Information about Asia and the Pacific Asia y el Pacífico
Journal Issue

A Framework for Assessing Fiscal Sustainability and External Viability, with an Application to India

Karen Parker, and Steffen Kastner
Published Date:
October 1993
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

I. Introduction

The achievement of balance of payments “viability” is a central objective of most Fund-supported adjustment programs. A viable balance of payments is one in which the current account deficit can be financed, on a sustainable basis, by capital inflows on terms that are compatible with the development and growth prospects of a country. An externally viable economy does not rely on exceptional financing, be it through the use of Fund resources, concerted lending, rescheduling, or the accumulation of arrears. Rather, external financing needs are met through spontaneous capital inflows, including from private and official sources (see Guitián, 1981).

Until recently, assessments of whether an economy was externally viable were based almost exclusively on medium-term balance of payments and external debt projections. These were used to provide a profile of future debt service obligations and to estimate prospective financing gaps. Provided the economy as a whole was expected to generate sufficient resources to meet its external obligations, it was regarded as viable. The domestic resource transfers—usually from the private to the public sector—that were required to meet those obligations received less attention. The links between medium-term fiscal policy, private savings and investment, and the balance of payments often were not made explicit. The prospect of public insolvency seemed remote since governments face a “soft” budget constraint. If necessary, the monetary authority could issue currency or domestic debt so that the public sector could meet its external obligations.

Yet balance of payments projections alone cannot identify whether a country is on a viable path. The experience of many countries during the 1980s showed that policies which channel a large share of private savings toward financing government deficits carry a high cost in terms of inflation, interest rates, and growth. A satisfactory assessment of a country’s medium-term outlook must therefore entail a careful examination of the public-private resource flows implied by the external, fiscal and monetary targets. At issue is the sustainability of fiscal policy and its consistency with other macroeconomic objectives.2/

Over the past decade a large literature has developed around the issue of fiscal sustainability. Intertemporal models have identified the conditions for government solvency and have highlighted the consequences of untenable fiscal strategies (see Barro (1974), Sargent and Wallace (1981), Buiter (1983, 1985), Spaventa (1986), Eaton (1987), Calvo (1988, 1989), Horne (1988, 1991), Guidotti and Kumar (1991), Buiter and Patel (1992), and Barrionuevo (1992)). Since private agents recognize the government’s intertemporal constraint, the authorities may find it difficult to influence real output through expansionary fiscal policies. On the contrary, a rapidly rising public (or publicly-guaranteed) debt stock could set in motion perverse expectational dynamics as individuals demand increasing premia to compensate them for the risk of repudiation through inflation and other taxes. Without adequate fiscal correction, the government could have to incur increasing amounts of debt simply to meet interest obligations.

While the literature on government solvency underscores the costs of postponing adjustment, it leaves unanswered a number of practical questions which confront authorities in setting fiscal policy: What are the short-run tradeoffs between adjustment and deficit financing? What is the appropriate mix of borrowing and financing from seigniorage? What are the implications of alternative fiscal strategies for growth and external viability? To address these questions, one must explore the relationship between fiscal sustainability and macroeconomic performance using empirical tools.

This has been the goal of several recent econometric models, including Anand and van Wijnbergen (1988, 1989), van Wijnbergen, Rocha and Anand (1989), Easterly (1989a, 1989b), Bhandari, Ul Haque and Turnovsky (1989), Dooley et al. (1990), van Wijnbergen (1990), Ul Haque and Montiel (1992) and Montiel (1993). These studies have attempted to measure the sustainable fiscal deficit for a number of countries, and have examined the long-run relationship between fiscal policy, financial reform, and external balance.

A drawback of some of these approaches is that they do not easily lend themselves to operational use. Frequently the model structures are opaque and not easily adapted to alternative country circumstances and policy experiments. The data requirements may be extensive. Moreover, the focus often is on long-term, or “steady state” relationships, which may be of limited use for programming purposes. On the other hand, operational work too often relies on a static, loosely joined accounting framework. The intertemporal relationships between fiscal policy, private sector behavior and the balance of payments may not be well specified. It is thus difficult to assess the sensitivity of the projections to alternative assumptions about exogenous variables and policy performance.

This paper attempts to move beyond the standard accounting framework by incorporating dynamic elements into the analysis of fiscal policy. It is a step toward bridging the gap between the literature on fiscal sustainability and the demands of operational work. Clearly there is a tradeoff between analytical rigor and the usefulness of econometric models for policy purposes; in an attempt to balance these concerns, we propose a simple yet comprehensive model which can be used to study the macroeconomic implications of alternative fiscal strategies. It is not intended as a precise forecasting tool, nor does it incorporate the latest technical innovations. Rather, it provides the basic analytical scaffolding for the detailed sectoral analyses commonly used in program work—placing these in a dynamic, medium-term context. The framework is sufficiently flexible to accommodate a variety of country circumstances and policy scenarios.

The centerpiece of the model is the multi-period fiscal balance sheet which identifies the sources of financing to the consolidated public sector. Potential financing from seigniorage is given by money demand equations. Net private savings, projected on the basis of estimated equations for consumption and investment, are added to net public savings to determine the current account balance. Real interest rates are determined endogenously, and depend in part on the public debt-to-GDP ratio. This allows for higher domestic risk premia in response to an unsustainable buildup in government debt, forging a link between fiscal policy, growth, and external viability.

The model generates annual estimates of public spending compatible with identified targets for growth, inflation, revenue, and domestic and foreign borrowing. The difference between the spending path consistent with these targets and spending based on current projections is the fiscal adjustment required to meet the authorities’ macroeconomic objectives. The framework also can be used to assess the implications for inflation, interest rates and debt of an assumed spending path. Finally, it allows one to study the impact of financial reform on private savings and the balance of payments.

The principal output of the exercise are medium-term fiscal, savings and investment, and balance of payments projections. These would be accompanied by sensitivity analyses to evaluate their robustness to alternative assumptions about macroeconomic policies, exogenous variables, and model parameters. We have applied the model to India in order to illustrate the types of simulations that may be conducted.

