By Dan Kervick
Is The United States a Monetary Sovereign?
I have set out a simplified model of a monetarily sovereigngovernment. But near the end of theprevious section, I began to suggest that the United States government is indeeda monetary sovereign by this kind. Thereader might now suspect that I have yielded my rational mind over to a simplisticfiction of my own creation. And by thispoint, the reader is probably thinking that however interesting it might be toimagine this fictional entity, the so-called monetary sovereign, such fictionshave nothing to do with the complexities of the real world, because actualgovernments maintain accounts that are indeed constrained by the amount ofmoney in those accounts and by the external sources of funding to which theyhave access. After all, can’t a governmentdefault on its debt? What about the recentdebt ceiling debate in the US? Whatabout what is happening in Europe with the sovereign debt crisis? Also, if a government like the United Statesgovernment was a monetary sovereign of the kind I have described, theconsequences would seem to be enormous. Surely if a democratic government possessed this kind of power, we wouldmake much more use of it than we do. In short, monetary sovereignty as describedseems both too simple to be real and too good to be true.
These skeptical intuitions are reasonable, so they need tobe addressed. First, let’s consider thequestion of whether monetary sovereignty is toosimple to be real.
I will argue that the government of a country like theUnited States is much closer to the ideal of monetary sovereignty than thetypical citizen recognizes. To theextent the model is overly simplified, that is due entirely to choices we have made about how ourgovernment should be organized internally. The financial and monetary operations that occur in our actual governmentare not carried out by a single operational center, but rather involve severalparts of the executive branch, most prominently the Treasury department. Congress is involved as well, as is the FederalReserve System. These branches of thegovernment are subject to various legal restrictions and constraints. But these are all constraints that thecountry’s legislators have chosen toimpose on the government’s financial operations. They are to that extent voluntary and could thereforebe altered.
Congress has chosen, for example, to make the US Treasury,and even Congress itself to some extent, function as a currency user rather than a currency issuer, and has attempted to assign to theFed all primary responsibility for direct decisions over the increase anddecrease of the money supply. Ultimatemonetary authority obviously resides in Congress, but Congress has delegatedmuch of that authority to the Fed, and has been reluctant to exercise theauthority directly by engaging in direct monetary operations on behalf of thepublic it represents.
These restrictions have been implemented in several ways: The Treasury Department can only spend ifthere are sufficient points on its monetary scorecard – that is, if sufficientdollars are credited to its bank accounts. Its accounts are held at the Fed and administered by the Fed. It isforbidden from overdrawing its accounts at the Federal Reserve, and the Fed hasno authorization to credit those accounts directly and unilaterally. So theTreasury can’t create money itself by a direct act, in the course of itsordinary operations, nor can the Fed create it directly for the Treasury. IfCongress has authorized some spending by the Treasury Department, the Treasurycan only carry out that spending if the combination of tax revenues and borrowedfunds currently supplying Treasury accounts constitute sufficient funds for thespending. If tax revenues areinsufficient, then in most cases the Treasury Department will sell bonds to theprivate sector, and raise funds in that way. However, Congress has alsoimposed a debt ceiling on Treasury borrowing, so the Treasury’s prerogative inissuing bonds is capped.
The Treasury Department does possess, through its operationof the US Mint and as a result of certain loopholes in existing authorizationsto mint coins, a potential source of direct control over monetaryoperations. But taking advantage ofthese loopholes would be highly unusual and politically controversial. And if Congress remained determined to keep delegatedmonetary authority with the Fed, then the loopholes would probably be closedquickly by legislation.
Also, the Treasury Department is forbidden from sellingbonds directly to the Fed. So while the Fed is permitted to create moneyand use it for making loans to banks in the Federal Reserve System, or for thepurchase of financial assets from private sector owners of those assets, itcannot purchase bonds directly from Treasury. And thus the Treasury cannot borrow directly from the Fed. The two departments must instead follow amore roundabout method. The Treasury cansell bonds to private sector dealers in an auction, as it ordinarily does. The Fed can then, at its discretion, purchasethose bonds from the private dealers in separate auctions. Treasury ends up with some amount ofborrowed funds, but also with a liability to pay the Fed the principle on theloan. Any interest payments on the bondswill be returned to the Treasury, since the Fed is not permitted to collectinterest from the sale of Treasury bonds. So the Treasury ends up in a better position than if the bonds werestill owned by the private sector dealer. But the Treasury still owes the Fed the principle. How these loan payments are funded is then ultimatelyup to Congress to decide.
