Arnold Kling  

In Which Winnie-the-Pooh Uses a Bank

Mark Thoma's Question... EconLog Welcomes Guest Blogger...

In an ongoing dialog with Mencius Moldbug, I changed from my favorite fruit tree metaphor.

My favorite Winnie-the-Pooh story is the one where Pooh and Piglet try to catch a Heffalump. They dig a pit for a trap, and Pooh puts a jar of his honey in the pit, in order to lure the Heffalump. That night, Pooh wakes up hungry. Realizing that his only jar of honey is in the trap, he goes to retrieve it. Comedy ensues.

The analogy with banks is that banks that make long-term loans are vulnerable if their customers wake up hungry. Is unsound banking natural, or does it arise only with government? Is it ok, even if it is unsound?

Here's the set-up. We're on a honey standard, with jars of honey as the medium of exchange and the store of value.

A depositor can put a jar of honey in the bank, claiming the right to withdraw it any time. The bank says, full disclosure, that it plans to take some of the honey jars on deposit and lend them to entrepreneurs who plan to trap heffalumps. If nothing goes wrong, the entrepreneurs will catch the heffalumps, exchange them for honey, keep a profit, and pay back their loans with interest. The bank will keep some of the interest as profit and give the rest to depositors, who get their honey back, plus interest.

If something goes wrong, though, I may lose some or all of the honey I put on deposit. One of the things that can go wrong is a lot of us wake up hungry and ask for our honey before the heffalump traps ever mature.

Do I make the deposit? The bank says that it has the law of large numbers going for it. The chances are really small that lots of depositors will wake up hungry at the same time. There may be only a one in a million chance that I won't get my honey back. Maybe I'm willing to take that chance.

I'm not convinced that this sort of bank is what would naturally evolve. Instead, we might see a bank that offers two types of deposits. One would be honey deposits that bear no interest. You store your honey with the bank, and you can withdraw it at any time. It stays in the vault until you ask for it.

The other deposits are long-term deposits with penalties for early withdrawal. These do not all stay in the bank--some of them are lent to heffalump entrepreneurs. The penalties would decline over time. They would rise as the bank's supply of honey on hand falls. The penalties would ensure that everyone who waits until the heffalump traps mature will get their honey back, with interest. There is no basis for a run on the bank based on pure fear.

This bank does not pretend to offer a free lunch. You cannot make an interest-bearing deposit unless you are willing to pay a penalty for early withdrawal. The bank can use the law of large numbers (the fact that it is unlikely that most people will not withdraw at once) to specifiy that low penalties will prevail normally. But if lots of people wake up hungry at once, the higher penalties will kick in, because the bank's reserves start to drop.

If you want an interest-bearing deposit, you know that you are taking a risk that when you want to withdraw, if it happens that other people want to withdraw also, your penalty will be high. The more confident you are that you won't get hungry earlier, the more certain you can be of getting your honey back with interest.

Suppose government comes along and says, "Gosh, there sure are a lot of honey jars sitting idle in non-interest-bearing accounts. If we offered deposit insurance, then people would put their honey into interest-bearing accounts, and banks could fund more heffalump traps." Does that make society better off?

Well, yeah...except, where does the government get the honey to cover its deposit insurance obligations?

Instead, government can force everyone (or trick everyone, by making false promises) to make interest-bearing deposits. In many scenarios, people are better off. Once in a blue moon, though, a lot of people get hungry at once, and the government has to ration the honey that people want to withdraw. That is, it reneges on its promise.

Think that government reneging on promises is unrealistic? If you're 40 years old, come back in 30 years and tell me how that Medicare thing is workin' for ya.

Again, people might be better off most of the time if they trust banks (or government) and ignore the possibility that they could get shafted. Most of the time, they won't get shafted.

Mencius thinks that this might work differently with fiat money instead of honey. It is true that if everybody wakes up wanting to touch something green with the picture of a President on it, the government can just print some. It doesn't have to ration green paper the way it has to ration honey. But I think it amounts to the same thing--once all the green paper is printed, each piece is worth less, which is just like getting back a jar of honey that is missing a few spoonfuls. So I don't see how fiat money makes the scam any more or less transparent.

