.

According to the Rule of 72, GDP would have to increase 7.2% per year for it to double in 10 years. In a “reasonable” world that is pretty darn difficult since the U.S. economy would be lucky to manage half that growth rate consistently.

It’s possible that the Fed will continue monetary stimulus which will continue to inflate multiples in the stock market, right? Oops, that’s not their priority at the moment. :wink:



The QQQ doubles more than every 10 years.

The difference in an index being the best rise to the top. It is based on the GDP but in a concentrated sort of manner.

The worst down side for the QQQ? A guess? 200 or 175. But right now the QQQ might have an upside.

Tech has had bad news for more than a year now. Instead it is the financials now having the bad news. The markets are looking to see how ubiquitous the financial misdeeds are.

The WSJ story without the paywall:


This is much higher risk than and the commentary understates the risk. These institutions have $1 to $4 trillion in assets. To have $500 billion in US paper down $100 billion give or take is problematic. Worse the depositors are running for the exits. Going to higher paying money markets at brokerages. The brokers often having similar asset bases.

The question is will a money market fund go under?

Do money markets have bonds?

In general, money market funds invest in six types of securities. The money fund may invest in some or all of these security types: U.S. Treasuries. **Interest-paying debt securities—bills, notes, and bonds—issued by the U.S. government.**Feb 27, 2023 Google result

snippets from PF’s linked WSJ article

U.S. commercial banks’ holdings of U.S. government securities surged 53% over the same period, to $4.58 trillion, according to Fed data.

next snippet

Bank of America and its megabank peers can afford to part with a lot of deposits before they are forced to crystallize those losses.
Most of SVB’s liabilities—89% at the end of 2022—are deposits. Bank of America draws its funding from a much wider set of sources that includes more long-term borrowing; 69% of its liabilities are deposits.
And, unlike SVB and Silvergate, big banks hold a range of assets and serve companies across the economy, minimizing the risk that a downturn in any one industry will cause them serious harm.

1 Like

Dealbook by Andrew Ross Sorkin NYT column no link.

This is about the doubts that caused the SVB bank run. This is in the air now. If it subsides all will be well within about a month. But if the FED keeps upping the FF rate we are more likely to see bank runs.


The collapse may have been an unforced, self-inflicted error: The bank’s management chose to sell $21 billion of bonds at a $1.8 billion loss, in large part, it appears, because many of those bonds were yielding an average of only 1.79 percent at a time when interest rates had risen drastically and the bank was starting to look like an underperformer relative to its peers. Moody’s was considering downgrading its rating. The bank’s management — with the help of Goldman Sachs, its adviser — chose to raise new equity from the venture capital firm General Atlantic and also to sell a convertible bond to the public.

It isn’t clear if the bond sale or the fund-raising, at least initially, had been made under duress. It was meant to reassure investors. But it had the opposite effect: It so surprised the market that it led the bank’s very smart client base of venture capitalists to direct their portfolio clients to withdraw their deposits en masse.

The bank and its advisers may have also made a tactical mistake: The General Atlantic equity investment could have been completed overnight, but the bank’s management also chose to sell convertible preferred stock, which couldn’t be sold until the next day. That left time for investors — and, more important, clients — to start scratching their heads and sow doubt about the firm, leading to an exodus of deposits.

My comment, the major clients are not any different at the other major financial organizations. If risk elevates they will exit.

1 Like

Siegel’s constant from 1802 to 2022 data is 6.7% in real terms. This figure was first measured in 1992,and then revisited in 2022,with no change. See knowledge at Wharton database. A double in ten years in real purchasing power might be a bit of a stretch, but it is not far off the historical average.
P.S. I googled siegels constant. Still can’t cut and paste a website on this new tablet.


An S&P 500 index fund returning 9-10% CAGR is a long term average.

VFINX, over a period from 1985 to today, has a historical CAGR of very nearly 11%. There are periods where it is not so good, periods where it is stellar. But I stand by my comment that a double in 10 years in reasonable times is a reasonable thing to expect. You must not be expecting the next decade to be reasonable.


I don’t think the stock market (the subset of corporations that are publically traded) is a good analog for GDP, or vice versa.

So much of our economy is based on small companies that will never be represented on a stock exchange.


Historically, stock market growth has ultimately paralleled corporate profits, which have averaged about 7% + inflation.

So assuming a reasonable starting valuation, doubling in 10 years–while by no means a sure thing–isn’t outlandish. Current NASDAQ P/E is about 23. The average over the last 30 years is about 40.


One thing to keep in mind is that the next 10 years is going to differ in one potentially very important way from the last 50. Age demographics.

Young people tend to buy stocks. The elderly tend to sell. The retirement of the massive Boomer generation could put significant downward pressure on equity valuations.

Those who believe in the power of demographics have concluded that:

  1. Equity values are closely related to the age distribution of the population, with values declining as the population ages.

  2. The rise in the stock market from 1980 to 2000 can be attributed primarily to age demographics, with the Boomers entering their prime stock buying years.

  3. Modeling for the US population project virtually no change in P/E ratios for the next 80 years (from 2020).

This means that the last 100 years of stock performance when America was a relatively young country with a rapidly growing population may not be predictive of the next 100 years when the population will be older and slow-growing.

1 Like

The generalization has no weight when you look at the data.

About 50% of American’s own stock. Of that 50%, about half of all stock is individually owned vs owned by a corporation.

But, most importantly, as of 2021, nearly 90% of all stock is owned by the wealthiest 10%, with half of all stock owned by the top 1%. The last two groups are very unlikely to be selling their stock to buy bonds in retirement.

The small percentage of retirees that will be selling will be massively dwarfed by the wealthiest 1% who will still be accumulating.