ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846
-A A +A

How Much Public Debt Is Too Little?

Pronab Sen ( is with the International Growth Centre, India.

Virtually, the entire literature on public debt is on determining “how much is too much,” beyond which it becomes a systemic threat to the economy. On this basis, about 80 countries, including India, have fiscal rules designed to steadily reduce public debt. This article argues that there is a minimum stock of public debt, below which it is also a systemic threat, and outlines some of the considerations which should be taken into account. It further argues that the composition of public debt is equally important, but has been totally neglected. Both the level and composition of public debt, therefore, should be taken into account while framing fiscal rules.

All too often, in governments, the left hand does not know what the right hand is doing. A decision taken by one ministry may not entirely take into account its repercussions on the other ministries.1 The net result is policy dissonance, which, at best, dilutes policy synergy, and, at worst, can lead to systemic crisis. This paper examines a particular instance of the latter form of policy dissonance—the decision on the level and issuance of public debt. The irony is that the policies that can potentially be impacted by this decision are all, in one way or another, related to the government’s Ministry of Finance itself; or, to put it more bluntly, even the right hand does not seem to know what it is doing.

In an article in 2017,the author had examined the recommendations of the review committee on the Fiscal Responsibility and Budget Management (FRBM) Act2 through the prism of the debate contained in the note of dissent by Arvind Subramanian,3 one of the members of the review committee, and the rejoinder of the committee to this note of dissent (Sen 2017). What was striking was that despite their differences on a number of important counts, both the protagonists had implicitly agreed that a secularly declining public debt to gross domestic product (GDP) ratio is unambiguously a good thing, and hence recommended fiscal rules that would lead to precisely such an outcome.4 Intuitively, this view is seriously problematic, despite the fact that it is entirely in consonance with the established view of the economics profession. Almost the entire literature on this subject treat public debt as an unavoidable evil—sometimes, but rarely, condescending to consider it a “necessary” evil.5 Virtually all the analytics, therefore, is on determining how much public debt is too much, beyond which it becomes a systemic threat to the economy.6 The discomfort with this perspective primarily stems from the fact that government debt is the only interest-yielding risk-free asset in any country,7 and is therefore central to a wide range of key economic variables and decisions in a modern economy. Unless these aspects are explicitly taken into account while assessing the “optimal” level of public debt, the analysis would be seriously flawed, and indeed perhaps dangerous.

Of late there is some questioning about whether public debt levels are too low,8 but these are context-specific and have arisen in the aftermath of the global financial crisis and the painfully slow recovery in many countries, despite the interest rates being close to zero and/or even negative in some cases. The question being asked is whether higher public debt has therapeutic value in a situation where private balance sheets are broken. This inquiry, on the other hand, is about the prophylactic role of public debt—that is, whether public debt lends a degree of stability to an economic system which cannot (or rather, should not) be arrogated to private debt.

The purpose of this paper, therefore, is to outline some of the considerations which should be taken into account while determining the desirable stock of public debt and of its flow counterpart—the fiscal deficit. The context is obviously India, but most of the conceptual elements have wider applicability.

Monetary Considerations

In all modern economies, national currencies are backed by some form of sovereign debt. Central banks, such as the Reserve Bank of India (RBI), issue currency on the basis of their holdings of sovereign bonds and sometimes of gold. In an autarchy, therefore, the minimum level of public debt held by the central bank would be equal to the value of the national currency in circulation minus the value of gold holdings. In India this would amount to roughly 14% of GDP. In an open economy, however, this tight relationship between currency and public debt can be loosened by the central bank holding sovereign assets of other countries—that is, foreign exchange reserves.

As things stand, the rupee is backed almost entirely by foreign assets as per the RBI’s balance sheet.9 There is nothing intrinsically wrong in this, especially in a context where the balance of payments has by and large been in surplus for more than a decade, almost entirely due to surpluses in the capital account. For a country with a non-convertible currency, such as India, building up foreign exchange reserves is desirable both for precautionary purposes and for preventing an unwarranted appreciation of the rupee.10 Nevertheless, the RBI always needs to hold a certain amount of central government debt instruments11 in order to carry out its monetary management responsibilities in a credible manner.12 There is no hard and fast rule governing how much public debt a central bank should carry in its books; but as a rule of thumb, if the currency is not convertible, then the more open the country is to foreign portfolio flows, the higher should be the quantum. At present the RBI holds 15% of the stock of central government securities, which is 10% of all government securities. This does not seem excessive for the roles that the RBI has to perform.

