Post #1813: Won’t be hiring ChatGPT as my AI staff writer, either.

 

AI: It’ll always give you an answer.  It might give you a very convincing answer.  And it might sometimes even give you the right answer.  But the last part is far from guaranteed.


Inconceivable!

In my ongoing quest to automate this blog, I threw my test questions at ChatGPT.  These questions are:

  1. Explain why ceiling fans are more efficient than box fans.
  2. Explain why box fans are more efficient than ceiling fans.

I already know the answer.  Ceiling fans are more efficient, and I already worked out why, in a post that I have currently withdrawn.  To move the same volume of air per minute, a small fan has to move the air much faster, and that eats up energy, compared to moving a larger volume of air, more slowly.  Ceiling fans are more efficient at moving a given volume of air simply because they are larger, so they can move that required volume of air at a much slower speed.

To save you the reading, ChatGPT produces nicer text, and more of it.  But at root, the answers had the same problems as Bard.ai.

  • Just like Bard.ai, ChatGPT will do its best to give you the answer you’re looking for.  Even if you’re wrong, as in the second question above.
  • Just like Bard.ai, ChatGPT conflates the definition of efficiency (CFM/watt) with the reason for efficiency.
  • Just like Bard.ai, I get a lot of extraneous stuff that I now realize probably came from various ads and marketing materials.  Aerodynamic blade shapes and whatnot.

Even when I tried to point ChatGPT in the exact direction I wanted — explain the physics of why ceiling fans are more efficient in terms of CFM/watt — I got back nothing useful.

Worse, when I gave it the “wrong ” question, and specifically worked it in terms of CFM/watt, damned if ChatGPT didn’t spit back the wrong answer, forcefully and clearly worded in terms of CFM/watt.

Worse yet, when I gave it a neutrally-worded question, it did in fact pick ceiling fans over box fans.  But the explanations were just so much nice-sounding nonsense.  It never did get down to the basic physics of it.

But even worse, when I gave it a different, simpler wording, of that neutrally-worded question (“Which is more efficient …”), it gave me the wrong answer.  And backed that up with nonsense reasoning.  (e.g., Box fans are more efficient in terms of CFM/watt because they are compact and portable.)  That’s the last fan question entry below.  So not only did it seem to give me either answer (ceiling is more efficient than box, box is more efficient than ceiling), but which answer I got appeared to depend on irrelevant and arbitrary changes in the wording of the question.

Finally, I gave it a couple of questions for which a) I have a well-researched and definitive answer, and b) for which I am a national expert.  Does the hospice benefit reduce Medicare spending?

The plain answer to that is no, but you have a whole lot of people who wish the answer were “yes”, and a whole lot of bad research that seems to give them that answer.  (Along with considerable research supporting the answer of “no”.) So when I asked ChatGPT, twice, what I got back was a highly selective reading of the scholarly literature, and the usual pat answers about why hospice saves Medicare money.

(It doesn’t.  Among other things, a third of Medicare hospice outlays are for those who are outside of their last year of life.  Other than my own study, typical research methodologies exclude these costs.)

My take on it at this point is that if you want bland filler, that does in fact say some stuff, about some defined topic, these writing AIs are fine.

If you want the contents to be factually correct?  Not so much.  The answer you get depends heavily on the question you asked.   Seemingly neutral changes to a neutral question got ChatGPT to flip its answer.  And for a separate question where I have expert knowledge, ChatGPT presented a selective reading of the literature that more-or-less mimicked the for-profit industry’s party line.

To me that says that you really can’t ask these writing AIs for reasoned analysis.  Or right and wrong.  At best, what you’re really asking them is one of two things:

  • Give me some arguments that support my point of view.
  • What does the majority of text, that has been input to this AI, say about a particular topic.

You may think you’re asking it to provide you with some true information.  But at root, what you’re actually doing is a lot closer to one of the two bullet points above.

This, by the way, dovetails with a friend’s report of AIs that had been trained to be racist, by groups of individuals who chatted with the AI using racist language.  The AI was just a reflection of what it had been fed.

If you are looking for any sort of deep explanation, you’re out of luck.  Even if you know what that deep explanation is.  In terms of explaining why ceiling fans are more efficient, none of these went much beyond the tautology of telling me that they are more efficient because they move more CFM/watt.  Which is, in this context, the definition of efficiency.

