From Central Bank Independence to Democratic Public Finance

By Dan Kervick

We have reached that point in the United States.  The present system of public finance in the US is inefficient and antiquated: its fusty architecture hampers the capacity of the national government to respond to economic fluctuations and crises in a timely and effective manner; its byzantine operational complexity thwarts democratic governance and generates pervasive public confusion about the full range of public policy options; and its over-reliance on government bonds means that wasteful and unearned profits flow to some of the most affluent members of US society, as they are paid service fees for intermediating what ought to be routine operations of the government.

The anchor of the existing US financial system is the Federal Reserve System: a central bank that possesses forms of operational and political dependence that have become a model for central banks around the world during the neoliberal era.   But it is this high degree of independence that is the main source of financial inefficiency.  The central bank anchor that was intended to produce stability serves instead to drag the country down when it most needs to move briskly.  Central bank independence – at least in its current form – is an idea that has failed the test of crisis, and should be discarded.

The necessary correlate of central bank independence is the self-imposed dependency of fiscal authorities.  Many modern democratic governments, despite the fact that they are the issuers of their own national currencies, have established public financial systems in which their own fiscal operations are bound by formal constraints that eschew the full benefits of a national currency.  These constraining rules require the national treasury to build up account balances through taxation and debt-issuance, and then to spend only within the limits imposed by the sizes of those balances.   The authority to issue currency, an inherent power of sovereign national governments, is then delegated to the central bank alone.  The point of this practice is presumably to prevent a democratically elected legislature from spending the country into monetary instability.  But the resulting system is authoritarian and paternalistic, an artifact of the capture of public finance by wealthy patrician elites, powerful financial institutions and anti-democratic reactionaries.  This paternalistic system is not a necessary condition for monetary stability.  It creates an artificial and burdensome institutional division between monetary and fiscal policy, and needlessly lames our democracy in the pursuit of the public interest.  It should be abolished.

To make my case for these reformist claims, I will begin with a simplified model of streamlined public finance as practiced by a hypothetical currency-issuing national government without an independent central bank.  I will then compare that simple model of financial governance to the more complex and rigidly structured system based on central bank independence.  Finally, I will conclude with some thoughts on ways of combining the more efficient and powerful public finance of the simplified model with the monetary policy discipline that is supposed to be provided by independent central banking.

To begin, imagine an extremely simple form of national government, one without a lot of separate agencies, divisions, branches and departments.  You can imagine an all-powerful czar if you like.  But for the purposes of this thought experiment, a unicameral and democratically elected legislature that is responsible for all aspects of national governance works just as well.  This government enacts laws and enforces the laws it enacts.  It taxes and it spends.  It also issues and manages the currency the public uses for transacting all domestic business.   And it issues government bonds as well: a variety of financial instruments carrying different maturities and rates of interest.  These bonds are government IOUs: promises for the payment of specific amounts of the national currency at or over pre-specified periods of time.

Nothing important hinges on how our imaginary government issues currency.  It can issue the currency by manufacturing and spending physical notes and coins; it can issue currency by crediting electronic balances to its own account and then transferring those balances to private sector accounts via the electronic payments system it supervises; or it can issue currency by crediting electronic balances directly to private sector accounts, without first crediting the balances to any government account.  The precise system used is not essential, and reflects only a decision of technological convenience.

Since the government can issue the currency whenever it wants, it doesn’t need to obtain the currency first either by taxation or by bond sales in order to spend.   So if the government doesn’t need to tax to obtain the currency, why does it tax?  It taxes for three main reasons: one is to help preserve the stability of the currency.  Since the public is constantly using the currency to buy things and sell things, there is a public interest in maintaining generally stable prices for goods and services given in exchange for the currency, to the degree at least that the real costs of producing those goods and services haven’t changed.   Stability can be managed to some degree by coordinating the interrelated operations of taxing, spending, and issuing currency.

