Part of the What gives money its value? series

Great article written by over at: Origin of Specious

The idea is that whatever you back money by, it will quickly reach full-capacity, thus, the fact that money is backed by employment, means that if people recognized this more, we would have full employment all the time.


I vaguely touched on an idea in my debt book that I didn’t develop. I didn’t realise there was a controversy around it. I am now aware of the controversy, and so I think I’d better develop the idea.

The controversy concerns the nature of fiat money. Fiat money is money, issued by a state authority, that isn’t officially backed by any real commodity.

On one side are those, including the MMT school, who say that fiat money is effectively government debt. On the other are those, generally of a more orthodox persuasion, who deny this; for them, money is an intrinsically (nearly) worthless token, which possesses value only because it is socially accepted in exchange for goods and services.

The view that money is government debt, I call the push theory of fiat money: money is ‘pushed’ into having value by what backs it, namely the full faith and credit of the state authority that issues it. The other view I call the pull theory: money is an intrinsically worthless token ‘pulled’ into value by society’s habit of accepting it in exchange for real goods and services.

You can see this controversy played out in the comments on a very interesting blog post by Randall Wray, which I highly recommend. But I think there’s a lot of talking at cross-purposes here, and I wonder whether my idea could help towards finding a point of agreement between the push and the pull theories.

Eric Lonergan argues that the push theory rests on a logical fallacy. Push theorists argue that since fiat money can be used to settle debts to the state – mostly tax debts – it must therefore consist of debts from the state. Semi-formally, the fallacious argument looks something like this:

  1. Anything that is a debt from X to Y can be used to settle an equivalent debt from Y to X.
  2. Money can be used to settle a debt from citizens to the state.
  3. Therefore money is a debt from the state to citizens.

The argument is indeed fallacious; there is an undistributed middle. From 1 we can infer that anything that is a debt running state -> citizens – call this an SC debt – is something that settles a debt running citizens -> state – call this a CS settlement. We can infer that anything that is a SC debt is a CS settlement. But we cannot infer the converse, viz., anything that is a CS settlement is an SC debt.

What we would need to make the argument above valid is an extra premise: only an SC debt can be a CS settlement. But what would justify this extra premise? Certainly we don’t want a general principle holding that only an XY debt can be a YX settlement. Suppose that we are both potters, and you make a pot for me today, while I promise to make you a pot tomorrow. My pot is then used to settle my debt to you, and on the general principle proposed here, we should conclude that my pot is a debt from you.

If the state were still on a metal standard, the push theory would be easier to defend. We could agree at least that currency, as such, is government debt. Each banknote (and each entry in a bank’s reserve account) is the promise of a certain amount of precious metal. It is important to recognise that this does not mean that the currency has value only because of its capacity to be redeemed for precious metal. The currency can be exchanged for goods and services. And it can be used to cancel tax debts. But there is nothing wrong with saying that it is debt, so long as we are clear on what we mean: the state owes precious metal to each and every holder of currency, if she chooses to redeem it. We can call the metal the item owed, or we can call it the collateral against which the loan is securitised; either way, we are clearly talking about a debt.

Lonergan would not, I think, find any fallacy in the following argument:

  1. A note from X, legally entitling Y to be provided with something other than another note of the same sort, is a debt from X to Y.
  2. Currency (on the metal standard) is a note from the state, legally entitling the bearer to be provided with something other than a note of the same sort.
  3. Therefore currency is a debt from the state to the bearer.

It is only when currency stops being redeemable for metal that the trouble begins. The structure looks very similar to how it looked when the currency was backed by metal, which gives succour to the push theorist. And yet the currency appears not to be backed by anything, which gives force to the arguments of the pull theorist.

My idea is as follows. The currency is backed by the public services the government provides. While the central bank issues currency, it is redeemed at the tax offices. What do you get when you redeem currency by paying your taxes? You get all the public services the government provides to its taxpaying citizens: national defence, infrastructure, health services, education, etc. etc.

Here the pull theorist might object. On the metal standard, people have a choice whether to redeem their credit notice, hold onto it, or exchange it with somebody else. If fiat currency is debt redeemed in taxation, then people have no choice but to redeem it. Normally with a debt, the creditor has the power to forgive the debt. But the earner of currency has no choice about tax payments. This could be the basis of a new distinction between currency and debt.

It is, however, not true that earners of currency have no choice about tax payments. Tax evasion is always an option, even for those who can’t afford to hire dodgy accountants. It’s just that if you don’t pay your taxes, the state might haul you off to prison. So we can include not hauling you off to prison as among the services provided by the government in redemption of tax payments. You might object that this is not a service, being an omission rather than a commission, and I am sympathetic to this line of thought. Nevertheless, it is an omission for which taxpayers are very willing to redeem their currency.

This version of push theory is immune to Lonergan’s critique. It also has some fairly important consequences, of which I’ll mention two here.

No Free Market

First, it means that in an economy with a fiat currency, there is no free market. There is nothing even remotely resembling a free market. Free markets exist when there is competition and prices are a function of supply and demand. When prices are set by a single authority then there is no free market.