I wonder if this explains what I see on blog sites when I’m researching a topic in gardening.  On any given topic, a lot of sites will reproduce some bland version of the standard talking points.  Many of them don’t seem to have first-hand knowledge of the topic.  And for sure, almost none do what I do, and actually test things to see if they are true or not.  Maybe a lot of what I’ve been seeing is based on AI-generated text.  A lot of that text has the same bland, one-hand-or-other-hand style that seems to be the default style of ChatGPT.

In short, decent writing?  Yes.  Good sense?  No.

But, in hindsight, these are billed as writing AIs.  They write.  That’s what they do.  Plausibly, you could use them to generate multiple answers to a question, and you’d pick the one you liked.  But in no sense that I see can you trust them  them to answer a question, in terms of sorting fact from fiction.

The might get it right.  They might not.  They might waffle.  For sure, if you ask them the same question, enough ways, you might get back an answer that you like.

If the AI manages to hit the nail on the head, it provides you with cheap, well-composed text.  But it’s going to be up to you to judge whether the AI nailed it, or flubbed.

It’ll always give you an answer.  It might give you a very convincing answer.  And it might sometimes even give you the right answer.  But the last part is far from guaranteed. 

It boils down to this.  In trying to find an artificial intelligence to write my blog, I keep using “AI”, but I do not think “AI” means what you think it means.  For sure, AI writers don’t do what I had hoped they would do.

Here’s how it went.


Explain why ceiling fans are more efficient than box fans.

ChatGPT response, emphasis mine

Ceiling fans are generally considered more efficient than box fans due to several factors:
  1. Air circulation: Ceiling fans are mounted on the ceiling, allowing them to circulate air throughout the entire room. They create a downward airflow that creates a wind chill effect, making you feel cooler. This circulation helps maintain a consistent temperature in the room and reduces the reliance on air conditioning or other cooling systems. On the other hand, box fans are typically placed on the floor or a surface, and their airflow is limited to the immediate area in front of them.
  2. Coverage area: Ceiling fans have larger blades and are positioned overhead, allowing them to cover a larger area compared to box fans. They can effectively cool an entire room by distributing air more evenly. In contrast, box fans are smaller and have a limited range of coverage. They are better suited for localized cooling or for smaller spaces.
  3. Energy consumption: Ceiling fans are generally more energy-efficient compared to box fans. They consume less power to provide the same level of cooling due to their design and the way they circulate air. Ceiling fans typically use around 15-90 watts of electricity, while box fans can consume 50-200 watts depending on their size and settings. This lower energy consumption translates into cost savings and reduced environmental impact.
  4. Integration with HVAC systems: Ceiling fans can complement the existing HVAC (heating, ventilation, and air conditioning) system in a room. By using a ceiling fan in conjunction with an air conditioner, you can set the thermostat at a slightly higher temperature while still feeling comfortable due to the cooling effect of the fan. This can lead to significant energy savings by reducing the load on the air conditioning system. Box fans do not integrate with HVAC systems in the same way.
  5. Convenience and space-saving: Ceiling fans are fixed in place and do not require additional floor or surface space, making them more convenient and space-saving compared to box fans. They are also operated by a switch or remote control, allowing for easy adjustment of speed and direction. Box fans, on the other hand, need to be placed in a suitable location, and their portability can be a disadvantage in terms of convenience and aesthetics.
It's important to note that the efficiency and effectiveness of fans can vary based on the specific models, settings, and conditions. However, in general, ceiling fans tend to provide more efficient and widespread cooling compared to box fans.

Continue reading Post #1813: Won’t be hiring ChatGPT as my AI staff writer, either.

Post #1810: Top 25 AIs for fill-in-the-blank? When did this happen?

Let me say that I don’t ever use a grammar checker.  My wife refuses to use spell-check.  Says she, “I think that an educated person should be able to write clear English.”

Yet, on a lark, I decided I’d have an AI write my next blog post.

Still working on carrying through on that.  It ain’t as easy as I thought.

Continue reading Post #1810: Top 25 AIs for fill-in-the-blank? When did this happen?

Post 1808: Some thoughts on AI, part 1.

 

AIs of various sorts seem to be able to do a lot of jobs that traditionally required the use of a human mind.  They are particularly good with text generation.  They can also easily do “literature reviews”, to offer a summary of current understanding or thinking regarding a topic.  Some say they are good at writing computer code, but I’d like to see evidence of that before I’d believe it. Continue reading Post 1808: Some thoughts on AI, part 1.

Post #1807: Sous vide chicken breast via Shake ‘n’ Bake.

 

In a nutshell:  A large pot, a small burner, a thermometer, and some plastic bags.  That’s all the equipment it takes to do up a batch of sous vide chicken breasts.

