100-percent reserve banking and state banks

Today we examine two propositions: (a) 100-percent reserve banking; and (b) national government spending without taxation and debt issuance. Believe it or not the two propositions have been related in the debates over many years. Modern monetary theory (MMT) is agnostic to the first proposition although individuals within the paradigm have diverging views. However in the case of the second proposition it is central to MMT that a currency-issuing government has no revenue-constraint and should not issue debt to match net spending. Further, taxation is an effective tool for attenuating overall agggregate demand rather than raising revenue for a government that can spend regardless.

Before reading this blog I suggest you consult some earlier blogs – Operational design arising from modern monetary theory Asset bubbles and the conduct of banksFunctional finance and modern monetary theoryBreaking up the banks – as background material.

The discussion on my blog in recent days has, in part, focused on what to do with banks and specifically whether we should have a Full-reserve banking system which is described as a banking system:

… in which the full amount of each depositor’s funds are available in reserve (as cash or other highly liquid assets) …. [and] … when each depositor had the legal right to withdraw them.

Full-reserve banking is a.k.a a 100-percent reserve banking system or a Giro system (the latter being from the old days of the Bank of Amsterdam). The Postbank in the Netherlands still runs a Giro bank!

A related proposal was advanced by James Tobin in 1985 (20-21) where he proposed a new “deposited currency” which would be payable to anyone on demand by either the central bank or the commercial banks. In the latter case, the deposit-taking banks would have to buy government bonds with the deposits before they could issue “deposited currency”. This would eliminate the need for reserves and deposit insurance.

It is juxtaposed with the so-called fractional reserve system where bank loans create deposits which are not backed 100 per cent by reserves.

This debate overlaps with an arcane and long-standing debate within the even more arcane Austrian school of economics about who is right – the “100 percenters” or the “Fractional Reservers”. Austrians all think they support freedom but define this to mean absence of coercion (principally by government).

In this context, the “100 percenters” consider – in their representation – that a fractional reserve system is fraudulent because it promises to pay amounts in excess to what actually exist and that fraud is coercion. So the fractional reserve system is the anathema of freedom.

Bring on the “Fractional Reservers” who say that freedom means doing what you like and so why stop banks and borrowers doing what they like. Its all a free exchange after all! They claim that people will understand that some banks are less likely to remain solvent than others and the market will take care of that.

Anyway, I am happy for these characters to spend their time debating among themselves – sort of running around in circles chasing their tails as they get increasingly agitated about which scheme best promotes freedom. This way they reduce the time they have left to bother the rest of us.

You get a feeling for the emotions that the Austrians hold in this regard, when you read the work of noted (misguided) US libertarian, the late Murray Rothbard. In the October 1995 edition of the Austrian rag, The Freeman – he called fractional reserve banking “a gigantic scam” (see Reprint):

… the idea, which most depositors believe, that their money is down at the bank, ready to be redeemed in cash at any time. If Jim has a checking account of $1,000 at a local bank, Jim knows that this is a “demand deposit,” that is, that the bank pledges to pay him $1,000 in cash, on demand, anytime he wishes to “get his money out.” Naturally, the Jims of this world are convinced that their money is safely there, in the bank, for them to take out at any time. Hence, they think of their checking account as equivalent to a warehouse receipt. If they put a chair in a warehouse before going on a trip, they expect to get the chair back whenever they present the receipt. Unfortunately, while banks depend on the warehouse analogy, the depositors are systematically deluded. Their money ain’t there.

The old “Giro” or “100 percent reserve” banking system operated by people depositing “specie” (gold or silver) which then gave them access to bank notes issued up to the value of the assets deposited. Bank notes were then issued in a fixed rate against the specie and so the money supply could not increase without new specie being discovered.

Rothbard juxtaposes this against the “swindling or counterfeiting” system which “is dignified by the term “fractional-reserve banking”” – whereby banks now create deposits out of thin air – “which means that bank deposits are backed by only a small fraction of the cash they promise to have at hand and redeem.”

The Austrians then claim that governments are parties to the “inflationary counterfeit-pyramiding” because in a free market the banks operating under “fractional reserve banking” would go broke virtually immediately as soon as other banks demanded cash from a bank that had extended a loan with money they didn’t have.

The only way this “embezzlement” can persist is because government supports the cosy cartel via the creation of a central bank because:

Banks keep checking deposits at the Fed and these deposits constitute their reserves, on which they can and do pyramid ten times the amount in checkbook money.

He then describes the process of an open market transactions (central bank buying government bonds from banks in return for a government cheque) as the “beginning of the inflationary, counterfeiting process”. The rest of his rant is an exposition of the familiar money multiplier process which means he is just perpetuating the mainstream myth about how the banking system works.

Please read my blog – Money multiplier and other myths – to see why Rothbard fails to grasp the essence of modern banking.

Anyway, Rothbard says:

Thus, the Federal Reserve and other central banking systems act as giant government creators and enforcers of a banking cartel; the Fed bails out banks in trouble, and it centralizes and coordinates the banking system so that all the banks … can inflate together. Under free banking, one bank expanding beyond its fellows was in danger of imminent bankruptcy. Now, under the Fed, all banks can expand together and proportionately.

He then attacks the “lie of “bank deposit insurance”” as an “arrant hoax” because it just backed by the US federal government rather than any pre-existing “insurance fund”:

Suppose that, tomorrow, the American public suddenly became aware of the banking swindle, and went to the banks tomorrow morning, and, in unison, demanded cash. What would happen? The banks would be instantly insolvent, since they could only muster 10 percent of the cash they owe their befuddled customers. Neither would the enormous tax increase needed to bail everyone out be at all palatable. No: the only thing the Fed could do, and this would be in their power, would be to print enough money to pay off all the bank depositors. Unfortunately, in the present state of the banking system, the result would be an immediate plunge into the horrors of hyperinflation.

So very emotional language, which is a common characteristic of the “sound money” (Austrian libertarian) gang. They seem to get so het-up whenever they are talking about economics that you get the impression they think the sky is about to fall in on them any time soon. Even the private militias they support won’t help them if the sky ever does fall in.

But all this freedom nonsense really amuses me. I know I have not grown up in the US where the term is bandied around relentlessly. But the free marketeers never consider the coercion of the market when you are poor and starving and the rest of it.

More importantly, it is also a pity that their emotional positions are not backed by an understanding of how the monetary system actually works. They hark back to the gold standard which placed governments in a financial straitjacket and prevented them from reacting effectively to crises of the sort that has nearly crippled the world economy over the last few years.

Their depiction of the fractional reserve-money multiplier world exemplifies this misunderstanding of banking operations.

The idea that the monetary base (the sum of bank reserves and currency) leads to a change in the money supply via some multiple is not a valid representation of the way the monetary system operates even though it appears in all mainstream macroeconomics textbooks and is relentlessly rammed down the throats of unsuspecting economic students.

These students become functionally illiterate in the ways of the economy from day one of entering their studies and by the end of a typical undergraduate program are reduced to babbling meaningless trite one line statements like those that the Austrians regularly litter their writings with.

The money multiplier myth leads students to think that as the central bank can control the monetary base then it can control the money supply. Further, given that inflation is allegedly the result of the money supply growing too fast then the blame is sheeted home to the “government” (the central bank in this case).

The reality is that the reserve requirements that might be in place at any point in time do not provide the central bank with a capacity to control the money supply.

We have regularly traversed this point. In the world we live in, bank loans create deposits and are made without reference to the reserve positions of the banks. The bank then ensures its reserve positions are legally compliant as a separate process knowing that it can always get the reserves from the central bank.

The only way that the central bank can influence credit creation in this setting is via the price of the reserves it provides on demand to the commercial banks.

Even Rothbard’s depiction of open market operations as a central bank vehicle for increasing the money supply is erroneous. The argument is that the central bank purchases government bonds from private banks for “cash” which then expands via the money multiplier (the fractional reserve model).

However, what really happens when an open market purchase is made is that the central bank adds reserves to the banking system. This will drive the interest rate down if the new reserve position is above the minimum desired by the banks. If the central bank wants to maintain control of the interest rate then it has to eliminate any efforts by the commercial banks in the overnight interbank market to eliminate excess reserves.

One way it can do this is by selling bonds back to the banks. The same would work in reverse if it was to try to contract the money supply (a la money multiplier logic) by selling government bonds.

The point is that the central bank cannot control the money supply in this way (or any other way) except to price the reserves at a level that might temper bank lending.

The fact is that it is endogenous changes in the money supply (driven by bank credit creation) that lead to changes in the monetary base (as the central bank adds or subtracts reserves to ensure the “price” of reserves is maintained at its policy-desired lavel). Exactly the opposite to that depicted in the mainstream money multiplier model.

The other fact is that the money supply is endogenously generated by the horizontal credit (leveraging) activities conducted by banks, firms, investors etc – the central bank is not involved at this level of activity.

Please see the following blogs – Deficit spending 101 – Part 1Deficit spending 101 – Part 2Deficit spending 101 – Part 3 – for further information about all this.

Now back to 100-percent reserve banking proposals.

If you require all institutions to hold liquid reserves of equal value to their deposits then the fear of a bank run is eliminated. In the current recession, the run on Northern Rock in the UK would have been avoided. Banks runs are rare but disruptive.

The proponents also argue it would eliminate the need for the central bank to lend reserves to banks who are deficient at any point in time. So banks would always have enough “liquid” reserves to meet any need. The elimination of the central bank, of-course, satisfies the Austrians hatred of government.

The other claim is that a 100-percent reserve banking system would allow the government to control the money supply and hence maintain price stability. This is because all money would be created by the national government in this system. Banks would not be able to create deposits by extending loans.

So the idea is that the banks would have a 100 per cent reserve-deposit ratio which ensures that the money supply is 1-for-1 multiple of the monetary base.

Any firm that wanted to borrow funds would have to issue a convincing asset as collateral.

However, a hybrid model of the 100-percent reserve banking system recognises that financial intermediation is necessary for a smooth functioning production system. This means that many small savers can deposit with a bank who “spreads the risk” and on-lends to firms who need the funds for working capital.

The more reasonable advocates of full-reserve banking recognise that you need separate institutions that will take risk and provide credit.

So banks would still lend but only for each dollar of currency they hold (can raise). Depositers could formally forego their right to withdraw their funds for some fixed-term and in return they would receive interest. This would stop bank runs and allegedly ensure that at the end of the period the funds would be available to the depositer.

They claim also that with a controlled money supply, productivity growth would see prices fall and wealth increase. They allege that there would be no shortage of money to feed growth because less money would have to be used (given lower prices). It is a crazy argument.

There might even be a two-tier banking sector (this is similar to Tobin’s idea). Some banks would become deposit-taking institutions who hold government debt and manage the clearing house system (for the daily reconciliation of cheques between commercial banks).

But then the credit creation becomes the domain of other institutions to ensure that the real economy has access to working capital to facilitate production. To ensure financial stability, these institutions will either still need access to central bank discount facilities (overdrafts) to ensure the clearing house system doesn’t fail or they would be required to issue their own liabilities that were attractive to prospective lenders.

In this regard, regular commentator JKH noted that:

I’d always assumed that the (“theoretical”) answer was that overdrafts wouldn’t be permissible, since as you say allowing them would not accomplish anything – it effectively contradicts the purpose of 100 per cent reserves on demand money. Therefore I’ve assumed credit granting institutions would have to issue liabilities first in order to warehouse demand deposits (that are 100 per cent reserved by the deposit issuing bank), before extending new credit.

Yes this is the position taken, for example, by Henry C. Simons (Chicago School economist) who formulated the so-called Chicago Plan which you can read about in more detail HERE. Most of the Austrian School also supported it (particularly Ludwig von Mises).

It was expressed clearly by Milton Friedman in a famous presentation to the US House sub-committee (US House, 1975, 2156-57). Friedman said:

I have long believed that the most effective way to reduce regulation is to separate the monetary functions of the commercial banks from their credit conscience. The way to do this would be to require all institutions offering demand deposits to keep 100-percent reserves; make them depositary institutions in fact and not, as now, simply in name. Free entry into this industry could be permitted. The institutions could compete for customers financed by the interest received from the Government and by service charges to customers. The lending and investing activities of today’s commercial banks would be carried out by new institutions created by them which would raise their funds through time deposits or debentures or stock. These institutions would then be freed almost entirely from regulation. This is not a new proposal. It dates back over 40 years. It was supported in the 1930’s by Henry Simons at the University of Chicago. It was proposed in a book by the great Yale economist, Irving Fisher.

Friedman later changed his position somewhat but that is another story.

As an aside, some people have tied Minksy in with the 100-percent reserves camp. He was definitely not in that camp. There was an article in 1991 by Charles Whalen (Stabilizing the Unstable Economy: More on the Minsky-Simons Connection, Journal of Economic Issues, 25(3), 739-763) where this matter was considered in relation to the similarities and contrasts between Minsky and Henry Simon’s (a Chicago School economist who taught Minksy).

Whalen said that:

Simons’s banking reforms are more drastic than those proposed by Minsky. For example, Simons recommended the following changes: (a) “complete separation, between different classes of corporations, of the deposit and lending functions of existing deposit banks;” (b) “legislation requiring that all institutions which maintain deposit liabilities and/or provide checking facilities (or any substitute therefore) shall maintain reserves of 100 per cent in cash and deposits with the Federal Reserve banks;” and (c) “displacement by notes and deposits of the Reserve banks of [private-bank credit and] all other forms of currency in circulation, thus giving us a completely homogeneous national circulating medium” … Nonetheless, the banking recommendations of these men are linked by an important similarity – both desire a “radical decentralization” that would facilitate “new [bank and non-bank] enterprise” and the multiplication of “small and moderate-size firms”

However, Minksy did argue, as Whalen quotes (page 750), that the economic system:

… would be more stable if the banking system were decentralized, with many small and independent banks serving an industrial and commercial structure of small and medium-size firms … [and later] … a major necessary reform is for the Federal Reserve to shift from the open-market technique to discounting.

But Minsky stills see a crucial role for the central bank in maintaining financial stability. The Giro bank lobby sees no role for the central bank.

How does this fit into modern monetary theory (MMT)?

Note that the current practice is that loans create deposits. Clearly, under a 100-percent reserve system, all credit granting institutions would have to acquire the funds in advance of their lending.

There would be the equivalent of a gold standard imposed on private banking which could invoke harsh deflationary forces. Further, the 100-percent reserve banking system does not eliminate credit risk so that crises could still occur if there were significant defaults under the “fixed-term” variation.

State banking

What about state banking? This has overlaps with the debate about 100-percent reserve banking.

There has been some discussion in recent days in this blog about the proposals by Ellen Brown in her recent book the Web of Debt. For those who don’t wish to read this book you can see a good summary of the principle propositions in her December 5, 2008 article – A Radical Plan for Funding a New Deal.

The article poses the question (remember it was written as the crisis was intensifying in 2008):

How can the new President resolve these enormous funding challenges? Thomas Jefferson realized two centuries ago that there is a way to finance government without taxes or debt. Unfortunately, he came to that realization only after he had left the White House, and he was unable to put it into action. With any luck, Obama will discover this funding solution early in his upcoming term, before the country is declared bankrupt and abandoned by its creditors.

MMT tells you as a matter of first principles that taxes or debt do not fund government spending in a fiat currency system. The reality though is that national governments act as if they are still operating under a convertible currency (gold standard) where taxes and debt issuance were used to defend the gold reserves of the nation (that is, finance spending).

Brown notes in relation to the US that:

It had long been held to be the sovereign right of governments to create the national money supply, something the colonies had done successfully for a hundred years before the Revolution. So why did the new government hand over the money-creating power to private bankers merely “pretending to have money”? Why are we still, 200 years later, groveling before private banks that are admittedly bankrupt themselves? The answer may simply be that, then as now, legislators along with most other people have not understood how money creation works.

I agree with some of this sentiment. The fiat currency system frees the government from its gold standard revenue-constraint. But Brown is presenting an all or nothing scenario and hints at the 100-percent reserve banking approach.

I don’t think that a recognition that the national government is not revenue-constrained leads logically to the second proposition that private bank credit creation is undesirable. See more on this later.

I agree however with her view that “(n)ot only are banks merely pretending to have the money they lend to us, but today they are shamelessly demanding that we bail them out of their own imprudent gambling debts so they can continue to lend us money they don’t have.”

It is deeply disturbing that the banking sector has become so powerful in our modern economies. A properly regulated banking system would have prevented that. But that is not the same thing as abandoning private credit creation.

Brown then considers “3 ways to fund the “New” New Deal” in the US, which clearly has relevance to all sovereign governments.

