Among the many interesting comments on my previous post about free banking was one by Philippe which provoked an extended (and perhaps still ongoing) exchange between Philippe and George Selgin. Referring to this assertion of mine,
if free banking were adopted without abolishing existing fiat currencies and legal tender laws, there is almost no chance that, as Hayek argued, new privately established monetary units would arise to displace the existing fiat currencies
Philippe made the following comment:
In “The Denationalization of Money” Hayek argued that you should be able to pay taxes with privately-issued currencies. However, this would in effect turn those ‘private currencies into’ de facto state currencies, or forms of government ‘fiat money’. For some reason Hayek chose to ignore this massive contradiction in his argument. Essentially, what he was actually arguing was that private corporations should be granted special state powers, i.e. the power to issue money backed by the state’s legal powers of taxation.
I thought that this was a very insightful observation by Philippe, though its significance for me may be somewhat different from its significance for Philippe. In my criticisms of Hayek’s free-banking position – I call it a free-banking position even though free banking may be a misnomer inasmuch as Hayek advocated banks’ creating new currency units not just allowing banks freedom to create a complete menu of liabilities denominated in existing currency units – my argument was that newly created currency units would be worthless unless the banks made them convertible into some outside asset not under their control. Or in Nick Rowe’s helpful terminology, Hayek advocated free-alpha-banking, while conventional free bankers advocate free-beta-banking. The reasoning behind my argument is that the value of a pure medium of exchange depends entirely on its expected value in exchange, so if a currency unit is not defined in terms of a commodity providing a real, valuable, service apart from being used as money, people will eventually realize that its value must go to zero. Any positive value that it may temporarily have is just a bubble, and, like every bubble, it will burst.
However, unlike the private issuer of a currency unit, a sovereign issuer can impart a real value to a fiat currency by making the currency acceptable for discharging tax liabilities, creating a real demand for the currency distinct from its use as a medium of exchange. Using this argument, I have suggested that bitcoins are a bubble, though it is possible that are some techie reasons that I don’t understand why bitcoins could provide real services that would allow them retain a positive value. At any rate, the point made by Philippe — that if governments were to accept newly created private currency units in payment of taxes – they could retain their value just as fiat currencies issued by governments do – is a point that had escaped me in my criticisms of Hayek. So, well done, Philippe.
However, Philippe seems to carry this valid point a bit too far, accusing Hayek of a massive contradiction in arguing that private corporations be granted special state powers. The problem with that argument is that it begs the question what is special about money that confers the sole power to issue money to the state. I actually once wrote a paper trying to answer that question (once again relying on an argument that I heard from Earl Thompson) published in a volume called Money and the Nation State. I summarized the argument in chapter 2 of Free Banking and Monetary Reform.
The short answer is that currency debasement may be necessary as a means of emergency taxation when a sovereign is faced with a hostile military force threatening its survival. To profitably debase a currency, you have to be the monopoly supplier. Ergo, sovereigns that monopolize the mint or the supply of currency have a better chance of surviving than sovereigns that don’t. Hayek was a staunch anti-communist, cold warrior, so the defense argument, at least on some level, would have appealed to him. But otherwise, it’s not clear to me why Hayek could not have said that, apart from certain national-defense functions, there really are no government services that may not be provided by private enterprise. After all, we do allow various services that were once exclusively provided by the government to be provided by private enterprise. I don’t say that this is always a good thing, but it doesn’t seem to me inherently unreasonable to believe that the burden of proof is on the one claiming that the state has an exclusive right to discharge a particular service, not on the one who questions that such an exclusive right exists.
Having said that, I will also say that it also seems perfectly reasonable for a government to say that a tax obligation that it legitimately imposes – I freely admit that I am now begging the question where this legitimate power comes from, but only libertarian fanatics dispute that power – can only be discharged in terms of a currency unit that the government itself specifies. And implicitly or explicitly, George Selgin and other free bankers — i.e., free-beta-bankers — seem to be perfectly OK with the government specifying the currency unit in terms of which tax obligations may be discharged. In other words, the government may impose a tax obligation on me that is specified in dollar terms. To discharge it, I have no choice but to pay the government the requisite number of dollars, either delivering the government’s own currency or delivering a private (beta-bank) money denominated in dollar terms.