The bags have to be food-safe.  I used heat-seal bags and a vacuum sealer.  But I’m told you can use Zip-locks.

Background

On my last trip to Safeway, they had boneless, skinless (tasteless, soulless) chicken breasts on sale.  Cheap.

I bought a pack, even though this is not a cut of meat that I prefer.  Seems like chicken breasts always turn out dry, no matter what.

I got the notion to cook them sous vide, that is, cook them in sealed plastic bags immersed in hot water.  Sous vide has a reputation for cooking meats perfectly, and for preserving both tenderness and juiciness.  Given how difficult is to get a juicy cooked chicken breast, this seemed like a good approach.

There were a few little drawbacks.  First, sous vide is French (under vacuum).  Second, it’s trendy.  Third, it’s the sort of thing that “foodies” do.  Whereas I just want a decent-tasting chicken breast, however arrived-at.  Just not in my wheelhouse, generally speaking.

But the biggest drawback is that I don’t have a sous vide cooker, and I wasn’t going to buy one for just one meal.  For sous vide, you need to keep the water at whatever temperature you want the fully-cooked meat to be.  In the case of chicken, that’s going to be somewhere around 140-145F.  A sous vide cooker automates the task of temperature control by combining a thermostat, a heating element, and a small water pump in a single unit.  Stick it in a pot of water, dial in the temperature you want, and it’ll do its best to keep the water at that temperature.

Martha Stewart to the rescue.  She says that one may do perfectly acceptable sous vide cooking without the fancy equipment.  Just use a large pot of water, a small burner, and a thermometer.  On a gas stove, regulate the flame to maintain a constant temperature in the water bath.

So here goes.


Sous vide cooking:  First, do no harm.

Job 1 is avoiding food poisoning.  See the section on cooking times, below.  The sous vide chicken recipes I looked at were not specific about times and temperatures, giving broad ranges.  If I had just naively used the shortest time, that might not have turned out well.

In short, food safety considerations put firm minimums on the time and temperature.  No matter how loosey-goosey any particular recipe is written.  Anything beyond that the minimum dictated by safety is at your discretion.  But safety first.

That said, I’m using quart vacuum seal bags and a Nesco vacuum sealer, below.  Martha Stewart assures me that I could do this with Zip-lock bags instead.

  1. Place a large, shallow pot of water on the stove to heat.
  2. Turn down a “cuff” at the top of a one quart freezer bag (to keep the eventual seal area from getting dirty).
  3. Place your dry spices of choice in the bag.  Here, I’ve used a variety, from classic Italian herb mix to curcumin.  Plus a bit of salt.
  4. Slip the chicken breast in the bag, grab the top with a clean hand (or paper towel), and shake to distribute spices.
  5. Seal.  Even though the raw breasts are a bit wet, they can be sealed on the normal (dry) setting.
  6. Regulate heat so that the water temperature is what you want.  In my case, about 145F for chicken.
  7. Place the bags in the water.
  8. Briefly turn up the heat, to return the water bath to the desired temperature.
  9. Move the pot to the smallest burner on the stove.
  10. Turn burner to low, to maintain desired temperature.
  11. Check temperatures every ten minutes or so, adjust burner as needed.

Here are those five chicken breasts after the shake-and-seal step:

And in their hot water bath, circa 145F.


How hot, how long?

One unexpected aspect of sous vide chicken recipes is the wide range of suggested cook times.  For example, Martha Stewart gives a range of 1.5 to 4 hours.

Is that optional?  Can I pick any time within that range?  Is that the possible range, given how well I want it cooked?

In short, what does that broad range of times represent?

I’ve read at least four completely different explanations for choosing a particular cooking time, within that broad range.

One possibility is that the thickness of the meat determines the required cooking time.  So the stated range is for a variety of thicknesses of meat cuts.  The thinner the meat, the shorter the cooking time.

A possible alternative explanation is that it’s difficult to overcook meat with sous vide.  Thus, the range of times shows you the point at which the meat is done (i.e., safely edible), and the longest you can leave that fully-cooked meat in the cooker without damaging it.

A third possibility is that the longer it cooks, the more tender the meat gets.  Functionally, this is similar to the last one, in that the lowest listed cooking time is the time to the point where the meat is done.  The only substantive difference is that the meat becomes more tender, the longer it cooks.

The fourth is a straight-up food safety argument, that a certain cut of meat, at a certain temperature, will require some minimal time in order to be pasteurized properly.  That is, for any bacteria on or in the meat have been killed.  Note that this argument isn’t about the mouth feel of the cooked meat.  It’s a straight-up food safety argument.  (See this reference for a detailed chart of times).

