key: cord-0696053-ngwuo4ep
authors: Bajaj, Ayushi; Datt, Gaurav
title: Financing of fiscal response to COVID-19: a pragmatic alternative
date: 2020-08-07
journal: Indian Econ Rev
DOI: 10.1007/s41775-020-00090-6
sha: 9f79ed09606581026605f34d4abe862f5c795529
doc_id: 696053
cord_uid: ngwuo4ep

As governments around the world, including the Indian government, mount a fiscal response to the Covid-19 crisis, the question of how to finance it has risen to prominence. We argue that the option of the central bank monetizing the additional government debt and then writing it off offers a pragmatic way out.

around the world is a foregone conclusion. 1 Governments in country after country are frenetically announcing an expanding slew of rapid response measures-to augment resources for front line first-response agencies in the health sector, to provide immediate relief to afflicted populations whose lives and livelihoods have been disrupted, if not destroyed, and to prop up as much of the economy as possible. The fiscal price tag on these measures is massive and growing rapidly. 2

There is now an active discussion about how can this unprecedented fiscal response be financed, whether it will lead to unsustainable fiscal deficits and government debt, and whether it will have to be financed through higher tax burdens in the future. The discussion has gained further impetus as governments consider whether they can afford a second or third round of fiscal injection once the initial one expires. The main argument of this short paper is that none of these concerns around sustainability or affordability is an issue. This is because there is an alternative way of financing the fiscal response. The simple alternative mechanism is for the central bank to buy government debt equivalent to the size of the fiscal response, and then to write it off.

And T-Bills) 65%

SecuriƟes against Small Savings 31%

Other receipts (incl. State Provident Funds) 3%

External debt 1% 2 As of June, 2020, the global fiscal support is estimated at $11 trillion (14% of world GDP) with a direct budget support (additional spending and foregone revenues) component of $5.4 trillion and a similar amount in liquidity support (loans, equity injections and guarantees) (IMF 2020b). 1 According to the International Monetary Fund's (IMF) World Economic Outlook Update of June 2020, the global economy is projected to contract by 4.9% during 2020, while the World Bank's Global Economic Prospects of June 2020 projects a contraction of 5.2 % (IMF 2020b; World Bank 2020). This is an order of magnitude worse than the global financial crisis of 2009 that was associated with a global economic contraction of 0.1%. Per capita incomes in 2020 are projected to shrink for over 170 countries, including advanced, emerging and developing economies (IMF 2020a).

We will illustrate our key ideas focusing on the Indian case. However, our arguments are more general and of wider relevance to many other countries that have launched or are in the process of launching a significant fiscal package in response to the Covid-19 pandemic.

To explain the mechanism, note first how governments typically finance their fiscal deficits. As an illustration, the fiscal deficit of the Government of India for FY 2019-20 (as per the revised estimates) was Rs. 7.67 trillion, equivalent to 3.8% of GDP (Government of India 2020). Figure 1 shows how this deficit was financed. About 65% of the deficit was financed by sale of government securities in the market (private sector) and another 31% was through issuance of government securities to the small savings fund (household sector). Though the relative share of market borrowings has generally been higher in earlier years, the key point is that nearly all of the fiscal deficit is financed by issuing government securities to the private or household sector.

With the looming recession, the appetite and the ability of the private sector for holding additional government securities could be limited. In these uncertain times, the private sector will also be looking for a reasonable rate of return on government securities, which could put an upward pressure on interest rates. Issuing government securities against the national small savings fund may be similarly compromised. But, there is an alternative mechanism. The Reserve Bank of India (RBI) could directly buy government bonds (call them Corona bonds, if you like) to the tune of the fiscal response, offering money or more pertinently, crediting an equivalent amount to the government account. This will inflate government debt on the one hand and expand RBI's balance sheet on the other. But, this need be only temporary. Having thus monetized the additional government debt, the RBI could soon write off the government debt and shrink its balance sheet. There is also no need to finance the fiscal response through higher taxes in the future.

Such monetization of government debt is an anathema to many economists. Of course, you do not want to resort to such a mechanism as the normal mode of operation, in the interest of preserving the independence of the central bank and to discourage unscrupulous government spending. The normal procedures for financing fiscal deficits exist for a reason. However, these are not normal times. Writing off government debt incurred in (and limited only to) the exceptional circumstances of supporting urgent relief effort and pulling the economy out of a potentially disastrous slump is justified.

