Wednesday, December 9, 2003
Does India face a public debt crisis?
High real interest rates explain the recent rise in central government debt. How worried should we be, asks Suman Bery
Public debt has triggered several of the financial crises in large emerging markets over the past decade. Indeed, these crises have been dubbed 'capital account crises'1 , as they reflected vulnerabilities in balance sheets of the major sectors of the economy (particularly government and the corporate sector), rather than the more traditional concern of excessive domestic spending.
Public debt was not a major factor in the Asian crises of the late 1990s, but was a feature in the crises in Argentina, Brazil, Mexico, Russia and Turkey.
In the case of India, our traditional concerns have been with fiscal deficits (both Centre and state) and with the size and maturity of the country's external debt. Somewhat less attention has been paid to the level, cost and structure of India's overall public debt, both domestic and external.
A recent paper by the Chairman and one of the members of the Twelfth Finance Commission (Rangarajan and Srivastava; henceforth R-S)2 provides valuable historical and analytic perspective on these issues. The authors review the growth of the central government's liabilities over the past fifty years.
They partition the change in public liabilities (as a share of GDP) over various time periods into two components. These are the primary deficit (the fiscal deficit less interest payments), and the differential between the real growth rate and the real interest rate.
As is well understood, if the real growth rate of the economy exceeds the average real interest rate paid on government debt, the effect is to lower the ratio of debt to GDP, for a given level of the primary deficit.
In order to achieve such a partitioning, there has to be a consistent set of accounts linking the deficits (flows) with the debt (stocks). This is not usually possible with figures as presented in the Receipts Budget.
Accordingly, R-S use figures of outstanding liabilities given by the Controller and Auditor General (CAG) (which provide a more complete picture of central government liabilities) and derive a fiscal deficit (and primary deficit) series consistent with the change in these liabilities.
In calculating the real rates, the authors use the implicit price deflator for GDP at market prices. The interest rate on government debt is calculated by dividing interest payments during a financial year by the stock of liabilities outstanding at the beginning of the year.
Their findings are interesting, both over the long haul, and for the last decade. At the end of 1950-51 the ratio of central government liabilities to GDP was 28.84 per cent. By the end of 2001-02 this ratio had just about doubled, to 55.36 per cent of GDP.
Surprisingly, almost all the increase had taken place by 1989-90. Taking the last decade as a whole, the debt : GDP ratio has barely moved: it was 55.31 per cent at the end of 1989-90 and 55.36 per cent at the end of 2001-02.
More revealing, though, are the movements over the decade. Between 1990-91 and 1996-97 the ratio of debt to GDP fell sharply (to 49 per cent), but this improvement was completely reversed thereafter. The difference between the two periods does not lie in any major change in the primary deficit, but instead in the widening gap between the real interest rate and the real growth rate.
In turn, the truly striking development has been the enormous swing in the effective real interest rate paid by the central government. This has gone from - 2.53 per cent in 1990-91 to + 6.77 per cent in 2001-02, a swing of almost a thousand basis points. While the nominal effective interest rate on government borrowing has risen steadily, the real change is in the inflation rate, which has steadily declined from 13.8 per cent in 1991-92 to 3.5 per cent in 2001-02.
These numbers raise several intriguing analytic and policy issues. The first is why real interest rates have risen so much in the past few years, and how long they are likely to remain high.
The effective cost of borrowing by the government reflects the weight of past rates, and is expected to adjust only sluggishly to lower inflationary expectations and monetary easing. What is striking is the very long period (almost without break from 1962 till 1994) over which the real interest rate facing government remained negative.
While in the short run real interest rates can be influenced by nominal interest rates (including the impact of administered interest rates) over the long run equilibrium real interest rates ought to reflect deeper forces in the economy, such as the perceived level of risk and the productivity of investment.
The price level should adjust to reconcile the nominal interest rate with the 'natural' long run interest rate. With the reduction in financial repression and the increased productivity of capital in a more competitive economy, however, positive real rates are likely to remain the norm for some time to come.
Second, some analysts point to the fact that India has had high fiscal deficits over extended periods of time without suffering unduly. The R-S analysis demonstrates how the impact on debt accumulation has in the past been offset by the gap between the real interest rate and the real growth rate.
Indeed over the fifty years studied only 18 per cent of the cumulative primary deficit has led to an increase in the ratio of debt to GDP; 82 per cent of the potential increase was neutralised by the fact that the real growth rate was higher than the real interest rate.
Much thus will depend on the growth of the economy. Indeed the main failure of the Latin economies was to convince the holders of their debt, whether domestic or foreign, that they would return to rapid growth.
At least in terms of sentiment India is in a rather different position today. But it is the imperative of growth that has raised questions on the wisdom of sequestering foreign savings in the build up of reserves, and raising nominal interest rates through sterilisation.
The link between fiscal policy and growth is also much disputed, not only in India, but also in the U.S and Europe. At the extreme there are those who argue that, with unemployed resources available, the government needs neither to worry about the stock of debt or the size of the deficit.
My own belief is that Indian growth would be stimulated by credible evidence of an improvement in the primary fiscal balance (what is sometimes called an expansionary fiscal contraction). This was, after all the experience of the early 1990s.
It is important, though, not to be too mechanistic. There is no magic associated with a given debt : GDP ratio. The important issue is one of government credibility. India has a deserved reputation for honouring its debts and for low inflation. This leads to a greater willingness by its population to hold government liabilities voluntarily. The problem is that it is not easy to forecast when and where the limit will be reached, nor what the end-game will look like.
A combination of higher primary surpluses and faster growth therefore is indicated to stabilise the debt : GDP ratio and signal sustainability of the debt position. Perhaps we knew it all along, but it is valuable to benefit from research that lays out the case so clearly.
1 International Monetary Fund, Independent Evaluation Office. 2003. The IMF' s role in Capital Account Crises. Washington D.C.
2 C Rangarajan and D K Srivastava, 'Dynamics of Debt Accumulation in India'. Economic and Political Weekly , Vol 38, No. 46 (November 15-21, 2003, pp. 4851-4858. Mumbai. http://www.epw.org.in sbery at ncaer.org
(The author is Director-General of the National Council of Applied Economic Research, New Delhi. The views expressed here are personal)
Business Standard Ltd.