The peculiarity in Hayek’s argument is that he was proposing that governments impose a tax liability in, say, dollar terms, and then accept payment in some other currency unit without specifying any method by which an obligation specified in dollars would be discharged in terms of another currency unit. Any creditor is free to specify at the time an obligation is created the terms on which the debt will be discharged (subject of course to legal tender laws, but for purposes of this discussion I am ignoring legal tender laws which are not the same as tax acceptance). The government is not just any creditor, but there doesn’t seem to me to be any compelling reason why a government should not be entitled to say we have created this obligation in terms of dollars and it must be discharged in terms of dollars. And, if I am right in asserting that acceptability in payment of taxes is a necessary condition for an inconvertible fiat money to retain value, there does seem to be something funny about Hayek’s argument for the creation of private fiat moneys, even if it is not a flat-out contradiction as Philippe claims.
What is funny is the degree to which the viability of a Hayekian private fiat currency is dependent on its being accepted by the state as payment for taxes. Moreover, Hayek’s argument was that there would be a discovery process in which many competing currencies would vie for acceptance with the market eventually choosing one or a few currency units as somehow being the most desirable. Hayek thought that the currency unit with the most stable value would eventually capture the largest market share. There are lots of problems with the argument, especially that it ignores the network effects that tend to produce an entrenched monopoly, and the extreme path dependence of such outcomes, but on a practical level, it seems almost unimaginable that a government would, or could, allow any number of distinct competing currency units to be simultaneously acceptable in payment of taxes.
I do not mean to be overly critical of Hayek, for whom I always have had the greatest admiration, but he had an unfortunate tendency to get carried away with certain utopian ideas and proposals, for example his idea of separating the law-making power from the governing function of parliaments into two distinct bodies, going so far as to propose a method for selecting members of the law-making body under which people at the age of 35 would each year elect a number of their contemporaries to serve a 15-year term in the law-making body, the law-making body being composed entirely of people between the ages of 35 and 50. He presents the idea in volume 3 of Law, Legislation and Liberty, a wonderful book of great philosophical depth and erudition. But it is amazing that Hayek felt that such an idea could ever be implemented. I don’t like to think so, but it occurs to me that his toleration for certain dictators might have had something to do with his imagining that they could be persuaded to implement his ideas for political and constitutional reform. His Denationalization of Money was a similar flight of fancy, based on some profound insights, but used as the basis for practical proposals that were fantastically unrealistic.
Uneasy Money 
There are some interesting arguments relative to this in a quite neat volume by Arthur Nussbaum in his book A History of the Dollar (1957). They’re interesting because there was lots of monetary innovation in the 13 colonies, and the British took varying approaches to it. They closed down a Massachusetts mint on the grounds that it usurped the prerogative of the crown. But all 13 colonies created bills of credit that effectively circulated as money for significant periods of time, and in some cases, they specified a ratio between sterling and the bills. Well worth a look.
LikeLike
“The problem with that argument is that it begs the question what is special about money that confers the sole power to issue money to the state.”
State doesn’t have sole monetary issuance power, it simply has control over what it will accept for tax payment…… as it should. It is best to understand our monetary system as a hierarchy with State dollars sitting at the top. Something has to sit at the top but I think most think it is better that what ever sits at the top stays there for a while and its not a weekly or daily battle for superiority.
What would we gain by having a competition amongst moneys for top dog?
Why do we want to hold the money which sits at the top of the pyramid? Is it because it is redeemable for the most gold…… most oil……buys the best Army?
A good question to ask is why is the money which sits at the top of the pyramid (currently) the one that is used for tax payments and which comes first, its position on the pyramid making it the best thing to use for assessing taxes or its use as the thing for paying taxes being responsible for its position on the pyramid?
