Monetise public debt and deficits

The diminished role of monetary policy

The paralysis of financial intermediation today means that monetary policy (cuts in the official policy rates) have become largely ineffective in stimulating demand. Such cuts now appear to have little if any effect on either the marginal cost or the availability of external funds to non-financial enterprises and households. For individual open economies, the exchange rate provides a mechanism for stealing aggregate demand from the neighbours. If most of the neighbours are also in a situation of deficient aggregate demand, this redistribution of global effective demand – which is all that changes in exchange rates accomplish – robs Peter to pay Paul and is not part of a global solution at all.

The continuing importance of liquidity policy

Central banks are doing the right thing by massively expanding their balance sheets to put as much liquidity into the economy as possible. The balance sheets of the Fed and of the Bank of England have doubled since the crisis started and the balance sheet of the Eurosystem has expanded by about 50 percent. This has been accomplished through open market operations of various kinds, through which the central banks have increased their holdings of private securities by expanding the monetary base, mainly by increasing bank reserves with the central bank and other central bank loans.

Central banks will have to continue to do this. The private financial sector has to deleverage massively, but would (with credit markets and wholesale financial markets closed for business) do so in an unnecessarily destructive way if left to its own devices. The household sectors in the US, the UK and a number of other European countries have to deleverage (start saving seriously) on a significant scale. Left to its own devices, the short-run Keynesian aggregate demand fall-out from a necessary reconstruction of household financial wealth could be disastrous. So the public sector has to leverage up (borrow) at the same time the household sector is forced to deleverage.

This process of monetisation of private sector financial instruments can continue almost indefinitely. It is restricted only by the total stock of private financial instruments outstanding and by the central bank’s willingness to add private securities with higher and higher degrees of default risk attached to them to its balance sheet.

The central bank as the handmaiden of fiscal policy

With monetary policy almost powerless and with liquidity provision also subject to strongly diminishing returns, a traditional fiscal stimulus is the only show in town that has not yet been tried.

The central bank is central to the effectiveness of any traditional Keynesian fiscal stimulus through higher public spending or tax cuts whenever the fiscal probity of the government is questionable. This is the case in the US, the UK and a fair number of other countries. Unless your are a misguided supply-sider who believes in the reality of an incentives-based Laffer curve (tax cuts pay for themselves) or a misguided Keynesian-on-steroids (Kahn’s mighty multiplier has become mightier yet), tax cuts and public spending increases raise the public sector deficit.

In practice, when the government provides a fiscal stimulus to demand, the Treasury has to issue additional debt. If this debt is not bought by the central bank, it will have be be held willingly by the domestic private sector and the overseas sector. With the willingness and capacity to raise future taxes or cut future public spending compromised by years or even decades of profligacy, private absorption of the new debt issuance will only occur with higher sovereign default risk premia. It is interesting that even in countries like the UK and the US, where long-term government debt yields have been falling as the private sector has herded into the relative safety of public debt instrument, sovereign CDS spreads have been rising. The long risk-free rate has fallen as a result of the flight-to-safety, but sovereign default risk premia have been rising.

The only way to ensure that larger public sector deficits do not add to the sovereign default risk premia is for the central bank to buy the additional government debt. Assuming the central bank does not finance this purchase of public debt by selling private securities but instead by increasing the monetary base, the deficit is monetised and no financial crowding out occurs. The sale of Treasury debt to the central bank is an intra-public sector transaction that is of no significance as long as the issuance of additional base money by the central bank does more good than harm.

The public sector as a whole (the combined Treasury and central bank) can leverage up by borrowing through the issuance of interest-bearing debt or through the issuance of monetary debt – currency and central bank reserves. It is clear that for the leveraging up of the public sector to be effective in stimulating demand, it will have to occur through monetary issuance.

What’s special about central bank money?