More importantly, prudence requires that consideration should be given to situations where the balance of payments is not in surplus or is inadequate. The liquidity needs of the domestic economy cannot be held hostage to the availability of foreign exchange at any given point of time. One way to address this issue is to ensure that the flow of domestic sovereign debt (which is essentially the size of the fiscal deficit of the central government) is at least as large as the value of additional currency required by the domestic economy for its growth needs. In the Indian context, if it is assumed that the desired growth rate of nominal GDP is around 11.5% per annum,13 then a simple-minded application of Fisher’s equation14 yields a minimum central government fiscal deficit of 1.6% per annum, provided that the ratio of currency
held by the public to total currency in circulation remains roughly constant.15

Fiduciary Considerations

In any country, a large part of household wealth is held for precautionary purposes and for meeting post-work life consumption needs. For such investments, the return is less important than the security of the principal. By and large, countries with low risk thresholds and with poor or non-existent social security systems, such as India, will tend to place more importance to such assets and have a higher share of these in the total household financial wealth.16

All countries recognise this imperative and impose fiduciary status on institutions offering specific forms of assets.17 The common forms are life insurance, pension/provident funds, and certain types of mutual funds and asset management company products.18 In India, there is an additional asset class called small savings instruments for which the government itself is the fiduciary, that is, it bears 100% of the liability.19 This amounts to about 11.5 percentage points in the total public debt stock of 68% of GDP. The other assets which bear fiduciary protection comprise another 25% of GDP. The laws governing these assets specify that at least 50% of the value must be invested in public debt instruments, which include both central and state securities.20 Therefore, just for compliance with the law, the stock of public debt must be a minimum of around 12.5% of GDP on this count alone.

Although legally commercial banks are not fiduciaries, the perception of the depositors is usually quite different, and they tend to view bank deposits as a form of low-yield secure assets. Therefore, even if banks do not have legal fiduciary status, they certainly bear a moral fiduciary responsibility. Most governments recognise this tension between the legal and the moral/perceptual status of banks, and address it through “prudential regulations.” The more liberal forms of such regulations require that a specified proportion of a bank’s liability must be held in the “safest,” or highest rated, securities, whether public or private. In the United States, for instance, private AAA rated securities were eligible to play this role, and we all know how that has panned out.21 It is not yet clear whether the appropriate lessons have been drawn from the global crisis, especially on the distinction between “risk-free” and “almost risk-free.”22

In India, prudential regulations stipulate that 20% of the total net liabilities of banks (called the statutory liquidity ratio [SLR]) must be held in government bonds. This works out to 18% of GDP.23 This requirement has varied from time to time from a high of 25% to the present level of 20%. However, the experience has been that banks voluntarily tend to hold a significantly higher proportion of their assets in government bonds than mandated by the regulation.24

Therefore, if we add up the minimum amount of public debt required—by law—to meet fiduciary responsibilities in India, it comes to 42% of GDP, comprising 11.5% for small savings instruments, 12.5% for insurance, provident funds, etc, and 18% for the commercial banks. Thus, there is “excess” public debt of the tune of 26% of GDP.

Of this “excess” public debt, 3% accrues to the central government as external debt,25 6% to the centre and the states as reserve funds,26 5% as the RBI holdings of the central government bonds, 3.5% as the excess SLR holding of banks, 2.5% as the excess holdings of insurance/provident funds et cetera, and 6% as corporate holdings.27 Among these, the last four are the central determinants of the interest rate level in the country.

Interest Rate Considerations

The interest rate is one of the most important economic variables in any economy which influences a number of decisions taken by a wide range of economic agents. Savings and investment decisions are the most obvious, but interest rates also determine the choice of technology by firms, and play an important role in influencing the production structure of the economy.28

Sovereign debt instruments provide the anchor for all interest rates, since they are the only financial instrument with zero default risk. Theoretically, the interest rate on private debt of a particular maturity should be the interest rate on government bonds of the same maturity, with a premium reflecting the default risk of the private borrower.29 For it to effectively play this role, however, government bonds must be freely and actively traded so that their yield (which is the effective interest rate) accurately reflects the market risk and liquidity premia.30 It is, therefore, necessary that an active market should exist for the government bonds.

The market for government bonds in India is essentially a wholesale one, with no retail participation at all. The participants are banks, other financial institutions, foreign portfolio investors (FPI) and some non-financial corporates. As far as the volume is concerned, although the total stock of government bonds is nearly 48% of GDP, the volume “in float” is a much smaller 17%, since the statutory holdings by banks and other fiduciary institutions simply cannot be placed on the market. Thus, the market is limited to the excess holdings of these institutions (6%), the holdings of non-fiduciary bodies (6%) and that of RBI (5%). Thus, the range of participants and the float are large enough for a reasonably efficient market.

