Massive efforts to support health systems, meet citizens’ subsistence needs, and preserve business establishments in the face of the COVID-19 pandemic have caused government spending in many countries to skyrocket. While there is unusually broad agreement that aggressive fiscal policies are necessary (Baldwin and Weder di Mauro 2020), there is also growing concern over the consequences of the precipitous rise in already high government debt levels that may be slowing the adoption of further beneficial but costly interventions. Meanwhile, the pandemic has exacerbated an existing macroeconomic concern – namely, the risk of continued below-target inflation, or now even disinflation, in the face of sharply curtailed demand.
Economic policymakers therefore face three interrelated challenges: (1) providing large amounts of debt-funded assistance quickly due to the pandemic, while (2) avoiding accumulating unsustainable levels of government debt and (3) heading off deflationary pressures.
For low-income countries, the first two of these challenges underlie the extraordinary call for a debt service standstill (Bolton et al. 2020). For advanced economies, we believe that monetising a portion of debt created to fund benefits for unemployed or furloughed workers could simultaneously address all three challenges, and that ultimately it would be the least costly and most palatable choice among the available alternatives.
Mechanics of monetisation
The sort of programme we have in mind would quickly put cash into the pockets of households with diminished incomes and at the same time avoid adding to the public debt. In one version for the US, Congress would enact legislation authorising the Treasury to create special issue bonds to be given to banks, whose value would be credited to individuals’ checking accounts in amounts determined by the programme’s rules. To direct the money where it is most needed, the deposit supplements could be based on a percentage of the individual’s reduction in average paycheque deposits relative to the three months preceding the pandemic, subject to a cap or other limits to ensure progressivity.
The special issue bonds would be zero coupon perpetuities and therefore would not obligate Treasury to any future payments. The legislation would require the Fed to buy these bonds from the banks at par and treat the resulting reserves as non-interest bearing (effectively increasing required reserves by the value of the bonds). The bonds would then remain on the Fed’s balance sheet indefinitely. This monetises the special issue bonds.
This particular mechanism amounts to a directed helicopter drop of money, as also advocated by Gali (2020). The legislative and accounting gymnastics built into our proposal have several important purposes. The first is to put the decision to monetise a portion of emergency spending into the hands of elected officials rather than the monetary authority. That would avoid, to the extent possible, the perception that the central bank is overstepping its authority by taking fiscal actions, thereby helping it to protect its future independence. The second is to underscore that the policy is in response to a specific and unusually problematic constellation of events, and that debt monetisation should not enter the standard toolkit of policymakers. Its more mundane purpose is to ensure that balance sheets of all participants in the transactions in fact remain balanced, and to allow the Federal Reserve to monetise the debt without having to report a negative book equity position, which we believe would cause no real problem but could be more controversial.
A similar approach may be just as useful in the euro area, where fiscal authorities are still debating area-wide fiscal policies and joint funding mechanisms. Creating pan-euro area zero coupon perpetuities for purchase by the ECB may meet less resistance than common borrowing would, although recent rulings by German courts suggest there would be questions about the legality of such measures. Importantly, the same three factors are apparent in the euro area as well: the pandemic is devastating livelihoods and economic activity, public debt is a problem, and inflation has persistently been below target.
Government debt monetisation has been taboo for a long time, and for good reason. It has too often led to epochs of dangerously high inflation that, once set into motion, were destructive and hard to stop. However, at this juncture almost all advanced economies are persistently missing their inflation targets from below.
If inflation were to rise above the target by a small margin for a while as a result of the proposed policy, we expect that many central banks will quietly (or vocally) welcome it. And if inflation and inflation expectations begin to climb too much, then central bankers have an array of standard policy tools to combat it. Small amounts of excessive inflation can be dealt with opportunistically, letting the next recession drag on a bit longer. And if the policy were to trigger a more serious rise in inflation than we anticipate, a more costly but effective alternative is to create a recession as a monetary policy choice, as in the Volcker recession.
The question, then, is whether it is better to incur a higher human and economic cost of the pandemic by not intervening forcefully now for fear of incurring too much debt, or to take more aggressive fiscal action now, monetise some of its cost while inflation is below target, and risk having to fight some unwanted inflation in the future. To us, the answer seems clear.