II. Solvency, Fiscal Sustainability and External Viability

Most analyses of fiscal sustainability start with the intertemporal government budget identity, which identifies the sources and uses of financing to the consolidated public sector. This identity states that the public sector’s primary deficit (revenues minus non-interest spending) plus interest payments on the domestic and foreign debt are financed through a combination of domestic borrowing, foreign borrowing and seigniorage. The net worth of the public sector can be estimated by adding up a sequence of single period identities over a finite horizon to derive the present value budget constraint. Solvency requires that the public sector’s net worth be non-negative; the discounted value of expected primary surpluses plus seigniorage must equal or exceed the initial stock of government debt.3/

Although straightforward in principle, the concept of solvency is of limited practical value in defining a sustainable fiscal path. First, it is difficult in practice to determine whether the public sector is solvent due to the uncertainty of future income and expenditure streams. Few have attempted to estimate comprehensive present value budget identities for specific countries. 4/ Second, and more fundamentally, solvency is a weak criterion with which to establish fiscal sustainability. A multitude of fiscal scenarios—and debt to output ratios—may be consistent with a given intertemporal constraint. Yet not all of these are sustainable.5/ The solvency condition relates to long-run relationships and is of little use in identifying appropriate spending programs or debt targets for the medium term. Finally, the sustainability of fiscal policy must be evaluated in the context of the authorities’ overall macroeconomic objectives and constraints; solvency is only one element of this broader picture.

It may not be possible to define a critical debt to output ratio beyond which solvency clearly is violated. 6/ Yet economists and policymakers do care about the size of the public debt, for a number of reasons. First, a rising debt burden is liable to crowd out non-interest spending; public investment is often the first to be cut. Second, increasing indebtedness raises the risk of default, leading to higher real interest rates. These not only increase debt servicing costs but could inhibit private investment and growth. 7/ Third, as the public debt grows, confidence in the government’s ability and willingness to make necessary fiscal adjustments in order to restore solvency erodes. A loss of credibility may result in inflationary expectations, a shrinking of the monetary base and reduced scope for seigniorage (see Sargent and Wallace (1981) and Cagan (1956)).

Absent a debt criteria that is defensible on theoretical grounds, several studies have recommended setting fiscal policy so as to stabilize the debt-to-output ratio at the base-year or, alternatively, some targeted level (see Buiter (1985), Anand and van Wijnbergen (1989), Horne (1991), and Ul Haque and Montiel (1991)). One rationale for this approach is that if the government has not yet defaulted, the cost of servicing its debt must be lower than the expected cost of repudiation. This argument, advanced by Cohen (1985, 1988) considers the government’s willingness, as well as its ability to repay its obligations in assessing fiscal sustainability.8/

There are obvious problems with this ad hoc approach to evaluating fiscal strategies. The base-year debt ratio may or may not represent a sustainable—much less an optimal—debt burden. The simple fact that a government has not yet defaulted does not mean that the present fiscal path is appropriate. The authorities may already be relying on financial repression, confiscatory tax schemes and inflation to extract resources from the private sector. Such policies are likely to inhibit investment and growth, aggravating the government’s fiscal difficulties. Moreover, under this criterion, countries with low initial debt stocks would have to subject themselves to the same fiscal discipline as those which are heavily indebted. Targeting an alternative debt ratio raises the same analytical problems cited above.

Given the inherent difficulty of measuring the public sector’s net worth or targeting an appropriate debt ratio, a practical concept of fiscal sustainability is needed. One can think of sustainable fiscal policy as one in which the intertemporal budget identity is satisfied ex ante over the entire projection horizon. In other words, the government’s spending program is consistent with its ability to raise revenue and/or finance its deficit while meeting its macroeconomic objectives and servicing current and prospective obligations. This concept of sustainability involves an important time dimension. It is not enough that the authorities be able to meet their current financing needs; they must do so in a way that does not jeopardize their ability to fulfill future commitments. This concept also underscores the importance of evaluating fiscal sustainability in light of overall economic performance.

Of particular concern is the relationship between fiscal policy and the balance of payments. As noted above, a country’s external performance depends critically on the fiscal stance adopted by the authorities. Even so, a sustainable fiscal strategy is not a sufficient condition for balance of payments viability. The latter depends also on private savings and investment, which are influenced by fiscal and monetary policies.

The link between fiscal policy and the balance of payments is evident from the national income identity: private plus public savings minus investment equals net foreign savings, or the current account balance. Balance of payments viability requires that this identity be satisfied ex ante for each period over the projection horizon; foreign savings must be adequate to close the gap between domestic savings and investment. The availability of foreign savings depends upon a number of factors, including market conditions overseas and the government’s policy stance.

Like the intertemporal budget constraint, the national income identity always is satisfied ex post; a shortfall in the balance of payments eventually is met through exceptional financing (including the use of Fund resources, rescheduling or arrears), administrative control of imports, or currency depreciation, External viability requires that there be no prospective financing gaps.

An analogous concept of solvency exists for the balance of payments. An economy could be considered solvent if the net present value of its future trade surpluses is expected to equal or exceed the outstanding stock of external debt. To ensure solvency, domestic production of tradeable goods must eventually exceed absorption in order to service the debt. As Anand and van Wijnbergen (1988) point out, the adjustment is greater for debtors whose production is dominated by non-tradables. Thus, it is common to look at debt-to-export ratios in assessing external viability.

Yet debt-to-export ratios are only a short-run indicator of capacity to repay. They understate countries’ long-run debt service capacity since resources can be shifted from the non-tradables to the tradables sector. Also, debtors might be inclined to depreciate their currencies in order to stimulate exports and thus lower debt-to-export ratios. For these reasons, Cohen (1988) has suggested using a weighted average of output and exports, which he terms a country’s resource base. One chooses the weights so that real depreciation has no effect on the debt-to-resource base ratio. 9/ This ratio is then used as a criterion for assessing external viability.

Though intuitively appealing, Cohen’s index suffers from a number of limitations. For the reasons cited earlier, estimating the limits of a country’s external debt by means of a solvency condition is, for all practical purposes, impossible. Like the other debt criteria, it does not indicate how fast or how far a country’s external debt may grow without threatening insolvency. An economy which appears externally viable over the projection horizon could in fact be creating an unsustainable debt burden.

For these reasons, it is unwise to project a medium-term balance of payments without specifying the underlying savings and investment flows. One should evaluate alternative fiscal strategies in terms of their macroeconomic implications, not in terms of deviations from an arbitrarily-specified debt ratio. What is needed is an integrated framework which allows one to study the public-private resource transfers implied by the authorities’ fiscal and external debt plans. As Masson (1985) stresses, the concept of fiscal sustainability rests on present and future policy settings, taking into account the overall economic environment.