The market maker, to all intents and purposes, is the RBI. However, the RBI’s functioning in the market is different from that of any standard private market makers in the sense that its objective is not to maximise profits from arbitrage and trading fees, but to attain the desired level of the interest rate. In order to carry out its mandate, therefore, the RBI necessarily has to have a target rate around which it can work. Most central banks have a base value of the real interest rate which is publicly stated, and monetary management is essentially about deviations from this base value depending upon cyclical factors. In India, neither the RBI nor the governments have articulated the “desirable” level of the real interest rate during the normal times for more than a decade now.31

Conceptually, the minimum (that is, risk-free) long-term real interest rate, especially in a developing economy, should be at or around what is termed as the “social rate of time preference,” which is a measure of the value a society places on consumption at present relative to consumption in the future.32 Typically, poor countries will have a significantly higher rate of time-preference than the richer ones. In India, the social rate of discount (which is the nominal counterpart of the rate of time preference) used to be fixed by the Planning Commission.33 This figure stood at 12.5% for nearly 25 years beginning from the Fifth to the Ninth Five Year Plans (FYPs).34 Subsequently, it was reduced to 9.5% from the Tenth FYP onwards to reflect the increase in incomes, the surge in the savings rate and the reduction in anticipated inflation.35 As things stand, NITI Aayog has given no indication of what this figure could be.

One way to get around this problem is to assume that the coupon rate on treasury bills reflects the finance ministry’s take on this.36 At present, the coupon rate on the 10-year treasury bills is slightly above 7%. Assuming that the target inflation rate is 4%,37 this yields an implicit time preference rate of around 3%, which is too low by any yardstick. In order to judge whether the voluntary holdings of government securities are too high, the actual market-determined yield should be compared to the desired rate.

If the yields are higher, it implies that there is an excess of government securities in the market. If, on the other hand, the yields are lower, it means that there is an excess demand for government bonds, and any reduction in their supply will lower interest rates even further, thereby distorting the various decisions that are contingent on the interest rate.38 At present, the yields on 10-year treasury bills are roughly similar to the coupon rate. Therefore, the voluntary holdings of public debt amounting to 12% of GDP are by no means excessive, and may even be too low.

Minimum Public Debt

If the voluntary holdings of public debt are added to the mandated and the RBI holdings, the minimum public debt stock comes to at least 58% of GDP. However, this is only the current position (as of 2016–17). The minimum debt ratio will almost certainly change over time depending upon developments in the economy. Therefore, before any fiscal rule is adopted, there must be some understanding as to how the minimum debt ratio is expected to evolve in the future.

In view of the government’s push towards greater “financial inclusion” in the form of bank deposits and insurance coverage, it is likely that this ratio will trend upwards in the foreseeable future.39 Therefore, a certain amount of cushion needs to be provided for contingencies. Seen in this light, the FRBM Committee’s recommendation of a target public debt ratio of 60% seems eminently sensible, not as a ceiling but as a floor.

Level of public debt: The main impact of insufficient public debt will show up primarily in the interest rate on government securities being lower than conceptually desirable. While this may improve the government’s fiscal position, it could be disastrous for the economy on a number of counts. First, the earnings of fiduciary institutions, including banks, from the legally mandated assets would go down sharply, thereby jeopardising their financial viability. In such a situation, they would be forced to invest their other funds in higher yielding that is riskier assets, which would increase the systemic risk in the economy.40

Second, since the overall level of interest rates would be much lower, it would tend to reduce domestic savings and increase investments. All else remaining constant, this would lead to a sharp increase in the current account deficit (CAD), thereby increasing the external vulnerability of the country. It should be noted that the exchange rate may not depreciate to correct this imbalance since the portfolio flows may more than offset the CAD. Any sudden withdrawal of FPI funds could then trigger a crisis.41

Third, the lower interest rates would encourage the adoption of more capital-intensive technologies across all sectors and boost the growth of capital-intensive sectors relative to the more labour-intensive ones. Both these will have serious adverse effect on the growth rate of employment, which is particularly a concern at this time when the country is trying to cope with its much-vaunted “demographic dividend.”

The destabilising effects of excessive public debt are well known, but insufficient public debt can be just as damaging to an economy. The former works through loss of government credibility and reduced investment demand, while the latter through impaired viability of the financial sector and reduced supply of savings. Therefore, it makes sense for countries to adopt fiscal rules which would maintain the public debt ratio somewhere between these two extremes.