To keep things clear for the following discussion, it will be useful to distinguish between three variants of deficit (or debt) monetization:

(i) Direct monetization: this involves the central bank financing fiscal deficit by buying government securities directly from the government in the primary market; (ii) Indirect monetization: this involves government borrowing from the market by selling government securities and the central bank in turn buying them in the secondary market (through open market operations); (iii) Direct monetization with debt write-off: (i) accompanied by writing-off of government securities held by the central bank from the asset side of its bal-ance sheet (or equivalently by directly buying government securities that are explicitly non-redeemable).

Our main arguments relate to the third variant of deficit monetization though the first two variants have also been actively proposed and debated in recent discussions. 3 In the rest of this article, we consider a range of potential issues or counterarguments to the proposed financing mechanism.

If the RBI monetizes government debt, this will certainly inject more liquidity into the economy. Whether it impacts inflation will depend on the supply response. There are reasons to believe that this risk is small. First, with the looming recession the economy is well below its potential and may remain so for quite a while. Second, though there has been a surge in inflation in recent months reflecting the disruption in supply chains since the lockdown (Fig. 2) , this is expected to moderate with the easing of restrictions. This will be further helped by low global oil prices. Third, if a good part of the fiscal response is specifically directed towards maintaining supply chains and facilitating firms (especially small and medium enterprises) to reopen without an overhang of unpaid debts, rents and other obligations which would otherwise put them out of business, it will go a long way towards generating a supply response. By playing this supportive role, government spending can in fact crowd in rather than crowd out private spending. We also note that maintaining supply chains is a priority in any case irrespective of the size and mode of financing of the fiscal response, and should continue to be so. Fourth, there are clear indications that the Covid-19-induced supply shock is already morphing into a large demand shock as people's livelihoods and incomes plummet, creating a deflationary overhang. A sizeable fiscal response can stem this tide without a major risk of fuelling inflation.

To be sure, there could be some legal impediments to such a mechanism as the Fiscal Responsibility and Budget Management (FRBM) Act (2003) prohibits the RBI from operating in the primary market for government securities. 4 However, such legal impediments need not be insurmountable as amendments to the FRBM Act are possible that can allow direct purchase of government securities by the RBI under "exceptional" circumstances. Note that the fiscal deficit targets of the FRBM Act were already breached when the Union Budget of February 2020 invoked the FRBM escape clause for a revised fiscal deficit of 3.8% of GDP (for FY . This was before the onset of Covid-19. Since the pandemic, they will be breached further both on account of the fiscal package announced by the government as well as the revenue shortfall due to the pandemic-related disruption of economic activities. Thus, the impediments imposed by the current provisions of the FRBM Act need not be a binding constraint.

A large measure of the opposition to direct monetization of the fiscal deficit stems from what could be described as the "slippery slope" argument. The argument goes that if the RBI directly finances government deficit, it will lead to fiscal dominance of monetary policy, compromise the independence of the central bank and open the door to unrestrained future spending by the government that will exploit this financing instrument indiscriminately. Compromised monetary policy, the argument goes, will stymie the RBI's ability to control future inflationary spells. While the accountability of government spending is a real issue, it is unclear if central banks are in a position to enforce such fiscal discipline. By the same token, it is not inevitable that monetization of deficit will open the floodgates to fiscal profligacy. Some safeguards are possible, and the slope need not be slippery if monetization provisions are explicitly limited to the current exceptional circumstances as a matter of conscious policy agreed upon by the RBI and the government.

Is there a risk of downgrade by rating agencies and capital flight by foreign investors? Recently, Moody's has already downgraded India's sovereign ratings, though this "was not driven by the impact of the pandemic", but by "vulnerabilities in India's credit profile that were present and building prior to the shock" (Moody's Investor Service 2020). The month of March 2020 also witnessed a significant net outflow of capital by foreign portfolio and foreign institutional investors, accompanied by some depreciation of the Indian Rupee (Fig. 3) . Would monetization (with or without a write-off) of government debt risk further rating downgrades, depreciation and capital flight? Several points are notable in this regard. First, while the recent capital flight was not limited to India but part of a more general phenomenon of flight to safety afflicting many emerging market economies, the capital outflow from India (as also from other emerging markets) appears have subsided during May and there are signs of reversal during June (Fig. 3) . Second, there has been some depreciation of the Rupee mostly during March and early April, from about 72 INR/USD at the end of February to just under 77 INR/USD around mid-April. Since then the Rupee has stabilized and during May and June has mostly remained within the 75-76 INR/USD band. A modest depreciation is not necessarily a bad thing; apart from helping exporters, it serves to increase the surplus of the RBI which can also help with the fiscal bottom line. Third, it would be naïve to believe that the rating agencies care only about the fiscal deficit and how that is financed. For instance, the announcement of ratings downgrade by Moody's (on June 1, 2020) also noted: "Although a rating upgrade is unlikely in the near future, Moody's would change the outlook on India's rating to stable if outturns and policy actions were to raise confidence that real and nominal growth will rise to sustainably higher rates than Moody's projects." (Moody's Investor Service 2020) The counterfactual must therefore also weigh in the prospective growth performance of the economy. Holding back on the fiscal response due to potential concerns with its financing will almost certainly come at the expense of growth performance, which is unlikely to yield more favourable sovereign ratings by credit agencies.