LikeLike
Legal tender laws appear to me to be a red herring. I have never heard of anyone being charged by them. Here in Canada, it is probably easier to spend U.S. dollars in retail establishments than it is to spend Canadian $50 and $100 bills. Even though the stores post signs announcing their policies, nobody is ever charged.
The reason why legal tender laws are redundant are income and value-added taxes. All transactions have to be translated into national currency units in order to be taxed, and taxes withheld.
If someone pays me in hypothetical private Canadian dollars, but I have to pay taxes using another type of dollars, the first type of dollars had better be worth par versus the type I am paying tax in.
And if the government accepted private money at par, it is effectively treating the issuing entity as having no credit risk. This is a subsidy. The government better have done some credit analysis before doing so. Banking regulations are just how the government does that credit analysis.
LikeLike
The flaw in both Phillipe’s and your argument is to assume that the taxing authority would need to set a fixed tax liability in terms of a private currency and then accept that currency at a fixed rate. For example, people owe a 10 ducat tax and 10 one ducat notes, issued by the ducat issuing private bank, will always settle that tax obligation.
Most importantly, the tax authority could accept a variety of these private fiat currencies to pay taxes, but only at the current exchange rate. Once received, the government could sell them for whatever currency it wanted to hold and spend. Is it really that hard to imagine that the U.S. government would let people pay taxes in euros or pesos at the current exchange rate?
It would be possible to require taxpayers to do their accounting in one currency, and then let them pay the taxes in whatever currency they want.
If people were given the right to use general rules to calculate their tax liability using whatever fiat money they want, then either the tax system would have to be pretty simple, like a flat tax on consumption or wages, or else indexing would be needed as part of the rules. That is certainly feasible.
I think that letting people use what currency they want to pay taxes provides very little benefit to any particular bank. That is because people can still abandon it for its rivals. If the value of a bank’s currency falls to zero, then it cannot be used to cover any tax liability.
If you recollect the BFH system, and consider payment of taxes, some of these issues will be a bit clearer.
In Hayek’s system, the banks are making nonenforceable promises to keep the purchasing power of their currencies stable. Will they break their promise? In BFH, there is a legally enforceable contract to keep bank issued monetary liabilities tied to the common unit of account. While the BFH contract is a type of convertibility, we can imagine other sorts of contracts to bind banks that are not like that at all.
Finally, I think that no one, including the government, could be obligated to accept any type of money tendered. Actual money issued by a solvent bank? The point of the restriction, ,however, should not be to allow whoever received the money to hold it forever as an investment, but rather to exchange it for the money they prefer to hold and then spend.
LikeLike
“Hayek advocated free-alpha-banking, while conventional free bankers advocate free-beta-banking.”
I would say, David, that Hayek’s isn’t a version of free banking at all, my reason being that the institutions he calls “banks” aren’t really banks. Banks are intermediaries that fund their investments using transactable debt instruments. Hayek’s “private fiat money” isn’t debt at all. The closest analogue is something like bitcoin. Banking has nothing to do with it. The confusion of Hayek’s scheme with “free banking” has been a frequent source of misunderstanding of both subjects.
LikeLike
Sorry for my failure to correctly close my italics–as I meant to do after “debt.” I didn’t mean to shout!
LikeLike
“It is possible that are some techie reasons that I don’t understand why bitcoins could provide real services that would allow them retain a positive value”
They are cheaper to use than Visa/Mastercard as a universal exchange constant.
It’s a meta-intermediary. A real intangible asset that can be used as an intermediary form between two currency exchanges – effectively decoupling those processes.
Effectively taking the place of the US dollar which has a tendency to take that role outside of the US state.
(Any fiat currency outside its currency area is really just an intangible asset in the hands of the holder).
LikeLike
We need to think like accountants for a change. When a government issues a paper dollar, that dollar is added to the liability side of the government’s balance sheet, and that dollar is backed mostly by “taxes receivable”, which are on the assets side of the government’s balance sheet.
If I pay $1 of tax using a paper dollar, then the government’s liabilities fall by $1, while “taxes receivable” falls by $1,
But if I hand the government a $1 check drawn on BoA, then the government’s liabilities are unaffected. Its assets rise by the $1 check, while taxes receivable falls by $1.