Composition of public debt: While the stock of public debt has been the subject matter of considerable research, the composition of the debt stock has practically received no attention whatsoever. The entire literature on the sustainability of public debt and the dynamics between the fiscal deficit and the debt stock ratio assumes that the nature and composition of the debt holdings does not matter. Technically, this is absolutely right for this specific line of enquiry where the focus is only on the financing role of public debt. But unfortunately, things are not quite as simple as that when the other functions played by public debt in any economy are taken into consideration. In such cases, the fact is that the composition does matter, and indeed matters a great deal. Any analysis which ignores this is seriously flawed.

In such cases, a basic distinction has to be drawn between the debt instruments that are tradable (namely, government securities) and those that are not (such as, all non-securitised public debt instruments like postal savings accounts, public provident funds, external borrowings, and similar other deposits). It is the former which are held by fiduciary institutions and also determine the market yields, and thereby the interest rate structure of the economy, while the latter has no direct role to play.42 This distinction then raises the very disturbing possibility that a country can simultaneously face an unsustainable level of total public debt and an insufficiency of appropriate debt instruments to manage macroeconomic policy objectives.

This possibility is directly related to the share of non-traded public debt to the total—the higher the proportion, the greater the chances of such an occurrence. This may not be particularly relevant to developed countries that have most of their public debt in the form of tradable securities. But it is of consequence for the developing countries that tend to have a very significant part of their public debt stock in the form of negotiated loans, whether from multilateral and bilateral aid agencies or from financial institutions. Moreover, in countries where the commercial banking sector is relatively weak and/or has limited geographic reach, postal savings accounts tend to form a high proportion of household financial assets, which automatically become a part of the government’s non-tradable debt stock. In all such cases, the compositional issue makes fiscal management considerably more complex.

This is no mere speculation. India may actually be on the brink of facing such a situation. As things stand, of the total public debt of 68% of GDP, tradable instruments account for 47.5 percentage points and the non-tradable account for the remaining 20.5 percentage points.43 Thus the non-tradable to total debt ratio is 30%. In comparison, as has been assessed, of the minimum public debt of 60% of GDP, the tradable component should be 48 percentage points and the non-tradable 12%—that is a non-tradable share of 20%.44 Given these proportions, it is entirely possible to have a public debt ratio well above the minimum and yet run the risk of systemic disruption. It is clear, for instance, that the supply of government securities is already below its optimal level.45

Although this is not an immediate problem, since the gap is small and may well be due to estimation errors, it should not be allowed to persist. The real problem may emerge as the country moves from the current debt stock ratio to the recommended if the bulk of the reduction is obtained from slow growth of government securities. At present, the RBI has sufficient stock of government bonds to tackle any immediate mismatch problems, but if the gap increases from 0.5% of GDP to say 5%, it will simply run out of instruments. This and other related issues are taken up in the next sections.

Fiscal Deficits and the Flow of Debt

The stock of public debt is only one part of the story. Consideration must also be given to its flow—namely, the fiscal deficit—which is the annual rate of generation of public debt. There is little point in having a desired level of public debt to GDP ratio if the addition to this stock is not consistent with maintaining the ratio over time. This is where the FRBM Committee went wrong. Having determined the desired level of the public debt ratio at 60%, it recommended a fiscal rule which would restrict the fresh inflow to 4.5% per year. As Subramanian correctly pointed out in his dissent note, consistency demanded that the fiscal deficit should have been specified as 6.2% per annum in order to stabilise the public debt ratio at the desired level.46

In its defence, the committee has stated that the 60% public debt ratio is in fact a ceiling, and it can fall below that over time. This position of course squares the circle, since there is, then, no lower limit to which the ratio can be allowed to fall.47 However, if 60% is the floor, as has been argued above, then the problem resurfaces, since any consolidated fiscal deficit below 6.2% could lead to breaching the floor.48 On the other hand, if nominal GDP continues to grow at 11.5% per year as assumed by the committee, a 6.2% fiscal deficit limit will lead to the public debt ratio converging to the 60% floor in 15 years and continuing thus thereafter. This does not seem such a bad outcome, unless there are other, and as yet unstated, reasons for wanting to reduce the fiscal deficit by a larger magnitude.49

It should be noted that a 6.2% target consolidated fiscal deficit requires no fiscal correction at all since the present levels of deficit are 3.5% for the centre and 2.7% for states. Moreover, it is more or less consistent with the FRBM Act currently in force, with only minor adjustments both in the total and in the centre and states targets. The demand for reducing the consolidated fiscal deficit further is almost always justified by appealing to global norms, but nobody seems to ask the question of whether the global norm is optimal in any sense. The argument that something should be done because everybody else is doing it is very dubious logic at best, especially when the country has been performing far better than the comparators.