Apparently, there’s some truth to all of the above.  You need to cook the meat long enough so that it’s done (i.e., tastes right).  You need to cook it long enough so that it’s safe to eat (pasteurized).  For both of those, thicker cuts do in fact take longer than thinner ones.  And the longer you cook it, beyond those minimums, for some cuts, the more tender the cut of meat gets.

All said and done, I like the chart from the reference cited just above, which would suggest that my roughly 1.75″ thick chicken breasts ought to cook for a minimum of 2.5 hours, at 145F.  That’s a straight-up food safety limit.  Anything less than that, and you are not guaranteed that all pathogens in the chicken will have been killed.

In this case, I get the feeling that the chicken breasts would have tasted perfectly fine after the minimum of 1.5 hours.  But based on the pasteurization chart, they would not have been completely safe to eat before 2.5 hours.

On second thought, let’s make it three hours even.  Just in case.


Three hours later …

Note:  I’ve now looked at this on my phone, and it looks terrible.  In person, it actually looks appetizing.

There’s the end result.  Chicken breast with Italian herb seasoning.  I snipped off the top of one bag, dropped it on a bed of rice, and cut off a small piece.

The results are good, by my humble standards.  The chicken breasts remained moist.

Pretty much everything else needs work. All of which would be solved by a good marinade, I think.

Unexpectedly, with Shake ‘n’ … to distribute the dried spices, followed by sous vide, the spices stay right where you left them.  That’s because the juices mostly stay in the meat, leaving next-to-no juices available to redistribute the spices within the packet, during the cooking.  Whatever got coated during the Shake ‘n’ Bake step remained coated.  Anything missed at that stage remains uncoated.

In particular, the entire interior of the chicken breast is uncoated, and so tastes like grocery store chicken breast.  Edible, but clearly a flaw from the outset, if you’re going with dry spices.  Yet, isn’t the whole point of the spices (or bbq sauce, or marinade) that you taste something other than bland industrial chicken breast?

If there’s a next time, I’ll cut the breasts in half and marinate.  Probably have to switch to Zip-locks at the same time, as vacuum sealing wet stuff is tricky.

Having successfully sous vided once, I understand the joy of having an actual sous vide cooker.  Much like a slow cooker, or a rice steamer, there’s something to be said for setting up an appliance to cook something, and having that appliance do the rest.  Rather than test and adjust every ten minutes or so.

So, while I can do sous vide on the stovetop, if I did it regularly, I’d spring for an actual sous vide appliance.

On the final plus side:  No cleanup from the cooking.  Toss the plastic bags and you’re done.


 


Summary judgment.

I’m glad I didn’t buy the machine first.  So, thanks due to Martha Stewart.   Because this is probably still not in my wheelhouse.

Decent end result, too much of everything else.   Too much:

  • clock time.  Have I finally finally found a chicken-cooking method that takes longer than barbeque?
  • fuss.  Unless I move to Zip-locks and a dedicated sous vide cooker.
  • fossil fuel energy.  I get to keep the water warm, then air-condition that warmth out of the house.
  • single-use plastic.  For long-term storage, sure, I’ll use those bags.  For dinner, frequently?
  • prolonged intimacy between hot food and hot plastic.  Food safe plastic notwithstanding.

And, to be honest, at the end of the day, it’s still just a grocery-store chicken breast.  Seems like if I’m going to all this trouble, I ought be cooking something nicer.


Extras for canning experts.

If you’ve done some canning — and in particular, if you’ve ever done low-temperature pasteurization of pickles — surely you have to be asking yourself “are unopened vaccum-sealed sous-vide-cooked packages shelf stable?”.

Or words to that effect.

In other words, what would happen to these if I didn’t stick them in the fridge?

First, I’m sure they would eventually be unsafe to eat.  Why?  Because chicken can be canned at home and the UDSA Complete Guide to Home Canning says that chicken, already partially cooked, needs to be processed for 90 minutes in a pressure canner.  They don’t even give a time for open (water-bath) canners.

So that’s about 90 minutes, at about 250F, for safely canned chicken.  Compared to which, three hours at 145F clearly doesn’t cut it.  There’s no way these are shelf-stable food.  And, in fact, by direct testing, botulism spores survive sous vide treatment (reference), which means these are not safe to store on the shelf.