She notes that a sound strategy would begin with the nationalisation of failed private banks instead of bailing them out:

Accumulating a network of publicly-owned banks would be a simple matter today. As banks became insolvent, instead of trying to bail them out, the government could just put them into bankruptcy and take them over … As in any corporate acquisition, business in the banks nationalized by the government could carry on as before. Not much would need to change beyond the names on the stock certificates. The banks would just be under new management. They could advance loans as accounting entries, just as they do now. The difference would be that interest on advances of credit, rather than going into private vaults for private profit, would go into the coffers of the government.

I definitely see the creation of public banks (or in Australia – the return to public banks – which were all privatised in the halcyon days of neo-liberalama!) as a way forward in a new banking system.

A public bank can discipline the cost structure of the private banks. For example, in Australia commercial banking is run as a cartel and they screw customers with all sorts of charges that having nothing to do with their own costs. A public bank with access to sovereign funds could offer very low charges.

The free market lobby will argue this is unfair competition – because the government has the backing of its currency issuing capacity. True enough but the same lobby are always reminding us that the government is about to go bankrupt. Moreover, if all the public bank was doing was restoring some balance between banks costs and consumer fees then that is addressing an externality arising from a non-competitive private sector market.

I also would not have bailed one financial institution out. Rather, as Brown notes, the government could have just taken over the operating concerns, repaired the capital bases and carried on offering services that advanced public purpose.

Within this context, Brown considers the “three ways government could fund itself without either going into debt to private lenders or taxing the people”:

(1) the federal government could set up its own federally-owned lending facility; (2) the states could set up state-owned lending facilities; or (3) the federal government could issue currency directly, to be spent into the economy on public projects. Viable precedent exists for each of these alternatives:

The first option would see a federal bank established to make credit available to infrastructure developments and the like. I can see advantages in a public bank helping strategic private sector investments (which promote public interest). But if the proposal is a way to overcome some perception that there are financial constraints on government spending then the idea is inapplicable to a modern monetary economy.

If it is to provide cheap credit to the non-government sector to advance public purpose then the idea is fine. It would also give control back to the government in determining the type and pattern of public infrastructure developments.

During the neo-liberal period, we saw the rise of the ludicrous public-private partnerships as the principle method of “financing” public infrastructure development. The reality was that the governments became so “debt-timid” that they allowed the private equity groups to determine how public infrastructure was provided and there are countless examples that have now emerged with demonstrate the pattern of development doesn’t serve public purpose at all. Sydney’s toll roads are a good example.

The second option is a creature of a federal system and allows states in the US to issue “low-interest credit on the fractional reserve model”. In Australia, state government-owned banks were common until they were privatised.

They provided service to the retail segment of the sector and until the neo-liberals started to deregulate the financial market (which pushed these banks into the wholesale segment – that sent them broke) these banks served a great purpose.

However, if this service was to provide finance to low-income house hunters then I think it would be far better for the public sector to provide the housing and eliminate the credit risk – which reached massive proportions in the sub-prime market. There is nothing wrong with the public ambition to house all its citizens. But there are good and bad ways of doing this. Fannie and Freddie of later days exemplified the bad way of doing it.

The final option of a government-issued currency is totally consistent with MMT.

Brown says:

A third option for creating a self-sustaining government would be for Congress to simply create the money it needs on a printing press or with accounting entries, then spend this money directly into the economy.

The US government already more or less does this. It just obfuscates that it is doing this by placing a series of voluntary constraints on itself – most notably its debt issuance program.

It would be much more efficient if the national governments around the world eliminated those operations and sent the officers who manage the debt-issuance off to do something productive – like caring for environment or sick and old people.

Brown notes that the “usual objection to that alternative is that it would be highly inflationary” but also says that “if the money were spent on productive endeavors that increased the supply of goods and services – public transportation, low-cost housing, alternative energy development and the like – supply and demand would rise together and price inflation would not result”.

So this is clear – if nominal demand grows at the same pace as productive capacity then inflation will not occur. The mad ravings of Rothbard that all money base and/or private credit creation is inflationary is a dogmatic assertion. It might be under circumstances where nominal demand growth outstrips real capacity to deliver new goods and services.

Regular commentator, Tom Hickey notes in this regard that:

All of the spending on public goods results in increased production of goods and services, some which are provided by the private sector, since the government would function mostly as a contractor hiring subcontractors. This would increase real output capacity, GDP, and national prosperity (by lowering the Gini coefficient). Automatic stabilizers could be used to stabilize … [nominal aggregate demand] … relative to real output capacity to maintain price stability by altering spending and taxation appropriately.

Apart from spelling things with “z” I agree with this summary of the inflation issue :-).

I note that some people claim that inflation can occur when there is high unemployment – so-called stagflation. It can but that comes about from cost pressures rather than demand pressures. In these cases, inflation targetting strategies are useless anyway.

Further, it is also true that specific asset classes can inflate before full employment is reached – the so-called asset price bubble. But an asset bubble is not inflation, which is defined as continuous increase in the general price level.

It is not sensible to address a bubble in a specific asset class (say housing) by deflating the whole economy. More targetted measures are required and fiscal policy is better served to deliver effective solutions in this context. Please read this blog – Asset bubbles and the conduct of banks – for more on that topic.

Where I stand

I do not support a 100-percent reserve banking system. It is the work of a lobby that hates and distrusts government. I have no objection to major reforms of the financial system as specified below but I also consider there is nothing intrinsically wrong with private credit. The private sector should be able to borrow and banks create credit under the strict conditions noted below.

The debt explosion that has brought the World economy to its knees was not the fault of endogenous money creation. It was the result of lax regulation; criminal activity; and a neo-liberal obsession that national governments had to run surpluses (and hence squeeze private sector liquidity and wealth).

With this obsession dominating public policy over the last few decades, economies could only grow (mostly) if the private sector took on increasing debt levels. The rise of the financial engineering sector – with the elimination of regulations that might have reasonably controlled its errant tendencies – guaranteed that the households would take on this debt … on increasingly dubious grounds.

My specific proposals for banking reform are dealt with in some detail in these blogs – Operational design arising from modern monetary theory Asset bubbles and the conduct of banks – – Breaking up the banks.

I start from the proposition that the only useful thing a bank should do is to facilitate a payments system and provide loans to credit-worthy customers. Attention should always be focused on what is a reasonable credit risk. In that regard, the banks:

  • should only be permitted to lend directly to borrowers. All loans would have to be shown and kept on their balance sheets. This would stop all third-party commission deals which might involve banks acting as “brokers” and on-selling loans or other financial assets for profit.
  • should not be allowed to accept any financial asset as collateral to support loans. The collateral should be the estimated value of the income stream on the asset for which the loan is being advanced. This will force banks to appraise the credit risk more fully.
  • should be prevented from having “off-balance sheet” assets, such as finance company arms which can evade regulation.
  • should never be allowed to trade in credit default insurance. This is related to whom should price risk.
  • should be restricted to the facilitation of loans and not engage in any other commercial activity.

So this is not a full-reserve system. The government can always dampen demand for credit by increasing the price of reserves and/or raising taxes/cutting spending.

The issue then is to examine what risk-taking behaviour is worth keeping as legal activity. I would ban all financial risk-taking behaviour that does not advance public purpose (which is most of it).

I would legislate against derivatives trading other than that which can be shown to be beneficial to the stability of the real economy.

So I support bank nationalisation? Probably, but that is for another day. Having a strong public banking system to compete against the private banks will achieve mostly the same ends and is more likely to be politically acceptable in the current climate.

That is enough for today!

Postscript

I forgot to mention in the main body of the post the following point. While I think MMT is agnostic to the concept of 100-percent reserve banking there are strong reasons why, for consistency, an MMT proponent would not want to support it.

A major proposition is that the banks will store the deposits they attract as risk-free government bonds. This allows them to earn a modest interest which helps them to attract depositors to their business and ensures they have liquidity on demand as per the logic of the proposal.

This, of-course presupposes that there are risk-free government bonds being available in quantities sufficient to support such a scheme. MMT shows the futility of issuing government debt. Further MMT leans towards a zero interest rate policy with fiscal policy then performing the counter-stabilisation (which it is a more effective tool than interest rate adjustments anyway).

So there is no need to issue to debt to manage bank reserves and thereby allow the central bank to maintain a non-zero interest rate in the fact of budget deficits.

In that sense, how will a 100-percent reserve banking scheme store the deposits safely?

This Post Has 64 Comments

  1. Bill, good post. I handy link to have to deflect the inevitable attacks by rabid libertarians that must be endured from time to time!

    I find it odd you do not much mention the issue of maturity transformation. I consider this rather vital to the functioning of a modern economy and although you do allude to it the skimpy treatment it gets here leads me to think you don’t consider it as one of the top problems with the full reserve camp.

  2. Hi Bill,

    Excellent! Have you considered backing up your blog in atleast 10 places ? Offline, solid state drives, I guess are the more safer ones, though flash drives may do the job for you.

  3. Hey Bill,

    I was wondering what your thoughts were on proposals for full-reserve banking for ecological reasons? The argument goes that fractional-reserve banking necessitates growth and thus has negative environmental (and social) consequences. Herman Daly, for instance, argues this. Other arguments can be found at Feasta’s website and as presented in the Money as Debt video. A lot of the concern has to do with the charging of interest as well. Is this too a misunderstanding of fractional-reserve banking?

    Also, I would love to see a post about your thoughts on alternative currencies.

  4. POSTSCRIPT Wednesday, January 13, 2009

    I have added an additional point to this post this morning about the stance MMT should take to 100-percent reserve banking.

    It is at the end of the main body of text.

    best wishes
    bill

  5. Bill, thanks for addressing full-reserve banking(FRB here).

    I stumbled on your blog some months ago through the UMKC connection, mainly through Dr. Hudson, and was immediately impressed with your grasp of how money works, and I have spent a great deal of time poring over the historical MMT writings. Being an all-my-adult-life monetary reformist, I have to admit there are several areas of non-understanding yet with MMT, and the question of full-reserve banking was a gnawing uncertainty.

    Thanks again for the lengthy presentation, but I remain unconvinced of the substance of your stance. Rather than deleting the BillyBlog link, I thought I’d give my opinion here. I don’t think you gave Murray Rothbard his due on the subject of FRB, and I am a progressive, people-first activist. But I leave the libertarians to defend Murray.

    While I can see the difficulty in matching up full-reserve banking with other aspects of MMT, I never saw it as a conflict, more of a rough fit that would require some re-thinking. Having read Minsky’s Foreword to Dr. Ronnie Phillips book on the Chicago Plan and New Deal Banking Reform, I also feel that your separation of Minsky from FRB leaves a lot unsaid.

    But what struck me as inadequate were your reasons for not supporting it, even conditionally and I know that my disagreement with your position stems from my agreement with the position of Simons, Douglas, Fisher and Soddy that the government of any sovereign nation should create ALL THE MONEY. So, to me, I was looking for the WHY NOT?

    And I didn’t see anything convincing in your concerns.
    Something about there being a need for liquidity or financial flexibility that the government can’t provide. But, never looking for a solution to whatever that problem is. Like I could only support FRB and government-issue of all the monies IF ……
    I say there needs to be a fully-founded public money supply capable of meeting all the needs of the economy and for whatever shortfall comes up, we can solve it.

    You decry the so-called progressive economists who line up behind the deficit hawks as both ignorant and immoral, and I agree. But to me it is as simple as an American to say that “Only the Congress shall create ALL THE MONEY”, and then let’s have it as to whether bank credits are money.

    I’m sorry but the tradition of private credit creation is not a family value in my family. If it is to be defended, it cannot be because we have always done it thus(in our lifetimes). And I don’t see it as a progressive value. But there we go again…

    Those who historically have supported the FRB option ALWAYS gave the rationale as monetary and economic stability – it naturally being counter-cyclical as to inflation and deflation. You can attempt to replace that stability through draconian regulation with some success, of course, only bringing about another deregulation cycle for the grandkids to deal with – as we are here.

    So, again, to me looking at FRB from a historic perspective not having anything to do with free-banking, I say that you have not really provided any substantive discussion as to why you are opposed to it. Fisher’s Chapter 10 proposal in Ritter’s Money and Economic Activity – Readings in Money and Banking makes a damned good case for FRB. I remain a supporter of this concept.

    But you still have the best economic blog going.
    Respectfully.

  6. “Brown says:

    A third option for creating a self-sustaining government would be for Congress to simply create the money it needs on a printing press or with accounting entries, then spend this money directly into the economy.

    The US government already more or less does this. It just obfuscates that it is doing this by placing a series of voluntary constraints on itself – most notably its debt issuance program.

    It would be much more efficient if the national governments around the world eliminated those operations and sent the officers who manage the debt-issuance off to do something productive – like caring for environment or sick and old people.

    Brown notes that the “usual objection to that alternative is that it would be highly inflationary” but also says that “if the money were spent on productive endeavors that increased the supply of goods and services – public transportation, low-cost housing, alternative energy development and the like – supply and demand would rise together and price inflation would not result”.

    So this is clear – if nominal demand grows at the same pace as productive capacity then inflation will not occur. The mad ravings of Rothbard that all money base and/or private credit creation is inflationary is a dogmatic assertion. It might be under circumstances where nominal demand growth outstrips real capacity to deliver new goods and services.”

    Wow, this is true utopian dreamland stuff.

    What possible basis do you have for believing that all money would be spent productively, and output would increase perfectly in line with spending?

    For example, if the Australian govt funded it’s AUD 300bn budget in this manner, instead of through taxation and borrowing, do you truly imagine output would increase 30%?

    The reality is that debt-funding of deficits is a mechanism that imposes a restraint which helps to ensure that government spending is kept productive.

    In your world, people would pay for government services indirectly through persistant, high levels of inflation, rather than directly through taxation. Taxation is open and explicit; inflation is a hidden and insidious.

  7. Dear lturner

    Wow, this is true utopian dreamland stuff.

    What possible basis do you have for believing that all money would be spent productively, and output would increase perfectly in line with spending?

    There is a difference between “spending productively” and spending to increase production. You seem to be confused about that. I also assume you mean “real” output here. Nominal output will increase $-for-$ in line with government net spending by definition. But the issue is what will happen to real spending.

    I would gather from your hostility that our definitions of what is “productive” will differ. For example, it is highly productive, when considered on an intertemporal (intergenerational) basis just to have adults working each day. The gains come from the learning that their children do watching their parents go to work. The empirical evidence is overwhelming that children who grow up in jobless households have inferior outcomes in their own adult lives – lower investment in education, lower incomes and the rest of the pathologies of disadvantage. You won’t find that concept of productivity and dynamic efficiency in one of the mainstream textbooks that your rhetoric seems to mimic.

    Further, I have never assumed (or said) that “output would increase perfectly in line with spending”. I actually think that real output will rise by a greater factor than the dollar injection of net public spending. The robust studies of the value of the expenditure multiplier, especially when there is so much excess capacity, points one to that conclusion. Have you studied that literature in detail? I doubt it from the tenor of your remarks.

    I also have never advocated zero taxation and you will not see that in the blog – so the statement that “if the Australian govt funded it’s AUD 300bn budget in this manner” is erroneous and a misrepresentation.

    In MMT, taxation has a very important function although that function has nothing to do with “funding” national government spending.

    You say:

    The reality is that debt-funding of deficits is a mechanism that imposes a restraint which helps to ensure that government spending is kept productive.

    That is your un-researched opinion. It doesn’t bear scrutiny. Can you outline the mechanisms whereby bond markets (and the auctions conducted by the AOFM) determine projects that are allegedly “productive” against those that are not?

    Can you show me once when the bond auctions have failed because the investors thought the government was “wasting money”?

    Given that the auctions never fail, perhaps by your own logic, you are thinking that all net public spending to date has been productive.

    The proposition that bond markets bear witness to the “productivity” of net government spending is completely far fetched. They do nothing of the sort. The proposition also raises the issue – again – of what we are going to call “productive”. However, that debate is never conducted within the realm of a government bond auction.

    So your “reality” is just your ideological prejudice – nothing more.

    Further, you said:

    For example, if the Australian govt funded it’s AUD 300bn budget in this manner, instead of through taxation and borrowing, do you truly imagine output would increase 30%?

    Finally, you once again misrepresent my position as advocating no taxation. The only statement I made was that net positions would not be matched by debt-issuance if I was running the show. Further, you do not understand the way in which the fiat currency system operates – taxes do not “fund anything”. Taxation merely takes purchasing power out of the hands of the non-government sector so they don’t spend as much.

