Fiscal space: Not if but how

In the first part of a series, the authors argue the means must be found to avert potential economic and social collapse post-Covid

Extraordinary circumstances ca­ll for extraordinary responses. The Covid-19 crisis is nothing if not extraordinary. Re­sponses are imperative to deal with the health crisis, cushion the hundreds of millions workers and households that have lost livelihoods, and protect firms against the collapse of business in virtually all non-essential sectors. With the private sector collapsing, the government’s role will be pivotal. But does the Indian government have the fiscal me­ans, especially when revenues are also declining? In truth, this is a rhetorical question: One way or the other the means must be found to avert potential economic and social collapse. The only question is how. We suggest a menu of options.

Start with some illustrative, even cru­de, magnitudes. Say, conservatively, non-essential sectors, comprising 50 per cent of GDP, have to be locked down for two months. This is a loss of one-mo­nths’ GDP or 8¼ per cent of the annual GDP. Assume one-third of that loss in GDP is made up by way of cushioning household income and firms’ costs, while health-re­lated expenditures amount to an ad­d­i­tional 2 per cent. Just that (and this is surely an underestimate), yields ad­d­i­tion­al expenditure — centre and states combined — of close to 5 per cent of GDP. Assuming India’s nominal GDP will be roughly Rs 200 trillion (current prices), additional expenditures will be Rs 10 trillion ($135 billion).  

In principle, there are five ways of financing additional expenditures over the next 12 months or so: 

• Reduction in other expenditures (Rs 1-1.5 trillion)
• Foreign borrowing, from official sources and non-resident Indians (NRIs; Rs 1-1.5 trillion)
• Public financing by issuing g-secs (including to banks and LIC) (Rs 5 trillion)
• Monetary financing or “printing money” (Rs 1-1.5 trillion)
• Mobilizing additional resources via raising taxes and cutting subsidies (Rs 1-1.5 trillion)

Central and state governments incur non-discretionary expenditures (interest, wages and pensions, defence etc.) which cannot be cut easily. But others (for example, new projects) can. In the Covid-19 context, there will be planned expenditures that cannot be spent because of social distancing and the associated disruptions. Total government spending is about 26 per cent of GDP. Less than a quarter, say 6.5 per cent, is discretionary. If 15 per cent of these expenditures are cut that would release financing of about 1 percent of GDP. 

Of course, central and state governments must decide expenditure cuts but some principles can guide them. First, cut recently initiated projects and fund those near completion: economic returns from the latter will be quick and the latter delayed. Second, all public sector "revival" projects, especially with low returns, need to be shelved: rehabilitating old fertilizer plants in Bihar, reviving MTNL/BSNL, etc.  These never made sense and now is the time to bury them.

 

 

Raising money from private capital markets is an option but the large outflow from India and other emerging markets renders such financing difficult, costly, and volatile. So, the two promising foreign sources would be multilateral institutions and NRIs.

  

While multilateral banks (World Ba­nk, ADB etc.) are another source of che­ap and long-term financing, they cannot sharply raise overall lending to India in the short term unless they be­come more bold and creative. What they can do immediately, how­ever, is to repurpose existing loans. An­nual disbursements from multilateral banks are around $10 billion, and re-purposing even half wou­ld yield ano­ther $5 billion. India should also make a contingency plan for see­king quick disbursing fu­nds from the IMF if needed.

Tapping NRIs is ano­ther possibility as India has done on four previ­ous financial emergen­cies: $1.6 billion in the 1991 BOP crisis (“India De­ve­lop­ment Bonds”); $4.2 Billion in 1998, post-Pokhran nuclear blast (“Resurgent India Bo­n­ds”); $5.5 billion in 2000 (“India Mi­l­l­e­nn­ium De­posits”); and then $26 billion in the 2013 “taper ta­ntrum” episode, via foreign currency non-resident bank account deposits. In the current crisis, since foreign exchange reserves are comfortable at nearly half a trillion dollars, and yields globally at record lows, India can raise around $8-10 billion relatively cheaply (100 basis points above US Treasuries).