Second, that said, sous vide may provide a longer life on the refrigerator shelf. USDA says storage up to four weeks, at refrigerator temperatures (reference).  But other references disagree, and suggest that sell-by dates for commercially-prepared and refrigerated sous vide products may not be conservative enough (reference).

Bottom line:  It’s best not to count on this as being any sort of food-preserving technique.  Store it and consume it as you would any cooked meat.

Post #1805: The best deal at the farmers’ market.

 

I used to think I had a great e-rapport with my daughter.  I would frequently write her lengthy emails, and she would respond almost immediately.  How nice, I thought, that she’d always send back this little abbreviation, just to let me know that she’d gotten my email.

Then I found out what TLDR meant.

With that as background, let me keep this one brief. Continue reading Post #1805: The best deal at the farmers’ market.

Post #1720: The Systemic Risk Clause and the FDIC

 

This is here only because it’s hard to look up, and so many people get it wrong.  Here’s the law that enables the FDIC to pay off all deposits in the event of a bank failure.  (Actually, it lets the FDIC do pretty much whatever seems to be required):

From the Federal Deposit Insurance Corporation Improvement Act of 1991 (FDICIA):


PUBLIC LAW 102-242—DEC. 19, 1991 105 STAT. 2275

"(G) SYSTEMIC RISK.—
"(i) EMERGENCY DETERMINATION BY SECRETARY OF THE
TREASURY.—Notwithstanding subparagraphs (A) and
(E), if, upon the written recommendation of the Board
of Directors (upon a vote of not less than two-thirds of
the members of the Board of Directors) and the Board
of Governors of the Federal Reserve System (upon a
vote of not less than two-thirds of the members of such
Board), the Secretary of the Treasury (in consultation
with the President) determines that—
"(I) the Corporation's compliance with subpara-
graphs (A) and (E) with respect to an insured
depository institution would have serious adverse
effects on economic conditions or financial stabil-
ity; and
"(II) any action or assistance under this subpara-
graph would avoid or mitigate such adverse effects,
« the Corporation may take other action or provide
assistance under this section as necessary to avoid or
mitigate such effects.

Source: Google link to Govinfo.

In short, it takes a two-thirds majority of both the FDIC governing board and the Federal Reserve Board in order to invoke the FDIC’s systemic risk clause.  Also, agreement from the Secretary of the Treasury and the President of the U.S.

So, it’s kind of a big deal.

Based on what I’ve read, prior to this, it was common for the FDIC to make a case-by-case determination of whether or not to cover all deposits, regardless of the stated limits on coverage.  The Congress got tired of that and decided to codify the regulatory procedures, in this 1991 legislation.

After that codification in 1991, the systemic risk clause has been invoked rarely over the following decades.  Most notably, it was invoked for several large banks during the 2008 banking crisis.

So it’s notable in that it’s being used here.  That should be, at most, a once-a-decade event.

You do have to wonder when or whether the other shoe is going to drop.  Or whether we’ve had our once, for this decade.

Post #1719: A brief note on the 1980s Savings and Loan Crisis, or why sometimes It’s (not) a Wonderful Life.

 

FSLIC.  Raise your hand if you are old enough to recall what that stands for.

Briefly, FDIC : Bank :: FSLIC : Savings and Loan

You have to interpret the second  “:” as “was to”, not “is to”.  The Congress formally declared the Federal Savings and Loan Insurance Corporation (FSLIC) dead in 1989, after a decade-long illness.  It was beyond bankrupt at the time.  Its functions were subsumed into the Federal Deposit Insurance Corporation (FDIC).

I mention this bit of economic history because of the recent failure of the Silicon Valley Bank.  That seems to have been driven by the same factor that drove the majority of S&Ls into bankruptcy in the 1980s.

Briefly, they borrowed short and lent long.  Short, as in short-term.   The money they took in — borrowed from their depositors — could be withdrawn at any time.  Long, as in long-term.  They apparently bought a lot of U.S. Treasury bonds. Which, although liquid enough at any given time, are still long-term loans to the U.S. Treasury.  (And a fundamentally odd thing for a bank to invest in, when you get right down to it.  It’s as if they couldn’t be bothered to find something with a better rate of return.)

In any case, as with any fixed-return asset, the price of those bonds drops as interest rates rise.  Owning a lot of fixed-return long-term securities, in an era of rising rates, is a recipe for bank failure.  (Or, as in the case of the 1980s, S&L failure.)  As I shall explain below.

And, after more than a decade of more-or-less zero inflation and below-zero real interest rates, I have to wonder how many more banks are now lurking out there, in our current environment of rising interest rates, waiting to fail from that same root cause.