    And … there is a logical gap here that you haven’t quite worked out – if there was no taxation then the federal budget would be very much closer to zero than the figure you state.

    best wishes
    bill

  8. >I’d always assumed that the (“theoretical”) answer was that overdrafts wouldn’t be permissible, since as you say allowing them would not accomplish anything – it effectively contradicts the purpose of 100 per cent reserves on demand money. Therefore I’ve assumed credit granting institutions would have to issue liabilities first in order to warehouse demand deposits (that are 100 per cent reserved by the deposit issuing bank), before extending new credit.<

    Hi Bill / JKH and all others,

    If I understand correctly, a 100% Reserve Banking system would effectively, if thought through, outlaw credit as we know it (as has been the case in previous centuries and as is still demanded by the Sharia e.g.)? As JKH says above, that would be the end of demand money because all credit money would – per definition – have to be in circulation prior to the lending act. While I assume that this could have a sobering effect on bankers' bonuses, I also presume the (remaining) credit system would either grind to a halt and/or just shift to illegal loan-sharking or what not (is that one of your objections?).

    My other question is, where would new money enter the system to offset deflation in a growing economy? The only mechanism I see would be through active government spending because the direct – and obviously sometimes corrupting – link between banks and the currency issuing agency would effectively have been capped. So is it true that the Austrians, and the other 100% RB proponents, are actually – if unwittingly – proposing massive, ongoing, fiscal interventions? And is this in any way coherent with their demands for a balanced budget? Or are they just too ideologically blinded to see the problem? The productivity argument you mention can't possibly account for things like population growth, can it?

    Or am I getting this all wrong?

    Regards, Oliver

  9. Bill

    Excellent post. I’ve never been clear on 100 per cent reserves. It’s always been a confusing, amorphous Rubik’s cube of risk considerations and institutional possibilities. Discussions are typically clouded on issues of liquidity risk, credit risk, “maturity transformation”, and macroeconomic money supply/inflation considerations. They’re often vague on the specification of the substance of the reserve and the domain of reservable deposits. Assuming a 100 per cent scenario, is the reserve substance gold? Is it government bonds? Is it current central bank reserves? Are checkable deposits reservable? Are all deposits reservable? What is the proposed institutional configuration? Is it a government system or part of the private banking system? Is it a two tier banking system, or a banking system with two tier banks (100 per cent reservable deposits and other liabilities)? Is there still credit created money? Is the motivation for 100 per cent reserves an erroneous interpretation of the multiplier? Does the “solution” to the “problem” pertain to the liquidity risk of depositor funds, or control of money supply/inflation, or both? The potential permutations and combinations seem endless.

    I think 100 per cent reserves is a very complicated subject, because it’s at the intersection of so many different issues. In that sense, it’s somewhat understandable that the official MMT position on it is agnostic.

    It seems to me that the top down issue is the question of the substance of the reserves, which in turn reflects institutional design for the location of deposits and reserves. If deposits are fully reservable in government bonds for example, then whatever the balance sheet configuration, the result is effectively a risk free segregation of credit and liquidity risk with respect to bank deposits. That compares to today’s mechanism, where risk management issues are dealt with by the combination of bank capital, FDIC deposit insurance, regulation, and self-imposed risk policy for a heterogeneously composed balance sheet. That’s supposed to be enough to result in prudent levels of deposit liquidity and safety without requiring 100 per cent reserves.

    There is a difference between the two systems, though. If 100 per cent reserves were held in government bonds or in existing central bank reserves as in today’s system, they would become an ongoing source of deficit financing, as per MMT interpretation with limited reserves (normally) as they exist today. But if the FDIC requires money to cover losses on deposits in excess of its own existing fund for this purpose, it can mean an increase in the deficit. From this standpoint, 100 per cent reserves are closer to an imposed form of “self-insurance” in terms of the immediacy of the reserves that are set aside. The question is whether the deficit “tail risk” is manageable by comparison, and whether or not there are other advantages to the existing system that are foregone in a 100 per cent reserve system. A reasonable answer is that the deficit tail risk is manageable provided that first loss private capital is set at a prudent level.

    The idea that one would specify government bonds as reserves rather than the present form of central bank reserves seems silly. They’re both fiat creations. If the reserve is government bonds, why shouldn’t it be any instrument of fiat creation, including central bank reserves in their existing form? An ideological hatred of central banks isn’t a particularly good reason per se to insist on government bonds rather than central bank reserves as they exist now.

    Conversely, if the argument from Rothbard et al is that 100 per cent reserves are the panacea for a banking system run wild on the “deposit multiplier”, then they don’t know what they’re asking for, because they don’t know what they’ve got. That the multiplier model is a fallacy is clear from MMT, which is a description of accounting reality, among other things. Moreover, there is a rich irony in the misunderstood belief in the deposit multiplier mechanism. Because the central bank supplies reserves in response to deposit growth, the specified reserve ratio has no direct effect on the pace of credit created money. I.e. the direct effect is the same, whether the reserve requirement is 10 per cent (US) or 0 per cent (Canada). This fact has allowed a number of countries to move to a zero reserve requirement. What this means is that a more accurate description of any supposed “multiplier” mechanism is that it is closer to infinite, at least in theory, because no reserves are required to expand deposits. The multiplier believers have missed out on being wrong in an even more spectacular way than with allegations of mere finite multiplier power.

    This irony compounds further as the multiplier believers attempt to solve the “multiplier problem” by recommending 100 per cent reserves. That’s because it is still possible to have 100 per cent reserves in a system of credit created money, provided that the institutional configuration allows it, and provided that institutional forces are allowed to control the pace of reservable deposit growth by issuing non reservable liability alternatives (NRLs) – e.g. qualifying term liabilities, perhaps including time deposits if they are deemed not reservable.

    Thus, it is possible to have a 100 per cent reserve requirement in a system of credit created money. However, the higher the specified reserve ratio, the more the overall balance sheet result depends on bounds attributable to macro level trends in customer selection of banking system liability mix. I.e. the bounds on reservable deposit expansion and reserve expansion in a 100 per cent reserve system will depend on the success with which banks can issue non reservable liabilities in order to arrest the pace of growth in credit created reservable money.

    This view may differ from your interpretation, Bill, because I think you impose some assumed institutional constraints on a 100 per cent reserve system – i.e. that banks presumably are not allowed to create money from the process of lending. I agree with your explanation, but I believe that institutional specification is a choice between two alternatives for money creation, as I shall explain further below.

    One of your points:

    “Note that the current practice is that loans create deposits. Clearly, under a 100-percent reserve system, all credit granting institutions would have to acquire the funds in advance of their lending.”

    That’s what I originally assumed, as per your quote of my comment. Thinking about it further, that may or may not be the case. Let me play devil’s advocate. As an extreme example of a credit created money system with 100 per cent reserves, consider the hypothetical transformation of the current US banking system to a 100 per cent reserve system:

    Here are some recent deposit numbers for the US banking system:

    $ Trillions

    Domestic checkable deposits .7
    Small time and savings deposits 4.9
    Large time deposits 1.8

    Total 7.4 trillion

    Now suppose the reserve regime is changed in a very aggressive fashion to stipulate that this entire $ 7.4 trillion deposit base requires 100 per cent reserves. Then assume the Fed supplies the newly required reserves by purchasing virtually all of the open market treasury debt.

    The immediate effect is that the pre-existing $ 7.4 trillion deposit base becomes 100 per cent reserved, but that the banking system also suddenly has an ADDITIONAL $ 7.4 trillion in what are INITIALLY the demand deposits created as the result of the Fed’s debt repurchase.

    Now assume the banking system issues new non reservable liabilities (NRLs) in exchange for those new demand deposit liabilities. The outcome of course depends on the willingness of the public to redeem its demand money for such NRL’s. But there would be a considerable propensity for them to do so at the right price, since the public has just sold its term treasury debt at what is presumably an acceptable price, and it would simply be a matter of pricing the terms on which it would replace the term structure it just lost with somewhat higher risk bank term NRL’s instead.

    The effect of such an aggressive expansion of both the scope of deposits subject to reserves and the required reserve percentage results in a massive increase in aggregate required reserves in this example. This requires a matching injection of actual reserves by the central bank. The central bank manufactures that injection by purchasing an equal amount of the government’s outstanding market debt. The net result is that the banking system becomes the new primary conduit for cumulative deficit financing, up to the level of required reserves. In this example, the initial required injection of reserves approaches if not exceeds the total level of outstanding market debt.

    Interestingly, the overall result of this banking system transformation is not dissimilar to would follow from the sudden implementation of the MMT idea of zero interest rates with elimination of debt issuance, which would also leave the money and reserve impact of cumulative deficit spending residing in the banking system.

    Although the $ 7.4 trillion example is extreme, it illustrates that the MMT observed function of credit created money can continue in this institutional structure, notwithstanding the fact that the banking system is 100 per cent reserved against the most broadly defined deposit base.

    Here is a micro explanation of the same dynamic:

    Consider a new loan of 100. It creates a new deposit of 100. The central bank then supplies new required reserves. In doing so, it also creates a second reservable deposit of 100. The temporary position is loans 100, reserves 100, deposits 200, and a newly created shortfall of 100 in reserves. This iterative process apparently has the potential to be endless, since reserve injections create new deposit liabilities that create more reserve requirements, with a 1:1 repetition. The commercial banks break this chain by issuing 100 in NRLs (non reservable liabilities). That extinguishes 100 in reservable deposits, and the system is in balance. The stability of the system would then depend on the ability of the banking system to issue NRLs at the margin, displacing new reservable deposits at the margin.

    At a macro level, the stability of the system would depend on the trend liability composition of the banking system, as between reservable deposits and non reservable liabilities. The mix in my example is $ 7.4 trillion each. The Ponzi type risk to which I referred would include the case where the public simply refused to pare back its initially transformed position of $ 7.4 trillion in new reservable deposits; i.e. it refused to redeem them for new bank NRLs. The constraint is that banks will seek to issue NRLs (provided that they are allowed institutionally) in order to “drain” reservable deposits and thereby arrest the dynamic process by which their money creation would otherwise cause a cascade of additional reserve requirements and additional injections of reserves and reservable deposit creation into the banking system. It is in theory a potentially infinite iteration of reserve requirements until arrested as a process by proactive bank liability management that also serves the interest of maintaining suitable liquidity and maturity transformation standards with respect to its credit portfolio and supporting liabilities.

    The example of $ 7.4 trillion in reservable deposits seems wild. It also approaches being problematic in terms of the natural supply of fiat backing. What would happen in theory if the specified reservable deposit base with 100 per cent reserves exceeded the cumulative budget deficit? Then the government would have to purchase non government assets in order to manufacture the required fiat backing. As it happens, the $ 7.4 trillion number wouldn’t necessarily be a problem, given the cumulative budget deficit and its prospects. The constraint is that the composition of liabilities must stabilize to some reasonable degree as between reservable deposits and non reservable instruments. Otherwise, there is a potential Ponzi like growth in deposits and reserves.

    However, this aggressive example would result in an important change in the risk profile of the financial system. The existing profile is one of $ 7.4 trillion in deposits, which are mostly insured by FDIC (or could be insured), plus an additional amount of outstanding treasury debt which currently just happens to be near the same order of magnitude as the $ 7.4 trillion in deposits. So the existing system is fairly close to $ 14.8 trillion in (gross) risk free assets available to the non-government sector. But once the government debt is eliminated, and the deposits that replace it converted to NRLs, the government supply of gross risk free assets is cut in half. The banking system would roughly double in size, but half of the new liability profile ($ 7.4 trillion in NRLs) would be risky, as compared to the risk free treasury debt it replaced. Given the risk on NRLs, the willingness of the public to replace its risk free term treasury debt holdings with risky bank term liabilities depends on pricing that risk.

    The example is extremely aggressive in the defined scope of the reservable deposit base, including all checkable, saving, and time deposits at $ 7.4 trillion. A more benign example could restrict the reservable scope to $ 700 billion in checkable deposits. The dynamic of credit created money feeding back into 100 per cent reserve requirements would be more moderate as a proportion of the banking system balance sheet. An intermediate example might exclude time deposits. This is a logical possibility, given the deferred nature of any liquidity risk on such deposits.

    FDIC insurance now in place is effectively a put option on deposit risk. A hypothetical shift to 100 per cent reserves would convert this put option to a balance sheet equivalent; i.e. deposits backed by risk free fiat reserves. This converts an option derivative arrangement into a risk free balance sheet arrangement. It seems intuitive that because “loans create deposits” works under the existing FDIC put option arrangement, it might also work under the derivative to cash transformation I’ve described.

    100 per cent as a reserve requirement is one number from a continuum of such choices [0, 10 … 100]. The question becomes, where on this continuum does a floating rate fiat system suddenly become a fixed rate system? It depends more on the institutional configuration than the number. A fixed rate system really requires that banks be prohibited explicitly from creating money from credit. Otherwise, as the example shows, the continuation of a floating rate system is at least theoretically feasible. The freedom to create money is separate from the level of reserves required against that money. There is no natural dividing line between a 99 per cent reserve requirement and a 100 per cent reserve requirement.

    Bill, the points listed in your conclusion summarize potential institutional structure in terms of permissible risk scope. The way I interpret it, the banking system remains dedicated to direct GDP credit support, following prudent risk management guidelines and financial activity involvement in doing so. The casino asset trading system is banished from the banking system. This general directional emphasis seems fairly broadly favoured among the PK/Chartalist/MMT group.

    While I’ve attempted to demonstrate that credit created money can continue in a 100 per cent reserve system, I agree that 100 per cent reserves aren’t necessary. Moreover, it seems to me that favouring zero per cent reserves is a natural position for those who understand the true role of reserves (i.e. settlement balances) in today’s system, in the sense that anything more than that is redundant in a system that could be structured efficiently otherwise with respect to risk bearing.

  10. bill, thanks for your reply.

    There is no confusion for me between real and nominal output. You clearly stated that prices would not increase, so these are one and the same.

    Your theory pays scant regard and shows little understanding of the problems of price and currency inflation and how they are related. This is the main objection that I have, and general hand-waving assertions such as: “So this is clear – if nominal demand grows at the same pace as productive capacity then inflation will not occur” amount to nothing.

    Perhaps you might consider a simple investigation between the rate of inflation of consumer price inflation and currency inflation (adjusted for real growth). Use the currency measure provided by the RBA in their statistical tables – I think you would be surprised to see the stability of the relationship.

    There have been numerous examples governments where unable to continue to fund themselves through the bond markets. This has generally occurred following a period of persistant budget deficits combined with some monetization of this deficit. At some point, these governments were faced with a choice between default or monetization. At this point, monetization has in some cases led to hyperinflation.

    To be fair there are two examples which run counter to this argument – namely Japan and the US, both of which are now running large deficits, partly funded by printing money. The Japanese situation has persisted for a long period of time without fostering inflation or a bond market revolt. The US situation is playing out before our eyes. If the US treasury begins funding their deficits directly through issuing bonds to the Fed, and inflation remains contained, then your theory will receive some validation.

  11. Great post, Bill. Thanks for elucidating a subject that I had sensed that sound money advocates had not sufficiently thought through, but could not put my finger on in detail.

    However, I had never put much attention on either 100% reserve banking, or the gold standard, etc., that the sound money folks harp on because it is not feasible politically. Moreover, a proper understanding of MMT makes sound money arguments rather irrelevant today in achieving a stable system capable of meeting public purpose and “providing for the general welfare.” Politically speaking, a 100% reserve requirement has about as much chance as a return to the gold standard in the US, in fact, probably even less. This would involve changing the commercial banking as it presently exists in radical ways, when the problem that the US faces is putting a leash on financial oligarchs, which seems to be beyond its reach.

    I think that the pressing issue is dealing with credit money in relation to Minsky’s financial instability hypothesis and debt deflation. The US and world are where they are now because we are now experiencing the end of a financial cycle and the consequences of the Ponzi finance that characterizes it. The sound money solutions in this regard are basically to eliminate commercial banking. However, according to MMT a principal reason that debt rises in relation to income is insufficient net financial assets of non-government, caused by too little deficit spending (which creates NFA) or too much taxation (which reduces NFA). If government fiscal policy matches nominal aggregate demand with real output capacity, then this imbalance need not arise in the first place.

    Moreover, there is little evidence I see for the claim that either the FRB or fractional reserve lending resulted in the present crisis, or in the onset of the Great Depression. It seems to be grounded in fitting “evidence” selectively to ideology to arrive at a conclusion.

    Rather, in a late-stage financial cycle, Ponzi finance dominated, and risk was not only underpriced but also misrepresented. “What is underpriced is over-used.” The crisis arose not because of any inherent failure of the financial system itself, but because the system was subverted by special interests. It is ridiculous to blame home buyers for lying on their applications, when this could easily have been checked, or borrowers who were lured into deals they couldn’t carry. This is predatory lending and fraud, regardless of rates, and both are illegal. Similarly, it wasn’t fractional reserve banking, low rates, or CRA, Gramm-Leach-Billey, or repeal of Glass-Steagall. It was playing loose. Laws and regulations were in place to prevent it. They were placed in abeyance due to neoliberal ideology and special-interest influence. Not that other regulation would not have helped. Elizabeth Warren recently said that if there were adequate consumer protection legislation, regulation, and oversight in place, the crisis could have been averted (not “might have been averted,” as she is sometimes erroneously quoted).