Boy, does this stuff give me a headache.

Comments and Sharing

COMMENTS (19 to date)
MattYoung writes:

I think the government takes an interest in the heffalump industry, and makes monopoly promises.

The future outlook for the heffalump industry becomes volatile because the senior aggregator can change heffalump projections with fewer, larger transactions.

Government eventually exits the heffalump industry because heffalump consumers demand more market clarity.

Luis M writes:

Your analogy is spot on. No headache here, although I am thinking of needing some honey for me tea.

Normal people would not willingly place there money in an account and then be told "oh by the way I am going to loan your savings out and make some money from it, but no worries it will be there when you need it" 'wink wink'.

As always leave it to government to legislate "behavior" in the interest of society.

I wonder if one where to look at the costs of supporting a fractional banking system for a society are they worth it?

In normal times no doubt, but do the black swan events wipe out those costs?

How much damage will be done in order to prop up our fractional reserve system.

- The inflation damage for guaranteeing the lost capital gambled away.
- The damage done to the confidence of the financial sector of our economy
- The increase deficit, the risk transfers from financial institutions defaulting to a potential government default.

It would appear in under the last 80 years that we have 20 years of damage and now we are starting at maybe another 10 years.

It would seem that fractional reserve banking is a shell game where the saver i.e depositor always loses.

In the non-insured past they literally lost, now they are slowly bled via the inflationary costs as government is forced to devalue (print more currency) to make up the illusion of the guaranteed deposits. In addition to the the costs of allowing the government to print money in the first place which they use to manipulate other aspects of government finances.

I would say that instead of a few spoonfuls of honey missing you put in 1 jar of 100% honey and you get back on withdraw 80% honey and 20% corn syrup.

The problems come when all you get is corn syrup.

M writes:

Great analogy, great movie (full disclosure: father of a four-year-old). :-)

mensarefugee writes:

Its not just that we would need two different accounts for honey. We would need two different types of honey for the analogy to be correct.

Translated to the real world. We cant have the same currency for Non-Interest bearing accounts that have no risk, and Interest bearing accounts that have risk. Because if it was the same, when the interest bearing accounts tank, and get bailed out - the value of the non-Interest bearing accounts shrinks too. (Grisham's law of honey?)

A lot of talk about this goes on in libertarian circles, with the illuminati being blamed at the extremes.

At any rate - Im glad it gives you a headache, now I know its weird enough to confuse even trained libertarian economists.

Misery loves company :)

Mencius writes:

Even when the game was played with gold, there were a thousand and one ways to effectively dilute the gold supply.

Example: your bank issues notes that say "one ounce of gold, if you get hungry." Your government, which is in cahoots with your bank - probably because the former is borrowing from the latter - says, "you can pay your taxes with gold or with bank notes, accepted at par." Result: a considerable lessening of redemption pressure, a considerable expansion of money supply.

Eventually (after World War I) the dilution got too great and the old tricks no longer worked, which is how we got to a fiat currency. But the transition was gradual. The only sharp line is between systems that don't engage in credit expansion and those that do. The Anglo-American banking system crossed this line in the 17th century, so it's a little difficult to blame W, Barney Frank, etc.

Mencius writes:

I can't resist quoting my favorite passage from Condy Raguet's 1840 Treatise on Currency and Banking. From page 37:

Such being the theory of this branch of my subject, I have the satisfaction to state in regard to the practice under it, upon the testimony of a respectable American merchant, who resided and carried on extensive operations for near twenty years at Gibraltar, where there has never been any but a metallic currency, that he never knew during that whole period, such a thing as a general pressure for money. He has known individuals fail from incautious speculations, or indiscreet advances, or expensive living; but he never saw a time that money was not readily obtainable, at the ordinary rate of interest, by any merchant in good credit. He assured me, that no such thing as a general rise or fall in the prices of commodities, or property was known there; and that so satisfied were the inhabitants of the advantages they enjoyed from a metallic currency, although attended by the inconvenience of keeping in iron chests, and of counting large sums in Spanish dollars and doubloons, that several attempts to establish a bank there were put down by almost common consent.

Marcus writes:

Concerning the 'borrow short to lend long' business model, I believe it would occur even without any government influence because there is an interest rate spread between short term and long term rates to be exploited.

