Thursday, March 18, 2010

Reserves? We Don't Need No Stinkin' Reserves

EconomicDisconnect pointed me toward a story last night from the Economic Collapse Blog regarding Ben Bernanke's comments on the need (or lack thereof) for reserves in the fractional reserve banking system.

Fractional reserve banking mean that when you deposit $100 in the bank, the bank can lend out, say, roughly 90% of that money - they may keep 10% as reserves.  MikeyMortgage takes a loan from the bank ($90 of your $100) and buys a house with it.  The home seller, HarryHomeseller, receives MikeyMortgage's $90, and deposits it in the bank, who then lends out roughly 90% of it again, this time to CarlCarbuyer, who borrows $81 of HarryHomeseller's $90, etc etc etc. 

Some people scream and yell about how fractional reserve banking is a scam because if everyone wants their money back at once, the bank doesn't have it.  That's not quite true, at least in theory - if you want your initial $100 back, the bank doesn't have it in the above example - they've lent it out, but they do theoretically have assets that they can sell that are WORTH $100.  Assuming the bank has made good loans, they can sell MikeyMortgage's mortgage note and get the money to pay you back.

This brings us to the next point - it makes perfect sense to have some reserves.  First, as we've proven over the past few years, the value of this collateral (the loans) can drop - we need some cushion.  Second, if you have reserves on hand, you have cash to pay back to depositors who want their money back without necessitating that you liquidate assets.   Which brings us to the Economic Collapse Blog:

"Up until now, the United States has operated under a "fractional reserve" banking system.  Banks have always been required to keep a small fraction of the money deposited with them for a reserve, but were allowed to loan out the rest.  But now it turns out that Federal Reserve Chairman Ben Bernanke wants to completely eliminate minimum reserve requirements, which he says "impose costs and distortions on the banking system". At least that is what a footnote to his testimony before the U.S. House of Representatives Committee on Financial Services on February 10th says. So is Bernanke actually proposing that banks should be allowed to have no reserves at all?

That simply does not make any sense. But it is right there in black and white on the Federal Reserve's own website....

The Federal Reserve believes it is possible that, ultimately, its operating framework will allow the elimination of minimum reserve requirements, which impose costs and distortions on the banking system."

Now, it's been brought to my attention that there are several countries who currently operate banking systems without reserve requirements.  The real question is, is Bernanke suggesting that we don't need reserve requirements because in a rational free markets world, banks would hold adequate reserves anyway, and thus don't need more expensive restrictions imposed on them?  I think that claim can be easily refuted by saying "See:  U.S. Banking System 2007-2009."   The alternative, then, is that Bernanke really believes that banks don't need to hold reserves at all.  
Ben Bernanke's expertise in banking is greater than mine.  That much is given - although I wasn't riding shotgun next to Alan Greenspan as the metaphorical car that is the U.S. Financial System was driven off a cliff.  However, I literally do not understand how zero-reserve fractional reserve banking can work.  If, in reality, banks lend out every dollar of deposits, with no allowance for depositor redemptions or decline in collateral (read: LOAN) value, isn't the system guaranteed to fail?

Anyone care to explain this?  Anyone?  I'm looking for someone who can explain to me the rationale, even in theory, for how banks could operate with no reserves.  My specific questions:  most importantly:  if you have no reserves, and the price of your assets (loans you have made) declines, aren't you instantly insolvent?  Also, if you have no reserves, how do you handle depositor requests for redemptions?  Is the answer simply that the Federal Reserve is there to backstop insolvencies arising from these two situations?  That's not really an answer.  Maybe the answer is that the Fed backstops "temporary" insolvencies until they can recover and right themselves - until the value of the assets "comes back."  Extend and pretend!  What if the value doesn't come back though? 



Anonymous said...

Super easy: banks borrow reserves from the interbank market if they need. And FED lends them reserves as part of OMO and interest rate policy

Banks do not need reserves to lend.

Kid Dynamite said...

yes anon - but how about to remain SOLVENT? how do they magically repay these loans they have taken out in the interbank market or from the Fed if the value of their assets (loans they have made) declines?

Anonymous said...