More importantly, the level of the consolidated fiscal deficit ratio must take into account the compositional inconsistency that has been discussed in the previous section. It can be shown that preventing a growing mismatch between the available stock of government securities and their requirement has an important implication for financing of the current fiscal deficits: government securities must account for at least 5% of GDP in the total financing of the consolidated fiscal deficit for the immediate futureUnless this condition is met, the gap between the actual and desired stock of government securities will continue to widen to a point where fiduciary institutions will be under stress both from being unable to meet their legal requirements and from reduced income flows from their holdings of public debt instruments.

Clearly this requirement cannot be met if the fiscal deficit is reduced to 4.5% of GDP. However, at first glance, if the fiscal deficit ratio is held at 6%, this may not seem to be a particularly demanding requirement considering that in recent years the issue of government securities has been somewhat less than 4.5% of GDP. But it does require a change in the manner of thinking about the financing of fiscal deficits. At present, the issue of government securities is a residual after taking into account the receipts from the non-tradable forms of debt and the target fiscal deficit. Given the compositional problem, however, this has to change, and the annual issue of government securities has to be a target in itself.

This creates an entirely new fiscal dilemma. Since, by definition:

Fiscal deficit non-tradable debt receipts + receipts

from securities

the government’s ability to meet both a fiscal deficit target and a target on minimum issue of government securities depends entirely upon its ability to control the non-tradable component.

The non-traded forms of public debt can be categorised into four broad sub-components: (a) external debt (3%); (b) reserve funds and other deposits (6%); (c) provident fund of government employees (5%); and (d) small savings deposits of the public (6.5%). Of these, (c) and (d) are determined by the portfolio decisions of savers, and the government has little direct control except by way of changes in the applicable interest rates.

There has been a substantial downward revision in these interest rates over the last 20 years—around 400 basis points over the period.50 As a consequence, the growth of small savings has been somewhat slower than the growth rate of GDP leading to a gradual decrease in its share. The provident funds of government employees, however, have shown a sharp increase in recent years. This has been driven mainly by the shift from a defined-benefit pension system for government employees to a defined-contribution system since 2004.51 There is no reason to believe that this pattern will change dramatically even if the interest rates are reduced further to say 7% in keeping with the T-bill rate. But these rates are politically extremely sensitive and difficult to change.

External debt used to be extremely important at one time as a valuable source of foreign exchange, but its importance has waned over the years as private capital flows surged. Moreimportantly, it has dropped steadily as a proportion of GDP for the last one-and-a-half decades and is now a mere 3%. In any case, currently the annual inflow as a percentage of GDP is in the second place of decimal and can be ignored. Over the longer run, repayments will probably become larger than the fresh additions and the stock will gradually reduce.52

The category “reserves and other deposits” has been rising, but it is not clear how much control the government can exercise over its future movement since it comprises a number of different forms of liabilities.

On the whole, therefore, there is considerable uncertainty about the government’s ability to control the non-tradable component of debt receipts. In such a situation, there is always the possibility that a choice may have to be made about whether the fiscal deficit target should be maintained or the target for issuance of government securities. In any event, the government needs to carefully track the evolution of the minimum stock of government securities and then to take hard-headed decisions about how it can be managed. This is not currently a practice that is followed by the Ministry of Finance or the RBI.

If, however, for whatever reason, the government is adamant about reducing the public debt stock and its fiscal deficit without running into the compositional problem, a politically difficult but administratively simple expedient would be to discontinue the small savings schemes and/or shift government provident funds to an institution. In both cases, the minimum debt stock requirement will reduce and the non-tradable component will drop.53

Pressures and Policy Choices

The central message of this paper is that the public debt stock and the fiscal deficit have dimensions which go beyond, and are perhaps more important than, the mere financing of government’s expenditure. This is hardly an original thought, but, in recent years, it appears to have been drowned out by the fiscal policy wisdom purveyed by international finance capital and its hand-maidens—the rating agencies. The deluge has been going on for long enough that most policymakers, including economic technocrats in the government, have come to believe that public debt, fiscal deficits and interest rates are objectives in themselves rather than instrumentalities that they actually are. As a result, the current discourse on fiscal management almost completely ignores these dimensions.

This is not a problem unique to India. In recent years, fiscal rules have been implemented in country after country. According to the International Monetary Fund (IMF), 80 countries had such rules in place in 2014,54 many of them developing countries. It is more than likely that many of these countries have not adopted these rules on the basis of any reasoned analysis of the likely future consequences, but on rule-of-thumb prescriptions based on policy decisions taken in very dissimilar contexts.