III. A Framework for Assessing Medium-Term Viability

This paper presents a medium-term macroeconomic simulation model which can be used to evaluate alternative fiscal strategies and their implications for growth and external viability. It attempts to bridge the gap between the literature on fiscal sustainability and applied economic and program work. The starting point of the analysis is the multi-period public sector balance sheet, similar to that used by Anand and van Wijnbergen (1988, 1989). It is based on the government budget identity, which identifies the sources and uses of financing to the non-financial public sector (including state-owned enterprises):

where D is the primary deficit (revenues minus non-interest spending, including investment), i and i* are domestic and foreign nominal interest rates, Bg and Bg* are the domestic and foreign public debt stocks, E is the nominal exchange rate (domestic/foreign currency units), and DCg is domestic credit to the government from the central bank. The superscript “.” indicates the change in a variable over time, whereas “∧” indicates percentage change over time.10/

The consolidated public sector balance is derived by adding the net profits of the central bank to equation 1. The central bank’s interest earnings on net foreign assets are assumed to equal the change in its net worth: NW=i*NFAg*E. There are assumed to be no profits or losses on its domestic operations. The change in reserve money, M is derived from the central bank’s balance sheet;M=DCg+NFAg*ENW11/ Therefore,

Equation 3 is derived by dividing both sides by the domestic price level and employing the definition of the real exchange rate: e = p*/p; real variables are denoted by lower-case letters. In a context of high inflation, it may be more appropriate to use an operational concept of the fiscal deficit, which subtracts the portion of nominal interest payments that constitutes real amortization of the public debt (see Tanzi, Blejer and Teijeiro, (1987)).12/ Employing the Fisher relation, subtracting pbg and p*(bg*nfag*) from both sides and rearranging yields 13/:

The budget identity states that the primary deficit plus interest payments on the domestic and foreign debt is financed through a combination of domestic borrowing (including from commercial banks), foreign borrowing and seigniorage. Seigniorage revenue is generated through increased demand for real money balances (m) by domestic residents and through the increase in nominal money demanded by the private sector (mp) to preserve a given level of real balances. Inflation may serve as a residual tax to bridge the gap between expenditures and financing from debt issue and taxes.14/

Equation 3 can be interpreted in several ways. In practical terms, it identifies the sources of financing for a given spending program. It is also a means of estimating the steady-state level of spending consistent with the authorities’ long-run inflation and debt targets (as in Anand and van Wijnbergen (1989)). As noted earlier, it may be interpreted as an intertemporal solvency constraint when the variables are expressed in present value terms (Guidotti and Kumar (1991) and Horne (1988, 1991)).

Clearly, the budget identity always is satisfied ex post, be it through spending cuts, tax increases, borrowing, monetization, debt rescheduling or repudiation. However, to ensure that fiscal policy is both sustainable and consistent with the authorities’ macroeconomic objectives, the budget constraint must be satisfied ex ante for each period over the projection horizon. An unsustainable fiscal policy could be indicated by rising inflation or an unacceptable degree of fiscal retrenchment within a short time frame. Financial repression, high risk premia on public debt, falling secondary market prices and a significant spread between short and long-term interest rates all point to an untenable fiscal stance.

The link between fiscal policy, private sector behavior and the balance of payments is evident from the national income identity:

where sp and vp are real private savings and investment (the former includes interest income from holdings of public debt), (x-m) is the current account balance, bp* is the stock of foreign debt held by the private sector, and nfap* are net foreign assets in the hands of domestic residents. The current account is the sum of net private savings and net public savings, or net external borrowing minus the change in foreign assets.

As noted earlier, balance of payments viability requires that the national income identity be satisfied ex ante for each period over the projection horizon. The demand for foreign savings—domestic savings less investment—must be consistent with the projected availability of such financing. That depends on a number of factors, including market conditions overseas and investor perceptions of the government’s macroeconomic stance.

In addition to the fiscal and national income identities, the model involves several behavioral equations: money demand, private consumption and investment, and the real interest rate on government debt. Whenever possible, these should be estimated using historical data. Alternatively, one can assume a standard specification and plausible parameter values. 15/ Financing from seigniorage is determined by the private sector’s willingness to hold non-interest bearing assets, embodied in money demand equations. Holdings of currency, demand and time deposits are estimated separately, and depend on income, interest rates, and inflation. Private consumption and investment vary with income, real interest rates, inflation and bank credit. The latter is derived from the monetary survey on the basis of estimated money demand and an assumed flow of credit to government.

Domestic interest rates are determined endogenously in the model. Despite free capital mobility, differential risks—of direct taxes, indirect taxes (e.g. inflation), and default—drive a wedge between domestic and foreign interest rates. The default premium is assumed to depend on the public debt-to-GDP ratio, as in Dooley et al. (1990). 16/ This formulation captures the impact on interest rates of an unsustainable debt strategy; it also allows for feedback effects of debt reduction on economic activity. The real interest rate on the government’s domestic debt is thus the sum of the foreign risk-free rate, a default premium, and a real depreciation premium. 17/ Nominal interest rates are derived from the Fisher relation.

To study the implications of alternative fiscal strategies for growth and external viability, one manipulates the set of policy target variables, which includes inflation (revenue from seigniorage), government spending, domestic and foreign borrowing, and reserve accumulation (Table 1). The remaining policy variables are determined residually so as to ensure consistency among the authorities’ macroeconomic targets. 18/ For example, if the authorities set inflation and borrowing targets for the medium term, government spending is determined as the residual. 19/ Alternatively, the government could choose spending and borrowing targets, relying on seigniorage (inflation) to close the fiscal financing gap. Finally, spending and inflation targets might be established, with government borrowing the residual. The sum of financing from these sources yields a measure of the consolidated public sector deficit from “below the line”, which is then used to estimate public savings. This, combined with private savings and investment projections, yields the external current account balance. In all three scenarios, private foreign borrowing is endogenous; it is assumed that the country is not subject to an aggregate external financing constraint.20/














A number of variables are assumed to be predetermined to the model. These include the path of the real exchange rate, domestic and foreign GDP growth, foreign inflation and interest rates, public sector revenue, and public investment. A number of financial variables, including reserve and liquidity requirements, also are exogenous and can be manipulated to simulate the effects of financial reform. The growth of exports and service credits is projected using WEO forecasts of import demand in partner countries, assumed movements in the real exchange rate, and estimated exchange rate elasticities. Transfers are exogenous and imports adjust to meet the current account projections.