That is a different question.

It is the domain of banks" equity, proper accounting/valuation and FDIC.

But banks are indiferent with respect to reserve requirements and the whole money multiplier concept is phantasy

Paulina said...

I am just an ordinary person with significant life insurance from my husband (God rest his soul) in about five banks, just in .5% accounts. I am trying to decide if this new Fed idea can just be interpreted as, "Give your money to the banks so they can blow it and give bonuses." Thank you for reporting this.

I remain more freaked about how to survive and raise my kids with these funds than ever. Your site is great. Love the dog and the syrup.

Taylor said...

This is the way I have to think about things when the gov. is involved...give a private sector like example. A casino's cage is basically a bank. Yes, they don't make mortgage loans and such, but they do lend money in the sense they take deposits and give you something of value in return. I realize they are under gaming laws that require them to hold a % of money played on hand, but what if they weren't? What if those laws were repealed? Would they hold no money on hand? No, they wouldn't because they would be responsible to payout when their "loans" were called. No gov intervention or bailout, just a bad name and reputation when they couldn't pay up and less business in the future because of it. What's the difference then between them and and bank? No gov. backstop. The banks would exploit this and run head first as fast as they could to the next crisis and then not hope, but expect a bailout, whereas a private enterprise with no history or expectations of gov. intervention would still maintain prudent practices.

Kid Dynamite said...

anon - i guess my point is this: i see that "practically" the Bank of KD can open up with no money. Johnny comes to me and wants a loan. I evaluate his ability to repay, and decide to lend him money, while simultaneously going to the Fed or the InterbankMarket and saying "hey guys - Johnny wants to borrow money - lend me some so that i can lend it to him, we'll use his mortgage as collateral."

the problem is that if i'm wrong about Johnny's ability to repay, i now can't make good on the loan i owe the Fed.

if this sounds eerily familiar to what just happened in our markets across a variety of asset classes (home loans, private equity, CDOs, MBS), that's intentional...

more reserves = more cushion against being wrong and borrower default.

ok - so maybe my previous comment question is a different question - but the question i care about is "by definition, don't lower (or no) reserves equate to a higher risk of insolvency" ???

Kid Dynamite said...

taylor - casinos are required to hold 100% reserves - they must have cash on hand to cover every outstanding gaming chip.

as i addressed in the post, the banking industry has PROVEN already that free market common sense solutions like the one you outlined won't work for them - they made loans until they were insolvent. it already happened!

DECREASING required reserves cannot make them MORE likely to stay solvent! right? it would have to make them more likely to become INSOLVENT

Kid Dynamite said...

paulina - it sucks for people living on fixed income that interest rates are so low right now, but no - i don't think that the goal of this idea would be so that the banks can steal your money and give it as bonuses at all.

the goal of low interest rates (and lower reserves) is to facilitate lending, borrowing and spending.

I am not a fan of ponzi finance, though, as should be well evident to my readers. I think that lower reserve requirements is quite similar to many characteristics of ponzi finance

Anonymous said...

no Kid, lower reserve requirements have nothing to do with default risk of your borrowers. These are completely independent processes.

Reserve requirements change the profitability of lending as often (like in the US now) but not always (as can be in the US in the near future) required reserves do not earn interest for banks.

But even if you were able to dig any correlation out of defaults and reserves it would be absolutely spurious

Anonymous said...

Kid, your cushion against borrower default is your equity.

Full stop here.

Do not confuse matters

Kid Dynamite said...

anon - i certainly didn't mean to imply any relationship between the reserves held by the banks and the default rates of borrowers - there is no reason they should be related.

what i meant is, if i, the bank, am required to hold more reserves, then, BY DEFINITION, i am more solvent - i can tolerate more borrower defaults before i am insolvent. right?

Anonymous said...

Kid, your cushion against borrower default is your equity.

Full stop here.

Do not confuse matters

Anonymous said...

Kid, your cushion against borrower default is your equity.

Full stop here.