The monetary liabilities of the central bank are liabilities only in the legal sense. They are not liabilities in the substantive economic sense. This is clearest with currency. A UK 10 pounds note carries the inscription: I promise to pay the bearer on demand the sum of 10 pounds. That’s good to know. All it means is that I can take my scruffy old 10 pound note to the Bank of England and exchange it for a new crisp 10 pound note. If I am lucky, I might get 2 five pound notes instead. But a given face value of currency does not give me a claim on the issuer for anything other than the same amount of currency. My 10 pound note is a claim on a 1o pound note. Because it is legal tender, and because it is also de facto accepted in payment for just about anything, it is wealth – an asset – to the holder. But to the issuer, it represents no obligation to provide anything else to the holder. It is not a liability in any substantive sense.

So currency is not just special because it pays a zero nominal rate of interest. There also is no principal to be repaid. As a store of value to the issuer, a currency note is like a zero coupon perpetuity or consol: a commitment to pay nothing forever. In ordinary finance theory, such an instrument should be worth nothing if the risk-free nominal interest rate is positive. This tells us how useful finance theory is, as currency notes have value (most of the time).

What is true for currency is not automatically true for the other monetary obligation of the central bank: commercial bank reserves held with the central bank. Most of the time these reserves now pay interest. And it also is not clear that the holder does not have the right to redeem them into something else – currency notes. This does not, however, change the picture fundamentally. First, the central bank could declare commercial bank reserves with the central bank to be legal tender (or to be legal tender for interbank transactions and other wholesale transactions). Second, the central bank could, at some cost in terms of inconvenience and the appearance of a large number of armoured vans on Lothbury, at the back of the Bank of England, redeem all reserves in irredeemable, non-interest-bearing currency.

So the public sector can turn its debt to the domestic private sector and to the rest of the world into monetary form. It can monetise the public debt. It can do the same with the consolidated financial deficit of the public sector. The general government (central, state and local), the Treasury, that is, would still be borrowing through the issuance of debt, but this debt would be acquired by the central bank who finances this purchase by issuing additional base money.

Will the central banks play ball?

Many countries, including the UK and the countries of the Eurosystem, have created operationally independent central banks. The Fed has a much lower degree of operational independence, both legally and substantively, than the ECB or the Bank of England. The part of this that matters is that an operationally independent central bank cannot be told by the government to buy (monetise) the government’s debt. It can choose to do so, it cannot be told to do so. The key question therefore becomes: would the central banks be willing to monetise the larger public sector deficits that will result from the proposed fiscal stimuli?

Using boil-in-the-bag monetary economic analysis, you might conclude that central banks already do so. No central bank controls the monetary base and lets the official policy rate be determined endogenously by the requirement that the exogenously chosen quantity of monetary base is willingly held by the private sector. Instead (to a first approximation), central banks set the official policy rate exogenously, sit back and watch the endogenously determined quantity of the monetary base that the private sector wants to hold, come out. Nice and a passing grade, but no merits.

First, it must be recognised that the value of the current official policy rate has a negligible effect on anything that matters – inflation, employment, liquidity or happiness.

In the boil-in-the bag model, the current official policy rate is the only interest rate we ever see. It is a static model. With inflation expectations either ignored or taken as given, nominal and real interest rates are the same. To get to a policy-relevant representation of reality we need to look at a properly dynamic model. In a dynamic model, deficits have to be financed. As long as there are deficits, financial asset stocks will be changing. Budget constraints, including the government’s budget constraint, have to be thought of dynamically – as intertemporal budget constraints. In addition to asset stock dynamics, price dynamics and expectational dynamics are important.

What this dynamic or intertemporal approach to economics means for monetary policy effectiveness, is that it is the entire sequence of past, current and contingent future official policy rates that matters for price stability and the real economy. It does so mainly through private sector expectations of these future contingent official policy rates. These then get priced into longer term debt instruments and other asset prices, such as the exchange rate and the stock markets. Past, current and contingent future liquidity management by the central bank will in addition affect the liquidity premia and even the credit risk premia on longer-dated financial instruments.

