By Dan Kervick
Modern Monetary Theory (MMT) emphasizes the central role of governments in sovereign monetary systems. MMT co-developer Warren Mosler has described the US dollar system, for example, as a “simple public monopoly.” L. Randall Wray has written that, “In the United States, the dollar is our state money of account and high-powered money (HPM or coins, green paper money, and bank reserves) is our state monopolized currency.” Sometimes this crucial MMT claim is expressed more broadly by saying the US government is the monopoly supplier of “net financial assets” to the non-governmental sectors of the dollar economy.
These claims continue to engender much confusion and resistance wherever MMT is discussed, and have been subject to several misconceptions – sometimes egregious misconceptions. I want to defend the MMT claim that the US government is the monopoly supplier of the dollar, and is therefore ultimately responsible for any net increases or decreases in dollars and dollar-denominated financial assets in the non-governmental sector of the economy. I will first describe the institutional and legal basis of this monopoly, and then turn to a discussion of what is and is not entailed by the claim that the US government runs a public monopoly over the dollar.
Of course, some people might have no problem at all in accepting the descriptive claim that there is a government-run public monopoly over the dollar, but might believe that the existence of this monopoly is a terrible travesty, a horrifying mutation of sound government into Big Government run amok. But in my view, the existence of such monopolies in monetarily sovereign countries is on the whole a very good thing, at least when the country in question has democratic institutions. Citizens of democratic countries should do everything in their power to achieve a democratically controlled currency monopoly if they don’t yet possess one, and should do everything in their power to preserve and improve their public currency monopoly if they are fortunate enough to have one.
Certainly, a public currency monopoly is not a panacea guaranteeing a healthy democratic society and a decent and prosperous life for all citizens. Nobody needs to be reminded about the outrageous inequality and economic oppression present in the contemporary United States and elsewhere. But a public monopoly over the dominant currency does give democratic citizens powers and political options they would not otherwise possess: powers to use their monetary system of behalf of public purpose and for the common good. If a nation loses its public currency monopoly, then its people risk deeper subordination to centers of concentrated wealth beyond their control.
We are seeing this sad process unfold in Europe. In order for a public currency monopoly to be fully effective, both monetary authority and spending authority need to be subject to democratic control, and should be married together in some degree so that they can be applied in a rational and coordinated manner toward the solution of public challenges. The tragedy of contemporary Europe is that in pursuit of their idealistic ambition to create a common currency and unified continent, Europeans in the Eurozone have divorced their spending authorities from ultimate monetary authority, and relinquished a substantial degree of democratic control over their monetary system. This is a backward step for democracy in Europe, and we are now seeing the ramifications. Weakly accountable technocrats in league with the continent’s financial masters are using their privileged positions to subordinate several of the continent’s governments to unelected, external control. They are using a brutal program of austerity to enforce this new system of undemocratic, centralized discipline; and their zeal to establish this new order is so great that they have plunged the continent into deeper mass unemployment and a second recession in the process.
The Consolidated Perspective
MMT writers frequently emphasize the distinction between currency issuers and currency users. This distinction applies to any monetary system whatsoever. Wherever there is a monetary system of any kind, there are some people or authorities with the power to issue new units of the currency, and others who simply use the currency that the currency issuers have brought into circulation. So part of what MMT theorists mean when they say the government in US-style systems is a currency monopolist is that the government in such systems is the sole authorized issuer of the currency.
But to understand the MMT model of the government as a currency issuer or monopoly producer of the currency, with the exclusive power to increase the non-government sector’s supply of net financial assets, it is necessary to look at the entire consolidated governmental sector. For the US system that means looking at the Treasury and Fed combined. The MMT picture only makes sense when one considers the government sector as a whole. If one focuses on just one part of the government, one will have a more difficult time understanding how the government can be a net creator of dollars.
The basic idea is that the government augments and diminishes the amount of money in circulation during some period of time whenever the monetary payments it makes to the non-governmental sectors of the economy exceed its monetary receipts from the non-governmental sectors of the economy. But again, one must consider all of the monetary payments the government makes and all of the government’s monetary receipts. Payments include spending by the US Treasury on goods and services, including payments to its employees. Also included are payments the Treasury makes to service its debts. But the Treasury is not the only part of the government that makes payments. The Fed makes payments when it buys securities from the private sector and when it pays interest on bank reserves. Similarly, both the Treasury and the Fed receive payments from non-governmental sector. The Treasury receives tax payments, payments for some of the public goods and services it provides and payments of money used to purchase its debt securities. The Fed also receives payments of various kinds: including payments for repurchases of securities, payments of interest on any debt securities it owns, interest payments for discount window borrowing, and penalty payments for reserve account overdrafts.
So how is this different from an individual, a household or a firm? Can’t some person or business in the private sector also spend more than they receive? Is this also money creation?