LikeLike
my argument was that newly created currency units would be worthless unless the banks made them convertible into some outside asset not under their control.
David, there is a real-world test of your argument being conducted right now in the form of XRP, the internal currency of the Ripple payment protocol. As JP Koning has noted with great insight, XRP is a purely private chartal currency–to participate in the Ripple system (e.g. open a user account, set up trust lines with “gateways” [deposit, withdrawal, and market making institutions], and make transfers), one needs to pay a small XRP “tax.” This requirement has already given XRP a positive value, despite the fact that XRPs are irredeemable.
This isn’t to say that XRP isn’t also a bubble that may crash like, I believe, Bitcoin will. But I believe the Ripple protocol’s potentially great advantages over the current payment system–such as instant real-time settlement of international payments–provide a much hardier basis for maintaining that positive value than Bitcoin possesses.
It should be an interesting experiment. If the system ever does take off with banks and XRP maintains a positive value for a few decades, wouldn’t that be strong evidence against your claim?
LikeLike
I found this recent Youtube presentation on Ripple to be quite informative regarding the system’s inner workings.
I feel that if the world’s monetary system were still based on redeemable commodity monies, irredeemable virtual currencies couldn’t exist. But in a world of irredeemable fiat, there might be a chance for a quasi-commodity cryptocurrency to become king (or at least a major player).
LikeLike
I think Selgin’s view of banking is too narrow.
Hayek’s vision is that the banks commit to maintaining the purchasing power of the monetary instruments they issue.
While a convertibility obligation is a pretty obvious way to enforce that, it isn’t the only way.
Suppose that if the bank failed to maintain the value of its monetary instruments, all of the stockholders and employees of the bank were subject to execution?
Those holding the money couldn’t demand that it be redeemed in anything, but if the money doesn’t buy as much as promised, they call the police, and the bankers are rounded up for punishment.
A single bank in a Hayekian system cannot depend on a growing demand for its monetary instruments. Simply limiting the issue won’t do. The demand for them can fall. To maintain their purchasing power, it is necessary to pay interest on them or else reduce the quantity in line with the demand to hold them. Both require that the bank hold assets.
Now, they could just let the purchasing power fall. And they could also just issue as many as they could right away, purchasing assets to divide among the stockholders.
Yes, those are the problems with the system. Hayek’s notion that the bankers will do their best to stablize the purchasing power and make normal profits in the long run from intermediation might not be sufficiently cynical.
But that is his system. The bankers are supposed to act like bankers. For for the system to work, these institution must be financial intermediaries.
LikeLike
Bill Woolsey writes, “I think Selgin’s view of banking is too narrow.
Hayek’s vision is that the banks commit to maintaining the purchasing power of the monetary instruments they issue.”
The question is, are the institutions Hayek describes in Denationalisation of Money “banks” according to any conventional understanding of that term? Do they make loans? Do they accept deposits? Do they intermediate? I believe the answer to all three questions to be “no.”
Of course, if one means by the term “bank” any institution that supplies currency of any sort, then Hayek’s private fiat money issuers are “banks.” But by that unorthodox definition so are mints and bitcoin “miners.”
Nor, by the way, have I ever really understood how a Hayek “bank” really works. The closest I get is that a “bankers” sets up shop, prints some notes (call them “Hayek’s), pledges to maintain their purchasing power, and then (here’s were things get fuzzy) goes shopping with them, for goods or securities, relying upon the pledge to convince sellers accept the Hayeks. Again, there’s no “banking” in the usual sense involved. Or such is my belief. If anyone can refer me to passages in Hayek that suggest otherwise, and coherently, I shall be all ears.
LikeLike
Bill Woolsey:
“Most importantly, the tax authority could accept a variety of these private fiat currencies to pay taxes, but only at the current exchange rate. Once received, the government could sell them for whatever currency it wanted to hold and spend….
It would be possible to require taxpayers to do their accounting in one currency, and then let them pay the taxes in whatever currency they want.”