Let’s conduct a thought experiment, and imagine how things mightwork if the Treasury could sell bondsdirectly to the Fed, and if Congress exercised more direct supervision over theFed’s purchases of Treasury dept.
Suppose the Treasury Department were permitted to issue aspecial class of bonds – call them “M-bonds”. These bonds could not be sold to privatesector purchasers on the open market, but could only be sold to the Feddirectly. Suppose that the bonds carriedno coupon payments and 0% interest, and matured in a year. In other words, if the Treasury sells a $1 billionM-bond to the Fed today, then the Treasury receives $1 billion from the Fedtoday, and next year they pay the Fed exactly $1 billion, with no interestpayments in between.
Suppose also that the Fed were not permitted to refuse tobuy M-bonds. Let’s imagine that Congresshas passed a law mandating that, if Treasury issues an M-bond and offers it forsale to the Fed, the Fed has to buy it. But let’s also assume that Treasury is still not permitted anyoverdrafts on its account at the Fed. Congress continues to mandate that any Treasury spending must be clearedthrough its Fed account, and that the only ways of funding that account are thoughtax revenues, sales of ordinary Treasury bonds to the private sector and salesof M-bonds to the Fed.
Now, finally, let’s suppose that the Treasury Department hasa standing policy of funding $100 billion of public sector spending each yearthrough the sale of M-bonds. It also hasa policy of issuing new M-bonds each year to meet the full costs of servicing its outstanding M-bond debt. Inother words, it always pays the debt it owes on its M-bonds just by sellingmore M-bonds. So, in Year One it sellsthe Fed $100 billion of M-bonds, and spends the proceeds. In Year Two, it sells $200 billion ofM-bonds, spending $100 billion of the proceeds and using the other $100 billionto pay off the Year One debt. In YearThree, it borrows $300 billion, spends $100 billion and uses the remaining $200billion to pay off the Year Two debt. Etc.
We can see that the portion of Federal debt attributable toM-bond issuance grows arithmetically by $100 billion each year. So the national debt continues to rise. But we can also see that that portion of thedebt is relatively meaningless. And itwouldn’t matter if M-bonds were not sold at 0% interest, but carried somepositive interest rate – say 10% or more. In the latter case, the debt due to M-bonds would not rise onlyarithmetically, but would rapidly compound. But itwould be just as meaningless, since the whole quantity of the previous year’sM-bond debt would be borrowed from the Fed each year, and then paid back thenext year with additional borrowings from the Fed. The Fed would be required to purchase thisadditional M-bond debt each year, so the rising debt places no rising burden onthe US Treasury or the American taxpayer.
It should be clear at this point that the entire functionaleffect of all that borrowing and repayment with M-bonds could be accomplishedby the following simpler alternative operation: Congress simply mandates that each year that the Fed must directlycredit $100 billion to Treasury Department accounts at the Fed. No bonds. No borrowing. End of story. While this might appear to be an entirelydifferent kind of operation, ultimately they are just too different mechanismsfor accomplishing exactly the same effect. Thus, the rapid arithmetical rise in M-bond debt in our thoughtexperiment is not functionally equivalent to a cycle of hyperinflationaryrunaway money printing. There is insteada fixed, modest annual amount of net money creation – $100 billion, which isjust a fraction of annual US GDP – and the ballooning debt payments are just anartifact of the convoluted M-bond method Congress has hypothetically prescribedin our thought experiment to accomplish this money creation. The M-Bond debt owed by the government to theFed – which is itself part of the government – has a fictional quality.
It is vital to recognize, then, that the third party privatesector involvement in the current borrowing relationship between the Fed andthe Treasury is entirely voluntary on the part of the US government. Congress could remove it at any time, simply bypassing the appropriate legislation.
Congress could also, at any time, direct the Fed to credit TreasuryDepartment’s accounts – their monetary scorecards – by any amount Congress seesfit. The recent debt ceiling crisis,therefore, is entirely the result of self-imposed, voluntary governmentconstraints. The government can never runout of money unless it chooses tosubject itself to various self-imposedconstraints.