To run the model, one first chooses a set of policy target variables—for example, inflation, domestic and foreign borrowing. Real and nominal interest rates adjust to the proposed debt and inflation profile. A program macro computes period-by-period seigniorage revenue from money demand equations whose coefficients are specified in a parameter input table. Seigniorage depends also on the assumed path of required reserves and the interest paid on these. The consolidated public sector deficit is measured from the financing side; spending is thus the difference between the overall deficit and assumed revenues.

A program macro computes private consumption, savings and investment on the basis of the inflation, real interest rate, income and credit projections. Net private savings are then added to net public savings to derive the current account balance. If the current account is constrained by available capital flows, the values of the policy variables can be adjusted in order to bring the deficit in line with projected financing.

The exercise produces fiscal, monetary, savings and investment, and external projections that are consistent with the authorities’ growth, inflation and borrowing targets. The difference between the model’s projected budget and current account deficits and those implied by current policies is the adjustment needed to reach a sustainable fiscal and external position. The adjustment can be made either in one of the policy variables, such as government spending, or in one of the “predetermined” variables, such as revenue. A variety of simulations can be conducted simultaneously, allowing for easy comparison of alternative scenarios.

The framework is contained in a spreadsheet format which is accessible and adaptable to a variety of country circumstances. 21/ The three-dimensional file consists of four sheets, three of which are identical in structure but which feature different sets of policy target variables. The fourth contains the exogenous variables, model parameters, money demand estimates and monetary survey, as well as private savings and investment equations, all of which feed into the policy scenarios. The projection horizon is five years. A program macro calibrates the intercepts of the behavioral equations so as to reproduce base-year values.

There are a number of limitations inherent in this approach. The most obvious is the potential for instability in the behavioral relationships, especially in a context of profound structural reform. It is, arguably, easier to estimate steady-state values for key economic variables than to forecast their evolution over the course of an adjustment program. 22/ Even so, there is a need for short- and medium-term projections to underpin macroeconomic policies; these should be as realistic as possible. The parameter values should be reestimated periodically to provide updated inputs to the model, particularly during periods of structural change. Moreover, the projections should be accompanied by careful sensitivity analyses to indicate the robustness of the estimates to shifts in policy, exogenous variables, and parameter values. It also is worthwhile to study the experiences of countries which have undertaken similar adjustment programs. The simulation exercise can be used to indicate the appropriate magnitude and focus of the adjustment effort, and to suggest a timetable by which the authorities feasibly can reach their medium-term objectives.

For the sake of transparency and flexibility, this framework incorporates a number of simplifying assumptions. The real exchange rate and output growth are treated as exogenous; the incremental capital-output ratio provides a rough consistency check between the investment and growth projections. It was felt that making these variables endogenous would not increase confidence in the results but would make the model considerably more complex. During the course of structural reform both the productivity of investment and the determinants of the real exchange rate are likely to undergo dramatic changes. Rather than attempt to model these, it seemed more appropriate to examine the sensitivity of the results to variations in their assumed values. It would, however, be useful to incorporate feedback effects from the real exchange rate to relative prices, so as to make explicit the external transmission mechanism.

The framework also makes an assumption regarding the allocation of credit between the bank and non-bank private sectors. In many countries where credit is rationed—such as India—bank borrowing is an important determinant of investment; the investment projections are thus highly sensitive to this assumption. In such cases, it is preferable to project bank credit to the private sector on the basis of a portfolio-choice model. Further refinements along these lines may be incorporated at a later stage, or could be developed outside the model.

IV. A Policy Simulation Exercise for India

In mid-1991, India confronted a serious balance of payments crisis, at the height of which foreign exchange reserves declined to just 3 weeks of imports. The proximate cause of the crisis was the rapid withdrawal of foreign currency-denominated bank deposits held by non-resident Indians (NRIs) in response to turmoil in the Persian Gulf and the political crisis at home. The balance of payments was further undermined by the sudden surge in oil import prices and decline in workers’ remittances from the Gulf region, exacerbated by banks’ curtailment of short-term financing. The external crisis prompted the government to severely restrict imports, which weakened investment and growth.

However, the seeds of India’s balance of payments difficulties had been sown earlier, in the form of an unsustainable fiscal policy. The 1980s witnessed steadily increasing fiscal deficits, a loosening of previously conservative monetary policies, and a progressive worsening of India’s external position. Growth, while higher than in previous decades, did not fulfill the promise of India’s savings and investment rates, which are among the highest in the world. Although the productivity of investment increased due to reforms in a number of areas, incremental capital-output ratios remained significantly higher than in the 1950s and 1960s—well above those of India’s Asian competitors (Table 2).

Table 2.India: Selected Indicators




(Percentage change per annum)
Real GDP growth4.
(In percent of GDP)
Savings-investment gap-1.4-2.9-0.8-2.4
Public sector deficit 1/6.910.5

Consolidated public sector. Data for 1970/71-1979/80 begin in 1974/75.

Sources: Government of India, Economic Survey and IMF International Financial Statistics.

Consolidated public sector. Data for 1970/71-1979/80 begin in 1974/75.

Through most of the 1980s the authorities were able to finance large fiscal deficits through seigniorage and forced lending from the banking system. The strategy was sustained through high reserve (CRR) and liquidity (SLR) requirements combined with interest rate controls. From 1986 to early 1991, revenue from seigniorage averaged 2 1/2 percent of GDP and financed some 11 percent of total expenditure. By 1991 almost two-thirds of new bank deposits were channeled to the public sector, much of it in the form of bonds at below-market rates. 23/ Through these means, India largely avoided the pressures on interest rates and prices common to other heavily indebted countries. However, crowding out of private investment has been a serious problem (Sundararajan and Thakur (1980), Krishnaswamy, Krishnamurty and Sharma (1987), Pradhan Ratha and Sarma (1990), Sharma (1991)).

The Indian public sector debt has increased sharply over the past twenty years. Both domestic and external obligations of the central government more than doubled as a share of GDP, with most of the increase occurring over the past decade. The total debt of the consolidated public sector is estimated at more than 90 percent of GDP. 24/ At the same time, the government gradually has been forced to offer more attractive terms on its securities, reflecting in part concerns about the sustainability of fiscal policy. The rising debt and debt service burden further undermined the fiscal situation. By the late 1980s the symptoms of an untenable fiscal policy were evident: higher inflation, rising domestic and foreign interest rates, an increasingly onerous debt service burden, and a shortening of maturities. Economists warned of possible insolvency (Chelliah (1991), Buiter and Patel (1992)).