    The system itself does not need to be replaced with a step backward in time into an age that had its share of bubbles and panics, too. The system needs to be reformed so to insure that best practices are adhered to, preventing Ponzi finance, and there are adequate rules and a sufficiently independent watchdog in place to make sure of that. Credit money is like dynamite – a powerful tool in the hands of experts that know how to use it constructively and do, and extremely dangerous otherwise.

    My sense is that the private banking system is now entrenched, and the challenge is to insure that that is it operating in accordance with fiduciary responsibility and best practices, and serving public purpose instead of being subverted by special interests. My own political position is radical, so I would rather see things otherwise, but I realize that my best-case scenario is not going to happen barring a political revolution that I don’t foresee anytime soon. And if there were to be a political revolution, the outcome is far from clear. Realistically speaking, the US is a capitalistic democracy at present, not a social democracy, so trying to essentially change the private financial sector that is at the heart of US capitalism is a waste of energy. Better to work toward an understanding of how the monetary system works and craft a solution that is politically viable. A lot can be done if the ignorance is removed. But that is a huge task, given the inertia of the public mindset and the influence of self-serving opposition.

    The principal issue is serving public purpose. If we can insure that a private financial sector can do that without threat of being subverted, I can live with that. But it remains to be seen if even that can be achieved at present with all the money flying around Washington.

  12. What is the difference between a 100% capital requirement and 100% reserve requirement?

  13. From https://billmitchell.org/blog/?p=5098

    “The voluntary constraints, in turn, create political constraints on the government such that it has been pressured to maintain high rates of labour underutilisation for the last 35 years in most countries because they are unable to run deficits that are required to match the saving desires of the non-government sector.

    As a consequence, aggregate demand has been restricted and even undermined in recent decades by the pursuit of budget surpluses and the non-government sector has been pushed into dis-saving (and increased indebtedness).”

    Do you ever look at wealth/income inequality?

    I like to look at it like this. (savings of the rich)=(dissavings of the gov’t) plus (dissavings of the lower/middle class)

    It also seems to me that the rich/central banks use the interest rate to suppress aggregate demand to keep the labor market from tightening up so that the rich/central banks can exploit an oversupplied labor market to produce excess corporate profits (including banking profits) for themselves.

  14. JKH said: “Here are some recent deposit numbers for the US banking system:

    $ Trillions

    Domestic checkable deposits .7
    Small time and savings deposits 4.9
    Large time deposits 1.8

    Total 7.4 trillion”

    I think there is about $1 trillion of U.S. currency and at least half of it is held overseas. If so, what makes up the other “at least” $6.4 trillion?

  15. Oliver @ Wednesday, January 13, 2010 at 22:37

    One possibility is that the government creates new money by spending, resulting in cheques cashed at banks qualified to accept deposits, creating new deposit liabilities and accompanying bank reserves. The government decides on the level of money desired, and issues bonds to the public to drain any unwanted excess from the system. The credit or lending function then takes place in another set of institutions that must source such deposit money before lending it and who are prohibited from creating money by lending. This is comparable to a “fixed rate” monetary system. I believe this is close to what Bill described in his blog.

    An alternative which I attempted to describe in my comment above is that a single banking system continues conduct both lending and deposit functions, even with 100 per cent reserves. It continues to create money by lending, while issuing non-reservable liabilities to temper the growth in reservable deposits.

  16. Fed Up 12:44

    Those numbers are deposits in the banking system. Currency is not a deposit.

  17. Fed Up 11:41

    Reserves are an asset. 100 per cent reserve requirement means the level of required reserves is equal to the size of the deposit requiring reserves. The deposit is a liability. Asset size = liability size.

    Capital is on the right hand side of the balance sheet (e.g. equity). 100 per cent capital requirement means the level of required capital is equal to the size of the asset requiring capital underpinning. The asset is on the left hand side of the balance sheet. Asset size = capital size.

  18. “It also seems to me that the rich/central banks use the interest rate to suppress aggregate demand to keep the labor market from tightening up so that the rich/central banks can exploit an oversupplied labor market to produce excess corporate profits (including banking profits) for themselves.”

    Bill and other MMT’ers have addressed this issue in depth in books and articles. MMT shows that it doesn’t have to be this way since the government can use fiscal policy to create full employment (only frictional unemployment with no cyclical or structural unemployment) along with price stability, which neoliberals theorize is impossible.

    The neoliberal concept of NAIRU and monetary policy that supports it are chiefly ideological rather than empirical. This theory assumes a “natural rate of unemployment” of ~5%, even though frictional unemployment is only ~2%. It also posits that “inflationary expectations” increase below that rate. The reality is that these assumptions result in policy decisions that create a permanent stock of unemployment, which undermines the bargaining power of labor. Coincidence?

  19. Tom Hickey and others, does it matter if the gov’t deficit spends with currency or gov’t debt?

  20. Thanks for the replies everyone! I don’t think I have the whole currency, reserves, reserve requirement, and capital requirement correct in my head yet. I’ll have to do some more reading. Thanks again!

  21. JKH said: “Those numbers are deposits in the banking system. Currency is not a deposit.”

    What does the scenario look like if I take $100 in currency to the bank and then withdraw the same $100 in currency a week later? I think what I am asking is whether there is a separate entry for currency. Thanks!

  22. “What is the difference between a 100% capital requirement and 100% reserve requirement?”

    As I understand it, the difference between a capital requirement and a reserves requirement is essentially that a capital requirement is a solvency provision while a reserve requirement is a liquidity provision. Capital provision impacts leverage and risk, whereas reserve provision impacts loan pricing based on the interest rate as a cost of reserves. This is one of the factors banks use in pricing a loan.

    Only the central bank creates reserves, and reserves always remain in the interbank system – in the US, the FRS. Bank capital is the equity of a commercial bank. (Vault cash is used in computing required reserves, but generally speaking this is not material because most of a bank’s reserve requirement is met by bank reserves. In a 100% reserve requirement, 100% of loans could be held in vault cash, and some people envision the requirement in this way, doing away with the need for a CB. But that presents a problem for interbank clearing.)

    Banks do not leverage reserves, as is commonly believed. Banks neither “loan reserves” (in spite of Chairman Bernanke) nor loan against reserves (a misunderstanding of fractional reserve banking). The central bank provides reserves as needed at its current rate of interest.

    Bank capital is at risk when a bank loans, not bank reserves. They loan against capital, and they hold reserves for liquidity purposes. A 100% capital requirement would limit bank lending to bank equity, whereas a 100% reserve requirement would mean holding reserves at the central bank in the amount of loans. Banks would not be very profitable in either case, and there would be less credit money available readily available. This would reduce the velocity of money.

  23. Fed Up,

    Currency held by banks is an asset.

    You take 100 in currency to the bank:
    Bank asset 100 currency
    Bank liability 100 deposit
    Fed Up asset 100 deposit

    You withdraw 100 currency a week later
    Bank asset 0
    Bank liability 0
    Fed Up asset 100 currency

    US banks hold $ 53 billion in currency, compared to $ 7.4 trillion in deposits.

  24. “Tom Hickey and others, does it matter if the gov’t deficit spends with currency or gov’t debt?”

    The government spends by the Fed adding reserves on a spreadsheet to the Treasury account so that it’s checks clear in the reserve system when they are cashed at commercial banks. The government never pays with taxes or debt. It only looks this way because the US government is required politically, not financially, to offset spending $4$. This is a voluntary decision that is not a requirement of the modern monetary system, where the sovereign government is the monopoly provider of a non-convertible floating fx currency of issue. To think otherwise is to confuse the currency issuer with currency users (household, firms and in the US, states), or use principles that apply to a (now obsolete) convertible fixed rate currency system. The US government is not financially constrained and does not need to tax or borrow to spend, other than an arbitrary requirement in the name of “fiscal responsibility” as a political decision.

    This difference between issuer and user lies at the core of MMT. See A simple business card economy for an easily accessible explanation.

  25. Tom Hickey said: “The government spends by the Fed adding reserves on a spreadsheet to the Treasury account so that it’s checks clear in the reserve system when they are cashed at commercial banks.”

    Could currency be used to clear the checks? And if so, why isn’t currency used?

  26. A little off topic but…

    It seems to me that positive productivty growth and cheap labor produce price deflation.

    If the fed wants to maintain a small positive price inflation amount, they need to price inflate with currency, price inflate with debt, or price inflate some other way. They almost always choose to price inflate with debt which means they will probably need low interest rates, low reserve requirements, and low capital requirements.

    I’m trying to understand the debt creation process and if, in my scenario, that leads to excess savers and excess debtors. Thoughts?

  27. “Could currency be used to clear the checks? And if so, why isn’t currency used?”

    I assume you mean “cash,” money in circulation, or physical currency. Imagine the inconvenience of settling up that way on a daily basis among banks.

    As an aside, I was in the international gold reserve vault cellar of the Federal Reserve Bank of NY in the 60’s when the world was still on a gold reserve standard for international trade. Each country had a vault with its name over the door stacked with gold bars. There were a couple of guys (gnomes?) wearing steel shoes (gold bars are heavy) with a cart, transferring gold bars from one country’s vault to another based on the day’s settlement. Really, that’s how it was done in those days. Quite an amazing sight. But it’s a whole more convenient just to enter numbers in a spreadsheet by computer under the present non-convertible floating fx monetary system, instituted when President Nixon unilaterally shut the gold window on August 15, 1971, at Treasury Secretary Connally’s suggestion when there was a run on the dollar. What happened? Nothing, just a Connally predicted.

    BTW, when we speak of the government as “currency issuer,” and “currency of issue,” we mean coin minted by the Treasury, Federal Reserve notes, and bank reserves. Most of the “currency” issued by the government is in the form of bank reserves, which are just entries on a spreadsheet. Physical currency is sometimes even included in the notion of bank reserves because banks get physical currency by exchanging reserves for it when there is a public demand. With the advent of credit cards, a lot less physical currency is used in transactions. The government doesn’t spend with physical currency. The Treasury pays by check and the Fed make sure that the Treasury has the reserves to cover them. While the checks may be physical paper or electronic transfers, the Fed’s creation of reserves is digital not physical.

  28. “If the fed wants to maintain a small positive price inflation amount, they need to price inflate with currency, price inflate with debt, or price inflate some other way. They almost always choose to price inflate with debt which means they will probably need low interest rates, low reserve requirements, and low capital requirements.”

    Under normal conditions, the Fed attempts to manage “inflationary expectations” with interest rates in relation to actual and potential GDP, using some variation of the Taylor rule.

  29. Tom Hickey said: “BTW, when we speak of the government as “currency issuer,” and “currency of issue,” we mean coin minted by the Treasury, Federal Reserve notes, and bank reserves.Most of the “currency” issued by the government is in the form of bank reserves, which are just entries on a spreadsheet.”

    Sounds to me like a bank reserve could be defaulted on then. IMO, that makes it a debt instrument.

  30. Tom Hickey said: “Under normal conditions, the Fed attempts to manage “inflationary expectations” with interest rates in relation to actual and potential GDP, using some variation of the Taylor rule.”

    What if ‘expectations’ and reality are different?

    Is it possible to have too much debt and not enough currency?

  31. Bill does a good job in the above main post of listing and mocking some of the dafter claims made by the full reserve camp. But I still like the basic idea of FR (or perhaps 90% reserve banking).

    Bill claims that FR “would be the equivalent of a gold standard imposed on private banking which could invoke harsh deflationary forces” My answer: not if the state creates sufficient money. Indeed, if the state created too much, far from deflation, one would get INFLATION. Mugabwe has just proved that.

    I agree that FR, as Bill says, “does not eliminate credit risk”.

    Bill says “I do not support a 100-percent reserve banking system. It is the work of a lobby that hates and distrusts government”. My answer: the fact that a movement includes hate filled nutters, as the FR movement unfortunately does, is a weak argument. It’s the BEST arguments for any philosophy that one should always address.

    In this connection, the only claim for FR made by Friedman in his 1948 American Economic Review paper was that FR helps bring more stability. I agree.

    Bill’s answer is that better bank regulation and/or government intervention to dampen demand when required would be better than FR. That is correct in theory. He also says “. The government can always dampen demand for credit”. True in theory, but governments failed to do this in 2005-6. The reality is that the bubble leading up the recent credit crunch just wasn’t spotted.

    And another piece of brute reality is that the world economy has taken a trillion, trillion dollar hit (or whatever the figure is) which should have been enough of a jolt to induce a bit of “better bank regulation”! But what has actually happened? Absolutely nothing, far as I can see. Wall Street banks are up to all their old tricks again.

    The problem with “better bank regulation” is that it is complicated: banks will always outmanoeuvre politicians here. The reality is that Capitol Hill is swarming with bank funded lobbyists, cowboys, and other forms of bank funded low life.

    FR is a nice simple rule.

  32. “Is it possible to have too much debt and not enough currency?”

    No, the government as currency issuer is not financially constrained and can always issue the necessary to cover its obligations and commitments. Claims that the US is “running out of money,” is “going bankrupt,” becoming “insolvent,” or is “at risk of default,” are based on a misunderstanding (or misrepresentation) of how the modern monetary system operates. Such claims either take the government to be revenue constrained as a currency user when it the monopoly issuer, or else are couched in terms of a convertible fixed rate system when the US is now on a non-convertible floating fx system.

  33. Ralph: “FR is a nice simple rule.” Good analysis. But I have a practical question.

    Bill began by observing that MMT is agnostic about this. It is therefore not so much an economic decision as far as MMT is concerned as a political one, largely based on one’s attitude toward the financial sector. I am sympathetic to your argument, but to say that the banks cannot be controlled through reform and propose the political practicality of FR seems to me to be contradictory. I don’t see any chance of something like this actually happening in the US when even the most basic reform is being stymied, and there has been almost zero accountability or transparency. Maybe the new Pecora commission (not) will be able to reverse this, but it apparently doesn’t have any real clout.

    At bottom, I think that the problem probably lies almost as much with the American people as the banks. On one hand, people are clamoring for financial reform, but that would mean, on the other, the end of the loose credit that they have been enjoying and have become increasingly dependent upon. People hate debt but love credit. I suspect that the banks are “banking” on taking advantage of this internal conflict, just like they did at the onset of the crisis when the public was against a bailout, but was more against a “government take-over” of the financial system through resolution, which was demonized as “nationalization.” This politics of this is more complicated than the economics in that it involves the interaction of fear and anger, and greed.

    Other than Ron Paul, I don’t hear anyone in Congress loudly calling for FR. And Ron Paul on the right has about as much influence as Dennis Kucinich on the left. In, fact they sometimes team up where the libertarian position intersects with the radical position.

  34. Ralph

    What Mugabwe proved was that creating money to maintain aggregate demand when aggregate supply has crashed (because 45% of the farm land is no longer being used) than you get lots of inflation. The same thing happened in Germany in the 1920’s when France and Belguim used the Treaty of Versille to seize the coal mines of the Rhineland – Germany created money to provide an income for those displaced from the Rhineland and maintain aggregate demand but aggregate supply crashed because of the seizure of the Rhineland – France and Belguim appropriated any coal that was mined (using their own nationals) for their own domestic purposes.

    Similarily, governments creates inflation if they create money to increase aggregate demand and there is no increase in aggregate supply.

    Aggregate demand v aggregate supply, and the competition for profits between workers and firms determines inflation/deflation outcomes not the amount of money in the economy. However, the amount of money may influence the above factors, but the amount of money is not the sole driver.

  35. Bill-

    More importantly, it is also a pity that their emotional positions are not backed by an understanding of how the monetary system actually works. They hark back to the gold standard which placed governments in a financial straitjacket and prevented them from reacting effectively to crises of the sort that has nearly crippled the world economy over the last few years.

    I think the way to understand Austrian theory is as “the road not taken”.

    From 1870 through the early 1930’s the world was struck by numerous financial crises that turned into wide scale recessions or depressions. The keynesian/monetarist/statist response was, “Look, the private sector is inherently unstable, the gold standard is inherently bad, thus we need a fiat standard with government regulated/controlled banking”.

    The libertarian/Austrian response was, “The market is not inherently unstable, government policies broke the market. The government instituted a national policy of fractional reserve banking, which is inherently unstable. Second, it tried to maintain a gold to dollar conversion rate of $22/ounce, while printing far more greenbacks than could be backed by gold. If you do those two things, of course you are going to get a crash. If the government simply stops breaking things, we can get a fully privatized financial sector that will be perfectly stable without any systematic crises”

    The proper Austrian response to the Great Depression would be to do a monster devaluation and then go back to the gold standard at a sustainable ratio. I think this could have worked. Instead, the government decided to go off the standard altogether.