Congress has not provided itself with any institutionalizedmeans for conducting monetary operations directly, and has imposed on bothitself and the Executive Branch – the two political, elected branches of thegovernment – a system that requires both branches to act as though they are themere users of a currency that is controlled by the Fed. Congresshas thus imposed a quotidian accounting constraint – to use a term introducedearlier – on the political branches of government. The Fed, on the other hand, is effectively permittedto spend without a scorecard. But its spending options are limited by law: Itcan buy government bonds and other bonds on the open market. It can also lend funds to banks at a rate ofits own choosing. But it can’t buy abattleship, or hire 100,000 people to spruce up the national parks or build ahighway or rail line, or simply send checks to selected American citizens. Or at least if it tried to do these thingsit would likely be challenged legally for conducting operations that appear to exceedits intended legal powers. Just what theactual limits of those powers are, and how much Congressional spending power hasbeen delegated to the Fed, seems to be a matter of some controversy. But it is clear that the Constitutionalintention is that the “power of the purse” is supposed to reside with Congress. And thus any move by the Fed to begin conducting fiscal policy byspending money on all matter of goods and services would be extremelycontroversial to say the least.
It sounds a little bit strange, of course, to say thatCongress has imposed operational constraints or restrictions on itself in the area of monetarypolicy. After all, apart from thosesupreme laws that are embedded in the US Constitution, Congress makes thelaws. So in what sense can Congress beconstrained by laws of which Congress itself is the author and master? We might think here of the ancient Greekhero Odysseus, who had himself bound to the mast of his own ship to prevent hisship’s ruin on the rocky island of the Sirens. But the important thing to remember in this area is that while the USCongress might be bound by laws that Congress itself has created, these lawscan be changed at any time by the same Congress that enacted them. Congress can intervene in US monetaryoperations at any time, since US monetary power is constitutionally vested inCongress.
So the parts of the government that can actually accomplish a lot with their spending – Congress andthe Executive Branch – are presently required by law to act as mere currency users that must draw on private sectorfunding sources to carry out that spending, while the part of the governmentthat is permitted to act as a currency creator – the Fed – is subject to fairlystrict limits on what it canaccomplish and whom it can affect with that spending.
The whole system seems cumbersome and byzantine when viewedin this light. But perhaps theseself-imposed constraints have important policy justifications? Perhaps Congress in its wisdom has seen thatmonetary power is simply too dangerous for direct democratic governance, andthat even Congress itself cannot be trusted to carry out monetary operations inconjunction with spending and taxing operations, in a democratically influencedfashion? We will return to thisquestion later. But for now, let’s turnto the other instinctive reaction to the model we have developed of a monetarilysovereign government: that it is too goodto be true.
If the monetary sovereign is not subject to any operationalrequirement either to tax or to borrow in order to spend, and if the monetarysovereign has the power to create money at will, then isn’t that the ultimatefree lunch? Doesn’t that mean that agovernment of this kind can spend without limit either to purchase goods orservices for the public sector or to effect direct transfers of monetarybonanzas to private sector accounts?
We all know something is wrong with this suggestion, if weinterpret it in its most obvious sense. And where it goes wrong is in its loose use of the word “can”. Of course, in one sense the monetarilysovereign government can spendwithout limit. There is no operationalconstraint on this spending. The USCongress can authorize as muchspending as it desires, and of almost any kind. It can, if it chooses, permitthat expanded spending to go forward in the absence of any additional taxrevenues. It could remove the debt ceiling and authorize, or even direct,unlimited borrowing by the Treasury. Orit could direct the Fed to credit theTreasury Department account directly with some large amount of money. It couldeven eliminate the Treasury Department’s Fed account entirely, and simplydirect the Fed to clear any check issued by the Treasury Department, and alwaysmake a payment directly to the account of whatever bank presents that Treasurycheck to the Fed.
In the purely operational sense of “can”, our government cando all of these things. But we all knowthat under many circumstances, such actions could have very bad effects. In addition to whatever operationalconstraints do or do not bind government actions, there are also what we havecalled policy constraints. A policy constraint on government actions issimply a policy choice the government has made that cannot be effectivelycarried out if the government does not act within that constraint. And if the policies are sensible ones, thepolicy constraints are sensible as well.