Against this backdrop, in mid-1991 the government undertook a comprehensive program of macroeconomic stabilization and structural adjustment. A central objective of the reforms is the restoration of balance of payments viability. The authorities also sought to reduce inflation, which was running at over 16 percent, and lay the foundations for more vigorous and sustained growth. The centerpiece of the adjustment program is fiscal consolidation, which entails not only a reduction in the overall deficit, but also a strengthened revenue base and improved spending priorities. Such adjustments are expected to reduce the government’s demand for bank financing, thus freeing resources for private investment.

Fiscal consolidation is seen as a prerequisite for the financial reform measures articulated in the authorities’ latest development plan. These include: a gradual reduction in reserve and liquidity requirements, the creation of a voluntary market for government securities, increased reliance on indirect monetary instruments, increased flexibility of interest rates and reduced cross-subsidization of the priority sectors, enhanced financial competition, and strengthened prudential norms. It is expected that interest rates on government obligations gradually will be brought to market levels 25/, and that spreads on other financial instruments will decline. 26/ In addition to liberalization of the financial sector, the authorities have initiated a process of trade and investment reform and have begun to dismantle the country’s extensive industrial licensing system.

The Indian experience highlights the important links between fiscal sustainability, balance of payments viability and growth. As such, it makes an interesting case study for the framework proposed in this paper. The challenge which confronts the authorities is to identify a fiscal path that is both sustainable and consistent with their output, inflation, and balance of payments targets. As noted in the eighth development plan, economic policy must ensure that public sector investment rests on a sound resource base and that the current account deficit is limited to a level sustainable by normal capital flows. Achievement of that goal will require a substantial further increase in public sector savings.

To evaluate the implications of alternative fiscal strategies for India’s growth, investment and external prospects, we consider a number of alternative scenarios incorporating varying degrees of fiscal and financial reform. 27/ These involve different sets of policy target variables, illustrating the types of simulations that can be performed.

1. A full-adjustment scenario

The baseline, or full-adjustment scenario is one in which the authorities are assumed to set targets for inflation as well as domestic and foreign public borrowing; government spending is thus the residual policy variable. This scenario (Table 3) takes as given the government’s medium-term objectives for growth and inflation. In addition, the authorities have indicated that the current account deficit which can be sustained with normal capital flows is below one percent of GDP.

(in percent of GDP, except as indicated)
Real Output Growth1.
Inflation (period average)13.710.
Public sector debt88.090.593.
Domestic 1/59.860.963.063.962.960.6
Foreign 2/28.229.630.029.228.327.2
Public sector interest payments5.
External current account deficit-0.8-2.1-1.8-0.9-0.7-0.5
Overall fiscal deficit 1/-9.7-10.2-9.8-7.7-6.2-5.0
Domestic savings 3/23.220.621.723.625.026.5
Private sector21.519.420.120.520.821.3
Public sector1.
Domestic investment 3/24.022.723.524.525.627.0
Private sector14.514.715.
Public sector9.
Incremental capital-output ratio18.
Credit to Commercial Sector/GDP29.829.231.433.235.036.9
(in percent of exports of goods and non–factor services)
Total foreign debt service25.526.027.330.828.528.2
of which, public sector17.717.518.319.018.416.9
Interest payments on foreign debt12.
of which, public sector11.711.612.311.210.19.1

Consolidated public sector, excluding central bank.

Public and publicly-guaranteed medium- and long-term debt, excluding military debt.

The 1991/92 error term in the national accounts data is imputed to private consumption.

Consolidated public sector, excluding central bank.

Public and publicly-guaranteed medium- and long-term debt, excluding military debt.

The 1991/92 error term in the national accounts data is imputed to private consumption.

The overall fiscal deficit compatible with low inflation and a prudent debt policy is substantially lower than current levels—somewhere on the order of 5 percent of GDP. Reduction in the deficit would be achieved in part through revenue measures; the baseline scenario assumes a modest growth in public sector revenues of 2 percent of GDP over five years. However, real interest rates on government debt are expected to rise initially to around 5 1/2 percent as a result of financial reform. Non-interest spending would have to decline in order to offset the increase in debt service and to meet the deficit target. Over the longer term, interest spending is projected to fall along with the debt-to-GDP ratio.

The lower deficit could be financed with a combination of foreign borrowing (1/4 percent of GDP, net of reserve accumulation), domestic borrowing (around 4 1/2 percent of GDP in 1996/97, compared with 8 percent in the base year) and seigniorage of around 1 1/4 percent of GDP. Fiscal adjustment along these lines would allow for a gradual reduction of reserve and liquidity requirements. The decline in bank financing to the government would release significant resources for private investment; in the baseline scenario credit to the commercial sector is projected to rise by 7 percent of GDP over five years.

Private savings are projected to increase gradually from 1993/94 onward, after declining initially due to increases in administered prices, higher taxes and reduced transfers. The recovery of private savings, which is tied to the decline in inflation and higher interest rates, is likely to be more than offset by increased private investment. The expansion of private credit, along with a decline in lending rates and a more liberal trade and investment climate, may encourage firms to rehabilitate their capital stock in order to remain competitive. As a result of these reforms, private investment could increase by some 4 percent of GDP over the next four years. Even as production becomes more efficient, a rising share of private investment in GDP and in total investment will probably be needed to meet the authorities’ growth targets. 28/ If so, a greater fiscal correction would be required to reduce the current account deficit to a sustainable level.

Improvement in the current account will depend upon a strong export recovery. The volume of exports is projected to grow by 12 to 14 percent per year, supported by increased competitiveness, further trade reform, and industrial deregulation. Such growth rates are feasible; the collapse of exports to the former Soviet Union has run its course, while export growth to non-FSU countries has averaged 10 percent over the past five years. Imports are expected to increase in response to trade liberalization; import volume could rise by 8 percent annually over the next few years.

