Debt monetisation: are we already there? - Capital Economics
Global Economics

Debt monetisation: are we already there?

Global Economics Update
Written by Vicky Redwood

On the face of it, it might look like major economies have not yet reached the stage of debt monetisation. Where central banks are buying government bonds, it is on the secondary market, is supposed to be only temporary and the primary aim is not to fund government borrowing. But in terms of their macro-economic effects, debt monetisation and the current situation basically amount to the same thing. Accordingly, debt monetisation is not a magic tool which policymakers have yet to deploy and can whip out as a last resort.

  • On the face of it, it might look like major economies have not yet reached the stage of debt monetisation. Where central banks are buying government bonds, it is on the secondary market, is supposed to be only temporary and the primary aim is not to fund government borrowing. But in terms of their macro-economic effects, debt monetisation and the current situation basically amount to the same thing. Accordingly, debt monetisation is not a magic tool which policymakers have yet to deploy and can whip out as a last resort.
  • There is no formal, universally agreed definition of debt monetisation. But it is generally considered to have two defining characteristics that make it distinct from central banks’ purchases of government bonds through asset purchase programmes like quantitative easing (QE).
  • The first is that it involves the central bank funding the government directly, rather than just buying its debt in the secondary market. This can occur by the central bank buying new bonds straight from the government either directly or in auctions (the primary market); or by doing away with government bonds altogether and the central bank simply handing money to the government.
  • So far in this crisis, plenty of central banks have bought government debt in the secondary market via their asset purchase programmes. But only a few have funded their government directly. This includes the Indonesian central bank (which is buying government debt at auction) and the Philippines central bank (which has bought debt directly from the government under a three month repo agreement). Note that although the Bank of England extended a direct loan to the UK Treasury under the Ways and Means facility, this was essentially just a bridging loan.
  • The second supposedly distinguishing feature of debt monetisation is that it is permanent rather than temporary. In other words, the debt held by the central bank never gets sold back to the market/the government never has to repay the money given to it by the central bank. This equates to a form of so-called “helicopter drop”, meaning that government debt never rises. On the face of it, no major economies are doing this yet either, given that the government bonds bought in the asset purchase programmes are supposed to be sold back to the markets at some point.
  • But when it comes to their macro-economic effects, debt monetisation and the current situation in which asset purchase programmes are facilitating fiscal expansions amount to pretty much the same thing. Regardless of whether the central bank is operating in the secondary market or not, the result is the same. Government debt held by the private sector is lower than it would be otherwise and bond yields are lower than they would be otherwise. As for the permanence point, this may just be illusory. The “permanent” rise in the money supply could be reversed in the future (not least as central banks sought to quell the resulting inflation pressures), just as the current “temporary” asset purchase programmes may never get reversed and turn out to have been debt monetisation after all.
  • Admittedly, there is still a difference in the central bank’s motives in these two scenarios. When the central bank buys bonds in the secondary market, it generally does so to meet its own objectives such as maintaining the functioning of financial markets or meeting its inflation target. Indeed, we think that the US Fed will scale down its asset purchases significantly in the coming months as liquidity strains ease, even as the government’s debt issuance balloons. In contrast, direct financing is likely to be done with the sole intent of funding government spending. The latter would therefore appear to be a more dangerous blurring of the line between central banks and governments, setting the precedent for governments to ask central banks for money whenever they need it.
  • However, even under recent asset purchase programmes, this line has already blurred significantly. For example, some central banks (including the US Fed) have abandoned quantitative constraints on the amount of their asset purchases, while the explicit targets for government bond yields used by the Bank of Japan and the Bank of Australia have increased the overlap between monetary and fiscal policy even further. (Continued overleaf.)
  • Accordingly, debt monetisation and the current set-up are basically the same. This has two key implications. First, debt monetisation does not have to be a dirty word. The reason it has negative connotations is that it has often caused big problems in the past, with government spending getting out of control and a descent into inflation and hyperinflation (e.g. Weimar Germany, Zimbabwe). But this is generally because of the institutional breakdown that accompanied it. Such institutional failure does not have to happen now so long as central banks retain their independence and arrangements remain transparent. As Willem Buiter, a former UK MPC member has put it, “independence doesn’t mean having to say no to a request for direct monetisation. It means you can say yes or no.”
  • So in countries where central bank credibility is not an issue and inflation is not a big risk, a case can easily be made for the direct financing of governments where the debt market might not be able to digest the volume of debt needed in the short-term. This is especially the case for countries without QE programmes and those EMs which don’t have deep and liquid capital markets.
  • Indeed, we may well see more direct central bank purchases of government debt. Former RBI chief Raghuram Rajan has argued for direct monetary financing in India – indeed, the practice was common there until 1997. Reserve Bank of New Zealand Governor Orr has said that he is open-minded on the issue of the direct monetisation of government debt. And the Chinese government wants to issue special treasury bonds (which are used for funding specific projects), which it apparently wants the PBOC to take directly onto its balance sheet.
  • The other, less reassuring, implication of all this is that debt monetisation and helicopter drops are not magic tools which can be whipped out of the toolbox when all else has failed. (See here for our more detailed Global Economics Focus on helicopter drops.) To all intents and purposes, they would look the same and have the same results as the policies we are seeing now. In other words, central banks would “print” money electronically and governments could spend more freely because they would not have to issue as much debt to the market as otherwise.
  • Similarly, we don’t buy the argument that printing money to permanently write off public sector debt is a magic solution to the otherwise sharp rise in public sector debt that we will see as a result of governments’ measures to combat the coronavirus. For a start, this would not necessarily cut public sector borrowing costs. Although the government would not have to pay out any interest on government bonds, the central bank would have to pay interest on the additional reserves created by the permanent rise in the money supply. This would still generate a net saving for the consolidated public sector if the interest rate on reserves was less than the interest rate that the government would have paid on new government bonds, but in most countries there is currently little difference between the two. At least the government would be safeguarded against a sharp rise in risk premia on government bond yields at some point, but it could deal with that in other ways, such as financial repression.
  • Some people argue that a helicopter drop might make governments more willing to spend if their debt was not rising, while households would be more likely to spend their giveaway if they did not have to fear taxes rising further ahead. However, governments hardly seem to be letting high debt levels hold them back from providing assistance to their economies at the moment. And we have always been sceptical about how much the idea of “Ricardian equivalence” applies in reality.
  • Moreover, policymakers would still have to deal with any rise in inflation further ahead resulting from the permanent rise in the money supply, in the same way that policymakers will have to deal with any rise in inflation eventually resulting from current policies. This could be done – for example, by imposing controls on lending or raising interest rates. But such measures would impose costs on the economy and the financial system, and policymakers might not have the willpower to do it. Moreover, it might involve measures that would equate to the reversal of the initial helicopter drop, underlining our point earlier that the appearance of a permanent write-off of government debt could end up being illusory anyway.
  • So printing money to pay off the debt is not some magic solution that can solve the debt problem if all else fails. We shall continue to discuss in our research the other ways in which governments can deal with the rise in their debt burdens.

Vicky Redwood, Senior Economic Adviser, victoria.redwood@capitaleconomics.com

Feature HTML here