The second reason the government taxes is to ensure a continuing demand for the currency.  If individual people, households and firms were to grow less desirous of acquiring the government-issued currency, the public would lose much of its ability to achieve public purposes by buying goods and services with that currency.   The government and its pursuit of the public interest would then become dependent on other media of exchange, and on the private enterprises or foreign governments that happen to control those alternative media.  Tax obligations assessed by the national government in the national currency ensure a continuing demand for the currency, and serve to fight off invasive competitor currencies.

A third reason for taxation is to directly influence the distribution of monetary incomes.  High marginal income tax rates, for example, might be used to limit the growth of personal income at the top of the income scale in order to preserve the integrity of government, to build social solidarity, to minimize class antagonisms or to produce other desirable social effects.  And since price stability requires a prudent ratio of spending outlays to tax revenues, taxes on the affluent can also provide the policy space for spending directed to other purposes.

Now in the same way that our imagined government issues currency, it also issues a range of interest bearing securities of various maturities and payment terms.   We can call all of these securities “bonds” for short.  But why should such a government issue any bonds at all?  Why not just issue currency?  And why swap bonds for currency or currency for bonds?   The point of these transactions is to adjust the quantities and term structures of the financial assets held by the public.   A bond is like a time release capsule.  It represents an injection of currency into the private sector, but it is an injection that takes place over time, on a pre-determined schedule.  The issuance and manipulation of government bonds by the issuing government gives that government more control over the pace and volume of public accumulation of government-issued financial assets.  It also provides secure savings vehicles that can be useful in stressed economic environments when the public lacks confidence in, or options for, private sector investment.  And it also provides a tool for the management of interest rates, particularly the rates financial institutions charge one another as they borrow and lend their currency reserves.

The fundamental point here is that in its decisions about currency issuance and bond issuance our hypothetical government follows Abba Lerner’s Second Law of Functional Finance:

LL2: The government should borrow money only if it is desirable that the public should have less money and more government bonds, for these are the effects of government borrowing.

The use of the term “borrowing” is actually quite misleading in application to a currency-issuing government.   To say the government borrows money only means that it issues bonds and exchanges the bonds for currency that the public already holds, currency that was itself issued earlier by the government.   Lerner’s point is that the government should carry out these swaps only if it sees a public policy interest in swapping interest bearing bonds for non-interest-bearing currency.   Since the government issues the national currency, it can never have the problem of being unable to redeem its IOUs, since it can always simply issue the currency it uses to redeem the bonds.

That is not to say a currency-issuing government of the kind we are imagining can’t make monetary policy mistakes in its decisions about issuing or collecting currency and bonds.  The government can destabilize the economy, or make bad situations worse, if its decisions about the issuing, spending, redeeming and taxing of government-issued financial assets are not appropriate given the current nature of private sector economic activity, and given the current desire of households and firms in the private sector to accumulate or reduce savings of those assets.  But there can never be a concern about the solvency of the government’s IOUs.

Notice that for our hypothetical currency-issuing government the decision about whether to issue bonds and swap them for currency while running a deficit is strictly a policy option.  The government doesn’t need to borrow to run a deficit, and always has the option of simply spending more currency into the private sector than it taxes out of the private sector, in effect creating and issuing currency in the process.  It doesn’t need to finance the deficit by acquiring the additional money from some external source, since the government itself is the source of the currency.  If the private sector economy is stagnating, and running below capacity with high unemployment, the option of deficit spending without further borrowing might prove best.

Now suppose our imagined government wishes to run a deficit during a given period of time, and voluntarily makes the decision to issue new debt to accumulate back from the public sector a total amount of currency equal to the gap between spending and tax revenues.  We’ll say that the government is in that circumstance operating a bond-driven deficit.  With bond-driven deficit spending the net amount of currency held by the non-governmental sector does not change.  However, the total value of government-issued financial assets held by the public does increase by an amount corresponding to the value of the bonds that are issued.   Suppose instead, however, the government carries out its deficit spending in such a way that none of the gap amount is offset by bond sales – it simply spends more into the economy than it taxes out of the economy.  In this case we will say that the government is running a currency-driven deficit.  With a currency-driven deficit, the quantity of currency held by the public increases, while total value of the bonds held by the public does not.  Obviously, our imagined government can operate deficits that fall on a spectrum between the two extremes of a bond-driven deficit and a currency-driven deficit.