2. Financial reform in the absence of fiscal adjustment

The following simulations assess the potential impact on inflation, growth, and the current account of postponing fiscal adjustment while moving ahead with financial reform. These scenarios illustrate the potential risks cited by Sargent and Wallace (1981), Calvo (1988, 1989) and others of an unsustainable fiscal policy.

a. A high-inflation scenario

This scenario assumes that the authorities choose targets for government spending, as well as domestic and foreign borrowing. Seigniorage revenue (inflation) provides whatever residual financing is needed. Suppose that total spending as a share of GDP is maintained at the 1993/94 level throughout the projection period; revenues increase by 2 percent of GDP as before. The targets for domestic and foreign borrowing and reserve accumulation are as in the baseline scenario. Under these assumptions, the overall public sector deficit would decline only marginally to 8 1/2 percent of GDP by 1996/97 (Table 4). To close the financing gap, seigniorage would have to yield some 4 1/2 percent of GDP by 1996/97. If the authorities continue the process of financial reform by lowering reserve requirements, significantly higher inflation would be needed to generate the increased seigniorage. As prices are a determinant of money demand, the monetary base would shrink with progressively higher inflation rates, necessitating ever greater inflation to achieve the same revenue. Not surprisingly, inflation reaches 50 percent by 1996/97 under this scenario. 29/chart 1 depicts different spending paths and their impact on inflation over the medium term. 30/

(in percent of GDP, except as indicated)
Real Output Growth1.
Inflation (period average)13.510.56.518.032.050.0
Public sector debt88.090.593.086.875.060.8
Domestic 1/59.860.963.057.646.733.6
Foreign 2/28.229.630.029.228.327.2
Public sector interest payments5.46.28.612.913.712.1
External current account deficit-0.8-2.1-1.3-2.2-1.40.8
Overall fiscal deficit 1/-9.7-10.2-9.8-9.3-9.0-8.4
Domestic savings 3/23.220.622.220.919.217.6
Private sector21.519.420.619.417.815.8
Public sector1.
Domestic investment 3/24.022.723.523.120.616.8
Private sector14.514.715.214.612.18.3
Public sector9.
Incremental capital-output ratio18.
Credit to Commercial Sector/GDP29.829.231.429.124.017.9
(in percent of exports of goods and non–factor services)
Total foreign debt service25.526.027.230.929.128.8
of which, public sector17.717.518.319.118.517.2
Interest payments on foreign debt12.
of which, public sector11.711.612.311.310.39.4

Consolidated public sector.

Public and publicly-guaranteed medium- and long-term debt, excluding military debt.

The 1991/92 error term in the national accounts data is imputed to private consumption.

Consolidated public sector.

Public and publicly-guaranteed medium- and long-term debt, excluding military debt.

The 1991/92 error term in the national accounts data is imputed to private consumption.

Chart 1.Impact of Government Spending on Inflation

The increase in inflation rapidly erodes domestic (non-indexed) debt, which declines from 60 to 34 percent of GDP over four years. Real interest rates fall relative to the baseline scenario due to the decline in public sector liabilities. 31/ As is well known, the accelerated amortization of debt via the inflation tax appears above the line in conventional fiscal accounts—in the form of higher nominal interest payments. These rise from 5 to 12 percent of GDP over the same period, implying a severe squeeze on non-interest spending. However, in a situation of high inflation it may be more appropriate to consider an operational concept of the deficit. If one moves the portion of nominal interest payments that constitutes real amortization of the public debt below the line, it is evident that real interest payments decline and non-interest spending is maintained.

Even with a higher fiscal deficit relative to the baseline, the current account deficit is reduced under the high-inflation scenario. Net private savings increase despite higher consumption due to the severe contraction of private investment. 32/ The latter is due principally to shrinkage of the monetary base, which sharply reduces bank credit to the commercial sector. 33/ Other factors could contribute to a decline in investment: uncertainty, higher administrative costs, greater exchange rate risk, etc. These would be only partly offset by lower real interest rates.

The projected decline in investment would make it difficult for India to achieve its medium-term growth objectives. Absent dramatic productivity gains—which are unlikely to be realized in a high-inflation context—output growth would probably not exceed 4 percent over the medium term. Given their traditional aversion to inflation, the authorities would, if faced with such a scenario, probably postpone the financial reform and continue financing the fiscal deficit through reserve and liquidity requirements.

b. A high-debt scenario

This scenario assumes instead that the authorities set targets for spending, inflation and foreign borrowing. In this case the fiscal financing gap is closed with additional domestic borrowing (Table 5). As in the high-inflation scenario, spending is held constant as a share of GDP from 1993/94 onward. As a result, domestic debt rises by 7 percent of GDP over the medium term. Since these depend on the public debt burden, real interest rates also rise, reaching 7 percent by 1996/97. Interest payments rise correspondingly. This is a potentially explosive scenario of the type described by Eaton (1987) and Calvo (1988, 1989), where the real interest rate exceeds the growth rate of the economy. 34/

(in percent of GDP, except as indicated)
Real Output Growth1.
Inflation (period average)13.710.
Public sector debt88.090.593.094.795.395.0
Domestic 1/59.860.963.065.567.167.8
Foreign 2/28.229.630.029.228.327.2
Public sector interest payments5.
External current account deficit-0.8-2.1-1.8-2.4-2.9-2.7
Overall fiscal deficit 1/-9.7-10.2-9.8-9.3-8.9-8.4
Domestic savings 3/23.220.621.722.022.423.4
Private sector21.519.420.120.521.021.6
Public sector1.
Domestic investment 3/24.022.723.524.525.326.0
Private sector14.514.715.216.016.817.5
Public sector9.
Incremental capital-output ratio18.
Credit to Commercial Sector/GDP29.829.231.433.034.435.5
(in percent of exports of goods and non–factor services)
Total foreign debt service25.526.027.331.129.529.8
of which, public sector17.717.518.219.018.416.9
Interest payments on foreign debt12.
of which, public sector11.711.612.311.210.19.1

Consolidated public sector.

Public and publicly-guaranteed medium- and long-term debt, excluding military debt.

The 1991/92 error term in the national accounts data is imputed to private consumption.

Consolidated public sector.

Public and publicly-guaranteed medium- and long-term debt, excluding military debt.

The 1991/92 error term in the national accounts data is imputed to private consumption.

Considerable uncertainty surrounds the response of interest rates to financial reform. Nevertheless, Galbis (1993) found that numerous countries experienced protracted periods of high real interest rates when financial liberalization occurred in a weak macroeconomic and regulatory environment. In particular, an inappropriate mix of fiscal and monetary policies contributed to persistently high real interest rates in several countries.