Most advanced economies have resorted to leaning on central banks one way or another to accommodate their emergency spending needs, leading to a fear of monetization (Blanchard and Pisani-Ferry 2020). We argue that what has been done so far is not monetisation, but that to address the current emergency some strictly controlled monetisation is in fact desirable.
In general, monetisation occurs when a government funds its expenditures by issuing intrinsically worthless claims that the public is compelled to accept, effectively printing paper money. In our proposal, the special issue zero coupon perpetuities issued by Treasury and purchased by the Federal Reserve are doing exactly that. By contrast, issuing government debt in the open market creates an obligation for future repayments of value equal to that of the money raised; it is clearly not monetisation.
Where it gets trickier is when a government’s debt is purchased by its own central bank. The critical question is whether, and by how much, the central bank’s purchases directly reduce the value of claims by the public on the government.1 In the case of traditional open market operations there is generally no doubt that the government’s obligation to repay the debt has not been materially altered. This is reflected in standard measures of ‘debt held by the public’ that include the government debt held on central banks’ books.
The policy of quantitative easing (QE) employed during the Great Recession that followed the 2007-8 financial crisis massively increased the holdings of government debt on central bank balance sheets. However, like open market operations, those purchases did not in themselves extinguish government liabilities. With QE, central banks purchased government bonds in the open market, paid for with interest-bearing reserves. Banks voluntarily continue to hold those excess reserves as assets. The net effect of QE on the balance sheet of the public thus far appears to have been an equal value exchange: government debt for interest bearing central bank reserves.2 The flip side is that the consolidated liabilities of the government to the public – the sum of those of the central bank and the fiscal authority – also remained unchanged.
Indeed, many central banks – including the Federal Reserve and the ECB – are purchasing large amounts of government debt, but so far those actions have been akin to QE. Since early March of this year, the Federal Reserve’s holding of Treasury securities has expanded by about $1.5 trillion, paid for by issuing interest-bearing reserves. Those purchases are providing liquidity to the Treasury market and helping to support bond prices, but at this point they are not monetizing the debt. The Bank of England is allowing HM Treasury to run an overdraft account, but it hasn’t forgiven the obligation to repay.
Although not there yet, over time these policies may evolve towards monetisation. If parliament were to pass a law that forgives HM Treasury’s debt to the Bank of England, that debt will be monetised. Otherwise the expansion of the Bank’s balance sheet will reverse when the overdraft is repaid. Similarly, if the government bonds held by many advanced economy central banks are eventually sold back to the public, there will be a corresponding decline in outstanding reserve balances and the debt will ultimately be paid with tax revenues. However, if a government selectively defaults on its debt held by its central bank (assuming it legally can) or if its central bank is forced to roll over that debt forever, it will effectively be monetised. Note the use of the word “forced”. Under the mandate to maintain price stability, central banks would choose to sell debt back to the public were inflation to rise above target. Preventing such eventual sales would require a legislative change to central bank charters or some other mechanism to compel cooperation with the fiscal authority.
Hence, current central bank debt purchases, similar to those made in the aftermath of the global financial crisis, are not monetising debt. That fact helps explain the low levels of inflation that have accompanied the enormous expansion of central bank balance sheets. Only with an increase in reserves, unaccompanied by a corresponding increases in the value of future claims on government revenues, can a government obtain fiscal resources without adding to its liabilities. This is monetisation, and this is what we are proposing should be done in a limited way during these extraordinary times.
Authors' note: We thank Bill English for a very insightful discussion.
Baldwin, R and B Weder di Mauro (2020), Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes, a VoxEU.org eBook, CEPR Press.
Blanchard, O and J Pisani-Ferry (2020), “Monetisation: Do not panic”, VoxEU.org, 10 April.
Bolton, P, L Buchheit , P-O Gourinchas, M Gulati, C-T Hsieh, U Panizza and B Weder di Mauro (2020), “Necessity is the mother of invention: How to implement a comprehensive debt standstill for COVID-19 in low- and middle-income countries”, VoxEU.org, 21 April.
Galí, J (2020), “Helicopter money: The time is now”, VoxEU.org, 17 March.
1 This is separate from pecuniary externalities of monetary policy, such as treasuries being able to borrow at lower interest rates when policy is expansionary.
2 Reserves being interest bearing due to interest on excess reserves makes it easier to see this but what really matters is that this is a swap of two liabilities of equal value, reserves for debt. The analysis would not have been different in the absence of IOER.