The situations are quite different. A private sector economic agent, as a mere user of the currency, always has a finite stock of that currency. If the agent’s payments exceed the agent’s receipts, that stock is diminished. If the agent runs out of money completely, then in order to continue making monetary payments, the agent either has to raise more money by selling some non-monetary asset, or must obtain credit. The credit might require a new application for a new loan, but it might come in the form of an automatic overdraft on a bank account or line of credit. Either way it is credit. The private sector firm or household, to the extent it conducts its business in dollars, is subject to a financing constraint.
The government isn’t like that. As the issuer of the currency it does not possess any limited finite “stock” of money that imposes a financial constraint on its spending. It can never be in a position in which it must obtain credit from some outside entity to make payments in excess of its receipts. In effect, the government possesses a bottomless well of the nation’s currency, since it is the producer of that currency. The government has no financial constraint. The only constraints on how deeply it dips into this well are policy constraints: that is, its spending is constrained only by how effective the spending is in achieving the government’s various purposes. Usually, one of those policy aims is price stability, and so a government will want to restrain the size of the gap between spending and receipts if it is concerned that too large a gap will destabilize prices.
Again, this difference between the government and agents in the non-government sector can be missed if one focuses only on one part of the government and not the whole government. Current operational rules do require the US Treasury to behave something like a private sector household. If its spending exceeds its tax revenues, the Treasury must plug the hole in its budget with borrowing from the private sector. The Treasury is not permitted overdrafts on its account at the Fed and not permitted to borrow directly from the Fed. So in effect a Treasury deficit triggers an overdraft on the private sector, i.e. more debt to the private sector.
However, much of the debt the Treasury issues to the private sector is subsequently purchased by the Fed, and thus becomes an intra-governmental debt, both an asset and liability of the unified government at the same time, which represents neither positive nor negative value and can be rolled over indefinitely as a bookkeeping operation. As a result of this and other Fed operations, the combined actions of the Treasury and Fed together can result in a net injection of money to the non-governmental sector. The Fed, of course, does not collect taxes. Its purchases do not need to be financed by the collection of revenues from other sources. Rather, it spends simply by crediting the appropriate bank accounts of the people or organizations receiving its payments. The Fed does collect revenues since it owns financial assets that require people to make payments to the Fed, but its ability to spend is not constrained by the amount of revenue it has or will collect.
The government as a whole then, as currency issuer, can do something that currency users cannot do. It can run what I will call a pure deficit – an excess of spending over receipts that does not diminish the spender’s monetary stocks and does require taking on new debt liabilities. A pure deficit run by a currency issuer does not represent negative nominal value to that issuer, although it does represent positive nominal value to the totality of currency users in the broader economic world standing outside that currency issuer. A pure deficit thus accomplishes a net creation of financial assets. Currency users on the other hand can only run financial deficits: excesses of spending over receipts that diminish monetary stocks or add debt. That means the positive nominal value flowing to the world external to the currency issuer by virtue of the financial deficit is entirely offset by the negative nominal value accumulated by the currency users running the deficit.
Some people criticize MMT for taking this consolidated view of the government sector. They will point out that the Treasury and the Fed are not operationally integrated, but maintain a substantial degree of operational independence. The point is well taken, and it is good to bear in mind that when one makes general statements about what the government can and cannot do. The ability of the government to efficiently carry out in practice what it can do in principle depends on the nature of its institutions, and on how those institutions are organized and integrated.
But if our aim is to understand the role of the government as a whole in our economic system, the consolidated view is very fruitful. We can think of the economic world as a collection of balance sheets. Each household has its own balance sheet, but when we want to understand the household sector as a whole, we combine those balance sheets into one for the purposes of theoretical analysis. Similarly, if our aim is to understand the economic role and capabilities of the government as a whole, we will want to combine all of the government’s balance sheets into one.
Money, Credit and Central Bank Liabilities
It is sometimes held that all money is just a form of credit (or debt, depending on whose side of the credit ledger one is considering), and that the currency issued by the government represents a debt or liability of the government. Certainly, the money-as-debt view appears true when one considers certain forms of money. Commercial bank demand deposit balances, for example, are clearly liabilities of the bank. When you are in possession of such a deposit your bank owes you something, and you are then entitled to go to your bank and demand what is owed.
Every debt is a debt for something. So we can ask with regard to any debt what is owed, and what sorts of things would constitute payment of the debt. It is sometimes the case that a debt can be discharged in full with another credit instrument, after which the debtor no longer owes the creditor anything. If you have a contract with some firm entitling you to receive some corporate bonds issued by that firm, then you have been paid once those bonds are delivered. This is true even though the bonds themselves are a debt instrument. But what lies at the foundation of all these credit instruments?