Lower public sector savings relative to the baseline would be only partly offset by higher net private savings—coming almost exclusively from lower investment, rather than consumption. Private capital formation would rise by 3 percent of GDP over the next several years—one percent less than in the baseline scenario—due to higher real interest rates and reduced bank credit to the commercial sector.

Despite lower private investment relative to the baseline, it would be difficult to bring the current account deficit to a level that can be sustained through normal capital flows. Chart 2 illustrates the impact of different debt-financed spending paths on the current account. The higher current account deficit would be financed through private foreign borrowing, averaging 1 3/4 percent of GDP per year through the medium term. This borrowing sustains a high debt service ratio at 30 percent of exports of goods and services (compared to 28 percent in the baseline scenario). It is not clear, however, that adequate financing could be mobilized to support a current account deficit of up to 3 percent of GDP. If not, the authorities might postpone the financial reform and continue to extract high net savings from the private sector.

Chart 2.Impact of Government Spending on Current Account

A scenario of this type—a domestically-financed fiscal deficit of 8 1/2 percent of GDP and no further reductions in reserve and liquidity requirements—would allow only a small increase in bank credit to the private sector. As a result of this and higher real interest rates, private investment would rise by only 2 percent of GDP. Output growth would be correspondingly lower, possibly lower than real interest rates—which could lead to an unsustainable fiscal situation. Chart 3 depicts the impact of a lower SLR on private investment.

Chart 3.Impact of Financial Reform on Private Investment: Reduction of Reserve and Liquidity Requirements

These scenarios demonstrate that the manner in which the external target is achieved is a key determinant of viability. As Easterly (1989) notes, during the 1980s many debtor countries “relied heavily on implicit taxes on financial intermediation to domestically finance their deficits, which explains the [poor] performance of private investment and inflation. The non-crisis countries largely eschewed taxes on financial intermediation for domestic borrowing at market rates, with successful results. The policy conclusions are that larger deficit reductions—preferably implemented through tax reform and reduction of current expenditures—and less distortionary means of financing would lead to improved outcomes in the crisis countries”.

V. Conclusions

A satisfactory assessment of external viability should consider the relationship between fiscal policy, private sector behavior and the balance of payments. Whether or not a country’s external position is viable over the medium-term depends mainly on the sustainability of the public-private sector resource balance. Public sector deficits that absorb a large share of private savings may inhibit investment and growth, and thus an economy’s capacity to generate resources in order to meet its external obligations.

We present a medium-term macroeconomic framework which allows one to model these dynamics. The model, applied to India, highlights the importance of fiscal adjustment in a context of financial reform and limited external financing. A variety of policy scenarios are developed to evaluate the impact of protracted deficits on savings and investment, growth, and external balance. In contrast to other studies, interest rates are made endogenous so as to illustrate the impact on investment and growth of an unsustainable debt strategy. The model goes beyond estimates of steady-state fiscal deficits; it generates annual projections which can be used in program work.

There is a tradeoff between analytical rigor and the usefulness of modelling exercises for policy work. In an attempt to balance these concerns, this paper suggests a simple yet comprehensive framework to study the impact of fiscal policy on macroeconomic performance. It is not intended as a precise forecasting tool, but as a practical means of evaluating alternative fiscal strategies. The model focuses on the key relationships between fiscal policy, private savings and investment, and the balance of payments. It is sufficiently flexible to accommodate a variety of country circumstances and policy scenarios.

One should bear in mind that this approach does not focus on the composition of the fiscal adjustment required to ensure consistency with the macroeconomic targets. Recall that the analysis of fiscal policy is conducted entirely from “below the line.” Proposed spending cuts and revenue increases should be evaluated both in terms of their sustainability and efficiency. To properly assess overall economic performance, the simulations should be accompanied by sectoral analyses of fiscal, monetary and external policies.


    AnandR. and S.Van Wijnbergen.1989. “Inflation and the Financing of Government Expenditure: An Introductory Analysis with an Application to Turkey.”World Bank Economic Review. Vol. 3No. 1.

    • Search Google Scholar
    • Export Citation

    AnandR. and S.Van Wijnbergen.1988. “Inflation, External Debt and Financial Sector Reform: A Quantitative Approach to Consistent Fiscal Policy With an Application to Turkey.”NBER Working Paper No. 2731.

    • Search Google Scholar
    • Export Citation

    AnandR.A.Chhibber and S.Van Wijnbergen.1991. “External Balance and Growth in Turkey: Can They Be Reconciled” in A.T.Aricanli and D.Rodrikeds.Turkey in the Next DecadeCambridge: Mass.: Harvard University Press.

    • Search Google Scholar
    • Export Citation

    BarrionuevoJ.1992“Inflation and Fiscal Deficits: The Irrelevance of Debt and Money Financing”IMF Working Paper WP/92/102.

    BarroR.“Are Government Bonds Net Wealth?”Journal of Political EconomyNovember1974.

    BhandariJ.N.Ul Haque. and S.Turnovsky.1989. “Growth, Debt and Sovereign Risk in a Small, Open Economy”World Bank PPR Working Paper No. WPS 260.

    • Search Google Scholar
    • Export Citation

    BlejerM. and M.Khan.1984“Government Policy and Private Investment in Developing Countries”IMF Staff PapersVol. 311984.

    BlejerM. and A.Cheasty.1990“Analytical and Methodological Issues in the Measurement of Fiscal Deficits”IMF Working Paper WP/90/105.

    • Search Google Scholar
    • Export Citation

    BuiterW.1983“Measurement of the Public Sector Deficit and its Implications for Policy Evaluation and Design”IMF Staff PapersVol. 30No. 2.

    • Search Google Scholar
    • Export Citation

    BuiterW.1985“A Guide to Public Sector Debt and Deficits”Economic PolicyVol. 1No. 1.

    BuiterW. and U.Patel.1992“Debt, Deficits and Inflation: an Application to the Public Finances of India”Journal of Public EconomicsVol. 47.

    • Search Google Scholar
    • Export Citation

    CaganP.1956. “The Monetary Dynamics of Hyperinflation” in Studies in the Quantity Theory of Moneyed. byMiltonFriedman.University of Chicago Press.

    • Search Google Scholar
    • Export Citation

    CalvoG.1988“Servicing the Public Debt: The Role of Expectations”American Economic ReviewVol. 78No. 4.

    CalvoG..1989. “Controlling Inflation: The Problem of Nonindexed Debt” in EdwardsS. and F.LarrainDebt. Adjustment and Recovery: Latin America’s Prospects for Growth and DevelopmentLondon: Basil Blackwell.