Economists are in the habit of drawing a sharp theoretical distinction between monetary and fiscal policy.  But notice that for our imagined government monetary policy and fiscal policy are seamlessly integrated.  If the government runs a currency-driven deficit, for example, it is easy to see that the decision is simultaneously an act of monetary and fiscal policy.   It is an act of monetary policy because the combined impact of the spending and taxing operations affects the quantity of currency held by the public; it is an act of fiscal policy because it does involve spending and taxing.

The system described so far gives the government a great deal of flexibility in using its authority to issue currency and bonds as part of the means to achieve its purposes.   But now let us suppose our hypothetical government voluntarily adopts the following somewhat more rigid system, which we can call the bonds-first system.   For accounting purposes, it divides its spending into two categories: (i) ordinary spending and (ii) bond redemptions and repurchases.  It then resolves that the total quantity of ordinary spending during some standard fiscal accounting period must always be less than or equal to the total quantity of currency that it obtains from the public during that same period via tax receipts and bond sales.   The government subsequently redeems or repurchases the bonds it has issued, and it allows itself the choice at that time of either issuing currency to carry out these operations, or paying for operations in the same way it conducts ordinary spending.  Thus, the only way that the quantity of currency held by the public can increase is through the issuance of bonds and their subsequent redemption or repurchase.

What are the effects of this policy?  Well here is a very important one: every government deficit – that is, every excess of ordinary spending over tax receipts – leads to profits for people who hold extra currency, and who are able and willing to purchase bonds that the bonds-first system mandates must be sold to enable deficit spending.   The purchasers of the bonds will not all be wealthy, no doubt.   But in general, those with large amounts of surplus cash to invest come from the more affluent parts of the community.   The bonds-first system creates a class of wealthy investors who are, in effect, stockholders in the business of government.  The rich get richer by investing surplus cash in routine government operations.

Finally, let’s imagine the system is made even more rigid.  The financial operations of government are divided among two separate agencies, each with its own account and balance sheet, and each granted a great deal of independence over the operations it supervises.  One part of the government issues the currency, the other half issues bonds.  The part of government that issues the bonds also taxes and carries out ordinary spending.   The part that issues currency carries out bond repurchases at its discretion.  It can use also make loans to banks, and to pay interest on some other government-issued assets such as commercial bank deposits at the central bank.  In less conventional times, it can make emergency purchases of privately-issued financial assets.   Bond redemptions on the other hand are carried out by the bond-issuing part of the government, the same part of government that collects taxes and does ordinary spending within its self-imposed budget constraint.

One effect of this system is simply to solidify the bonds-first system in place, and to place full responsibility for whether or not currency will be issued in the hands of the currency-issuing branch.  But is has the further effect of guaranteeing that the government never directly redeems its bonds via currency issuance, but must either tax or issue more bonds to do these redemptions.  Currency is always injected first into the private banking and financial sector, and then impacts the ordinary budget only by being taxed back or loaned back to the government.   Aside from the fact that this process funnels possibly pointless interest income to bond purchasers, the continual financing of interest-bearing debt with interest-bearing debt can create long-term sustainability issues that require restrictions on ordinary spending that would not otherwise be in place.  It’s not that there is any issue about the government’s long term solvency.  But the piling up of interest payments may create pressures for long term price stability management, pressures that are relieved by reductions in other kinds of useful government spending.

This is the institutional setup we effectively have in the United States.  The currency issuing part of the government is the Fed.  The bond issuing part of the government is the US Treasury, whose fiscal activities are directed primarily by Congress, and the President.  In the US public financing family, the Fed wears the monetary pants.

Why in the world would any government decide to have a system like this?   I believe those who are most drawn to the present system and have endorsed central bank independence for genuinely public-spirited reasons, and not simply because they personally benefit from the present bonds-first system, base their support on the belief that central bank independence promotes monetary policy discipline.  They imagine the Congress as Odysseus, the legendary Greek warrior who decided to have his crew members tie him to the mast of his own ship as he sailed past the island of the Sirens.  The independent central bank system is supposed to work something like that.  Congress gives itself a budget.  Spending requires draining money from the private sector in the form of either tax revenues or payments for the purchase of bonds.  In this way, it is felt, the government is able to guard against the temptation to destabilize its own monetary system by injecting too much currency into the private sector through politically popular spending while failing to drain an adequate amount of currency through politically unpopular taxes.