I think the role of government has been a mostly futile attempt to make this inherently unstable business model more stable.

Todd writes:

Arnold, I thought that was your most lucid explanation yet. I didn't get a headache, but as Mencius points out, as long as we have a government with as much power is ours has, there is little reason to hope that even a gold standard would be able to withstand the inflationary pressures of the state. The operation of the system seems plain enough, does your headache come from trying to decide if government intervention makes people better or worse off over the long haul? I can see how that question is more difficult to answer, though I think I side with Luis M and think we would be better off without government guarantees, since what they guarantee is actually impossible to achieve.

Tim writes:

"...where does the government get the honey to cover its deposit insurance obligations?

Bees? (Fed?)

fundamentalist writes:

Great analogy! What you have described is 100% reserve banking, which Austrians have demanded for decades. The trouble with today’s banking system is that it loans honey it doesn’t have. That’s the fractional reserves system. To make the Winnie-the-Pooh analogy fit fractional banking, you would have to have some way for the bank to create synthetic honey with the wave of a wand and loan that out.

scott clark writes:

Don't club fractional reserves so hard. Lots of businesses overbook as a matter of course. Parking garages oversell their spaces 'cause they "know" not everyone is going to show up at the same time, airlines oversell their seats 'cause they "know" not everyone is showing up to take that ride. Restaurants do it too. And when the numbers turn against them, they try to cut a new deal, until things balance out. Banks oversell their deposits. The problem comes in if they over-oversell their deposits, and everybody over-oversells their deposits. So I don't think you can blame the fractional reserve system, that's a particular way of doing business that emerges in a free market, and as long as everyone knows the score, i.e. full disclosure, the problem is giving government guarantees, lenders of last resort, regulatory apparatuses that make customers complacent and fractional reserve bankers more risk loving.

Rolf Andreassen writes:

In the banking system you describe, what does the bank gain from offering demand deposits? It's not making any profit from honey that's sitting in its vaults.

fundamentalist writes:

scott clark: "Lots of businesses overbook as a matter of course."

I guess it's a matter of degrees. I don't know of any business that operates with the leverage that banks use, except home owners. I tend to agree with Hayek in "Monetary Matters and Trade Cycles" that fractional reserves are a natural outgrowth of free banking and we can't do anything about it. However, as Hayek wrote they are the chief instigator of the cycle because of their extremely high leverage.

DWAnderson writes:

The example misstates the results absent deposit insurance if everyone wants their honey back.

Instead the results would be that the bank enters bankruptcy proceedings and depositors would have their debt (deposits) turned into equity in the bank, with the old equity holders being wiped out. The despositors would then collectively own all of the loans that their deposits had funded. They would certainly suffer a loss of liquidity (which they might effectively pay a premium to recover by selling their new stock for less than the PV of the loans owned by the bank), but not necessarily any other loss of value.

This is the contractual system in place for money market funds, into which people freely deposited trillions of dollars without coercion or a government guaranty.

It does not seem inherently implausible that they same system could work for banking if people were willing to accept liquidity risk when there was a run. Indeed it is likely that a variety to contractual mechanisms might be put in place to mitigate that risk, e.g. a five day waiting period for certain types of acocunts or amounts.

The system I have described is fractional reserve banking and there is certainly no swindle involved.

It is certainly true, however that the system of deposit insurance and regulation has crowded out these alternatives in the US banking system.

falcon writes:

""But if lots of people wake up hungry at once, the higher penalties will kick in, because the bank's reserves start to drop.""

As the reserves start to drop, I'd imagine the bank would start offering up a nice interest rate to new depositors in order to replenish their safes. The jars of honey in the non-interest bearing accounts would be enticed to move some of their honey into interest bearing accounts - injecting new honey to be directed to the biggest and best heffalumps.

I'm starting to get a headache, but as long as there are huge heffalumps to be trapped, the penalty rate should allow for the existing traps to trap, and the new interest rate should allow great traps that should be built to still be built.


Derek Christiansen writes:

Gold is not money.

Money is a physical manifestation of the psyche (ideas, emotions, visions of the future, etc.)