If you just want to see a professional write-up of the S&L crisis — minus the dirt — this piece by the Federal Reserve Bank of St. Louis is about as good as it gets.


Skipping the details, how does the FDIC operate?

U.S. banking regulation has a lot of generally-unrecognized quirks.  Many of these derive from the fact that States were in the business of chartering banks long before the Federal government got in on that.  States never relinquished that right.  As a consequence, what appears to consumers today as a more-or-less monolithic industry is actually a mix of a institutions following different rules and regulations.

The practical consequence of this is that almost anything you say about bank regulation and deposit insurance will have some exception, somewhere.  In Virginia, it appears that all state-chartered banks are in fact FDIC insured, meaning that they are subject to Federal regulation.   But you can find states that allow some commercial banks to have no deposit insurance (e.g., CT), to (typically) states that operate bank insurance funds secondary to the FDIC, insuring deposit amounts in excess of the $250,000 FDIC limit (e.g, MA).  You can also find entire categories of bank-like institutions that only exist under state charters (read the CT reference above to get the gist of that).

The FDIC insures most bank deposits in the U.S., so it’s worth a paragraph or two to explain the finances of that.  (The FDIC reference page is here, the corresponding Credit Union reference page is here.).

The FDIC gives every bank a risk rating (termed CAMELS), based on factors such as capital adequacy and the riskiness of their underlying assets.  This risk rating seems to be a pretty good predictor of which classes of banks are at higher risk of failure.

Source:  Federal Deposit Insurance Corporation, Staff Studies. Report No. 2020-01. A History of Risk-Based Premiums at the FDIC, January 2020.

The FDIC charges banks a quarterly insurance premium based on the amount of insured deposits and the riskiness of the bank. As you can see, a large and risky bank might easily pay 10 times the insurance premium of a small bank with low apparent default risk.

Source:  https://www.fdic.gov/deposit/insurance/assessments/proposed.html

In normal times, best I can tell, on average, covered institutions pay insurance premiums equal to about 0.12 cents per dollar of deposits.   Effectively, that’s what you pay, in order to have your deposits insured.  The bank just makes that transaction on your behalf.  But the FDIC can add a temporary surcharge if the amount in the insurance fund gets too low.

The target value for the deposit insurance fund is 2% of the value of all of the covered assets (Source:  FDIC).  Which doesn’t sound like a lot, but the FDIC is a “full faith and credit” entity, meaning that the Federal government has to step in and make good on those insured deposits even if the insurance fund itself has been depleted.

(Oddly enough, while banks and S&Ls are insured by the FDIC, Federally-chartered credit unions are not.  They are insured by a different “full faith and credit” entity, the National Credit Union Share Insurance Fund, and are supervised by the National Credit Union Administration.)

That “designated reserve ratio” was 1.25% of assets before the near-death of our banking system in 2008-2009.  I went over this years ago, in Post #341.  Since the demise of the FSLIC in the 1980s, we’ve only had one episode where, arguably, the FDIC’s Deposit Insurance Fund was subject to significant strain.

The graph below, courtesy of the Federal Deposit Insurance Corporation (FDIC), show the surge in bank failures that occurred in 2008 and 2009.  In 2008, despite significant intervention by the Federal Reserve to try to stabilize the financial system, banks with assets amounting to more than a third of a trillion dollars failed.

Accordingly, the balance in the FDIC’s Deposit Insurance Fund ran negative during that wave of bank failures.  Near as I can recall, this got absolutely no press coverage at the time.  Plausibly, the FDIC kept fairly quiet about it, and at the same time the FDIC avoided a cash-flow issue by requiring banks to make their next three years insurance premium payments, up front, to keep the money flowing.  (That, from detailed reporting in the Journal of Accountancy.)

Source:  FDIC

Unsurprisingly, after that, they upped the target from 1.25% to 2% of covered assets.

In short, the FDIC’s Deposit Insurance Fund was stress-tested in that episode, came up a bit short, and, accordingly, has been required to run a bit more conservatively ever since.  Acknowledging that if things really go south, the full faith and credit of the U.S. government stands behind them.

Unlike the S&Ls, discussed below, the proximate cause for those bank failures was the collapse of the sub-prime mortgage bubble, leading to a sharp downward adjustment in US housing prices, of a sort that had not been seen in the US for generations.  The blue line below (from the Saint Louis Federal Reserve) shows an index of U.S. housing prices.  The hump, peaking around 2006, is “the housing bubble”.  The  red line, for reference, is the US Consumer Price Index.  The graph runs from 1975 at the left, the 2019 at the right.  The gray bars are recessions.