The core of this now dominant narrative is that rapid growth can only take place through high levels of private investment; and that higher levels of public debt, fiscal deficit and interest rates all retard private investment in specific ways. This may indeed be true under certain circumstances, but rational policymaking demands the alternative objectives must also be evaluated before any decision is taken. Moreover, recent empirical research strongly suggests that private investment may not always be quite as growth enhancing as it is made out to be.55 The relationship between growth and private investment has been shown to be non-linear, and the growth rate is likely to start falling as private credit rises into the range of 80% to 100% of GDP. India is already perilously close to this range. Thus, the findings of this paper suggest that India may be at an inflexion point where a number of trade-offs need to be carefully considered before decisions are taken.

Take, first, the fiscal deficit ratio. It is commonly used as an indicator of “crowding out” of private investment, and thereby retarding growth, due to government profligacy. India has been consistently projected as an outlier on this count and penalised through a lower rating than justified by other performance criteria. The simple fact of the matter is that while this is an important consideration for determining debt sustainability, it is an inappropriate metric for assessing the degree of crowding out. The correct metric in fact is the fiscal deficit as a percentage of non-government savings. On this metric, India no longer appears as an outlier.56 The decision on whether the fiscal deficit ratio needs to be reduced, therefore, should be based on other, and more fundamental, considerations such as debt sustainability or meeting the need for government bonds.57

The public debt stock presents a more complex challenge. On one hand, as has been assessed, some slack exists even at present, but a relatively rapid reduction of this slack runs into the problem caused by the compositional requirement of public debt. On the other hand, both the compositional issue as well as the minimum public debt stock requirement can be addressed by a stroke of the pen—simply reduce the SLR and/or broaden its scope to include private debt instruments.58 These measures require no legislative action or even approval of the government. The RBI, as the regulator, is perfectly competent to take these decisions on its own.59 Given the ease of taking such a step, this may well be the next arena in the battle for minds between the interests of international finance capital and of low risk-threshold domestic savers. It should be clearly recognised that in effect it represents a trade-off between a supposedly rapid, but high-risk, growth strategy and a more stable, sustainable growth path.

The interest rate argument will attract even greater adverse pressure since both finance capital and large domestic corporates will array themselves against it. Unfortunately, this is an unequal battle. While the gains from a lower interest rate to these entities are obvious and immediate, the drawbacks are diffused and macroeconomic in nature. As a result, there is likely to be little countervailing pressure. However, it should be made amply clear that the social discount rate, and hence the yield on central government securities, is a political decision and not a technical one. In taking such a decision, the government and the RBI must ask themselves one fundamental question: are we really a capital-surplus, low poverty developed country?

If not, interest rates cannot be left to the market alone. But this is only the beginning of a process. The next step is to incorporate some of the considerations outlined in this paper in any new version of the FRBM Act.60 Finally, these should become integral parts of the models used for assessing public debt dynamics, and eventually into the larger macroeconomic models used for both fiscal and monetary policy formulation.


1 For major decisions there is usually some form of consultation between “concerned” ministries, but these are generally limited in their scope.

2 The committee was appointed by the Ministry of Finance. It is popularly known as the N K Singh Committee.

3 Arvind Subramanian was the former chief economic adviser to the Ministry of Finance, Government of India.

4 The main justification for a steady reduction in public debt appears to be the views of international rating agencies. Why this should matter at all is not clear in view of the excessive foreign portfolio investments that India has received despite not too favourable ratings. In particular, the chief economic adviser has publicly been critical of the rating agencies, and yet invokes their views to justify his position on this particular issue.

5 Public debt is deemed “unavoidable” because that is what governments, especially politicians, do. The “necessary” part is recognised only when there is a serious contractionary shock to the economy.

6 The bulk of the extant work is on sustainability of public debt, which is a perfectly legitimate enquiry in view of the periodic debt crises that have gripped various countries over the years, mainly in Latin America but more recently in the PIGS (Portugal, Ireland, Greece and Spain). The very recent literature on “optimal” public debt is substantively no different although the term “optimal” gives the impression that some level of public debt can actually be good. For the most part, the results are a revalidation of the “crowding out” hypothesis of yesteryears.

7 Government securities of course do carry some market risk, in that changes in inflation can alter the real returns and that their prices can change according to demand-supply changes, but not default risk.

8 See, for instance, Kocherlakota (2015) and DeLong (2015).

9 In reality, roughly 30% of RBI’s assets are in central government bonds; the remaining 70% in foreign assets. However, the Issue Department (which is in charge of currency) is shown as having only foreign assets, and the entire holding of domestic public debt is shown to be with the banking department (which also holds about 40% of its assets in foreign reserves).