But the social disutility of this system should now be apparent.   Experience has shown that economies can get stuck for a long time in conditions of persistently low employment and sluggish performance.  These economic circumstances call for timely, currency-financed expansions of public investment and public enterprise, or currency-financed subsidies of private household consumption and private business investment.   The government should be able to carry out these expansions without counteracting its own efforts by draining more currency from the economy through additional taxes as a result of a self-imposed budget constraint.  An effective currency-issuing government should therefore reserve for itself the option of spending newly-issued currency directly into the accounts of households and businesses without boosting taxes and without the requirement that it makes additional and irrelevant injections into the financial sector bond market at the same time.  Since monetary stability and price stability depend on managing the total volume of government-issued financial assets held by the public, every dollar injected into the financial sector eliminates some of the space for direct injections into the real economy, where the spending is most needed.

Again, the problem isn’t that the additional bond liabilities can’t be paid off.  The problem is that these additional debt liabilities lead to an inefficient, wasteful misallocation of government financial resources.  If the government needs to purchase a collection of bridges, which represents both a valuable public investment in the future and a needed immediate-term injection of purchasing power for businesses and households, then it is illogical for the government to follow a policy rule that compels it to pump additional interest payments into the bond market in order to carry out the fiscal expansion.  It should not do this absent some additional, compelling public policy reason for supplying more dollars to the financial sector.  Even if the government does not collect a single cent in additional taxes, and rolls the bond debt over indefinitely, the result is a perpetual cash flow to financial institutions and private investors whose activities are completely irrelevant to the building of bridges and consumption by households.  This simply makes no sense.

An additional problem with or current public financial system is that its distribution of responsibilities and its complex institutional architecture – along with some antiquated nomenclature and quaint traditions of communicating through pious obscurities – leads to a great deal of public confusion about what is really going on in our own financial system, and to tragic ignorance about the full range of available policy options.  We see this tragedy playing out right now, as citizens across the country are being guided by partisan rhetoric and elite propaganda to contract the government deficit at a time when the economy very much needs large healthy deficits, all in a misguided wave of hysteria about federal government debt.  It would be much better to have a cleaner, simpler system in place, one in which currency driven deficit spending and debt service are carried out in a straightforward and integrated manner by a single arm of the government, so that the role of monetary policy in federal deficit spending is apparent.

Here, then, are my proposals:

1.  Move the supervision of the commercial banking system and the national payments system, and the authority to issue currency – that is, all of the current obligations of the Fed – to the Treasury Department.  The central bank should operate as a division of the Treasury subject to the direct oversight from Congress with frequent room for timely policy adjustments, legislative deliberation and lively democratic input.  Like other aspects of national economic policy such as budgeting, taxes, appropriations and public investment, monetary policy and financial system regulation should be determined by hands-on, publicly accountable, warts-and-all democratic governance.

2.  Consolidate the Fed and Treasury accounts into a single sovereign US Treasury account, consisting of a currency sub-account and a securities sub-account, through which the ultimate monetary authority of the United States is exercised under the direction of Congress and the President.   The initial dollar balance credited to the currency account is of no matter.  All that is important is that we are able to track the total deficit or surplus during any given period of time.   Expenditures are best seen as dollar creation and may be recorded as “liabilities” of the Treasury, and tax receipts are best seen as dollar destruction and may be recorded as financial “assets” of the Treasury.  But it should be born in mind that the very concepts of financial assets, liabilities and equity are only loosely applicable to a sovereign government that is itself the issuer of the ultimate means of payment for all financial assets and liabilities denominated in the government’s unit of account.  If the government owns a helicopter, it is in possession of a real asset.  But it hardly makes sense to think that the government has meaningfully acquired an asset when it receives a dollar, not when it issues dollars in whatever quantity it wants at virtually no cost.