One such policy which most governments seek to implement isa price stability policy. For goodreasons, governments seek to prevent prices on goods and services from risingor falling too much in a short period of time; or from rising or fallingsharply and suddenly, or in an accelerating fashion; or from behaving in anerratic and unpredictable manner. Priceinstability of these kinds can have an inhibiting, recessionary effect oneconomic activity, as the participants in the economy struggle to predict theoutcomes of their medium-term and long-term contracts and transactions. If a monetarily sovereign government suddenlyauthorizes the creation of excessively massive amounts of new money, and simplyspends that money into the private sector directly to make public sectorpurchases, or transfers it to individuals who in turn spend it, the effectcould be a sharp and sudden surge in the level of prices. High inflation and shortages of goods are thelikely result.
And yet the risk of runaway inflation as a result ofgovernment money creation is frequently exaggerated. Some commentators seem to assume that the merecreation of new money will always have a corresponding inflationary effect, nomatter how the new money is spent. Theyare constantly warning is that “hyperinflation” is just around the corner as aresult of government money creation. Butthis inference does not meet the test of either common sense or consideredexamination. Adding money to theeconomy only exerts pressure on prices if that money is in the marketplace, inthe hands of customers, competing with other potential customers for goods andservices to bid up the prices of those goods and services. If the money is inserted into the economy insuch a way that it mostly goes into savings or bank reserve buffers, it willnot contribute to price pressure. Suchappears to be the case with recent “quantitative easing” policies pursued bythe Fed.
But even if the money does accompany hungry customersstraight into the marketplace in pursuit of goods and services, it still mightnot exert much pressure on prices. Itreally depends on how and where the money is inserted. Consider an economy like the one we areenduring currently, with double-digit real unemployment and substantialunderutilized human and material resources. Many businesses are experiencing empty shelves, unused warehouse space,vacant office space, idle productive machinery and internal systems operatingwell short of their capacity. Inresponse to a surge in demand from new customers with money to spend, suchbusinesses can ramp up production rather quickly. They can hire workers from among the hugearmy of unemployed people hungry for jobs, put productive capacity back online, and fill up existing shelves or distribution facilities with very littleadditional cost per unit of output. Infact, with so much underutilized capacity, the cost per unit of output sometimeseven falls with additional production, as current capacity is used moreefficiently. So businesses would havelittle reason in these circumstances to raise prices on the basis of costpressures alone. At the same time, anybusiness that is even tempted to raise prices in response to the new demandwould face intense pressure from their competitors, who have been starved forcustomers throughout the recession, and who will be only too happy to keepprices low and reap increased revenues from boosted sales alone, with the same unitproduction costs, and without attempting to frost the tasty new cake with anuncompetitive price increase.
So, inflation fears vented over proposals for moregovernment deficit spending assisted by sovereign monetary power are oftenoverblown. An economy in a deeprecession like ours would likely benefit greatly from such a direct expansionof government spending.
In fact, not only is government spending in a recessionlikely to be beneficial, but the decision to throttle down government spendingand reduce deficits – that is, the decision to practice austerity – ispositively harmful in the same circumstances. That is because, in the absence of any change in a country’s currentaccount status with respect to its trade abroad, any decrease in the governmentdeficit corresponds to an aggregate worsening of private sector balance sheetpositions. If the government insistson pushing its own balance sheet into a position of surplus, it will likely pushthe private sector into a deeper deficit, which is precisely the wrong thing todo as the private sector struggles to deleverage, and as household and businessincomes fall. And in the context of aglobal recession, where virtually every country would like to increase exportssignificantly but few countries can do so because there are not enough foreignbuyers for their goods, the clear present need is for expanded public sectorspending.
But suppose our government chose to expand spending bymaking use of additional borrowing from the private sector? In that case, the additional deficitspending would drive up the national debt. Isn’t there great risk in these high debt levels? If the government’s debt goes to 100% ormore of our entire annual national product, isn’t that dangerous? Many pundits are warning these days aboutthe allegedly calamitous level of debt and the threat of ruin or bankruptcygovernments face as a result.
And private sector debtis certainly a big problem. As we havediscussed, individuals, households and firms – unlike monetarily sovereigngovernments – are mere users of debt instruments and monetary instruments theydon’t control, and operate under real and inviolable budget constraints. They can face insolvency if their debts gettoo large. And even if they are not inimmediate danger of insolvency, high debt burdens place serious limits on theability of private sector borrowers to spend their income on satisfying otherwants and needs.