When the central bank monetises public sector deficits (or the outstanding stock of public debt), it is therefore not enough to say that this means that the central bank simply fixes ‘the’ official policy rate and absorbs whatever amount of additional government debt is required to make that fixed value of the official policy rate consistent with equilibrium in the financial markets and the rest of the economy. We must translate this statement into a dynamic setting in which past, current and future contingent official policy rates matter for today’s economic performance.

In addition, we must recognise that the official policy rate is not set exogenously (even though it will be fixed or constant for a reserve maintenance period, if there are no unscheduled meetings). The sequence of past, current and planned/expected future official policy rates is set to achieve the objectives of the central bank. Given these objectives, in any given period, the official policy rate can be viewed as determined endogenously as a function of the information available to the monetary policy makers at the last rate setting meeting. In the UK the primary objective is price stability (defined as 2 percent inflation as measured by the defective CPI index). Subject to that, the Bank of England’s Monetary Policy Committee is to support the objectives of the government, among which are growth and employment. In the euro area, the ECB’s primary objective is also price stability (inflation below but close to 2 percent on the same defective CPI index). Without prejudice to that, it too can support all things bright and beautiful. The Fed pursues a more symmetric dual mandate involving employment and price stability. The Fed operationalises price stability through an incomprehensible mixture of the core and headline PCE deflators.

If an operationally independent central bank monetises a government deficit, this means either that the central bank believes that this monetisation is not in conflict with its objectives or that it has changed its objectives at least to some degree.

The Bank of England appears to have modified its operationalisation of the price stability objective in its latest inflation report. As I interpret its November inflation forecast, it now does not plan to return to the 2 percent inflation target over the standard, 2-year horizon it has always appeared to focus on in the past. Instead, as its central projection, it is predicting close to a 1.2 percent undershoot of the inflation target at a two-year horizon. The entire fan chart for inflation, reproduced here, shows that there is a material chance of negative inflation rates at a two year horizon.


If this were indeed their true central forecast (based on market expectations of future interest rates), this would mean that the MPC did not cut Bank Rate by enough at the last meeting- presumably because a cut of more than 150 basis points would have spooked the markets. Alternatively, and more likely, the MPC deliberately produced an excessively low inflation forecast, to permit it to cut rates very aggressively at the next few meetings in order to prevent the threatening inflation undershoot at a two-year horizon. When inflation in two years time turns out to be higher than expected, the MPC can blame a bad inflation forecast rather than an increase in the period over which they are pursuing a return of inflation to its target.

But I believe that an extension of the time horizon over which inflation is brought back from the current 4.5 percent rate to the 2.0 percent target from around 2 years to 3 or 4 years is exactly what has occurred. This is because I consider it extremely implausible that, with sterling having fallen to the extent it has (8 percent on the broad effective exchange rate measure since the November 2008 inflation forecast was made and around 25 percent since last year’s peak), the MPC would really believe that the most likely outcome is a significant undershooting of the inflation target at a two-year horizon. It looks as though the MPC felt they were faced with the choice of publicly admitting that they were extending the period of time over which they planned/hoped to return to the 2.0 percent target rate (thus risking their reputation for being serious about inflation) and extending the horizon over which they planned a return to 2.0 percent without publicly admitting this and producing an excessively disinflationary inflation forecast to ‘explain’ continued aggressive rate cuts (thus risking their reputation as competent forecasters). In my view they went for the first option.

What this means for the main issue under consideration – the monetisation of deficits – is that the MPC is likely to be willing to purchase additional public debt in amounts and at interest rates that do not imply a raising of their future path of official policy rates. For an interest rate setting central bank, this is what is meant by monetisation: a larger government deficit does not raise the bank’s planned contingent path for future official policy rates. Nor does it lead to a planned tightening of current and future liquidity provision, given the path of current and future official policy rates. This may, in practice, require that the central bank purchases part, most or all of the additional public debt that is issued.

Can the central bank make such a monetisation commitment and retain its inflation-targeting or price stability pursuing credibility? Only if, when stability and normal functioning are restored to financial markets, when banks are lending again and when the economy has regained a high degree of capacity utilitzation, the monetisation is undone. Under current extreme conditions, there is a massive increase in liquidity preference in the financial sector and throughout the economy. Additional real money balances are absorbed without inflationary consequences.