The government would be taking credit risk while it holds the private currency. On what basis does the government determine whether a private currency is a good credit risk? The government would need to put into place surveillance functions, which would end up replicating what the regulators are doing now.
And finally, the tax accounting in a multi-currency system would be a nightmare. Corporations would routinely arbitrage the rules for the determination of the value of transactions.
Considering that we already have a largely non-regulated shadow banking system, what purpose does all this added complexity achieve? Are there people walking around asking for a system in which they have to look up internal exchange rates before every single transaction?
LikeLike
“we already have a largely non-regulated shadow banking system.” Oh how I wish people would stop repeating this: (1) “shadow” banks aren’t banks; call them “investment” banks if you like, but recognize that that qualifier points to all sorts of activities having nothing to do with banking, of the sort that contributes to the money stock, which involves the intermediation of money deposits received in exchange for transactable IOUs; (2) it follows from (1) that even if shadow “banking” were entirely unregulated, it could not serve (as conventional banking does) as an important source of exchangeable substitutes for fiat currency, for the moment shadow “banks” started accepting deposits they’d cease to be mere “shadows” and would instead become the real things and, therefore, subject to the whole gamut of commercial-bank regulatory provisions–from state authorities, the FDIC, the C of C, the Fed, the Justice Department, the FBI–in short, the whole disaster; (3) the shadow “banks” were never “unregulated.” They were merely badly regulated, by the SEC. Nothing new here: people still repeat the myth that U.S. commercial banks were “unregulated” before the Civil War. Much easier than pouring over statutes.
LikeLike
Money market funds are a pretty effective short-term store of value, which is a core property of money. I believe that you van write cheques against them in the U.S. Therefore, anything that can issue paper to money market funds are effectively part of a shadow banking system. Sure, they do not issue “bank notes”, but the notes and coins in people’s wallets are not that important for the economy any more.
Note that I said “largely” unregulated. Anything in finance is covered by basic regulations that have been put into place to protect people from things like Ponzi schemes. There is no chance that any financial institution would ever be allowed to bypass such rules. Yes, this creates a barrier to entry. But if an entity lacked the legal capacity to navigate these barriers, no investor in their right mind would ever lend them any money in the first place.
But after that, there are very few rules governing things like capital etc. for entities like conduits that issue commercial paper. We just went through a rather nasty financial crisis that revolved around toxic waste that was stuffed into such conduits.
LikeLike
Having the government accept taxes in any currency sounds rather complicated and there sounds like there would be fair potential for abuse. Instead of having people use whatever currency to pay their taxes and then have the government convert it, why not just have people convert their preferred currency into the official currency at the going exchange rate and then pay their taxes?
LikeLike
David, you wrote: “my argument was that newly created currency units would be worthless unless the banks made them convertible into some outside asset not under their control.”
There is a real-world test of your argument being conducted right now in the form of XRP, the internal currency of the Ripple payment protocol. As JP Koning has noted with great insight[1], XRP is a purely private chartal currency–to participate in the Ripple system (e.g. open a user account and make transfers), one needs to pay a small XRP “tax.” This requirement has already given XRP a positive value[2], despite the fact that XRPs are irredeemable.
This isn’t to say that XRP isn’t also a bubble that may crash like Bitcoin. But I believe the Ripple protocol’s potentially great advantages over the current payment system–such as instant real-time settlement of international payments–provide a much hardier basis for maintaining that positive value than Bitcoin possesses.
It should be an interesting experiment. If the system ever does take off with banks and XRP maintains a positive value for a few decades, wouldn’t that be strong evidence against your claim?
[1] Check JP’s blogpost of March 3, 2013: “Birth of a currency”
[2] Currently about 1.35 cents at Ripplecharts (dot) com. Multiplied by the stock of XRPs distributed by Ripple Labs gives a total value of about $418 million.
(Re-posted without hyperlinks).
LikeLike
You can about the basics of the Ripple protocol from these quick tutorials:
https://ripple.com/knowledge-center/
LikeLike