Do not confuse matters

Kid Dynamite said...

sorry anon - i don't know why your comments have suddenly stopped showing up - hopefully they come back eventually...i am getting them in my email though, so you and I are having a conversation here where my comments are visible and yours are not.

you wrote "your cushion against borrower default is your equity."

true. but even before that point - don't reserves factor in? again: if i have a $100 deposit, and a $100 loan against it, every dollar decline in the loan value eats into my equity. If i have only a $90 loan against it, i have that extra ten bucks as "Cushion"

but WAIT - your point must be that, "NO - you do NOT have that extra $10 as cushion - because you owe that to the depositor - you still owe him the entire $100 when he wants it back"... right?

Kid Dynamite said...

anon - so you think i'm confusing reserve requirements with equity - right? i indeed may be...

Kid Dynamite said...

anon - let me try a different approach. if i have lower reserve requirements, i can make more loans, right? by definition, if i make more loans, don't I have a higher probability of having loans that default? and thus, don't i have a higher probability of having my equity impaired?

the hope, obviously, is that the profits on the loans that don't default make up for the extra loans that do...

the concepts are expected value and variance - with lower reserve requirements, don't we have higher variance (even if it's accompanied by higher expected value?)

Anonymous said...

yes, I do not know what happened to comments. Seems like a google issue (so I stop spamming you trying to repost and see them appear :) sorry for that

Anyways, the fact that banks can borrow reserves from the FED means that reserve requirements by definition are not capable of eliminating credit risk from the system as a whole.

So now lets get down to a single bank. Bank can borrow reserves in the _interbank_ market should it need to (fed funds, ie. transactions within the FED). If market does not want to lend this bank and FED does not want to lend to the market because interest rates are fine, then it seems like we have a problem.

When can a bank have problems borrowing short-term money? Only when it is clearly bust.

Will the previously made reserves help the bank in this case? Not a single moment. Go and try to tell other banks that you have 1bn of reserves at FED and just need 10k of cash to cover for losses. I am curious about faces you will get

Anonymous said...

Kid, reserve requirements are irrelevant for you ability to make loans. They only influence your margin. Banks can always borrow reserves unless they are already bust.

And yes, please make sure you distinguish reserve requirements from loss provisions which sometimes somewhere might pop up as "reserves". These are fundamentally different stories

Kid Dynamite said...

anon wrote "Will the previously made reserves help the bank in this case? Not a single moment. Go and try to tell other banks that you have 1bn of reserves at FED and just need 10k of cash to cover for losses. I am curious about faces you will get "

i guess my point is that without the reserve requirements, my $10k of losses would be $100k of losses because i would have made many more loans... right?

reserve requirements act as a brake pedal on the banking industry, right? in light of everything that we've learned in the past few years, i think we need MORE brakes, not fewer brakes...

Kid Dynamite said...

"reserve requirements are irrelevant for you ability to make loans" really?!?!? that seems impossible. perhaps i need to clarify. don't lower reserve requirements allow me to lend out more NOTIONAL?

if i have no reserve requirements, i can lend the same money out over and over again in full. Joe borrows it, buys a house from jane. Jane deposits it in my bank, and then i lend it out (in full!) to Jim. Jim spends it on a car he buys from Jack, and Jack deposits it in my bank again - again, in full... I keep lending out the full amount.

With reserve requirements, i'm lending out less each time. no?

Anonymous said...

:) Nope

Again, you need to understand that banks are not constrained by their deposits and reserves when they make loans.

Think of it this way. You come to your bank (where you have your account) and ask for a loan of 100k. Your bank approves and gives you 100k, ie. puts 100k into your current account (deposit). So now your bank has to go to the market and borrow 10k and put them into FED as requried reserves.

This is how life works. If your account is in a different bank then the issue gets a bit more complex but it is not important for the general idea

Kid Dynamite said...

anon - but if my bank has to go and borrow $10k and put it into reserves at the Fed, that $10k is a brake on the system!!! it's $10k that the "market" can't lend to someone else!

Kid Dynamite said...

ps - Anon - so my bank puts "$100k" into my account. now i withdraw it... what happens? maybe i just want to keep it under my mattress - i don't use it to buy a house or anything (i think this will be simpler)...

my bank goes to the Fed and says "I need $100k" ??? here's a promissory note from KidDynamite that he'll repay it?

something like that, right? of course, the Fed has $10k in reserves from me already... which is absolutely better than them having ZERO in reserves from me - from a risk management perspective, right? How can it not be?