But when the good times come back (and they will!), the additional central bank liquidity injected into the system will be burning holes in the pockets of bank loan officers and of lenders and spenders generally. To prevent an inflationary surge, the public debt will have to be demonetised again, by the central bank selling public debt to the private sector and the overseas sector. That is only consistent with sustainable public finances and the absence of high and rising sovereign default risk premia, if the government ensures that additional taxes and/or lower public spending will be forthcoming when the economy recovers.

That commitment by the central bank to demonetise the public debt in the future has to be credible today, lest there be an immediate increase in long-term nominal interest rates when the central bank monetises public debt and deficits, due to higher long-term inflation expectations. That commitment by the government to run sufficiently large primary budget surpluses (budget surpluses excluding interest) has to be credible today, lest there be an immediate increase in sovereign default risk premia as the government increases its borrowing.

There can be no doubt that, if public spending is not cut, future taxes have to be higher if the government borrows more now and permanent monetary financing is not an option. It is also virtually certain that tax rates will have to be higher. The government will get some help from the positive effect on the main tax bases (income, profits, consumption) of the expansionary fiscal measures, but not enough to prevent the need for future higher tax rates (taxes as fractions of the relevant tax bases).

Can the ECB monetise government deficits?

The brief answer is: of course they can! Yes! And most definitely!


As is clear from 21.1, overdrafts and any other type of credit facility with the ECB or NCBs are prohibited (that includes ways and means advances of the kind the UK still uses) and “the purchase directly from them by the ECB orNCBs of debt instruments”. They can, however, buy government debt in the secondary markets. And they do! The balance sheet of the Eurosystem (ECB plus 15 NCBs) of Nov 4 shows it held about € 37 billion of general government debt (out of total assets of over € 2 trillion). The ECB’s own balance sheet is very small. Most of the stuff is with the NCBs in the consolidated balance sheet of the Eurosystem, which is huge.

In the 2008 Convergence Report by the ECB, the ECB elaborates these issues on pp. 22-25; note that “This report does not cover Denmark and the United Kingdom, Member States with a special status that have not yet adopted the euro.” This is why the UK still has its ways and means advances.

So money financed deficits (helicopter drops) are possible in the euro area, but they require the consent and indeed active participation of the ECB. It is an insignificant minor operational detail, that the ECB cannot buy sovereign debt instruments directly from the governments involved. It can however, buy any amount in the secondary markets. From a macroeconomic management perspective therefore, the ECB can monetise government deficits and debt in the euro area. Whether it chooses to do so is a separate matter. My guess is that it will. Central bankers are committed citizens, but few of them are likely to think that their heads would look good on spikes. In addition, provided the double credible commitment can be delivered, there is no conflict between price stability and monetising the debt. The double commitment is, once more, (1) the central bank will demonetise again when liquidity preference shrinks, and (2) the government(s) will act countercyclically in good times as in bad times and will raise taxes or cut public spending as soon as the economy is back to normal, by an amount equal, in present value, to the addtional stimulus delivered today. It is the second commitment that is least likely to be credible. The heads of those responsibile for pro-cyclical fiscal policy during the past decade definitely deserve to be seen on spikes – or at best on the opposition benches.


Article 21

Operations with public entities

21.1. In accordance with Article 101 of this Treaty, overdrafts or any other type of credit facility with the ECB or with the national central banks in favour of Community institutions or bodies, central governments, regional, local or other public authorities, other bodies governed by public law, or public undertakings of Member States shall be prohibited, as shall the purchase directly from them by the ECB or national central banks of debt instruments.

21.2. The ECB and national central banks may act as fiscal agents for the entities referred to in Article 21.1.

21.3. The provisions of this Article shall not apply to publicly owned credit institutions which, in the context of the supply of reserves by central banks, shall be given the same treatment by national central banks and the ECB as private credit institutions.

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website