It is worth circling back to the basic problem we are trying to solve. Put simply, the key problem is how to engineer recovery from the massive economic downturn caused by the pandemic. There is near-universal consensus that a strong fiscal stimulus has to do the heavy lifting as monetary policy measures of cutting interest rates further are unlikely to significantly stimulate aggregate demand. 5 Governments around the world are already acting on this consensus. In light of this, the key issue is how to finance the fiscal response at minimal cost to the government and society at large. This is where monetization of fiscal deficit comes in as a potential solution.

With either direct or indirect deficit monetization, we end up with a situation where (additional) government debt is held by the central bank in the form of government securities. However, since the liability of the government is an asset for the central bank, monetization has no effect on the consolidated balance sheet of the government and the central bank. More importantly, the interest paid by the government on these securities shows up as additional income for the central bank and insofar as the surpluses of the central bank are fully transferable back to the government, other things being equal this is equivalent to the government financing its deficit at zero cost. However, as discussed by Rajan (2020) , we also need to consider a further consequence of the extra government spending. Once this spending occurs, it will increase the aggregate deposits held at banks and may lead to excess reserves within the banking system. If banks are unable augment their market lending, the excess reserves end up with the central bank. Currently, the RBI accommodates these excess reserves through its reverse repo facility offering the banks the going reverse repo rate of 3.35% (Fig. 4) . Under these conditions, as the reverse repo interest payments are ultimately a charge against the surplus of the central bank, debt monetization does not amount to a free lunch for the government.

However, there are two further qualifiers to the above sequence. First, to the extent the fiscal stimulus works, it should open up lending opportunities for the banking sector thereby providing a productive outlet for additional reserves due to government spending. Second, the reverse repo rate like the repo rate is a monetary policy parameter determined by the RBI. The RBI has already been cutting these policy rates in recent months (Fig. 4) and has the option of cutting them further in future. It is even arguable that this could be doubly helpful in not only reducing the cost of reverse repo operations but also in flattening the yield curve and lowering the bank prime lending rates. Thus, relative to borrowing through new market issuance of government securities at the going benchmark 10-year G-Sec yield of around 6% (Fig. 4) , direct or indirect monetization offers a potentially minimal cost option for financing the fiscal response.

In addressing this question, it is useful to look at the central bank's balance sheet. Historically, the size of RBI's balance sheet has been around 20% of India's GDP. Table 1 shows the RBI's balance sheet as of June 30, 2019. The total size of the balance sheet at this time was Rs. 41 trillion or about 21.6% of GDP. The main components on the liabilities side are: notes issued, deposits (of both government and banks) held by the RBI, and RBI's economic capital. The economic capital itself is comprised of total realized equity and revaluation balances (RBI 2019a). A part of the total realized equity consists of the RBI surplus that is transferable to the government after provisioning for risk buffers. The RBI's economic capital at this time was about Rs. 11 trillion, 28% of the balance sheet, while the total realized equity was Rs 4 trillion, about 10% of the balance sheet. The main components on the asset side are: gold (4% of the balance sheet), foreign investments (68%), domestic investments, primarily government securities (24%), and loans and advances, mainly to central/ state government and scheduled commercial banks (2%). Note that RBI's balance sheet has grown since June 30, 2019. As of May 29, 2020, the size the balance sheet was Rs. 52 trillion equivalent to about 26% of nominal GDP (RBI 2020). 6 However, RBI's economic capital relative to the size of the balance sheet has remained more or less unchanged at about 28% of the balance sheet. 7 Within the framework of the central bank balance sheet, a debt write-off implies a writing down of government securities on the asset side and a deduction in realised equity on the liability side. Thus, a write-off shrinks the size of the balance sheet back to what it was prior to its growth through the acquisition of government securities. This of course means a drawing down of the risk buffers; as Rajan (2020) puts it, it is akin to the government "selling the family jewels to pay for its spending". However, this need not be characterized as the government raiding the central bank. RBI's risk buffers exist precisely for a "rainy day" such as the one we are facing now. Any shortfall in the risk buffers now could be replenished through future surpluses of the RBI. How much of the new debt can be written off depends on the size of RBI's economic capital. The recent Bimal Jalan Committee recommended that the RBI should normally maintain its economic capital at 20.8-25.5% of its balance sheet to provide a reasonable risk buffer (RBI 2019a). The current economic capital of the RBI at 28% thus exceeds the lower bound of this recommendation by about 7% of the balance sheet or about Rs 3.8 trillion. This amounts to about 2% of GDP. However, this is while still maintaining the normal lower bound of about 21% of balance sheet as the overall risk buffer. In the current crisis situation, there is a case for further drawing upon these risk buffers (the buffers exist so that they can be drawn upon in a crisis situation), which could support a debt write-down of greater than 2% of GDP. Such a level of debt write-off is well above the estimated Rs. 2 trillion budget support component of the overall stimulus package announced by the government. 8 A write-off of government debt has two implications: one for public debt and the other for the government's fiscal deficit. While the write-off does not make any difference to the consolidated balance sheet of the government and the central bank, it has the effect of shrinking their individual balance sheets. In particular, the writeoff enables the government to carry a lower level of public debt and hence limit future interest payments on outstanding debt. In recent years, interest payments have claimed a quarter of total central government expenditure in India. At the same time, the write-off amounts to a windfall receipt for the government that effectively lowers its deficit. Insofar as the central bank and the government are separate entities, this matters in practical terms. Lower levels of government's fiscal deficit and public debt eliminate the need for additional fiscal consolidation through higher taxation or expenditure reduction. The latter if carried out would defeat the purpose of the fiscal stimulus in the first place.