Anonymous said...

Again no.

If demand for fed funds (reserves) is higher than current suply then interest rates will rise and force FED to intervene and provide more reserves through OMO. As long as you make sound loans and have good collateral and capital you will _always_ be able to get reserves you need.

However quality of collateral and capital is much more dependent on sentiment then any regulation or rules. Especially if collateral was kind of shaky in the first place. So make good loans and sleep well. Or regulate banks so that they do not do any subprime garbage.

I believe in the latter approach as "free" markets ideology has clearly failed. But reserves requirements is not solution

Anonymous said...

yes, this is what banks do. They go out and fund themselves based on maturity profile of their assets. Or not if they want to take interest rate risk

Funding department has no phone lines to loans department. These are independent functions. Moreover when any bank makes plans for the future they always talk about assets growth. However in the (conventional) logic of money multiplier they should be talking about deposits. This is not about the life outside and therefore (conventional) logic-assumption of money multiplier is false

Kid Dynamite said...

anon: I think your 11:15am comment really nails the problem. "If demand for fed funds (reserves) is higher than current suply then interest rates will rise and force FED to intervene and provide more reserves through OMO. As long as you make sound loans and have good collateral and capital you will _always_ be able to get reserves you need. "

that is exactly the problem - the Fed is now taking crap collateral. The assumption that we can just "have good collateral" is one that has ALREADY been proven false! in the theoretical world, as long as everyone makes loans that don't go bad, i understand what you are saying. in the real world, we've proved that loans go bad - and at rates well beyond what the Fed, any other regulator, or any lender can predict.

you then wrote:

"However quality of collateral and capital is much more dependent on sentiment then any regulation or rules"

that statement is certainly debatable. I won't go so far as to say it's outright false, but i think it may explains some of our differences in opinion here (or, my confusion at your responses)

anyway - i thank you for your attempts to educate me and others here. I'm going to move on with my day, as I think this is beginning to get into a semantic discussion and is getting away from reality.

EconomicDisconnect said...

Well I am at work and thus cannot be involved here to much degree but the conversation to me seems to boil down to theorhetical debates vs reality. In reality the banks have abused fractional lending and made poor loans that are so bad it has toppled the pyramid, which in theory could not be toppled. That seems to be Anon's point, nothing KD (or I) says about solvency has any meaning because it cannot happen, well except that it just did.

Furthermore, I thought getting loans from the FED was bad and had a stigma? Seems now the new theory is ultra low non-market based loans from the FED forvever. This is quite possible but I think though forevevr may not be so long.

Anonymous said...


and I also was a bit tough on that statement. I should have rather said "quality of collateral and capital is on sentiment (full point)"

EconomicDisconnect said...

Wow, just posted my comment and KD's is on the same page.

Anonymous said...


FED's OMO operations are not loans and have no stigma

Kid Dynamite said...

i also thing GYC's comment just now nails it.

i thank you all for your intelligent comments, and encourage continued discussion - but I must go figure out how to fertilize my lawn and prepare my garden, rather than sitting here inside with smoke coming out my ears pondering the theoretical viability of non-reserve banking.

Anonymous said...

final comment.

just by chance came across a good summery of the question at hand

Itamar Turner-Trauring said... has a relevant discussion:

Two hypotheses about the nature of money can be derived from the money multiplier model:

1. The creation of credit money should happen after the creation of government money. In the model, the banking system can’t create credit until it receives new deposits from the public (that in turn originate from the government) and therefore finds itself with excess reserves that it can lend out. Since the lending, depositing and relending process takes time, there should be a substantial time lag between an injection of new government-created money and the growth of credit money.


Testing the first hypothesis takes some sophisticated data analysis, which was done by two leading neoclassical economists in 1990.[3] If the hypothesis were true, changes in M0 should precede changes in M2. ...