    I think a fiat system can be made to work well, or a 100% reserve gold standard could be made to work well. Like programming languages, often it’s much more important to know how to use your tool well, than to waste time discussing which tool is best. The U.S. government in the 1920’s and 30’s did not know how to run a gold standard well. Now it does not know how to run a fiat standard very well. This tells me that maybe we should spend more time worrying about how we get better engineers, than worrying about what tools those engineers should use.

    I do not know if the Austrian claim that a fully free market, maturity matched banking system would work and be crisis free. The logic seems fairly plausible. But I do know that a) there has never been a free market banking system in the industrialized world and b) every major recession and depression has been grounded in the problems in the banking/credit system.

    Finally, I agree that the vast majority of Austrians have no clue how the modern financial system would work. Should they actually seize the reins of power, they almost certainly would leave the entire economy in a heap of rubble. There may be some theoretical path that could transition the economy from its current state to an Austrian utopia. But no one close to power has any clue how to find that path.

    That said, it’s not useless to study Austrian economics. It’s interesting to understand “the road not taken”. It’s also interesting because even though Federal Reserve banks do not maturity mismatch anymore, other parts of the banking sector do. If you want to understand what happened to the money market funds and the shadow banking sector, the Austrian insights can be quite helpful.

    The mad ravings of Rothbard that all money base and/or private credit creation is inflationary is a dogmatic assertion.

    When Rothbard talks of “inflation” he means “dilution”, which was the original meaning of the word inflation. That printing money creates dilution is a truism.

    If the government increases the supply of what I call “broad money” ( by broad money I mean all government backed paper – currency, bank deposits, CD’s, t-bills, etc), it dilutes the existing holders of broad money. Just as issuing new shares of stock dilutes the existing shareholders.

    Imagine counterfeiters printed enough money to dilute the broad money supply at 3% a year. Imagine productivity also grew at 3% a year due to advances in green energy. You get no CPI inflation. Yet the counterfeiters are still stealing from existing holders of broad money. Instead of enjoying increased purchasing power as productivity rises, they are stuck stationary while the counterfeiters gain purchasing power with their newly printed money.

    Similarly, if the government prints broad money and siphons it to political favored constituencies, then that is a seignorage tax. Holders of broad money are being taxed to support the government’s favored spending plans. Maybe Rothbard is wrong to get so upset about just another tax. But there is no denying that it is a seignorage tax. Just because the economy grows at 3% does not mean the tax does not exist. It just means it is camouflaged.

  36. JKH-

    Excellent post. I’ve never been clear on 100 per cent reserves. It’s always been a confusing, amorphous Rubik’s cube of risk considerations and institutional possibilities. Discussions are typically clouded on issues of liquidity risk, credit risk, “maturity transformation”, and macroeconomic money supply/inflation considerations. They’re often vague on the specification of the substance of the reserve and the domain of reservable deposits. Assuming a 100 per cent scenario, is the reserve substance gold? Is it government bonds? Is it current central bank reserves? Are checkable deposits reservable? Are all deposits reservable? What is the proposed institutional configuration?

    The first thing to understand about 100 percent reserves, is that it’s a policy for a radically different banking system. Trying to map our current rube goldberg banking system, to a theoretical Austrian, 100% reserve system, is a non-trivial task. Instead, I’ll try and answer your question from the perspective of a theoretical, blank slate system.

    The two key principles to a 100% reserve system are a) maturities must be matched and b) the backing asset should be what the depositor/lender believes it to be.

    If the customer puts US dollars in a demand deposit account, then the bank should just keep that money in a safe, to be available on demand. The bank should not lend out the dollars for any amount of time.

    If the customer places U.S. dollars in a time deposit account, where he must give 60-days notice for redemption, the bank should not lend out the dollars for more than 60-days.

    If the customer purchases a 3-year CD, then the bank should not lend out the dollars for more than 3 years. The bank in turn buy a 3-year corporate bond, a 3-year treasury bill, or something else. The bank should be up front to the customer about the risk profile of the loans it makes.

    If the legal tender of the land is gold coins, and the customer places gold coins in a demand deposit account, then the bank should actually store the gold coins.

    Under this type of banking system, there is never any “liquidity” risk. A bank is never in a position where it has to sell off a security before it has matured, in order to meet the demands of depositor. There still is credit risk. There is no private credit extension. If the government wishes to increase the money supply, it must actually print more currency (or mine more gold if it’s running a gold standard).

    There also could be bond mutual funds in a maturity matched system. The buyer of a share in a bond fund should be very careful about the average maturity of a bond fund. If he only intends to hold shares in the fund for a year, he should not buy a fund that has an average bond maturity of 10 years, because there is substantial risk of short term price drops.

    If you want to read more about a theoretical, Austrian banking system, I recommend this comment thread by Mencius Moldbug. He does a good job explaining to a modern financial professional how a maturity matched system would work.

  37. Devin,

    Thanks. That’s actually pretty clear the way you’ve written it.

    I’m interested mostly in trying to visualize the institutional configuration. Is the system integrated (credit and deposits in the same institutions) or is it bifurcated (credit and deposits in separate institutions)?

    Viewed this way, the issue of reserves is quite separable from the question of who creates new money. That’s notwithstanding the preference of some to conflate requirements for both characteristics (as well as that for maturity transformation) in the definition of a 100 per cent reserve system. That may be a definition of choice, but it doesn’t follow analytically. That is what I meant in the opening paragraph you quoted.

    E.g. it’s quite possible to visualize an integrated institutional system with 100 per cent reserves that is the same as the prevailing institutional system we have in every way with the exception of the magnitude of the required reserve ratio. That is exactly what I did with the extreme example in my comment whereby $ 7.4 trillion in deposits is 100 per cent reserved. You may object to such an example because it doesn’t refer to such characteristics as maturity transformation. But that’s exactly my point regarding the arbitrary configuration of multiple characteristics that is given the label of a single characteristic. My first order of business is examining the variable of 100 per cent reserves. I’m quite interested in the parameter of maturity transformation beyond that, but there is nothing analytical or logical that demands a particular association between the reserve ratio variable and the maturity transformation parameter in all cases.

    As another example, it is quite possible to visualize a bifurcated institutional system with 100 per cent reserves held against deposits in one set of institutions and credit extended against non-deposit liabilities in another set of institutions. However, there are important issues regarding payment system surrounding such a set of credit institutions. What is the institutional mechanism whereby this set of institutions effects payments across its members? In what account does it temporarily store the payment medium in order to settle such transactions? How does it clear and settle these payments? And most importantly, how does it handle a potential shortfall in payments in any one institution? Because it is extremely doubtful that such a set of institutions could exist indefinitely without experiencing any payment shortfall at some point. If perfectly matched, I suppose this is theoretically possible. But the idea that all credit is renewable at maturity only on the basis of a pre-existing renewed liability seems to be quite a tightrope to walk. It suggests massive uncertainty as a permanent structural feature in the supply of credit. Any breach in this assumption requires some lender of last resort. And whether that lender of last resort is a central bank or merely one of the depository banks acting as clearing agent for a customer credit bank, the jig is up in terms of maintaining a pristine maturity matched credit system. And the jig is up the first time it happens.

    I’m familiar with various Moldbug discussions going back several years on this issue, and others like Kling. Moldbug in particular is impressive. But I’m not convinced that the subject has received the full analytical vetting it warrants yet.

  38. Again, JKH, you’ve said exactly as I would in your second-to-last paragraph at 7:09. Perfect maturity matching doesn’t get rid of the liquidity risk issue; not even close.

  39. Dear Devin

    Thanks for the thoughtful comment. I agree that existing systems (whatever they are) can be made to work better if the policy makers really understand the possibilities.

    But that is the point – the possibilities presented a national government under a gold standard are limiting and may not allow it to achieve true full employment. It forces monetary policy to be working against fiscal policy if full employment is your objective. The conjunction of fiscal and monetary policy is achievable under a fiat monetary system. At present, the policy makers pretend they have a gold standard – or some hybrid and act as though they are as constrained as they would be under a gold standard. That is why we are getting such unsatisfactory responses to the current crisis and also why most nations have endured very high levels of unemployment (relatively) over the last 30-35 years.

    On dilution: I don’t use the terminology “printing money” to describe government spending. But that said, I disagree with the proposition that “printing money … [by which I will call government deficit spending not matched by debt-issuance] …. creates dilution is a truism”. Dilution can only occur if what you can buy in real terms per $ is reduced. That is not a function of the stock of “liquidity” – but what the liquidity can purchase in real terms. In economies with high degrees of capacity underutilisation and unemployment, increase public deficits can actually increase the amount of real goods and services at the disposal of the individual (for example, giving the unemployed a job) without reducing the real capacity per % held by others.

    Your last paragraph also confuses the role of taxation. No-one is taxed to pay for anything – whether it be broadly-spread spending or narrow pork-barrelling. That is the misconception of the Austrian paradigm (and mainstream economics as well). Taxes are levied to regulate overall purchasing power (and in some cases to reallocate resources away from “bads” such as tobacco taxes). So no-one is being taxed to support government spending – which occurs independent of the so-called revenue raising initiatives.

    best wishes
    bill

  40. Dear Devin (and JKH and Scott)

    If the point of the 100-percent reserve banking system is to reduce bank losses then I fail to see how it does that unless it prohibits credit creation at all. As soon as the bank starts making loans (even if the funds lent are fixed-term deposits already acquired) then credit risk is possible. Widespread failures (defaults) are still possible which just means that at some point in the future (when the fixed-term deposits expire) the “bank run” occurs.

    In that sense, I don’t see the point of it and so I start thinking about the other motives that are behind the suggestion – the Austrian “sound money” motives – which in my view disable the capacity of the national government to pursue and achieve public purpose – full employment, equity in opportunity, environmentally sustainable growth, and price stability.

    best wishes
    bill

  41. JKH-

    Trying to imagine what the banking sector would look like if you just change one variable ( ex. requiring 100% reserves) is a near impossible task. There are any number of directions it could take, some resulting in complete disaster, some being quite good, depending on the hoary details of the implementation. My own take is that switching to a 100% reserves should only be part of a complete switch to “temporal” based accounting, with maturity matching at every level. Just instituting 100% reserves now would do pretty much nothing, especially since deposit accounts are FDIC insured anyway. Exchanging a government backed “put” for a government greenback isn’t really changing anything.

    To respond to your second paragraph, I can conjure up at least one vision of how a port of our current banking system to an Austrian-like system could occur. This is of course pure fantasy, interesting as a fun intellectual exercise only.

    First, create an accounting of all government backed paper (FDIC insured accounts, treasury bills, AMLF backed money market funds, etc). Create enough dollars to back all government paper (on the order of $20+ trillion). Account for those dollars in a new software system run by the Fed. The Fed’s digital ledgers would have a database with a row for every dollar. The row contains the serial number, it’s current status, and the current owner (where the owner is basically a tax payer id). All deposit accounts, money market funds, etc, would be converted to direct accounts held at the Federal reserve. Instead of having $15K in a deposit at Bank of America, there would be $15K digital dollars held in my account at the Fed, with my social security number attached to them. Instead of having a treasury bond payable for $10K in 5 years, I would have a $10K in digital dollars held at Fed, with a “non-transferable” restriction on it for 5 years.

    The job of the banks would then to be as intermediaries for borrowing and lending. I would go to a bank and purchase a 15-year CD for $10K, that promised a repayment of $30K in 15 years. The bank would cash my check at the Fed, and the Fed would transfer money from my account to the bank’s account. My account at the Fed would be $10K less, and I would have an IOU from the bank. The bank would then lend out that money to a home buyer, who takes out a 15 mortgage. The home buyer deposits the bank’s check, and the Fed decrements the bank’s account and increments the home buyers account. The actual clearing of the checks is a trivial matter of software.

    So what happens if there is a shortfall you ask? The most likely reason for a shortfall is that the home owner defaults on their mortgage. Keep in mind that in the example above, the bank is not going to be matching up one CD to one mortgage, but rather a pool of CD’s to a pool of mortgages. The bank might have 100 mortgages due on Jan 1st 2015, and 500 CD’s it needed to pay off on Feb 1st 2015 (this is simplified as mortgages have a different re-payment structure than CD’s, but for now imagine their repayment structure is the same). The bank would expect a certain default rate on the mortgages. So the bank might offer a 5% interest rate on the CD’s, and make the mortgages at 7%. Part of the difference would be eaten up by expected defaults. The rest of the difference would be profits for the bank. If the bank makes exceptionally good lending decisions, it profits more. If it makes poor lending decisions, the defaults will eat up its profits. If the defaults eat up all the profits, the bank will have to either borrow against future profits, or sell stock to raise capital. If that doesn’t work, then the bank must default on it’s CD’s, and the owners of the CD’s would get slightly less than face value. If a set of mortgages are simply delinquent then the bank may issue short term bonds to third party, in order to get funds to payback the CD’s. The bank would want to avoid this situation, so it probably would always maintain a small buffer of cash to make any such desperate measure very rare.

    So in summary, defaults and/or delinquencies hit the bank stock holders first, and the CD holders second. I would consider both defaults and delinquencies cases of “credit risk” not “liquidity risk”. The described system certainly has credit risk, but I do not see how it has liquidity risk. I’m not sure what the “tightrope” walk you refer to is. Obviously the bank has to understand it’s risks very well, and keep a healthy margin for error, or else it will get wiped out. Shareholders do not want to get wiped out, so they would demand that the bank keep the spread between the interest on the CD’s and on the mortgages large enough to cover even unexpected defaults. Shareholders do not the bank walking a tightrope, because they are first in line to get wiped out. This is all a feature not a bug, as it keeps the incentives of the bank’s lending officers pointing the right way.

    The above plan is essentially the Austrian solution for “liquidity” crises. The PK’s solution is obviously very, very different. The advantage of the Austrian plan is that it does not require any sort of FDIC insurance or lender of last resort. Thus it does not have a moral hazard problem and thus does not require a central government regulator basically monitoring and approving all lending activities. The disadvantage is that it looks nothing like our current banking system, and nothing like the banking systems that have been running the world for the last three hundred years. So who know’s what it would it look like in practice, and certainly no current political authority is capable of implementing it.

  42. Credit is the backbone of a modern monetary system. Banks were created by the constitution as a great vehicle to have both capitalism and full employment. The private sector by itself cannot create full employment because of financial constraints and needs buffer for uncertainty and the government has to step in. Banks hold a unique position, else credit will be difficult and pricey. Banks have this unique ability to lend by expanding out its balance sheet. Preventing an easy expansion of the balance sheet will just increase the loan rates. The settlement balances at the Federal Reserve have just two roles – settlement and help satisfy the household need for currency notes. The central bank is also the lender of the last resort and can lend as much as possible. The whole system is designed wonderfully so that entrepreneurs, academicians, scientists, engineers, artists, advertisers, sales persons, managers, HR, unskilled laborers can make use of this and increase the quality of life for themselves and others.

    If the households permanently decide to use more cash, the central bank can help banks slowly reach that state where they keep more settlement balances, lending in unlimited quantities in the meantime.

    Its a misunderstanding of the system and poor economics which led to the crisis. Economic advisors to governments have no concept of “domestic private sector deficit” in their analysis. If they had noticed that the private sector was getting weak, they would not have allowed a credit boom to have happened. Instead they supported the boom and ignored reports showing scams and forgeries. It is neither low interest rates or low reserve requirerments which caused this crisis. Blaming the crisis for these two things is like blaming the invention of fire instead of blaming the fire prevention system.

  43. Devin,

    Too much worry about monetary aggregates will never be on Bill’s mind.

    The design you have mentioned is unachievable. Imagine the start of the world – government spends $100 and taxes $20. The private sector has $80 in their account at the central bank. The government cannot issue debt – it will take away those $80 from the central bank’s account and it will get replaced with government debt. So the government cannot issue debt $-for-$ for its deficit spending. Nothing wrong about it but Austrians will object.

    However, the most important flaw in the design I can see is that the amount of debt is totally upto household’s liquidity preferences. Households refuse to buy CDs and the interest rates shoot! The advantage of having a bank is that the banking system sets the price – the interest rates because of its ability to expand its balance sheet.

    “Loans create deposits” is a good thing!

  44. Devin,

    Thanks. That’s one of the clearer expositions I’ve seen for Austrian type architecture. It helps when probing the underlying ideas.

    In your constructed version of a 100 per cent reserve system, you blend all government backed paper, including what would have been FDIC insured deposits, into a big liability bucket of the Fed. There might be other ways of doing it, but using the Fed simplifies such an example. A separate non government credit banking system includes the usual loan assets etc., match funded by term liabilities of some sort. The credit banking system is not allowed to make loans without having the money in place first – i.e. without having issued liabilities to fund new loans, and then only on a matched maturity basis. That’s the pure interpretation as I understand it. I realize that there are deviations from perfection in the implementation, but I’ll focus on the pure model first. I’ll refer to the Fed deposit system, a separate credit banking system, and the combination of the two as the “bifurcated” system.