Politicians have recentlydrawn on these fears of private sector debt in the United States to elevatesimilar fears about the debts of the US government. We hear politicians and other nationalopinion leaders warn that the government faces “bankruptcy”. They say that it is “broke” or “out ofmoney”. And they are exploiting thesefears to pressure Americans to reduce the size of their public sector spending,and grant even more power to the private sector firms that helped steer us intoour current crisis. But the claimsbehind these warnings about government debt are often downright false. At best they are often wildly overblown, andbased on significant misunderstandings about how our government’s monetary systemoperates, and how any monetarily sovereign government relates to the world ofprivate sector finance with which it interacts. Hereare several facts to bear in my about federal government debt in the UnitedStates:
First, the US government, as a monetarily sovereign nationthat is the monopoly producer of the US dollar, can face no solvency risk otherthan a voluntary, self-imposedsolvency risk. The US borrows in dollars, a currency that theUS government itself controls and produces. The US government therefore simply cannotgo bankrupt and fail to pay its debts unless the US Congress chooses to prohibit the TreasuryDepartment from paying those debts, by choosingto prohibit the Treasury from making use of the inherent monetary power ofthe United States. Now this is in factwhat the US Congress threatened to do in the summer of 2011. That is not because the government faced anexternally imposed solvency crisis. Itis because some members of Congress chose to manufacture a crisis bythreatening a voluntary default, inorder to blackmail American citizens and other members of Congress intoreducing the size of public sector spending.
It is true that the Treasury Department is currentlyconstrained by Congress to sell its bonds to private sector lenders. But that is again an arrangement thatCongress has chosen. At any timeCongress could enact legislation permitting direct borrowing from the Fed –effectively creating what I called “M-bonds” – or direct the Fed to credit the TreasuryDepartment’s account by any amount Congress desires, including whatever amountmight be necessary to pay any existing debt liabilities. So there is simply no risk of US governmentbankruptcy other than the risk that the US Congress might, somewhat recklesslyand fanatically, choose to default onUS government debt.
The only real constraint that needs to be born in mind inthe area of government borrowing is the policy constraint of pricestability. Once economic activityreturns to full capacity, the need to preserve price stability will requirethat government debt liabilities to the private are met through processes that beginto remove compensating monetary assets from the non-governmental sector throughtaxation rather than processes that continually expand those monetary assetsthrough more central bank purchases of debt. Most of that transition will occur automatically. As economic growth returns and incomes rise,tax revenues will automatically rise along with the incomes.
Some worry about the size of the debt we owe to foreignlenders, including foreign governments. The Chinese government, for example, currentlypossesses over 9% of US treasury debt. Politicians use this fact to portray the Chinese as a potentiallyoppressive creditor that could choose to “call in our loan” and drive us intoinsolvency or crisis. These fears arealso overblown. When the Chinese or otherspurchase US treasury debt, they purchase it with dollars – dollars they alreadypossess. There are only so money thingsyou can do with a foreign government’s currency you possess. One ofthose things is to buy bonds from that foreign government (or save it with afinancial institution that is itself buying government bonds). A bond issued by the Treasury Departmentfunctions as the equivalent of an interest bearing savings account for peoplewith dollars to save. If the dollarholding foreign nation chooses not to put their dollars in “savings” by purchasingbonds either directly or indirectly, they will have to keep their dollars in“checking” by leaving them in bank accounts earning lower interest. Why would they do that?
Suppose the Chinese decided they no longer wanted topurchase US government debt. What wouldthey do with their dollars? Their onlyreal alternative would be to exchange those dollars for something else. That is, they could buy something with thedollars in markets where the dollar is accepted – primarily America. At that point, it is hard to imagine themedia screaming, “Crisis! Chineseseeking to buy massive amounts of American goods!”