 3.  The currency account, used for spending and tax receipts, should be permitted to carry a negative balance without formal limit as to quantity or time.  We can think of this as an “overdraft” system.  But an account overdraft in the private sector is a debt of the account holder to the financial institution that provides the account.  In the case of a monetarily sovereign government, the account holder and account provider are the same entity, and so whether the account is seen as being “overdrawn” or not is only a bookkeeping convention made relative to an arbitrary starting balance which the government itself determines.  The negative balance does not represent any kind of government debt to itself that must be repaid, nor is there any kind of penalty the government exacts on itself for having this negative balance.  It is rather more like a below-zero temperature on a temperature scale: a means of keeping track of changes in temperature relative to an arbitrary temperature designated as the zero point on the scale.

 4.  The Treasury would possess the authority to continue to issue securities as a public saving utility, and as a tool for managing the shape of the government deficit over time, and the pattern over time of private sector financial asset accumulation.   But there should be no requirement that deficit spending – an excess of expenditures over tax receipts – is always to be matched by a corresponding quantity of bond sales.

These are institutional changes that should be adopted to give the United States a more flexible, nimble and high-powered system of public finance.  But we can also ask what novel techniques could be employed for achieving public purposes within the framework of the new institutions.  How do we adapt our public financing practices so that, in addition to carrying out its routine fiscal responsibilities, our government is also able to respond quickly and efficiently to cyclical crises and novel emergencies calling for prompt action, and to deliver price stability, full employment, and the greatest achievable level of economic progress and well-being?

One thing to bear in mind is that, as a plain matter of accounting, the deficits or surpluses of the government sector, the external sector and the domestic private sector must sum to zero.   So if the government is running a surplus, the domestic private sector and external sector must be in a combined deficit.  And in order for the non-government sector to run a surplus, the government must necessarily run a deficit.  Furthermore, in a system in which financial assets and liabilities exist in a hierarchy, and are all ultimately grounded in the financial assets emitted by the government, a steadily growing private economy must be supplied with a steadily growing stock of government-issued financial assets just to maintain financial health and stability.

Note also that income that is saved is not spent.  So during periods of extraordinarily high private saving – which includes periods of high private sector debt service – the government must increase its deficit just to maintain total non-government spending at previous levels.  Increased government deficits can also help to liberate “animal spirits” that are suppressed by low levels of consumer and investor confidence, and help the private sector escape from prisoners’ dilemma situations in which firms have settled into a low-employment equilibrium.  Rather than thinking of a government deficit as an accumulation of burdensome debt that must be paid off in good times to counterbalance its being run up in tough times, it is better to recognize that the government should probably run a deficit at more or less all times, to accommodate or stimulate the continuous growth in the economy by the continuous provision of additional monetary assets.  Michael Hoexter makes the very compelling point that “deficit” is a very misleading term to use for an excess of spending over tax receipts by a currency issuing government, since that term connotes a lack or deficiency of some kind.   It would be better to describe this phenomenon as the government’s net contribution to the financial position of the non-government sector.

What is important, then, in thinking about new fiscal policy techniques is that by the adoption of these techniques the government be empowered to act quickly and effectively to expand its net contribution when needed, and in the ways that are needed, and is in a position to contract the net contribution just as quickly if a buildup of undesired inflationary pressures in a strong economy calls for a contraction.  There are a variety of tools we should think of adding to the fiscal policy kit that might serve us well here.  These are a few:

5. Flexible taxation – Rather than approving completely precise, fixed tax rates for specific groups of tax payers, Congress could pre-approve a set of tax ranges, and delegate adjustments within those ranges to the fiscal managers in the Treasury Department.   For example, a taxpayer with a current top marginal rate of 35% might instead have a top marginal tax range of 33%-37%.  If they are currently paying 35%, and the Treasurer determines that the government should expand its net contribution to the non-government sector, then the Treasurer could drop that rate to anywhere between 33% and 35%, without the need for additional Congressional action.  Paycheck effects would take place immediately.  All other rates would be adjusted accordingly, if Congress has decided on a policy of universal rate modification.  If the Treasurer decides there is a need to tighten up the net contribution, the rate could be increased to anywhere between 35% and 37%.  If more dramatic rate changes are needed, of course, Congress would then need to pass additional legislation.  And Congress would, as always, reserve the right to step in at any time to change the policies it had pre-approved.