Gold is now obtained by slicing pieces of mountains with bulldozers and then soaking the resultant dirt with potassium cyanide. Higher price of gold, more marginal mountainsides sliced with 'dozers.

That a healthy % of people see gold as money does not make it money.

Recent prices of gold reflect actual rarity in the world with demand for use in electronics as
semiconductor interconnection material.

What Arnold speaks of is a world where we should have banks that only handle "real" money.

Essentially we should only have one bank and then there could be no run.

ChrisA writes:

Good analogy and no headache. It is all sort of making sense for me now. I am still shaking my head over the realisation of the conflict between depositors and banks. Banks use their depositors money to speculate on all sorts of high risk ventures but the interest rate that banks pay on accounts just isn't high enough to justify any real risk of loss of principal. People only put their “safe” money into banks. But in reality a bank deposit is probably one of the riskiest investments you can make, I mean I personally would never make an investment that had leverage of 10 to 1; but in banking terms this is considered conservative! If I did make such an investment I would want huge returns to justify the risk, not 1 or 2%. Frequent failures are guaranteed with such leverage, (yes with a widely diversified portfolio this is usually not a problem except when failures are correlated, like today). Deposit insurance is only part of the reason that the current crazy structure is accepted, the whole regulatory apparatus gives the impression that the government is somehow keeping a check on things, we now know they weren't or were not capable.

It's no wonder that banking was so lucrative, bankers are getting the spread between 1 or 2% and the potential returns that 10:1 leverage should get. I remember a discussion on this blog last year on why investment bankers were so well paid, well it turned out to be organised rent seeking not value creation. Basically there was a bunch of rent being shared, the rent being the spread between government insured deposits and what a borrower had to pay. Obviously the participants wanted to share that rent among as few players as possible, which is why each one of them were so well paid. Unfortunately as every generation of bankers cashed out, the next generation had to leverage up some more to get the next chunk of rent, with the results that we see today.

My hope is that, as people begin to realise how misaligned the incentives are in banking, something will happen to banking just like happened to the mutual fund industry which has (partly) segregated into cheap internet brokerage accounts, ETFs and hedge funds, with each part of the industry offering different service/risk profile. Banks should segment into the different functions of banking; safe money storage accounts (with access to ATMs) and investment type accounts, perhaps with returns linked to the underlying assets. You can then divide up your funds according to your risk profile. Regulation (deposit insurance) was the reason this mess was able to keep going so long, so it will have to be a part of the solution. For the safe accounts government should create the ability for individuals to deposit money at the FED paying interest according the current FED rate (effectively they already do this with savings bonds). Principal is always returned. Banks could compete to provide ATM and checking services for these deposits for a share of the interest paid by the FED. There would then be no need for any federal deposit insurance system. The “private” credit industry would then have to rely on equity like deposits, perhaps with variable interest rates and/or clear risk of principal. Of course the interest rate would have to be much higher than today’s deposit system, but the reduced spread should come from bankers fees, not from the borrowers, since in this new system bankers would actually have to compete for deposits, and not rely on the deposit insurance system to funnel free money to them.

Gu Si Fang writes:

"Do I make the deposit? The bank says that it has the law of large numbers going for it. The chances are really small that lots of depositors will wake up hungry at the same time. There may be only a one in a million chance that I won't get my honey back. Maybe I'm willing to take that chance.

I'm not convinced that this sort of bank is what would naturally evolve. Instead, we might see a bank that offers two types of deposits. One would be honey deposits that bear no interest. You store your honey with the bank, and you can withdraw it at any time. It stays in the vault until you ask for it."

That's certainly correct. In addition, the slight differences between the two types of accounts would wider over time. If one is slightly more attractive than the other, a monetary-type self-reinforcing process would increase demand for this type of account.

What prevents this from happening? Essentially legal tender, which forces convertibility between the various types of monies (interest deposits, demand deposits, paper currency) and destroys the market incentives to chose between them.

DJH writes:

Banks demanding withdrawal penalties would be outcompeted by risk-taking banks willing to eschew penalties.

Even in a full disclosure environment where depositors understood the (miniscule) risk of ruin, many would accept the risk rather than accept penalties.

Comments for this entry have been closed
Return to top