In the United States, the paper money we carry is called a “Federal Reserve note”, and is regarded under law as a liability of the Federal Reserve system, in the same way that a note from a commercial bank is a liability of that bank. These notes are also, according to the Federal Reserve Act of 1913, supposed to be obligations of the US government, and “shall be redeemed in lawful money on demand at the Treasury Department of the United States, in the city of Washington, District of Columbia, or at any Federal Reserve bank.”
If that statement is right, then the bearer of a Federal Reserve note holds a debt of the US government, and the government owes something to that bearer. And that would suggest that there is no difference in principle between Federal Reserve notes – and perhaps bank reserve balances held at the Fed – and the demand deposit balances ordinary bank customers hold at their commercial banks. Each is a debt: a liability of its respective bank of issue. But if a Federal Reserve note is a liability of the Fed, what is it a liability for? Is it a debt for some of the supposed “lawful money” referred to in the Federal Reserve Act? What is that? Isn’t the Federal Reserve note already lawful money? And if there is some other kind of truly lawful money, and if money is always a credit instrument representing a debt of some kind, then what is that hypothetical lawful money itself a debt for – some even more awesomely lawful money?
The standard Fed line seems to agree with the view of money-as-liability. The official balance sheet promulgated by the Federal Reserve counts currency in circulation, along with bank reserves, as liabilities of the central bank. If the Fed purchases a Treasury security from a private dealer, the security is entered on the Fed balance sheet as an asset, and the money it paid for that security is entered as a liability.
But is this official stance economically accurate? Is the circulating dollar a liability of the Fed? If it is, it is a very peculiar kind of liability indeed. A liability is something that is owed; and to possess a liability is to possess something that represents negative value; it represents negative value to the person who owns the liability because it is a promise of a future payment. Ordinarily, the obligation to make a payment represents a future loss because the payment will diminish that person’s stock of wealth. But when some person possesses a Federal Reserve note, what else does the government then owe that person?
With all due respect to the official Fed story, I think the money-as-liability picture is seriously misleading to say the least. I want to argue that currency in circulation does not represent a liability of the government in any meaningful economic sense, and that registering it as such on the Fed’s balance sheet is merely an outdated accounting convention.
Imagine you are a breeder of milking cows. But these cows are very special. They are immortal. They never get old and they never get sick. Once fully grown, they produce the same amount of milk, every day, for as long as they live – which is effectively forever. The cows are perfectly identical; no cow’s milk is any better or any worse than any other cow’s milk. The cows are also very hard to breed, and unsurprisingly they are in very high demand. You can breed five per year, and you sell each one for a very high price. The sale of these cows accounts for virtually all of your income.
Now suppose you decide to build a new house, and decide to pay for the house with an IOU. The IOU is for one cow from next year’s stock. Once you have offered the IOU, and the builder has accepted it, the IOU becomes the builder’s asset and your liability. You owe the builder something of value, a cow worth one fifth of your annual income. When you deliver the cow, your stock of wealth has been diminished. You are one cow poorer. Hopefully, the house is worth it.
Now let’s suppose that with each cow sold, you generally issue to the purchaser a “certificate of substitution”. These certificates bind you to a commitment: If for any reason, the purchaser of the cow tires of the cow and wishes to trade the cow in for a new one, they are permitted to do so. You will accept the old cow back as a trade-in and give the purchaser a new cow.
Do these certificates of substitution represent a liability for you? Remember that we have assumed that these cows are perfectly identical. The old cows don’t get worn out; they don’t get older; they don’t get lazier or more reluctant to produce milk. Any one of the cows is just as good as another one. If you receive a cow back as a trade-in for a substitute, the cow you receive back is just as good as the one you give up. You can sell the trade-in again for as much money as you would get for selling any one of your other cows.
So no, the certificates of substitution that you have issued do not represent a liability; they do not represent negative value to you. They do represent an obligation, because by issuing the certificates you have obliged yourself to carry out a cow exchange if a purchaser requests one. But when you make such exchanges you do not end up any richer or any poorer than you were before you made the exchange. So these obligations are not liabilities.
So right away, we see that commercial bank demand deposit balances – which are IOU’s of commercial banks – are very different from the currency and bank reserves issued by the Fed. When you have a demand deposit, the bank actually owes you something, and if you demand payment of that thing, the bank’s assets are diminished when they make that payment. And the thing the bank owes is something whose production the bank does not control. Commercial banks are not permitted to print paper currency or mint US coinage. They are not permitted to create their own reserves at the Fed. They must acquire these things from the government, and to acquire them they must buy them.