    The Fed deposit system effectively eliminates liquidity risk on 100 per cent reserved deposits by virtue of its fiat powers of money creation. This eliminates credit risk at the strategic level, because the liquidity creating power of fiat currency issuers dominates potential concern about government solvency. It eliminates liquidity risk at the operational level, because fiat currency issuers can exercise this power at any time.

    The design of the credit banking system as I understand it is intended to eliminate liquidity risk through the combination of two characteristics. First, banks must have money before they can lend. Second, banks must match maturities of loans and deposits. The credit banking system still has credit risk, and non government capital is required against that risk.

    While the Fed system features 100 per cent reserves, the combined system requires dual reserves, in effect. You’ve noted that the credit banks would maintain bank accounts at the Fed. A credit bank by requirement must attract term liabilities before it can make a corresponding loan. Therefore, it must have funds in its Fed account, before allowing the drawdown of such a loan. It must show a “long” position in this account before advancing funds. Given this requirement, this is a reserve account in spirit. The credit banks hold these “lending reserves” with the Fed, which in turn holds 100 per cent reserves against all of its liabilities, including the reserves of the credit banks.

    Liquidity risk in its extreme form is a “run on the bank”. Essentially, depositors fear for the safety of their money due to concerns about a bank’s solvency. Whether those concerns reflect actual or perceived insolvency, the effect on liquidity risk is the same. Depositors want out – particularly those who aren’t insured. We know how that works for the conventional structure. How does it work in a bifurcated system?

    First, as noted above, the Fed deposit system is fully protected from liquidity risk, and fully protects the liquidity properties of its deposits. A government issuing a fiat currency doesn’t have to worry about liquidity risk on its own liabilities. The reason is that the government manufactures its own liquidity. Any concern regarding government solvency is superseded by its liquidity powers when the government is a currency issuer. This is a variation on the interpretation of the central PK MMT theme that governments are not revenue constrained.

    As you note, capital protection is featured in the credit system. Such capital protection combined with maturity matching should then mitigate liquidity risk in the bifurcated arrangement. But does it?

    I would say not. Liquidity risk and solvency are intertwined in market perceptions. But liquidity and solvency are very distinct characteristics at an operating level. (The term “solvency” is used in the sense of balance sheet solvency or net equity.) Indeed, because of this distinction, the “purest” version of the matched maturity system you have sampled is subject to an extreme form of liquidity risk. Consider the following example:

    Suppose a pure credit bank is set up just about as perfectly as you can make it on a matched maturity basis. Again, you’ve noted that there are practical limitations to perfect matching, but let’s just assume such a perfect matching situation is possible in concept. And suppose this bank has a capital position that is in fact quite adequate.

    Now suppose there is a credit scare. And, one of this bank’s loans goes bad all of a sudden. The loan comes up for maturity. At maturity, the bank declares it lost. It gets no cash back from the borrower. The bank writes the loan off. It deducts the loan loss from its capital. It still has plenty of capital though.

    This bank is perfectly matched. So it has a liability maturing at the same time as the loan.

    Assume that on the day prior to the simultaneous maturities of the loan and the liability, the market suddenly becomes fearful about the existence of more bad loans approaching their contractual maturity. The result of such fear is that the bank is unable to attract additional liabilities. There isn’t a “run on the bank” though, because liability holders can’t withdraw their money prior to their respective maturities. They’re locked in. Hence no run.

    What happens on the day both the loan and the liability mature? The bank adjusts its capital position to record the actual loan loss. That’s fine. But here is the essential point of distinction: the capital write-off has absolutely no bearing on the bank’s ability to repay that maturing liability. The capital write-off is a pure internal accounting wash: credit loans (down); debit capital (down). Effect on liquidity position: none.

    The bank’s capital position, assumed to be quite adequate in this example, is in a first loss position by virtue of its ranking in the liability structure. Credit bank liability holders are senior to capital. They do not absorb the loss while capital is in place to perform that function. They expect losses to be absorbed by capital. Anything to the contrary would make a mockery of the capital structure, and would essentially put creditors in a ramped up loss exposure simply due to the luck of the draw on maturity matching. Why would creditors expect to take a loss while plenty of capital was still extant? If they can’t get their money, it’s off to bankruptcy court or whatever wind-down resolution facility is in place. They won’t let the bank’s maturity matching scheme impede their access to their money, and the bank’s capital, as necessary.

    Notwithstanding the likely viability of its capital position, the bank has no money with which to repay the maturing liability. Moreover, the community of potential liability buyers has decided all of a sudden that in its view, rightly or wrongly, the viability of the bank’s capital position is at great risk. While there is no run on the bank, there is also no run into the bank. So if it can’t attract any more money to replace the liability, what happens next?

    Before answering that, consider the bank’s deposit account operation at the Fed (which I’ve called a reserve account).

    On the day of the deposit maturity, the bank has no money in its reserve account. It has none coming in from the maturing loan, and it has none coming in from new liability buyers due to the spreading credit scare.

    Apparently, it has two choices. It can fail on the payment and declare bankruptcy. Or it can make the payment and go overdraft on its reserve account.

    In either case, the choice for the maturity matched credit bank is not pleasant. How did it get into this difficulty, given such good intentions?

    Let’s assume this bank wants to go on living. So it goes overdraft on its reserve account.

    It appears to me that’s the end of the Austrian architecture right there. The crumbling logic is due to the requirement for an overdraft privilege with the central bank, which must exist in order to avoid failure in the purest of matched maturity models. And that is an earthquake of a contradiction in the overall Austrian architecture, as I see it.

    If the overdraft privilege exists, the discipline of “loanable funds” essentially vanishes as the underlying principle motivation for institutional bifurcation in the first place. If overdrafts are permissible, the constraint is no longer binding that funds must be available prior to lending, and there is no longer any reason for institutional bifurcation. We are back to the idea of an integrated system, even with 100 per cent reserves, as I described in the $ 7.4 trillion extreme case in my earlier comment in this thread. That system is a full fiat and “loans create deposits” operation, where banks continue to create money from credit, notwithstanding the balance sheet implications of 100 per cent reserves. 100 is just another number in that case – like 99, 10, or 0.

    It seems bizarre, but this most pristine of matched maturity examples leads to a catastrophic liquidity interruption, that is only solvable by government intervention. The bifurcated model exacerbates potential liquidity risk well beyond the normal dimensions that exist in the more conventional integrated system. Banks don’t normally fail when the first loan goes bad. But such failure is implied by the bifurcation architecture in its purest form.

    BTW, Scott Fullwiler has emphasized the critical nature of the overdraft question in several comments here and elsewhere in recent days. Bill makes a similar point above. The importance of the overdraft event per se is central to PK MMT analysis of the monetary system. It is one thing to say that new rules are intended to preclude overdrafts. It is another to say they cannot or will not occur.

    “Real world” modifications to this pristine example might include an element of delay in the process of crying uncle with respect to the premise of matched maturity protection. The logic crumbles quickly in the purist model; more slowly in a compromised model (more like real world banks in the current system). The credit bank reserve concept means that well run banks might tend to “stockpile” reserves of incoming term liabilities in order to offer some decent selection of maturities to customers coming in for loans. Such an approach would lead to some level of self-imposed reserves, created by an inventory of excess liabilities. This would position credit banks favourably in the short term for the type of credit/liquidity shock described in the “pristine” example, but the protection thus created would only be temporary, and the result only delayed, depending on the size of the liability stockpile and corresponding reserves.

    Given the system constraint that liabilities must be in place before loans are made, the maturity structure of credit in aggregate depends on the maturity structure of liabilities in aggregate. In addition to the bank being under this constraint, the banks borrowing customers are under it. Under today’s system, banks “cover” their liquidity requirements as a function of credit demand. Under the Austrian type “reserve constrained” maturity matched system, banks offer a credit menu on the basis of liquidity provided by liability holders. As a result, this is a highly constrained system for borrowers. Borrowers no longer have unconstrained options on term structure. They can only choose what’s available at the time. I guess that’s what “loanable funds” means under such a system. But it’s “loanable funds” not only on the basis of available quantity, but available quantity per maturity.

    In summary, the intention of 100 per cent reserve requirements is to ensure credit and liquidity protection for bank deposits and government liabilities. The maturity matching motivation is a discipline imposed on the credit banking side of the framework. Banks are not permitted to create money from credit. They must have money before they lend it, and they do this by issuing liabilities prior to lending. This requirement combined with maturity matching and capital discipline should make liquidity risk a non-issue. But a fundamental problem gets in the way of this objective. Credit banks in a bifurcated framework cannot avoid liquidity risk. Credit banks require accounts with the depository institutional framework in order to make and receive payments. In particular, they require such an account to make payment to holders of maturing liabilities. If a maturity matched liability coincides with a bad loan maturity, there will be no matching funds to make payment. The fact that capital is available to absorb the loan loss has no effect on operational liquidity. Liquidity and capital are distinct. The bank in that situation has two choices. It can fail on the payment and go bankrupt (or some other form of wind-up). Or it can go overdraft on its deposit account at the Fed. In order to make the second choice, it must have a pre-existing privilege to do so. If such a credit facility is in place, it means that the principle of pre-existing funding has been broken. In either case, the bank has definitely experienced liquidity risk, notwithstanding matched funding. The bottom line is that the combination of 100 per cent reserves and maturity matching provides no ultimate protection against liquidity risk. Liquidity risk is a function of perceived solvency risk. At the same time, the ex post balance sheet impacts of liquidity risk and solvency risk are operationally separate – because those impacts are recorded through entirely different accounts. It is that account separation that precludes capital from protecting realized liquidity exposure at the operational level. Finally, real world extrapolations of the “pristine” Austrian type model may add an element of buffer protection afforded by a self-imposed reserve of excess liabilities and corresponding deposits at the Fed. But the solvency/liquidity dynamic still operates, and will prevail according to the duration of the liquidity crisis and the time protection offered by such a liquidity buffer.

  45. JKH-

    THank you for the response. Some people design fantasy football teams, I design fantasy banking systems. Glad you find the exercise interesting.

    I would argue that the crisis that you call a “liquidity” crisis is really a “credit” crisis. The bank made a really bad loan, and thus cannot pay back it’s CD holders. That’s a credit problem, not a liquidity problem.

    I did make a big mistake in my description when I talked about “capital”. You are correct that “capital” in the balance sheet sense does not represent liquidity. I should have said that the bank would make up losses first with “cash on hand”.

    Every bank with sound management will have a cash “buffer”. This buffer is a cash desposit account, held in the bank’s own name at Fed. The buffer was originally funded and filled through either selling shares or retained earnings ( this makes it quite different than a reserve). If a home owner fails to pay their loan back to the bank, the bank will need to repay the CD holder out of the banks own buffer account. This will show up as a loss on the quartly earning statement. The bank will need to replenish the buffer by selling stock or retainined cash flow.

    How do we know this buffer will be big enough to cover losses in practice? Well, simple. As you point out, if the buffer is too small ( ie zero), the bank defaults at the slightest problem. Shareholders get wiped out. This is not good. The buffer therefore must be large enough to cover any reasonable variation in loan repayment. But if the buffer is too large, the bank is leaving idle money in an account that’s earning no returns for the shareholder. It is up to the shareholders and/or the management to figure out the appropriate size of the buffer. Maybe it’s 1% of desposits due in given year, maybe 5%. I’m not sure. But shareholders and management have a very strong incentive to get the buffer size right.

    So in your actual example of a failing loan, the process will go:

    a) Repay the maturing CD out of the buffer. Then, overtime the bank will need to replenish the buffer by either issuing more stock or retaining earnings.

    b) If the defaults are so great, they eat up the entire buffer, then the bank will need to take out an emergency loan. It will need to find a third party (not the Fed, but an ordinary private bank that has a better credit position) to make the loan. The loan would likely be made against the bank’s future profits or equity. The third party will make this loan if it thinks the bank is fundamentally healthy, and has just hit a temporary rough patch.

    c) If no third party is willing to make a bridge loan, then the bank is insolvent. It goes into default on the CD’s. Shareholders get wiped out. The owners of the CD’s are first in line to be paid off, but they will probably not get all of their money back.

    Under no circumstance can a “credit bank” run an overdraft at the Fed. As you correctly point out, this would defeat the entire point of the system.

    So I made a mistake in my original proposal in leaving out the part of the buffer. With the buffer, and with the ability to take out loans from third parties, a bank will only fail if it makes so many bad loans that it overwhelms all its existing cash, plus all its expected future profits. There is no failure due to a liquidity crunch, only due to major mistakes in making bad loans.

    What you did in your example is design a system pristine from the maturity matching perspective. But it was very impure from the credit risk perspective, and the bank in your example was making the unsound decision to have no buffer. Thus it failed instantly upon the first default. If the economy was “pristine” from a credit risk perspective, ie no defaults or deliquencies, then there would be no need for a cash buffer account. In the real world, there will be substantial credit risk, and thus all sound banks will keep a cash buffer account.

    The Austrian system I described does not eliminate credit risk. Banks will make mistakes, banks will fail, CD holders will occaisionally lose money. But incentives are aligned. Shareholders take losses first, so they have an incentive to makek good lending decisions and keep a sufficient buffer.

    This would position credit banks favourably in the short term for the type of credit/liquidity shock described in the “pristine” example, but the protection thus created would only be temporary, and the result only delayed, depending on the size of the liability stockpile and corresponding reserves.

    This is correct. In fact, a bank should never do this even as a temporary measure. If a loan goes bad, and it needs to repay a CD, it should never repay the CD with a “reserve” or from someone elses loan repayment. Those loan repayments are already marked for someone else. It should only repay out of the buffer.

    As a result, this is a highly constrained system for borrowers. Borrowers no longer have unconstrained options on term structure. They can only choose what’s available at the time.

    This is true. But in the age of the internet and global finance, I’d expect that borrowers would have a wide variety of choices available.

  46. Devin,

    “The bank made a really bad loan, and thus cannot pay back its CD holders. That’s a credit problem, not a liquidity problem.”

    We differ there. It’s a credit problem at origin, but the credit problem is absorbed by capital. The purpose of capital is to absorb losses. It’s also a liquidity problem. But the liquidity problem isn’t resolved in either situation I outlined without an overdraft from the Fed. If you allow overdrafts from the Fed, then you can define it as resolved, but that contradicts the point of setting up a separate credit bank system.

    “You are correct that “capital” in the balance sheet sense does not represent liquidity. I should have said that the bank would make up losses first with “cash on hand”.”

    Capital and liquidity are distinct but related. The purpose of capital is to absorb unexpected losses. The purpose of liquidity is to meet cash flow obligations. It happens that capital as a balance sheet item is also a source of liquidity (as funding), but it’s not the only source of liquidity in a leveraged institution.

    “Every bank with sound management will have a cash “buffer””

    Agreed – but it won’t protect a bank in the worst form of liquidity crisis due to insolvency.

    “The bank will need to replenish the buffer by selling stock or retained cash flow.”

    Banks issue stock and generate earnings to replenish their capital positions first. That also improves liquidity as per my previous comment, but improvement of liquidity is not the primary purpose of capital replenishment. There is a difference between a buffer against liability obligations (liquidity) and a buffer against losses (capital).

    “How do we know this buffer will be big enough to cover losses in practice?”

    The liquidity cushion doesn’t cover losses. That’s what capital does.

    So in your actual example of a failing loan, the process will go…”

    It doesn’t go that way because I specified a liquidity crisis that caused depositors to stay away. The bank may actually be insolvent, which is a condition of negative equity. But it only needs to be perceived as insolvent for there to be a liquidity crisis.

    Under no circumstance can a “credit bank” run an overdraft at the Fed. As you correctly point out, this would defeat the entire point of the system.

    Agreed.

    “So I made a mistake in my original proposal in leaving out the part of the buffer.”

    I anticipated that by briefly noting the buffer scenario at the end (I should have spent more time on it.) The point is that in the worst case, a finite buffer cannot protect against a liquidity crisis that is driven by actual or feared insolvency. It can only buy time, but time will run out in the worst case. That’s the point about the distinction between liquidity and capital. So the credit bank must either go out of business or get an overdraft from the Fed. The latter is a contradiction because overdraft privileges aren’t provided.

    “There is no failure due to a liquidity crunch, only due to major mistakes in making bad loans.”

    The root cause of the liquidity crunch is real or perceived insolvency from a capital perspective (expected losses due to bad loans), but the failure mechanism – the death spiral – is the actual liquidity crunch and failure to repay liability holders.