Raising money from the public by conventional bond issuance will have to be the biggest source of funding, probably up to 50 per cent of the total or Rs. 5-5.5 trillion (about 30-40 percent of the current gross market borrowing by the center and states combined). Given that g-sec rates have been rising recently (despite rate cuts), large increases in borrowing by the center and states could further intensify the pressures.  

 

 

But two counter-considerations should be no­ted. This crisis is distinctive in that the original shock was not bad or misguided policies by the government or others but purely exogenous. Hence, considerations of “moral hazard” — additional ex­penditures creating bad incentives or validating imprudent behavior — are misguided and markets should be willing to finance extra expenditu­res without undue penalties. If this assumption is optimistic, it will be apparent in g-sec rates.

Second, the pressure on interest rates can be limited by recourse to non-market borrowing (or “financial repression”) as the government is already doing. This would involve tapping into the NSS and allocating some of the new borrowing to public sector banks and LIC. About quarter to one-third of public financing can be via this route: not the best option by any means but the less-bad warranted by the circumstances.

A more controversial suggestion wou­ld be “printing money,” that is, by the RBI directly buying g-secs and state government bonds. The US Fed has been buying more than $1m of assets per second over the past two weeks. Should India do something similar? Two dimensions — economic and legal — are important.

Would printing money raise inflation? The Covid-19 shock is likely to be deflatio­nary, reflected already in a collapse of co­mmodity prices, attenuating the risks of monetary financing. Of course, aggregate supply, especially of essential goods, cou­ld contract and any resulting shortages may raise prices. Even so, price rises wou­ld be one-off, and monetary policy even in a pure inflation targeting regime is me­ant to look past transitory supply shocks.

A second concern is legal. In the last year or so, the RBI has de facto been buying vast amounts of g-secs to pump liquidity into the system. It has so indirectly through the secondary market. However, direct purchases of g-secs by the RBI may run afoul of the law. The historical bad practice of routine and ruinous inflationary financing was codified into law, a significant achievement which would be very costly to reverse. The only way to square this circle would be for the government and RBI to commit explicitly to direct buying as a one-off event, not to be repeated in the future.

If this can be ensured — and this government’s strong anti-inflation record warrant optimism — up to 15 per cent (1.5 trillion) of the total can be, responsibly, monetarily financed. This would amount to a not-too-high 3 per cent increase in the RBI’s current balance sheet, 5 per cent increase in reserve money and, assuming a decline in the money multiplier, a 15-20 per cent increase in broad money.

Reality is rarely as straightforward as opinion pieces and there are minefields aplenty that will have to be crossed.

Our strategy would increase domestic debt by about 2.5-3.0 per cent of GDP and external debt of about 1 per cent. These are unlikely to tip debt to an unsustainable trajectory. When assessing the risks of action for debt sustainability, we must not overlook the consequences of inaction which may well impact debt trajectories more adversely if economic activity goes into free fall.

Next, the crisis will require maintaining credit flow (over and above debt moratoria) to firms, large and small. Ideally, the RBI should provide liquidity to the banks to on-lend to them. But a potentially serious bottleneck looms: the infamous 4C problem — the fear of arbitrary scrutiny by Courts, CBI, CVC, and CAG, and now the Enforcement Directorate, ED — that has paralyzed lending by public sector banks. That fear will be exacerbated in present circumstances because lending to inherently loss-making situations will be extra-risky, making future scrutiny more likely.

A way around this would involve the RBI lending directly to firms, assuming the credit risk, as central banks in ad­va­nced countries have done. Of course, this course may simply transfer the 4C+ED scrutiny to the RBI unless explicit ex ante legal protections are provided. Going do­wn this route is tricky but needs to be amongst the set of less-bad policy options.

A final critical point. Given that states are in the frontline of the pandemic response, they must have access to resources quickly. In light of recent pressures on their borrowing costs, the RBI must intervene to ensure that spreads on state government bonds relative to g-secs remain modest. The center and the RBI must also make unambiguously clear that fiscal responsibility limits on state governments will not bind this year.  

But all these financing options may be inadequate. More resources might be needed. Where they might come from will be the subject of our next column.

This is the first of a three-part series. Read the second part — Creating fiscal space for the states, and the third part -- Solidarity today, new social compact tomorrow

Kapur is professor, Johns Hopkins University; Subramanian is former chief economic adviser, government of India