Their empirical conclusion was just the opposite: rather than fiat money being created first and credit money following with a lag, the sequence was reversed: credit money was created first, and fiat money was then created about a year later...

Anonymous said...

Canada pulls it off, but partly it's due to the small number of bank participants here, which then clear all interbank transactions through the BoC.

Here's a decent paper on how it's done up north:

Anonymous has it right, from what I've read. Default risk due to loans is different from reserve requirements. Also, collateral held at the BoC will have a haircut to deal w/ some of the problems you see.

Part of the current crisis (I would say a small part), is the creative accounting used by banks for reserve avoidance. The loss of transparency makes large banks (and small) more difficult to monitor and manage, possibly exacerbating the current crisis. The argument is that by making it simpler, it's easier for the central bank to monitor. Is this true? Hard to say for sure. Depends on whether you think the central bank is more effective at decoding accounting tricks, or managing internal accounts.


Anonymous said...

A slightly different take may be that the eventual plan is to eliminate reserve requirements as they currently exist, a fixed percentage of deposits, and control reserves through the discount rate, raising and lowering it to give a more granular control over reserves. Maybe that explains the surprise discount rate hike a few weeks ago, it was a test to see the impact on reserves held at the fed. Of course if the feds MBS assets begin to fall apart it begs the question where they get the money to pay the interest.


scharfy said...

Great Debate.

It is true that "reserve requirements" are always available at some price via the FED. Thus they only indirectly affect lending by virtue of their price relative to creditworthiness of the borrowers.

Banks never run out of money to lend. If they deem a borrower creditworthy, they make the loan of X dollars, and borrow 1/10X from the Fed at a given rate and call it reserves. In this manner, the PRICE of FED funds rate constrains and enables credit creation. (as well as equity - which is typically the stock price, plus liquid securities - that puts a ceiling on loans via Tier 1 capital ratios.)

But not the raw amount of deposits or reserves. Ever. They don't loan our money. They create it.

Thus, removing reserve requirements doesn't do much, given that "reserves" are actually borrowed funds from FED. Bernanke isn't loosening lending standards at all, he is just removing one link of the chain, as it it is unnecessary paperwork which doesn't constrain the Ponzi nature one bit.

Aaron Krowne said...

Sounds like banks would only be restrained by being "held" to Tier-1 capital requirements, per GAAP, Basel-II, etc.

This would not be a huge change from what happens in practice; but it is another step towards a "no real reserves" banking system. And it definitely is a case of our financial regulatory authorities shirking one of their core expected duties.

scharfy said...

Yes, having the Fed set money supply via the banks, and also being the regulator of such banks, while also being responsible for full employment and price stability - would make Stalin very proud. Central Planning baby.

You can feel the air sucking out of the asset bubble, while the FED flails helplessly to reinflate.

As per policy, they have the loan stimulator on full tilt, but credit (effective money supply) is collapsing. Hence less output, hence more defaults, and so on.

This in my view needs to happen, but they won't allow the reset button to be hit.

EconomicDisconnect said...

Good stuff everyone, and a good debate.

My only issue here is the assumption that the FED is now the permanent lending arm for the banking system from here to eternity and all that implies. If the system can only work at 0% rates and taxpayer backing for losses, how awesome is the system?

david said...

If you go to

you'll see two countries at the top of the list without reserve requirements - canada and australia. Neither was hurt during the crisis.

Btw, as John Hussman points out, the US in reality has no reserve requirement. He has some good discussions of the mechanism on his website, eg:

Anonymous said...

yes, change the system and make banks suffer a bit as well

Anonymous said...

Todays Keynesian clown bankers and their lucy goosey lending is exactly what got us where we are in todays mess. It would be insane to think these certifiable nutjobs can get us out of the mess that they got us into with more of the same.

Why, our elected officials, and these very same keynesian clown nut cases cannot and will not see this obvious conclusion is without a doubt very troubling indeed.

Personally, I can't wait for the trials and the clawbacks to begin;) It will most glorious.

Kid Dynamite said...

looking back, i think i thoroughly confused myself in this thread. i spent a lot of time talking about reserves against loans...

but the reserves we're talking about are against DEPOSITS.... the opposite...