    “If the economy was “pristine” from a credit risk perspective, i.e. no defaults or delinquencies, then there would be no need for a cash buffer account. In the real world, there will be substantial credit risk, and thus all sound banks will keep a cash buffer account.”

    My point was to demonstrate first a contradiction in the pristine case – that the credit bank required an overdraft. I then generalized it to a contradiction in the buffer case, where a sufficiently severe liquidity crisis always overpowers a finite buffer. I recognize that sound banks from a liquidity perspective carry a buffer, but that doesn’t prevent them from going down if real or perceived solvency (capital adequacy) is the underlying problem and that problem outlasts the finite liquidity buffer. That’s why a system with “credit banks” can’t exist without overdraft privileges or it simply goes down at the first instance of a liquidity crunch without the option of a temporary overdraft.

    “This is correct. In fact, a bank should never do this even as a temporary measure.”

    It’s finite by definition and therefore temporary by definition. No matter how large the buffer, it will not prevent a bank from going down due to the most severe solvency/capital real or perception problems. That’s generally how banks go down.

    The whole point of my piece was to present an example that juxtaposes a real/perceived credit problem against liquidity resources that are finite by definition – pristine or otherwise. The two finite cases are zero buffer (pristine maturity matching) or non-zero buffer (technically a form of favourable maturity mismatching). Both cases produce a contradiction in that the credit bank needs a Fed overdraft in order to avoid going down due to its liquidity crisis – it can’t repay its liability holders.

    Perception of credit problems is the cause of the liquidity crisis in the examples; exhaustion of finite liquidity is the mechanism. It always works that way in the real world, and it’s no different for credit banks as defined. That’s why the Austrian architecture fails. It can’t protect against the most severe liquidity crises, as in the real world.

    “But in the age of the internet and global finance, I’d expect that borrowers would have a wide variety of choices available.”

    Interesting point on information; but the internet information itself doesn’t produce the stockpile of liabilities that is required before lending.

    It’s a difficult subject. We may be talking past each other a bit. My general observation is that it’s important to understand that the purpose of capital is to absorb losses. Liquidity management is not about absorbing losses. It’s about meeting required cash flows. These are separate but related issues. Liquidity crises are generally about the perception and/or reality of balance sheet insolvency or expected balance sheet insolvency (i.e. negative capital), or at least enough of a capital risk problem that people want to get their money out and not take the risk of hanging around. Capital is also a source of liquidity in the sense of funding, but it is not the only source in a leveraged institution with liabilities. Those liabilities are not capital per se. They magnify the nature of a liquidity crisis, just as they magnify or leverage risk taking in the first place. Only when an institution is 100 per cent capital funded (i.e. equity) does a liquidity crisis blend perfectly into a capital crisis, because equity holders only expect to get the residual value of the firm out in the form of cash (i.e. liquidity).

  47. Bill-

    Thank you for the thoughtful response.

    But that is the point – the possibilities presented a national government under a gold standard are limiting and may not allow it to achieve true full employment.

    As I explained before, if you are running 100% reserve, gold standard, full employment is achievable if follow two rules 1) do not print more money that you have backed by gold at the fixed conversion rate and 2) if you break rule 1) devalue, devalue, devalue.

    If you run a fractional reserve system with a gold standard, then the central bank must follow Bagehot’s rules. The Bank of England followed Bagehot’s rules from 1870 to 1914 and had full employment and a tamer business cycle than we have now. When the Bank of England stopped following those rules, it ended in predictable disaster. When the government printed more money that it had gold, and then refused to devalue, it created ruin. It should have devalued in the 1920’s.

    Again, the gold standard is fine tool, that can be used to achieve full employment if used properly.

    At present, the policy makers pretend they have a gold standard – or some hybrid and act as though they are as constrained as they would be under a gold standard.

    Our problems today are analagous to those of the 20’s and 30’s. But the problem is not “gold standard” mindset. The problem is a refusal to devalue. Why is there a refusal to devalue? Popular outrage for one, and an ability to admit mistakes for second.

    That is why we are getting such unsatisfactory responses to the current crisis and also why most nations have endured very high levels of unemployment (relatively) over the last 30-35 years.

    Unemployment has certainly been a problem during recessions. And recessions certainly call for stimulus to maintain/restore employment. But I don’t see unemployment due to deficient aggregate demand as a chronic, ongoing problem. From 2003 to 2007, loose money led due to credit expansion led to pulling millions of workers from Mexico north into America to build houses.

    Dilution can only occur if what you can buy in real terms per $ is reduced. That is not a function of the stock of “liquidity” – but what the liquidity can purchase in real terms.

    This is simply incorrect. Dilution may be offset by an increased demand for money, or by productivity increases, but it’s still dilution. Dilution may be a good idea in some circumstances (such as an aggregate demand shock), but it’s still dilution.

    If you own stock in a company, and issues 3% more shares every year to payoff executives, do you consider this dilution “not to exist” if the economy also grows by 3% a year? If so, I have some stock to sell you … Like dilution of the money supply, dilution of a stock can be a good idea in some circumstances (such as to raise money from venture capitalists), but it’s still dilution.

    In economies with high degrees of capacity underutilisation and unemployment, increase public deficits can actually increase the amount of real goods and services at the disposal of the individual (for example, giving the unemployed a job) without reducing the real capacity per % held by others.

    This is true. But keep in mind the government doesn’t just dilute the currency during recessions. By any measure you choose, there was a very high rate of dilution during the 2003-2008 period. This dilution is an ongoing transfer of resources from the holders of government paper to the government’s favored constiuencies. A five year treasury bill bought in 2003 actually lost money in real terms, because the dilution rate was so high.

    Your last paragraph also confuses the role of taxation.

    It is correct to describe the system as one where government creates money to spend, and taxes to sterilize. You can also model the system as one where the Fed creates money, and the Treasury taxes to fund government operations. Technically, this is what happens, and this is how the government iteslf describes its operations. Either way, it’s a fiat currency, and in no theoretical sense is spending in dollars constrained by taxes.

    But the overall point, is that when the government extends credits ( or rather, gives banks a license to extend credit) real resources are transferred from the existing holders of money and government paper, to the people receiving the loans.

    If the point of the 100-percent reserve banking system is to reduce bank losses then I fail to see how it does that unless it prohibits credit creation at all.

    The point is to eliminate liquidity crises and bank runs. Not to eliminate credit risk. Credit risk will still be present. Almost all the major crises, from the panic of 1819 to the Great Depression to the run on the money markets last september have had maturity transformation at their heart.

    Widespread failures (defaults) are still possible which just means that at some point in the future (when the fixed-term deposits expire) the “bank run” occurs.

    A bank run is a totally different phenomena than a default. If there are widespread defaults in the maturity matched scenario, there is no bank run. When the desposits expire, the holder simply gets $.95 on the dollar (or whatever the losses were). A bank run is when a bank has made a lot of promises to redeem notes on demand, but does not have the liquid reserves to meet those obligations. As a result there is a race to the bank to get their hands on the last bit of cash. When it’s gone, the holders of desposits often end up with pennies on the dollar, which is why bank runs/maturity mismatching crises have been historically much more of a problem than default crises. This point gets obscured though, because a bank run that wipes out desposits will often result in defaults (the bank run causes banks to fail and the money supply to contract, the contracting money supply causes aggregate expenditures to fall, thus income falls, thus businesses cannot pay back loans).

    A bank run can occur even in the complete absence of any defaults. In most practical cases, defaults trigger the run, but the losses from the run far, far exceed the losses from the default. To understand this, there are number of articles, try here or here or read up on the here or read up on the Diamond-Dybvig model.

    In that sense, I don’t see the point of it and so I start thinking about the other motives that are behind the suggestion – the Austrian “sound money” motives – which in my view disable the capacity of the national government to pursue and achieve public purpose – full employment, equity in opportunity, environmentally sustainable growth, and price stability.

    Rothbard has a whole framework of his definition economic liberty, and its from this framework that he defines his idea of what banking should be. He often ignores evidence that does not fit his pre-existing framework. For instance, he does not recognize that once a deflation happens the government should step in and stop it.

    Other libertarians have a public choice perspective. They do not believe the government ever acts in the public interest, so they want it do as little as possible. When you look at the current response to the financial crisis, can you really fault them for this view?

    As for me, I note that markets are themselves creation of the government courts, so the whole free market versus government dichotomy never made sense to me. The government will do what the government does, I have no control over it. I prefer to analyze proposals from an intellectual stand point and figure out what they get right and what they get wrong.

  48. JKH-

    I think we are using different definitions of the word “liquidity”. When you say, “but it won’t protect a bank in the worst form of liquidity crisis due to insolvency.” you are using a different definition of the word “liquidity” than I was. Frankly, economists have abused the word “liquidity” to within an inch of its life. Keynes basically completely redefined the word in his general theory. It may be more fruitful to continue discussion without using the word, so that we don’t cross signals.

    It doesn’t go that way because I specified a liquidity crisis that caused depositors to stay away. The bank may actually be insolvent, which is a condition of negative equity. But it only needs to be perceived as insolvent for there to be a liquidity crisis.

    This sentence I do not understand. How does a perceived insolvency prevent the bank from meeting its cash flow obligations to people redeeming CD’s and deposits? Terms are matched, so as long as the loans are repayed, the bank has funds to redeem CDs. If loans do not get repaid, that’s not a perceived crisis, that’s an actual one crisis. Perhaps an example would help? (It may also be that we are using different definitions of the word insolvent. My definition of the insolvent is that a bank is insolvent if it does not have the cash to pay off people dedeeming desposits. My banking system does not use any sort of mark-to-market accounting, and so the entire concept of “negative equity” has no meaning. It might be more fruitful to imagine these banks using a temporal based, cash based accounting system, no such accounting system currently exists, but it would have to be invented for the above banking system to work).

    The root cause of the liquidity crunch is real or perceived insolvency from a capital perspective (expected losses due to bad loans), but the failure mechanism – the death spiral – is the actual liquidity crunch and failure to repay liability holders.

    I can certainly see how actual losses due to bad loans can cause a failure to repay depositors redeeming. I cannot see how simply “perceived” losses can cause a failure to repay despositors.

    Both cases produce a contradiction in that the credit bank needs a Fed overdraft in order to avoid going down due to its liquidity crisis – it can’t repay its liability holders.

    In my system, banks will fail if the losses on their loans exceed it’s cash on hand and ability to brorrow against future profits. Failure will hopefully be rare, but it will happen. Having some failure keeps any system healthy, it purges the banks with bad lending practices.

    That’s why the Austrian architecture fails. It can’t protect against the most severe liquidity crises, as in the real world.

    I think deserves longer comment, but I’ll wait for your responses above. The point of the Austrian scheme is to elminate crises due to bank runs/maturity transformation in way that does not introduce moral hazard and regulated lending. The Austrian scheme does not protect against all losses everywhere. It does not protect against banks failing and despositors losing money.
    It the Austrian view that losses due to bank runs/maturity transformation have been responsible for all major financial crises (not losses due to bad loans), and thus that is the problem to solve. Solve that problem without creating moral hazard, and the banking system will be stable and productive.

  49. Devin,

    I’m defining solvency as balance sheet solvency – positive capital or positive equity – which in my view is the correct definition. Cash flow matching does not define solvency. Solvency is defined as to whether there is any capital left to absorb more losses. Liquidity and solvency are different characteristics.

    “How does a perceived insolvency prevent the bank from meeting its cash flow obligations to people redeeming CD’s and deposits? … I cannot see how simply “perceived” losses can cause a failure to repay depositors.”

    This is fundamental to any liquidity crisis. I’ve described it several times in examples. In the case of a credit bank, a matched maturity loan goes bad. The bank can’t repay the depositor. Other depositors fear balance sheet insolvency. The bank can’t repay the first depositor by attracting a second one. Therefore the bank needs an overdraft or goes down.

    Negative equity doesn’t depend on marked to market accounting. It happens when the book value of liabilities exceeds the book value of assets. That can happen in the future and can be reflected as an expectation today.

    “The point of the Austrian scheme is to …”

    What I’ve depicted several times is that maturity matching alone can’t prevent a liquidity crisis. In fact, you need “inverse” maturity mismatch (liability duration longer than asset duration) in order to build a cash buffer.

    Lost in my comments perhaps is that I agree that maturity matching reduces liquidity risk. But it doesn’t prevent a bank going down if the market is determined to bring it down because of fears of insolvency. The only alternative is a Fed overdraft facility, which again is a contradiction.

  50. This is what the Austrian school is worried about …

    Taken from Rothbard’s The Mystery of Banking

    On the contrary, its liabilities-its warehouse receipts-are due instantly, on demand, while its outstanding loans to debtors are inevitably available only after some time period, short or long as the case may be. A bank’s assets are always “longer” than its liabilities, which are instantaneous. Put another way, a bank is always inherently bankrupt, and would actually become so if its depositors all woke up to the fact that the money they believe to be available on demand is actually not there.

  51. Very interesting commentary, taking on the structural question of FR, focused on the Austrian model proposed by Devin. While I appreciate all of these comments and a hard look at FR’s inner workings, I think there are larger cohort issues that just aren’t being addressed. Full-reserve banking doesn’t stand on its own. It can be either related to the Austrian approach or the Chicago Plan approach, IMHO.

    Bill said originally that he didn’t see FR working without government creating all the money, a la Henry Simons and the Chicago Plan. I never saw an Austrian model that embraced the Chicago Plan approach of government money-creation, so the tale being told here avoids that larger question.
    Scott has weighed in that FR doesn’t come close to solving the liquidity crisis, in and of itself.
    I think that is mainly because there is no provision for liquidity in FR banking.
    Given the Austrian or Chicago Plan approach, somebody has to FIRST create the new money.
    I agree with Ralphonomics on where the money-creation powers do and ought to lie.

    MMTers know that the Chicago Plan was a progressive idea that adopted the oft-debated FR concept as a part of the structure for monetary stability, that was in turn necessary to promote economic stability. My own position is that FR is superior to fractional-reserve banking, even just for its counter-cyclical qualities.

    Bill linked up with Ronnie Phillips paper on the Chicago Plan, certainly not a libertarian or Austrian proposal.
    So, to me, I can’t grasp the minutae of the FR discussion as it is presented in this fine repartee.
    I ask Bill and the other MMT community to consider whether the Chicago Plan for Monetary Reform is or is not a bonafide that ought to come up to the plate, and that FR banking only be considered for its contribution to such a comprehensive reform.
    Make that a plea for same.

    I believe it was the learned JKH who scoffed at such a radical approach(FR), given that we can’t even get Glass-Steagall to the fore. My position on all that is this.
    We ain’t there yet.
    The debt-money system is broken and broke.
    There has been no official recognition of that and as a result, no substantive proposals for reform.
    Remember the Chicago Plan proposal came out more than 3 years after the ’29 crash.
    And to me, we yet await the crash.
    MMT is definitely a partial fix for what ails us as its foundation is monetary sovereignty.
    But I am left with the question of whether a reformed Fed providing reserves to private bankers is superior to the Chicago Plan or that proposed by Douglas, Fisher, Graham et al in their 1939(?) paper – A Program for Monetary Reform.

    Bill, thanks so much for this whole discussion.

  52. It’s fair to point out that The Austrians don’t have a monopoly on the subject of maturity transformation. Maturity matching is not an Austrian mystery per se. The Austrian prescription in fact is simply a pure case of a technique that is commonly employed by banks today in varying degrees. Because it’s presented as a pure case in the Austrian version, or pure principle, it’s fair to ask the question what the institutional implications are. Moreover, when the hard questions are asked about implementation of the pure case, it turns out that the response is one of compromise, because the pure case is not viable. It then becomes clear that the pure case as presented is irrelevant, because it is not viable, and because real world banks already implement maturity matching to varying degrees. The fact is that real world banks implement both maturity matching and maturity transformation in different parts of the asset liability portfolio.

    As I illustrated in an earlier example, the pure case of maturity matching is conceptually dangerous from a risk management perspective. If a bank were perfectly matched to maturity, it fails when the first bad loan matures. It simply has no cash to repay the corresponding maturity liability, unless it has an overdraft privilege. Bad loans are matched to equity losses. They are no longer matched to maturing liabilities, because they can no longer produce the cash flow that is required for a perfect match.

    This is a matter of cash flow observation, as well as accounting for balance sheet equity. Perhaps Austrians are not as familiar with double entry book keeping as they should be. In order to protect against this contingent mismatch risk in the pure model, it is necessary to deviate from it by constructing a deliberate “inverse maturity mismatch” at the margin. This means that term liabilities must be mismatched against short term risk free assets. As described earlier, this means that the credit bank must build up a cash reserve with the depository for 100 per cent reserved deposits in the other part of the model. In other words, the pure Austrian matching model fails, unless it is tweaked to become a mismatched model.