10 The downside is that the “seignorage,” which should accrue to the Indian government, ends up with foreign governments.

11 State government bonds cannot serve this purpose since they are not sovereign.

12 One of the more important roles of central bank holdings of public debt is the sterilization of foreign capital inflows in excess of the amount needed for the desired growth of money supply. India has faced this issue on a number of occasions.

13 This is the base scenario of the FRBM Committee, and essentially translates to a desired growth rate of 7.5% in real GDP plus a target inflation rate of 4%.

14 Fisher’s equation is an identity which relates money supply with the total value of transactions. The common version of this equation, known as the Cambridge equation, is: MV = P y, where: M = money supply; V = velocity of circulation of money; P = price level; and y = real GDP.

15 This assumption is important in view of the fact that the present government has clearly indicated its desire to reduce this ratio significantly. This is one of the main components of the recent demonetisation narrative. If it happens, then the minimum central fiscal deficit requirement on this account will also go down. However, present trends suggest that this ratio is well on its way to returning to its pre-demonetisation level.

16 Physical assets such as real estate and gold are usually preferred when there is insufficient trust in the financial system. This has been the case in India for a long time, and its consequence has been a much lower level of savings available for productive purposes than would be the case otherwise. Building public trust in the financial sector must, therefore, form a key component of the government’s development strategy and the assurance given by public debt instruments may play an important role in this.

17 A fiduciary is defined as: “a person to whom property or power is entrusted for the benefit of another”. A fiduciary is required to act in the best interest of the client, and protection of the principal is a central objective.

18 Any fund or product which guarantees the principal falls into this category.

19 This class also includes the provident funds of government employees and postal savings.

20 Whether the 50% requirement is adequate for securing the fiduciary obligation is a question beyond the scope of this paper. However, any such rule has to draw a balance between security and a minimum acceptable level of returns.

21 The infamous collateralised debt obligations (CDOs), which triggered the global financial crisis of 2008, were by and large rated as AAA. And even after this we continue to take the pronouncements of rating agencies seriously.

22 It would be interesting to examine whether the impact of the global financial crisis was systematically related to the proportion of government bonds held by the banking sectors of countries.

23 Unfortunately, the RBI of late has been treating the SLR purely as an instrument of liquidity management rather than a means of fiduciary protection. This needs to be seriously debated and a view taken at the highest political level.

24 SLR holdings of banks have been as high as 32%, and are presently at about 24%.

25 The central government’s external debt is entirely concessional loans from multilateral institutions and some bilateral donors. There is no reason why it cannot be dispensed with, except for its lower cost.

26 This includes the market stabilisation scheme (MSS) bonds, which are the price the government pays for building up foreign exchange reserves. It is also a part of the much-maligned excess SLR holdings of the banks.

27 This includes holdings of government bonds by foreign portfolio investors (FPIs).

28 In theory, a relatively high interest rate encourages the growth of labour-intensive sectors vis-à-vis capital-intensive ones.

29 Under the assumption that the interest rate on the government bond fully captures the market risk premium (inflation risk and interest-rate risk) and the liquidity premium for that particular maturity. The default risk, on the other hand, for any single private entity can vary with the maturity.

30 The market risk and liquidity premium are the difference between the yield and the coupon rate.

31 The former governor of the RBI, Rahguram Rajan, in one of his speeches had indicated that the repo rate in India should be around 150 basis points above the expected inflation rate. However, he did not indicate what the rate on long-term government bonds should be. For this one would have to go into the issue of the desirable slope of the yield-curve, which is beyond the scope of this paper.

32 In developed economies, the concept that is currently in vogue is the “long-run neutral rate of interest.” In an interview with The International Economy Magazine in 2005, Janet Yellen, then chairman of the Federal Reserve Bank, San Francisco, had estimated the real neutral interest rate for the US to be in the range of 1.5% to 3.5%. In the mid-1980s, the consensus view was that it was around 5%. It is hard to imagine that the equivalent rate for India today would be less than what it was for the US then. Thus, the 4.5% social rate of time preference used by the Planning Commission is certainly not high, and probably on the lower side.

33 The social discount rate can be thought of as the social rate of time preference plus the expected rate of inflation. This rate is used as the discount rate for all social cost–benefit analysis and for appraisal of public investment. Private investment decisions of course are based on the market interest rate.

34 This figure comprised of a time preference rate of 6.5% and an inflation rate of 6%.

35 The time preference rate was reduced to 4.5% and the inflation rate to 5%.

36 Usually coupon rates are set as close as possible to the expected yield in order to minimise auction volatility. However, there is no indication of the considerations that have gone into
finance ministry’s choice of the coupon rate.