Under current arrangements, as we have seen, the TreasuryDepartment is constrained to sell bonds on the private market. So the fear might be that even if the Fed isprepared to buy up as much Treasury debt as is needed in order to supportTreasury spending operations, the Fed might not get that option if skittishprivate sector borrowers refuse to buy government debt at high prices. Again, the problem with this line ofthinking is that the entire world that does business in dollars has no otheroptions but to save its dollars in savings vehicles that are in one way oranother founded on government debt liabilities. The Fed exercises tremendous control overinterest rates through its open market operations. So realistically, there will always belenders ready to purchase bonds that the government issues, at the interestrates we desire, so long as the Fed stands ready to purchase as much governmentdebt as needed to set the interest rates its desires to set. Borrowing costs for the US government remainextremely low, despite the warnings of those who fear federal government debtis too high. Nor do people in other countries seem any lessinclined to save and do business in dollars. The dollar is currently very strong on world currency markets, despitepersistent warnings by the fear mongers that government money creation willlead to a hyperinflationary loss of value in the dollar.
Some of those who spread fear about dramatic inflation orhyperinflation resulting from government money creation point to the recentrounds of “quantitative easing”, in which the Fed purchased large quantities offinancial assets on the private market. Since the Fed effectively creates the money on the spot that it needs topurchase those assets, some fear that this massive program of purchases hasflooded the economic system with money, and the pressures from this deluge willeventually lead to a runaway rise in prices. But it is important to recognize that when the Fed buys financialassets, that purchase amounts to a removal of money from the economy over timeas well as an insertion of money in the present.
Suppose that some private sector entity A possesses a bondissued by some other entity B, where B can be either the Treasury Department orsome private sector lender. Supposethat the bond commits B to the payment of $10,000 to A over the next fiveyears, on some pre-determined schedule. Now suppose that the Fed offers to purchase this bond from A for $9000,and that A decides to sell the bond because A prefers the $9000 now to thedelayed receipt of $10,000 over five years. It is true that when the Fed makes the purchase it inserts an additional$9000 into the economy. But rememberthat $10,000 was originally supposed to move from B to A over five years. Now the $10,000 will flow from B to the Fedrather than to A. In other words, theFed has poured $9000 out of its infinite money well into the private sectortoday, but over the next five years B will pour $10,000 back down into thatinfinite money well. That amounts to anet removal of $1000 from the privatesector. All the Fed has done with itsbond purchase is swap out one financial asset- a bond – for a different asset –some money. It has adjusted theschedule of insertions and removals of money from the private sector withouteffecting a net increase in the amount of money inserted.
Finally, before moving on to a discussion of making the USmonetary system more democratic, it will be worthwhile saying a few words abouthow the current European monetary system falls short of the kind of monetarysovereignty – or near monetary sovereignty – I have attributed to the system inthe United States.
European governments are all part of a currency union – theEurosystem. Each government issues itsown bonds and each operates its own central bank. All of these transactions occur in Euros,the common currency of the Eurosystem. But those national central banks are subject to rigorous policyconstraints set by the European Central Bank. The individual countries themselves do not set their own monetarypolicies, and they borrow in a currency they do not themselves control. In effect this makes each government acurrency user rather than a monetarysovereign. If we think of governmentbonds as the equivalent of bank savings accounts, then each of the governmentsis in effect the equivalent of a savings bank that competes with the other governments in the Eurosystem to offerattractive interest rates to savers. This gives holders of Euros tremendous bargaining power to drive up bondyields and interest on government debt, because they can always take theirmoney elsewhere to other governments if they don’t get the yields theywant. And since the individualgovernments do not control their own monetary policies, they cannot maintainspending during periods of low revenues by selling debt directly to theirnational central bank and drawing on the money-creating power of that nationalcentral bank. Only the European CentralBank can alter those monetary policies, but the ECB is prohibited by treatyfrom buying government debt directly. TheECB also lacks the capacity to carry out a fiscal policy of central bankfinanced spending operations in Europe.
In effect, then, the technocratically-managed ECB runsEurope’s private banking system and the private banking system runs Europe’sgovernments. The citizens of Europehave turned their capacity for economic self-determination over to anundemocratic, continent-wide banking conglomerate. This is the worst kind of nightmare in thelong struggle between democracy and private wealth. It’s not as though the Europeans havesurrendered their sovereignty to be part of a larger sovereign democraticgovernment encompassing all of Europe. Rather, sovereign democratic governments have been transformed in thisinstance into something like mere business enterprises that are dependent onprivate wealth and financing for their operations. These governments now govern only at thepleasure of bankers.
Thisis Part Four of a six-part series. Previous installments are available here: One, Two, Three