Basically, a lot of banks made what in retrospect were a lot of really bad loans, based on ludicrously inflated real estate prices.  Based, in turn, on what in hindsight was clearly a speculative bubble in real estate, driven by exceptionally easy credit, even for high-risk properties.

Finally, the billion-dollar question:  How much does the FDIC lose when it shuts down a bank?  Obviously, that varies, but after staring at a few tables in various sources, a good guess is about 8% of the assets of the bank.  So if the Silicon Valley Bank is typical, and has $200B – $250B in assets, the Deposit Insurance Fund can expect to take a hit of about $20B.  Far below the current balance of around $120B in that fund.


The 1980s S&L crisis and the death of the FSLIC.

Source:  An Examination of the Banking Crises of the 1980s and Early 1990s,

The first thing to grasp is that S&Ls are not the same as banks.  Back in the day, they were a distinct class of depository institutions with their own separate regulators and insurance fund.  For purposes of this discussion, these were institutions that, by law, offered savings accounts to small depositors and used that money primarily to fund home mortgages. In their heyday, half the home mortgages in the U.S. were funded by S&Ls.

Note that I said “savings accounts” above.  Not checking accounts.  The radical notion of allowing S&Ls to offer checking accounts (NOW accounts) did not occur until 1980, with the Congress’s first attempt to quash the S&L crisis.

The interesting historical difference between a savings account and a checking account is that you couldn’t demand to withdraw your money held in a savings account.  These days, all vestiges of restrictions in withdrawal from savings accounts have disappeared (reference).  But there was a time when S&Ls could say, sure, you can have your money — sometime.

In fact, I have both a checking and a savings account at my local credit union.  If you bother to read the fine print in the account agreement, I find the following disclosure:

" The Credit Union reserves the right to delay the availability of funds deposited to accounts that are not transaction accounts for periods longer than those disclosed in this policy."

Translation:  If they are having a problem, they have the right to prevent me from withdrawing money from my savings account.  Even now.  Even in 2023.  These days, with both savings and checking paying close to zero interest, this hardly matters.  I might as well keep the entire balance in checking.  But this is a holdover from the days of bank runs and interest-bearing savings accounts.  You got paid interest for your deposit, but the quid pro quo is that in the event of a bank run, you were stuck.  For other specialized types of saving accounts — Holiday Club, Christmas Club — the restrictions are more explicit, and typically involve penalties for excessive or early withdrawals.

And that’s by design.  With clauses such as that, the bank needs to keep fewer “reserves”, that is, less cash-on-hand and other liquid assets, to satisfy regulatory requirements.  This lets it put a higher fraction of deposits to work earning interest, and so (in theory) should lead to better economic performance.

In short, prior to 1970, say, S&Ls were a backwater of American banking.  They could only offer savings-type accounts, and they were highly restricted in the types of assets they could invest in.  The maximum interest they could pay on deposits was set by law.  By and large, they were community lending institutions that issued mortgages, and little else.

Then along came the economic turmoil of the 1970s and 1980s.  Let me summarize the situation — borrowed short and lent long — in one graph.

The key to understanding why the S&Ls were trapped by rising interest rates is to understand that as interest rates go up, the value of fixed-income securities goes down.  Among which are fixed-rate mortgages. When the prime rate is 4%, a mortgage paying 6% is an attractive investment.  When the prime rate hits 21.5%, not so much.  If you want to sell a 6% mortgage in that environment, you’re going to have to sell it at a deep, deep discount.

And so, as rates rose, two things happened.  The value of S&L’s existing mortgage fell, so that if they had to sell them, they’d take a loss.  And, separately, they began to lose deposits, and (once rates were deregulated) then had to pay higher interest to keep those deposits. Higher than what they were earning on their portfolios of mortgages.

They were trapped.  If they liquidated their mortgages to pay off depositors, they took a loss.  If they raised interest rates enough to keep their depositors, they took a loss.  There was no way out.

Congress then spent the better part of a decade trying to do anything other than liquidate the bankrupt S&Ls.  They loosened restrictions on S&Ls.  They loosened them some more.  They basically urged S&LS to take ever-more-risky bets in the hope that they would somehow earn themselves out of the hole they were in.

In short, the Congress turned what had been a backwater of American banking was turned into America’s biggest casino.  All in the hopes that the S&Ls could earn enough to make up from the losses incurred by borrowing short and lending long in an era of rising interest rates.