37 This is assumed based of the mandate given to the RBI for inflation targeting.

38 The “investor community” and the rating agencies will of course welcome both the reduction in public debt and the lowering of the interest rate, but is this the constituency that the
government should be catering to?

39 Projecting the future trajectory of the demand for public debt is not a trivial exercise. It is broadly influenced by two factors. The first is the non-government savings rate and the
second is the distribution of these savings between different asset classes. Both of these are inextricably linked to the growth rate and to the size and class distribution of income in the country, and how these are expected to behave in the future.

40 Adverse selection is always a problem in banking. This will simply make it worse.

41 FPI holdings of central government bonds are only 5% of the total stock, but actually represent 20% of the demand for the floating stock, which is very high.

42 These will of course have an indirect role through a general equilibrium effect.

43 The break-up of the present stock is—Tradable: banks—21.5%; insurance, etc—15%; RBI—5%; corporates, FPI, etc—6%. Non-tradable: small savings, provident fund, etc—11.5; external debt—3%; reserve funds and deposits—6%.

44 The break-up of the minimum stock is: Tradable: banks—21.5%; insurance, etc—15.5%; RBI—5%; corporates, etc—6%. Non-tradable: small savings, PF, etc—12%.

45 This may be one of the reasons why in the last year, the yield on T-bills has dropped by 60 basis points and the rupee has appreciated.

46 This figure is derived from the standard equation that relates the steady-state value of the public debt ratio (d) with a constant fiscal deficit to GDP ratio (fd) and a constant steady-state nominal growth rate of GDP (g):

fd = d.[g/(1+g)]

If d is 60% and g is assumed to be 11.5%, then fd works out to 6.2%.

47 As it happens, if the consolidated fiscal deficit is pegged at 4.5% of GDP per year, the public debt ratio will fall to 41% of GDP in 15 years under very reasonable assumptions.

48 For instance, a 4.5% fiscal deficit will lead to the debt ratio going below 60% in just 5 years if growth remains at 11.5%.

49 The global rating agencies have been pressing for a 4% consolidated fiscal deficit, and have cited this as the main reason for keeping India’s sovereign rating at just above the “junk” status. Unfortunately, their view finds considerable traction in the government.

50 The movement in these interest rates has mirrored the changes that took place in the social discount rate with the rate on long-term instruments dropping from 12% in the mid-1990s to 8% at present.

51 Large increases in government salaries in 2006 and again in 2016 also contributed to this increase.

52 The process, however, will be excruciatingly slow since these loans have very long tenors, usually around 30 years, which means that the annual reduction in the debt ratio will be only about 0.15 percentage points on this account.

53 In the case of small savings, the government will have to continue to carry the existing stock on its books until they mature and thereby are extinguished. However, since the future stream of such savings will cease, the government will be able to issue securities of the same value for any given level of the fiscal deficit.

54 IMF Fiscal Rules Database, cited in the FRBM Committee Report.

55 See Arcaud et al (2015).

56 India has one of the highest savings rates in the world, barring China, and also happens to have negative government savings. Consequently, the non-government savings rate in India is above 30% of GDP, which means
that the government absorbs about 20% of these savings through its 6.2% fiscal deficit ratio. This is roughly at par with comparator countries.

57 This is of course apart from the most fundamental consideration of all—management of aggregate demand in the economy.

58 Every 1 percentage point reduction in the SLR, or allowing 1 percentage point of SLR to be held in private securities, reduces the minimum public debt stock by 0.9% of GDP and the necessary fiscal deficit ratio by 0.1% of GDP.

59 Something similar can be done by reducing the ratio of public securities required to be held by fiduciary institutions. This would, however, require legislative action and therefore much greater scrutiny and resistance.

60 The recommendations of the N K Singh Committee Report on fiscal deficit ratios should certainly not be accepted in their present form.


Arcaud, J L, E Berkes and U Panizza (2015): “Too Much Finance?” Journal of Economic Growth, Vol 20, No 2, pp 105–48.

DeLong, Bradford (2015):

Escolano, J (2010): “A Practical Guide to Public Debt Dynamics, Fiscal Sustainability, and Cyclical Adjustment of Budgetary Aggregates,” IMF Technical Notes and Manuals 2010/2,
International Monetary Fund, Washington,

FRBM Review Committee (2017): “Responsible Growth: A Debt and Fiscal Framework for 21st Century India,”

Kocherlakota, Narayana K (2015): “Public Debt and the Long-run Neutral Real Interest Rate,”

Sen, Pronab (2017): “When Windmills Tilt,”

Updated On : 12th Jul, 2019


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top