But the government completely controls the manufacture of the currency. It can produce paper and metal currency at a real cost that is always much, much lower than the current market value of the face-value quantity of the currency produced. And it can produce bank reserves at virtually no cost, simply by marking up accounts on computer screens. And the Fed can carry out these operations virtually any time it wishes, restrained only by its own monetary policy ends, and not by any inherent financial constraints. The Fed has no finite stock of dollars. When it issues some quantity of dollars, it hasn’t diminished its stock of dollars in any way.
There are other reasons that have been given for treating money as a liability of the government. It is sometimes said that dollars are the government’s liability because of the tax obligations the government imposes. The logic seems to be something like as follows: Suppose Peter owes Paula a debt for $1000, but Paula also owes Peter a debt for $1000. Suppose Paula’s debt to Peter is represented by a bond Paula has issued to Peter with a face value of $1000. Peter can use the bond to pay the other debt, the debt he owes to Paula. Having done so, both debts are discharged. They cancel out.
Now suppose Paula is the government and Peter owes Paula $1000 in taxes. He takes a $1000 bill to Paula, and the obligation is discharged. Doesn’t the similarity between these two cases suggest that the $1000 bill is something like a $1000 bond – a debt that the government owed Peter – and that the $1000 bill cancels out the tax obligation in the same way the $1000 bond in the first example cancelled out Peter’s debt?
I don’t think this is the right way to look at the $1000 bill. Consider another analogy: suppose you are a businessperson, the owner of a home and garden shop – and have been required by a court to deliver 100 bags of mulch to me. I have the court’s order in my hand, and it represents to me an asset worth 100 bags of mulch (BOM). Once the order has been issued by the court, I am better off. When the mulch has been delivered, your obligation will be discharged. But does that mean that the 100 bags of mulch you have in your possession before you deliver them represent a liability for me, because you can use them to cancel out your obligation? No, of course not. If they were my liability, then my 100 BOM asset represented by the court order would be offset by my possession of a 100 BOM liability of the same amount, with a net value for me of zero.
We can now see the difference between the first case – Peter’s debt to Paula – and the second case of the mulch owed by the store owner. Suppose for whatever reason, Peter’s debt to Paula is canceled by a court, but Paula’s debt is left intact. Then Paula still has a liability: she still owes something to Peter as represented by the bond Peter possesses. But in the second case, if the store owner’s debt is cancelled, then I have simply lost an asset. And I have no outstanding liability. I don’t owe the store owner anything.
I think we should view tax obligations more like the mulch obligation than Peter’s debt to Paula. The store owner possesses a debt owed to me, and he also possesses the means to discharge the debt. But the fact that he possesses those means does not entail that those means represent some liability I have. Similarly, if some citizen has a tax obligation to the government, and possesses the dollars needed to discharge that obligation, that does not entail that the dollars represent some liability of the government. If they did, and if a dollar were a sort of bond issued by the government, then if some court were for whatever reason to cancel the citizen’s tax obligation, the government would still owe the citizen something. But it doesn’t. Once the tax obligation has been cancelled, the books are clear between the citizen and the government.
Another reason that has been given for saying that dollars represent a liability of the government is the suggestion that the government has a general obligation to support the market value of dollars, making each of those dollars a liability. But this strikes me as a very ephemeral basis for attributing a liability to the government, since the obligation in this case is really nothing other than the government’s own voluntarily assumed policy choice.
And even if the obligation to support the value of the currency does represent a genuine liability of the government, it is worth noting that the cost of that liability is far, far less than the market value of the dollars in circulation. By analogy, suppose you are the custodian of a museum which houses a collection of paintings that has been entrusted to your care. Suppose the market value of the paintings is one billion dollars. Suppose also you are legally obligated by the terms of your custodianship to care for, and preserve the market value of the paintings. Each year on your books you enter the cost of preserving the paintings. How much is it? Certainly far less that $1 billion! So even if the government possesses a genuine liability to preserve the value of the currency in circulation, that liability is nowhere close to the market value of that currency. The government’s liability with respect to a single dollar in circulation is not one dollar, but a tiny fraction of one dollar.
So why does the Fed account for its dollars in this way? Why does it register currency in circulation as a nominal liability equal to the face value of the currency? I would suggest that Fed accounting practices are only outdated traditions that preserve the gold standard convertibility illusion of ye olden tymes by accounting for issued currency as a debt instrument or IOU of the Fed, just as it was in the pre-fiat days when government currency was only another bank credit promising redeemability for something else. But in the post-convertibility days of true fiat money, that accounting practice is nothing but a hoary convention with no real basis in economic reality. There is no meaningful sense in which the currency and reserves issued by the Fed represent claims on something that the Fed still owes the bearer of the currency. Nor is there any economically meaningful sense in which a negative equity position at the Fed – with Fed liabilities exceeding its assets – represents a state of bankruptcy, as it could if issued currency were a true liability. The Fed cannot go bankrupt.