6. Flexible spending – In addition to approving and appropriating the funds for specific kinds of government spending during a given fiscal period, Congress can pre-approve large packages of infrastructure spending, education spending, direct payments, subsidies and other varieties of useful public spending to be carried out over 5 or 10 year periods.  The Treasurer would then have the discretion to manage this spending over a longer period of time than a single fiscal year, and accelerate it or decelerate it as macroeconomic conditions warranted.  As before, Congress would always reserve the option to repeal its previous legislation, and step in with new initiatives.

7.  Negative interest bonds – The Treasurer could be given the authority to issue negative interest bonds and swap them for dollars held by large financial institutions or other designated parties.  Why would anyone voluntarily purchase a negative interest bond?  They wouldn’t.  Negative interest bond transactions would be mandatory.  A negative interest bond is just a type of tax.  Rather than imposing a one-directional tax that may or may not be offset later by some type of spending, a negative interest bond amounts to an immediate tax with a partial tax refund later.  This type of imposition might be useful in cases where the Treasurer determines it would be beneficial to reallocate exiting financial asset holdings rather than create more currency or other financial assets immediately, but believes it would be best if the new financial assets are then supplied gradually over time.

8. Negative interest lending – Either working through the banking sector or working directly with firms and households, the fiscal authorities could offer negative interest loans – in effect, partial subsidies for particular types of consumption and investment spending prioritized by Congress.  These subsidies would dovetail with other forms of direct spending on public investment to encourage mutually supporting economic activity in areas that the public, working through its representatives Congress, has deemed important.  As before, the decisions could be pre-approved and “banked”, to be carried out over periods of 5 years, 10 years or longer, and then put into action when economic conditions warrant an expanded government contribution.

Accustomed as most of us have become to the system of independent central banking and self-imposed government budget constraints, some will find these proposals radical, at least initially.  Skeptics have always argued that monetary policy and the currency system are too dangerous or too temperamental to be left to direct, hands-on legislative control, and that monetary policy must therefore be insulated from democracy.  Realism and historical experience tells us that the motivation of many of these putative skeptics is not to protect monetary policy and the national interest from the unruly and misguided mob; but to protect privately held money and those who have the most of it from the claims and obligations imposed by the public interest.   But some of the skeptics are honest and well-intentioned.   Regardless of the skeptics’ motives, the same skeptical claims were made in the past about the suitability of democratic legislatures for directing spending policies, or for setting tax policies, or for governing the military, or for determining the qualifications for offices, or for making the laws that keep the civil peace.   And yet now, after a few hundred years of expanded democratic control over all of these areas, carried out by some of the most successful governments in the history of the world, delivering broad and previously unimagined levels of prosperity to swelling masses of humanity, I think we can see that those earlier skeptics were wrong.  They will be proven wrong again on the score of democratic direction of monetary policy.

Democracies can do this; they can guide and direct their monetary affairs in a pro-active way toward the effective achievement of public purposes.  Obviously they will make many mistakes, as they do in all other areas.  But the benefit, as it always is with democracy, is that the mistakes and foibles of democratic government are more than offset by the good that comes from limitations on the corruption attending private concentrated power; by checks on the avarice and self-serving of elites; by reductions in the exploitation of the labors of the many for the benefits of the few; by the introduction of multifaceted legislative deliberation representing a large number of concerns and interests, and by the elimination of some of the social dysfunctions caused by secrecy and disinformation.   Democratic governments can run their own monetary policies without handing the country’s inherent monetary power over to an authoritative “dad” who keeps them in line, keeps his own counsel, makes most policies in secret with a limited number of peers, and grandly guides public expectations though cryptic and portentous pronouncements.    It’s time to end the era of the aloof, magisterial and serenely independent wizard banker, and hand monetary policy over to the grubby but marvelously effective public deliberations of wrangling democratic legislatures.