But the state clearly sets the relative price of the services it provides. If it doubles taxes while providing the same amount of public services, it has doubled the price of its services, not just in terms of money, but in terms of whatever you gave up – usually labour – to get that money. Textbook supply and demand analysis says that we should then consume less of the costlier services. But of course we don’t, since among those services is that of not hauling us off to prison, for which most people will pay whatever the price demanded.

It is also well-known by economists that as soon as a price is set in one market, the distortion carries through to every connected market. So once the state sets the price of its currency, every other price in terms of currency is affected. Worse still, there is no reason to believe that the distortion in other markets will be directly proportionate to the size of the market affecting them. The relative importance of that market matters. But the market for public services is, in a modern society, just about the most important market there is. We all have varying dependencies on public services, but there will always be people for whom they are the difference between life and death – or between prison and freedom. So even a small government with a small tax base is incompatible with the existence of a free market, wherever there is a fiat currency.

What if we went back to a gold standard? Once you recognise that currency is backed by public services, you realise it was never really backed by gold. It was always backed by public services, offered at a price set by the state; it’s just that once upon a time those services included the storage and delivery of gold. When that service stopped being offered, it mattered very little, since another service offered by the state – that of not hauling us off to prison for tax evasion – turns out to matter a lot more to people than easy access to gold.

Alright, well what if we privatised the currency, through some sort of ‘free banking’ arrangement? So long as there were still any sort of state at all, we wouldn’t have got rid of fiat currency. It would just look like we had.

In other words, where there is a state, there is no free market, and nothing like one. There is no ‘market mechanism’. There is no ‘law of supply and demand’. Or at least the extensions of these concepts are to be found only in the abstract models of economists.

Basic Income Doesn’t Work

The other thing this new push theory helps us to realise is that the basic income idea is largely a sham.

Push theorists and pull theorists can both agree that there are certain situations in which it can benefit just about everyone for the government to issue and spend more money into the economy without raising taxes. Whether we are in such a situation is always, of course, a very controversial empirical question. But few deny that it is always a theoretical possibility.

Such a situation will be one in which private spending is insufficient to purchase everything produced in the economy at current prices. The result would be deflation… if only prices were market-determined and thus a function of supply and demand. But since, with a fiat currency, prices aren’t market-determined, the real result is unemployment, unused capacity, and stockpiled inventories of unsold goods. There is also likely to be a piling up of debt, as businesses find that they can’t sell the things they took on debts to produce.

In this case, it makes sense for the government to issue and spend more money, so that people have more disposable income and increase their spending. The question is: what should the government spend the extra money on?

Push theorists tend to support the idea of a job guarantee: the government should spend its money hiring all the unemployed workers, at a living wage, to provide extra services (extra public services are always nice to have, even if they’re not strictly necessary). Once spending picks up again, the private sector can hire these workers back out of the public sector.

Pull theorists tend to support the idea of a basic income: the government should just hand everyone a cash payment. This has the advantage of not requiring a complex bureaucratic structure to determine how to employ a bunch of newly-hired public sector employees. It has the disadvantage of being a sham.

Various versions of the cash payment idea – ‘basic income’, ‘basic income guarantee’, ‘citizens’ income’ – have generated buzz in the media lately (here is a recent example). The job guarantee idea is almost entirely overlooked in the British media. Many very eloquent and knowledgeable people I know have pitched articles explaining it to various newspapers, and they’ve got nowhere.

I think this is because understanding the job guarantee proposal requires a far more radical revision of our economic understanding. The basic income proposal is grounded on a reversion to the comforting myth of the free market. Somehow, it is thought, prices have got stuck so that there is a surplus of goods and services and a shortage of money to buy them. The solution is to create and hand out money until we get back to the ‘market clearing’ price.

But a market that can clear is one that can adjust supply to demand by way of the price mechanism. And we’ve just seen that, where there is a fiat currency, no such mechanism can exist. Adding spending power is very likely to just push up the prices of goods and services already being sold while leaving the unsold goods yet unsold. Again, this would violate the law of supply and demand, which holds that a fall in the price of unsold goods relative to that of goods already being sold will shift spending away from the latter and towards the former. But we have seen that the jurisdiction of that law does not extend beyond the realm of economic abstraction.  (A tax-financed basic income is a different thing entirely – aimed at solving a different problem; the justification for it also depends on belief in the free market, but this is a topic for another post).

With the job guarantee, by contrast, the state buys the unsold goods and services and converts them into public services. It then ensures that people buy those services, through the tax mechanism described above. It makes the situation of unsold goods and services an impossibility rather than hoping that a non-existent market will render it improbable.

Most people focus on the question of whether the public or private sector should get to use existing resources. But the fact is that the private sector has no mechanism for preventing a huge volume of resources from going entirely to waste. Remember that the resources in question are human lives – stores of untapped talent, energy, and ingenuity that are born only once and will never live again.

If this were, as people say, a matter of market failure, then the solution might be to repair the market mechanism. But you can’t repair a mechanism that doesn’t exist. State planning is the only option. There Is No Alternative.

Read the article here with comments.