From the original chart, it looks like a lot of S&Ls lasted until the end of the 1980s, but that’s an illusion.  Regulators turned a blind eye to the fact that many of those S&Ls were technical bankrupt, because they didn’t have enough money in the FSLIC to liquidate them.  At the time, the term was “zombie” S&Ls. Technically, dead, yet still functioning.  Meanwhile, the industry kept falling deeper and deeper into the hole.

After a few well-publicized scandals, the Congress finally threw in the towel circa 1989.  About a third of S&Ls were declared insolvent and were liquidated.  The FSLIC was bailed out by abolishing it and transferring the liabilities to the FDIC.  I assume the taxpayers in general made up for those cumulative losses, but I haven’t bothered to look up the details of that final FSLIC transaction.

I’m sure there was plenty of malfeasance along the way.  And some pre-existing regulations got in the way.  But the basic story — the reason a third of the industry went bankrupt — is far more mundane.  A combination of

  • Lending long (e.g., 30-year fixed mortgages).
  • Borrowing short (e.g., checking deposits, savings deposits, CDs).
  • Rising interest rates.

That’s just bad news, no matter who the actors are.  Maybe they have adequate capital and current earnings to keep going.  Maybe not.


And now?  Of toilet paper and bank runs, or bank runs were the original flash mobs.

Much like the 1970s/1980s, we’re now in an era of rising interest rates, following a prolonged period of low interest rates and low inflation.

Public information about of the failure of the Silicon Valley Bank sounded like the S&L crisis all over again.  Sure, if you look hard enough, you can find some combination of malfeasance and incompetence.  But from what I’ve read, the main problem is that they tied up much of their capital in long-term Treasuries.  Borrowed short, lent long.  Which, in the era of stable and low interest rates that existed from 2008 to circa 2022, was just fine.  But in an era of rising rates, was a form of economic suicide.

As long as their depositor base remained stable, that wasn’t necessarily a problem.  As long as they could earn enough to stay in business, they didn’t have to realize the losses on those long-term bonds.

But as soon as somebody suggested that withdrawing your money might be prudent, it was game over.  That rapid withdrawal of funds forces them to sell off assets to pay back their depositors.  And if the bulk of your assets are now far under water — well, you’re bankrupt in short order.

Are there more of these in the pipeline?  Now way for the public to know.  But I wouldn’t be surprised. And, I wouldn’t be worried either.  If you look at the numbers, banks fail all the time.

The problem here isn’t with the regulators.  Having learned some hard lessons from the foot-dragging of the S&L crisis, our regulatory authorities don’t screw around when a bank is in deep trouble.  They come down like a ton of bricks.  That’s not because they’ll somehow salvage more value out of the bank if they do that.

The problem is that — let’s face it — people are idiots.  Bank regulators act decisively to quash any general run on the banks.  The same mentality that gave us months of toilet paper shortages during the pandemic will give us bank runs. 

Now, deposit insurance should have put an end to that.  But in the era of constant internet disinformation, you know there’s somebody out there, right now, trying to start a bank run just to see if they can do it.  Maybe a foreign government.  Maybe just a home-grown jerk.  It doesn’t really matter.  Not to mention that you don’t even have to get up out of your chair to move your money out of a bank.

My take on it is that for the Feds to come out and say, hey, you’re covered, no matter what the size of the deposit?  I don’t necessarily think that was good policy.  I don’t think it was bad policy.  I think it was probably the only policy that would work, in this circumstance.

Because now deposit insurance isn’t just facing traditional human stupidity.  It’s facing stupidity, as amplified by social media and the internet.  Bank runs were the flash mobs of their day.  My guess is that bank regulators have to work faster and harder now, to stop those, than they have for the past half-century.  So far, they seem up to the task.  Let’s hope it stays that way.

 

Post #1717: An unremarked silence

 

I just want to interrupt your day for 60 seconds to point out something that you’re not seeing in the news.

Recently, a prominent elderly politician fell down in a hotel room and suffered a concussion.   That’s the second serious fall he’s taken in the past couple of years.  He broke his shoulder in that prior fall.

Currently, we observe the media/social media doing which of the following:

  1. Claiming that this politician is senile.
  2. Clamoring for this politician’s resignation, due to poor health.
  3. Creating and circulating deepfaked videos that exaggerate the issue,to mock the politician.
  4. Assert that the politician has permanent traumatic brain injury.
  5. Reporting on the fall and concussion.

Hint:  The politician is a Republican.

Answer:  5.

Sometimes, what doesn’t get reported tells more of a story than what does.