Finally, money-financed debt has an international analogue too, in particular, in the form of a new issuance of the Special Drawing Rights (SDRs) by the International Monetary Fund. The IMF can create SDRs much like a central bank can "print" its own money. If a country is low on its foreign exchange reserves, it can exchange its SDRs for the needed currency. If not required, the country can simply continue to hold it in SDR form. There have been growing calls for new issuance of SDRs (Brown and Summers 2020; Birdsall 2020) . However, there are three potential issues with the deployment of new SDRs. First, the United States with its 16.5% vote share holds veto rights to new SDR issuance, which requires approval with at least 85% of the total votes held by IMF members. Thus, there is uncertainty if a new issuance can come through. Second, the newly-created SDRs are allocated in proportion to each country's shareholding in the IMF, which may not reflect their real needs. Nonetheless, low and middle-income countries would have a one-third share in the new allocation, thus offering them access to hard currencies at negligible interest rate at a time when their foreign exchange reserves are compromised by a collapse in exports and remittances. 9 Moreover, it is possible to improve upon this by voluntary reallocations (or donations) of the newly-issued SDRs from highincome to low and middle-income countries. Third, for the new allocation of SDRs to be truly beneficial to poorer countries, it will be important to supplement it with debt relief and forgiveness if the extra international liquidity is not going to end up paying off the bilateral and private creditors of several of the poorer nations.

No government should shy away from mounting the needed fiscal response to Covid-19 on grounds of limited fiscal space. The pragmatic though unorthodox alternative of the central bank monetizing and then writing off government debt offers a way out. (A similar pragmatic, though politically difficult, alternative also exists at the international level through new issuance of international money by the IMF coupled with debt forgiveness for poorer countries.) Though there are finite limits to the magnitude of debt write-off, the problem of financing the fiscal response in our view is a non-issue. Central banks have an important role to play in facilitating the fiscal response needed for economic recovery as well as for mitigating the adverse distributional consequences of the Covid-19 shock. Such support from central banks to fiscal policy in these exceptional times, and limited to these exceptional times, need not amount to a long-term abdication of their monetary policy objectives.

Annual economic report

New SDRs? that pesky 85 percent approval

Monetisation: do not panic. VoxEU.org

National governments have gone big. The IMF and World Bank need to do the same

Monetisation of government deficit may be inevitable amid coronavirus disruption: C Rangarajan

The effects of a money-financed fiscal stimulus

Mitigating the COVID economic crisis: act fast and do whatever it takes

Who's afraid of monetisation of the deficit? The Hindu

Moody s-downg rades -India s-ratin gs-to-Baa3-maint ains-negat ive-outlo ok--PR_42460 5#

Making the IMF's special drawing rights work for COVID-19 economic relief

Monetization: neither game changer nor catastrophe in abnormal times

India's lockdown: an interim report

Statement by Philip Lowe, governor: monetary policy decision, media release

Report of the expert committee to review the extant economic capital framework of the Reserve Bank of India, Reserve Bank of India

Reserve Bank of India annual

Reserve Bank of India bulletin

Falling real interest rates, rising debt: a free lunch

Global economic prospects