    And there is another critical area where the pure model of maturity matching fails. All banks that take risk, including the credit bank described in this thread, require equity capital. Like liabilities, equity capital is a source of funds. But unlike liabilities, equity capital has no nominal maturity date. Therefore, it is not possible to incorporate equity capital into a coherent maturity schedule for the entire bank, without deviating from the simply maturity matching formula of the Austrian prescription. Real world commercial and universal banks know this and allow for it.

    All this results from the portfolio effect of liquidity (the cash deposit with the 100 per cent reserved depository, plus other liquid assets) and capital (the equity that is intended to absorb losses in the credit risk bank). Commercial and universal banks are quite aware of such an interaction in their asset liability management. The functions of liquidity management (which includes the cash reserve and other liquid assets) and capital management (the cushion against losses) are quite separate (although partially intersecting) in these real world banks. The purpose of liquidity management is to protect the cash flow profile of the bank. The purpose of capital management is to protect the loss profile. The primary intersection of the two is that equity is a source of funding that doesn’t have to repaid with cash flow, unlike a liability. Therefore, in addition to providing loss protection, it improves the liquidity profile by alleviating cash flow pressures otherwise attributable to maturing liabilities.

    Commercial and universal banks are quite aware of these issues and typically have risk policies to deal with them. The fact is that no amount of maturity matching or other types of liquidity policies can protect against the ultimate risk of catastrophic capital loss. Catastrophic capital loss is not primarily related to liquidity; it is primarily a function of other types of risks taken – such as credit risk. There is no way that maturity matching can provide ultimate protection against such losses. Maturity matching can’t prevent the fact that some depositors won’t get their money back in the worst case for capital. It can only delay the day of reckoning in such a case. One gets the idea that the Austrians believe they’re the only ones that recognize the importance of the issue. I think it’s much closer to the truth that those who do recognize the issue reject the Austrian approach because their pure model is a naive treatment of a complex issue.

  53. JKH . . . all VERY well said, as usual. A few years ago (or more recently? can’t recall exactly) on Warren’s blog this topic was taken up and I also noted there what I’ve noted in the “when you’ve got friends like this” comments (can’t do FR w/o overdrafts or liquidity crisis) and also that maturity matching doesn’t help. Nobody took my side then. I’m glad JKH took the time to demonstrate this in detail here.

    Also, as I noted in “when you’ve got friends like this” comments and JKH and Joebhed have explained in more detail here, there are many different versions of FR. Minsky/Phillips and now (at least as of Oct. 2008 when I spent 3 days discussing re-regulating the banking system with him) Kregel support FR banking. HOWEVER, they recognize, consistent with JKH, Bill, me, and others, that this does not stop credit creation in which loans simulataneously create liabilities for the financial institution. Kregel at least sees the role of FR banking as only for the purpose of establishing savings/deposit accounts while doing away with deposit insurance. I’m agnostic on this point in theory, but would move alongside Bill and Randy in opposing if the effect is to eliminate the community banks. The broader point is that you really can’t compare the approach of Minsky/Kregel here with the proposals of Ralph and Joebhed–since the latter want to do away with private credit creation, the proposals are like night and day. The views of Minsky/Kregel are far closer to Bill, Randy, JKH, and me; in fact, they’re almost identical aside from the issue of deposit insurance vs. fully backed savings/deposits.

    Also, Joebhed’s suggestion that “FR is superior to fractional reserve banking” is not persuasive (to me at least), since real-world monetary systems aren’t fractional reserve systems anyway, as we’ve gone over many times before. I’ve also seen no convincing evidence that “the debt money system is broke and broken” beyond poorly designed regulation (and if not poorly designed, poorly implemented by regulators that have ideological biases against regulation) of said system. Warren, Randy, Bill, Bill Black, Eric Tymoigne and others in the MMT camp have laid out numerous proposals for re-regulating that we would argue are consistent with financial stability and continued existence of private money creation.

    I’m well aware that this won’t convince Joebhed or Ralph. I would, on the other hand, be interested in Joebhed, Ralph, or others who believe their version of FR is “progressive” could spell out in some detail how they think the system would work in combination with government money (or provide a link, perhaps, as maybe it’s already been done). Apologies if this request/suggestion appears naive. Until that happens, I think much of the “it will work–no it won’t” back and forth is about as far as we’ll get (assuming we want to get farther . . . some probably don’t, but it would be of interest to me at least).

    Best,
    Scott

  54. Scott/JKH,

    Not sure of a few things- though I think I understand Bill’s position completely. There are two scenarios:

    In Bill’s recommendation of zero interest rates and no debt issuance by the government, high amount of reserves occur naturally and not out of reserve requirements. Using “non-reservable liabilities” we can have the situation of 100% reserve requirement theoretically, though I don’t know what purpose the latter solves.

    In the scenario which actually is the world we live in, and where the government issues debt for every dollar of deficit spending, 100% RR is impossible. Maybe possible to demonstrate it on a black board, but I would imagine interest rates (on bank loans) hitting the roof in this arrangement.

    Scott, “when you’ve got friends like this” post was last week 🙂

  55. Scott,

    Thanks. The Austrian idea for 100 per cent reserves is effectively more about macro credit risk than it is about the holy grail of liquidity management, as the Austrians intend it. For example, the idea of segregating deposits against 100 per cent fiat reserves as per the earlier discussion in this thread makes them risk free from a credit standpoint. That combined with the government’s currency issuance powers makes them risk free from a liquidity perspective as well. It’s really a cash version of FDIC insurance, which is a derivative rather than a cash structure (i.e. a put option). (The step from fiat reserves to gold reserves is an additional dial, I think.)

    Conversely, credit risk is contained in a separate institutional structure, where the liquidity management dial is turned up full throttle to perfect matching (before compromises to the principle begin). But that Austrian matching discipline won’t necessarily prevent a catastrophic event, during which liabilities exceed assets at some point during a crisis.

    All liquidity crises are driven at the core by some form of non-liquidity risk (real or perceived), such as credit risk or interest rate risk. Credit risk as a core example is the origin of the crisis; liquidity is the transmission mechanism whereby the perceived equity situation dooms the normal functioning for the flow of funds (liquidity). The degree of maturity matching only buys time before an eventual failure in this case.

  56. I would, on the other hand, be interested in Joebhed, Ralph, or others who believe their version of FR is “progressive” could spell out in some detail how they think the system would work in combination with government money

    I’ll chime in. In my ideal world, here is the bank balance sheet:

    Have the liability side of banks to the private sector consist solely of bank capital (e.g. equity and possibly certain types of long term subordinated debt). Bank liabilities to the government are loanable funds lent from the CB at a government set rate. When a borrower wishes to borrow, the bank qualifies the borrower, sets aside risk capital according to some standard algorithm, and obtains all loanable funds form the CB. As the debtor repays the bank, the bank repays the CB with a regulated rate, earning money from the spread. There is no liquidity risk, only credit risk. Banks still can (and must) fail when loans are not repaid. No FDIC guarantees for any bank asset.

    Depositor services can be supplied by other institutions in exchange for fees — or the government can supply a minimal set of these services in the same way as it funds book depositories 🙂

    Here are advantages as I see them:

    * eliminate dangerous and wasteful short-term funding markets for banks, as banks have only equity liabilities to the private sector.

    * makes it harder for banks to mislead regulators about loan performance and liabilities.

    * makes bank funding costs much less dependent on the business cycle and market sentiment. In this case bank leverage allows a sharp increase in bank capital costs to translate into an insignificant increase in interest rates charged, effectively shielding banks from the private credit market interest rates.

    * No need to pay interest on bank reserves, or to sell bonds to drain reserves, as there are no bank reserves. Allows the government to separate the issue of managing the size of bank loans from managing the size of the monetary base

    * Forces banks to earn money from credit analysis rather than form yield curve arbitrage — this will add much more stability to the system over time, by changing banks’ operational focus.

    * Private entities (banks) with their own money on the line (bank capital) still determine whether someone gets a loan or not — you still have private credit creation.

    I would add:

    * LBOs, forex markets and the like can be handled by investment banks that should not be able to accept FDIC insured deposits.

    * Regulation should also shield the private credit markets from bank funding costs (e.g. only certain types of small business or consumer vanilla loans are allowed, with strict credit limits on the amount each entity can borrow). This is the main leakage (together with allowing banks to participate in the credit markets) that the bank regulation proposals that I’ve seen fail to address, so you still exacerbate the credit cycle by having banks earn arbitrage between government set funding costs and market set return expectations. As long as there is this arbitrage opportunity, you will have exaggerated business cycles and rentier profits in the banking industry. Of course these rentier profits are exacerbated if you pay banks interest on reserves.

    * The focus on credit-analysis, by removing yield curve arbitrage as a profit source, will advantage smaller banks and banks that lend to their own depositors, and this will in and of itself encourage banks to provide depositor services in an effort to datamine and then sell loans to depositors.

  57. I’m defining solvency as balance sheet solvency – positive capital or positive equity – which in my view is the correct definition. Cash flow matching does not define solvency.

    I was using the dictionary definition: “the ability to meet maturing obligations as they come due”. I don’t really care which definition is correct, just that we use the same definition, or else it’s confusing.

    I then ask JKH: “How does a perceived insolvency prevent the bank from meeting its cash flow obligations to people redeeming CD’s and deposits? I cannot see how simply “perceived” losses can cause a failure to repay depositors”

    JKH responds: “In the case of a credit bank, a matched maturity loan goes bad. The bank can’t repay the depositor.”

    But this crisis is not a result of “perceived” losses. It’s a problem of actual losses – the bank made a bad loan. If a bank makes a bad loan, it obviously only has two choices a) cover the loss itself, out of its own cash (or by borrowing against its own assets) or b) pass on the loan to the customer.

    And actual, the bank is not going to be matching one loan to one customer. If it was just doing that, there would be no point to the bank, the customer could just hold the loan in his own name. The point of bank is to pool risk. So it would pool 100 loans at 7% interest to 100 CD’s at 5% interest, and expect a 1% default rate. So if a few loans go bad no problem – that’s built in, it just takes a bite out of the banks profits. No mismatch or cash buffer needed.

    But yeah, if you match one loan to one CD, and have no cash buffer, that bank is going to be very, very fragile. I wouldn’t call that the “pure” version, that’s just an unsound bank. Austrian banking, broadly speaking, is way for designing banks that are very stable without the need for any lender of last resort. Any design that fails at that is a bad design.

    What I’ve depicted several times is that maturity matching alone can’t prevent a liquidity crisis.

    By your definition of the word “liquidity crisis” I agree. But it will prevent the worse kinds of liquidity crisis. All the worst liquidity crises, from the panic of 1819 to the Great Depression to the run on the Shadow banking sector in 2008 were due to maturity transformation.

    Lost in my comments perhaps is that I agree that maturity matching reduces liquidity risk.

    The question then becomes does it reduce risk enough that the banking system is no longer a danger to the economy as a whole? If so then you no longer need an overdraft facility, and you no longer need centralized regulations of all loans. I think that the Austrian scheme would reduce the risk great enough to make the banking system extremely stable. The reason is that in practice banks would pool loans and have a buffer. And since each bank is making it’s own loan decisions independently, I would not expect to have huge correlated mistakes in lending.

    Perhaps Austrians are not as familiar with double entry book keeping as they should be.

    I am likely guilty as charged here. But I strongly suspect that my Austrian bank would use nothing like the book keeping banks use today. Austrian banks use temporal accounting. Unfortunately, I have not explained this, and it really could have it’s own long discussion. Perhaps some other time.

    This means that term liabilities must be mismatched against short term risk free assets.

    I would not actual label this a “maturity mismatch” since the short term cash buffer is held in the bank’s own name and own account, it’s not actually matched against and CD libability. But I don’t want to argue the semantics. If you want, we can call it 105% reserve banking 🙂 The point is not to argue over whether the system is “pure”, but whether it is stable when all proper safeguards in place.

    Like liabilities, equity capital is a source of funds. But unlike liabilities, equity capital has no nominal maturity date.

    Pardon my ignorance, but then what is equity capital? Cash? Bonds? Stock? What does it represent? If there is a link that explains it well, I can read that. Is this something I can see on a balance sheet ( http://finance.yahoo.com/q/bs?s=BAC&annual )? What good is capital at absorbing losses if it can’t actual absorb the losses (ie be used to pay off depositors)?

    When I talk about a “cash buffer”, I’m talking about the cash item on the balance sheet. If the bank makes too many bad loans, and cannot pay off depositors, it would come out of the cash. If doesn’t have cash, it has to liquidate one of its other assets fast, or borrow from a third party, or else pass on the losses to the customer.

    Catastrophic capital loss is not primarily related to liquidity; it is primarily a function of other types of risks taken – such as credit risk. There is no way that maturity matching can provide ultimate protection against such losses. Maturity matching can’t prevent the fact that some depositors won’t get their money back in the worst case for capital.

    Agreed.

    Credit risk as a core example is the origin of the crisis; liquidity is the transmission mechanism whereby the perceived equity situation dooms the normal functioning for the flow of funds (liquidity).

    Here is the rub. Austrians do not believe that credit risk is the core origin of depressions. Depressions do tend to cause massive defaults, but they are not the cause. The cause of depressions is when a maturity mismatched structure collapses. So fix the maturity mismatching problem, you won’t have depressions.

    The key question to answer is: why did so many deposit holders lose their money in the Great Depression? There are two main candidates for answers:

    a) banks systematically made bad credit decisions

    b) a system of maturity mismatching broke down and a bubble in the price of 30 year money collapsed (a bank run is an extreme form of interest rate risk)

    If you believe a), then the Austrian scheme is a fix to the wrong problem. It doesn’t fix the real problem. If you believe b), the Austrian obsession makes a lot more sense. FDIC insurance and a lender of last resort is also a fix for b). But it requires centralized regulation of all loans. But Austrians, true to their libertarian roots, desire a scheme that allows for distributed decision making, whereby every bank is responsible for its own lending decisions.

  58. Remember …

    Exports create deposits

    Capital inflows create deposits !

    FX trading creates deposits!

    Banks buying back stocks create deposits

    Banks paying interests and the principal on their debt creates deposits

    Banks paying their employees creates deposits

    Banks buying a building/paying rent/buying computers etc. create deposits

    Bank proprietary trading creates deposits

    Banks distributing profits creates deposits and so do interest on deposits

    So the only solutions are : Continue the way it is OR set rates to zero and impose (preferably zero) reserve requirements 😉 (and implement proposals on regulations)

  59. @JKH/Devin

    It appears from here that the spot where you are disconnecting is the contractual status of the liability to the depositor. In a deposit bank, the deposits are demand deposits i.e. they can be retrieved by the depositor in full at anytime. It sounds to me like JKH is thinking that credit bank deposits a like demand deposits but with a time lag and interest, in which case I would agree, such a system could not work. As I understand it, the credit bank would be like a public investment fund. Money deposited in the fund attracts interest but both the interest and the principal are at risk (specifically credit risk). Consider the previous example:

    Bank is making loans @ 7% and paying depositors 5%
    Bank has $100 in deposits and has issued it all in loans.
    If all the loans are repaid normally, $105 is repaid to depositors and bank keeps $2 profit.
    If 1% of loans default, $105 is repaid to depositors and bank keeps $0.93 profit.
    If 5% of loans default, the bank keeps nothing and depositors get $101.65 repaid.
    If 10% of loans default, the bank keeps nothing and depositors get $93 repaid (they lose $7)

    There’s no liquidity risk here because the if the loan income is impaired, so are the liabilities to depositors. They can’t make a run on the bank, because it’s a term deposit, and the value of their investment at the end of the term is directly correlated to the performance of the loans during that time (so the money to repay is always there).

    No banking system can have guaranteed returns on investment AND be free of liquidity risk. One of them must be sacrificed.

  60. I think that you have missed the discussion on 100% reserve banking and narrow banking that went on in the UK. Most notably by John Kay, but also in the FT (Martin Wolf) and the Bank of England commissioned Kay to write a paper to make his case.

    There seems to be a desire to raise the reserve requirements to separate the financial casino of the City of London from the daily banking needs, because the UK has a very oversized financial sector in comparison to GDP. Similar reasons are now driving the Netherlands and Switzerland to alter regulation and push up reserve requirements.

    I did not see those considerations discussed.

  61. I would like to thank Devin as through his comments I finally got a grip on 100-reserve banking and how it could actually be a working one.

    The only weak link remaining imo is the big D, deflation. I think gold standard will lead to deflation almost by definition, even if often counter-cyclical way, and that will heavily disturb the economy. It seems to me that there are good theoretical arguments why the deflation is a self feeding disaster. Moreover deflation will always redistribute the wealth just like inflation. For me it looks that deflation is worse as it redistributes wealth mainly to rich and to those not taking risks or investing.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top