    • Search Google Scholar
    • Export Citation

    ChelliahRaja.1991. “The Growth of Indian Public Debt—Dimensions of the Problem and Corrective Measures”IMF Working Paper WP/91/72.

    • Search Google Scholar
    • Export Citation

    ClaessensS.D.Oks and S.van Wijnbergen.“Interest Rates, Growth and External Debt: The Macroeconomic Impact of Mexico’s Brady Deal”Unpublished PaperWorld BankAugust1992.

    • Search Google Scholar
    • Export Citation

    CohenD.1985“How to Evaluate the Solvency of an Indebted Nation”Economic Policy.

    CohenD.1988. “The Management of the Developing Countries’ Debt: Guidelines and Applications to Brazil”World Bank Economic ReviewVol. 2No. 1.

    • Search Google Scholar
    • Export Citation al. 1990. Debt Reduction and Economic Activity IMF Occasional Paper No. 68.

    EasterlyW.1989a. “Fiscal Adjustment and Deficit Financing During the Debt Crisis.”World Bank PPR Working Paper No. WPS 138.

    EasterlyW..1989b. “How Much Fiscal Adjustment is Enough? The Case of Colombia.”World Bank PPR Working Paper No. WPS 201.

    EatonJonathan.1987. “Public Debt Guarantees and Private Capital Flight.”NBER Working Paper No. 2172.

    GalbisV.1993. “High Real Interest Rates Under Financial Liberalization: Is There a Problem”. IMF Working Paper WP/93/7.

    Gil DíazJ. and R.Ramos.1987. “Lessons from Mexico.”Unpublished PaperBanco de MexicoDirección de Investigación Económica.

    GreeneJ. and D.Villanueva.1990. “Private Investment in Developing Countries: An Empirical Analysis.”IMF Working Paper WP/90/40.

    GuidottiP. and M.Kumar.1991. Domestic Public Debt of Externally Indebted Countries. IMF Occasional Paper No. 80.

    GuitiánM.1981. Fund Conditionality: Evolution of Principles and Practices IMF Pamphlet Series No. 38 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    HorneJ.1988. “Criteria of External Sustainability.”IMF Working Paper No. WP/88/60.

    HorneJ.1991. “Indicators of Fiscal Sustainability.”IMF Working Paper WP/91/5.

    KrishnaswamyK.K.Krishnamurty and P.D.Sharma.1987. Improving Domestic Resource Mobilization Through Financial Development: India. Manila: Asian Development Bank.

    • Export Citation

    MassonP.1985. “The Sustainability of Fiscal Deficits.”IMF Staff Papers. Vol. 21.

    MillerM.R.Skidelsky and P.Weller.1989. “Fiscal Deficits: Sustainability and Default Premia.”CEPR Conference on Capital Markets and Debt Management. Castelgandolfo, Italy.

    • Search Google Scholar
    • Export Citation

    MontielP.1993. “A Macroeconomic Simulation Model for India.”Unpublished PaperOberlin College.

    Pereira LeiteS.1992. “Coordinating Public Debt and Monetary Management During Financial Reforms.”IMF Working Paper WP/92/84.

    PradhanB.K.D.K.Ratha and A.Sarma.1990. “Complementarity Between Public and Private Investment in India.”Journal of Development Economics. Vol. 33.

    • Search Google Scholar
    • Export Citation

    SakrK.1993. “Determinants of Private Investment in Pakistan.”IMF Working Paper WP/93/30.

    SargentT. and N.Wallace.1981. “Some Unpleasant Monetarist Arithmetic.”Federal Reserve Bank of Minneapolis Quarterly Review.

    SharmaK.C.1991“Investment and Financial Behavior of Public Limited Companies.”RBI Occasional PapersVol. 12No. 2.

    SpaventaL.1986. “The Growth of Public Debt: Sustainability, Fiscal Rules, and Monetary Rules.”IMF Working Paper WP/86/8.

    SundararajanV. and S.Thakur.1980. “Public Investment, Crowding Out and Growth: A Dynamic Model Applied to India and Korea.”IMF Staff PapersVol. 27No. 4.

    • Search Google Scholar
    • Export Citation

    TanziV.M.Blejer and M.Teijeiro.1987. “Inflation and the Measurement of Fiscal Deficits.”IMF Staff PapersVol 34No. 4.

    Ul HaqueN. and P.Montiel.1992“Fiscal Policy in Pakistan Since 1970.”IMF Working Paper WP/92/97.

    Van WijnbergenS.R.Rocha and R.Anand.1989. “Inflation, External Debt and Financial Sector Reform: A Quantitative Approach to Consistent Fiscal Policy.”World Bank PPR Working Paper No. WPS 261.

    • Search Google Scholar
    • Export Citation

    Van WijnbergenS.1990. “External Debt, Inflation, and the Public Sector: Toward Fiscal Policy for Sustainable Growth.”World Bank Economic ReviewVol. 3No. 3.

    • Search Google Scholar
    • Export Citation
APPENDIX I Derivation of The Intertemporal Budget Constraint 35/

Equation 1 is the budget identity of the non-financial public sector, including state-owned enterprises. G is non-interest government spending, i and i* are domestic and foreign nominal interest rates, T are revenues, DCg is domestic credit to the government from the central bank, and B and B* are the domestic and foreign debt stocks. The “.” over variables indicate changes, and the “∧” rates of change. The primary deficit, D is non-interest spending minus revenues.

The consolidated public sector balance is obtained by adding net profits of the central bank to the non-financial public sector deficit. The central bank’s interest earnings on net foreign assets equal the gain in its net worth: i*NFAg*E=NW. Adding this term to both sides of (1) yields:

From the central bank balance sheet, M=DCg+NFAg*ENW, so

If there is net central bank lending to the private sector, an adjusted monetary base (M - DCp) is used; in this case MDCp=DCg+NFAg*ENW, where DCp is central bank credit to the private sector.

We deflate equation (3) by dividing both sides by the domestic price level; real variables are denoted in lower-case letters. By definition, the real exchange rate, e = P*/P. To derive an operational concept of the deficit, one subtracts that portion of nominal interest payments that constitutes real amortization (via the inflation tax) of the domestic and foreign debt: PbgandP*(bg*nfag*). Employing the Fisher relation and subtracting these terms from both sides of equation (3) yields:

By differentiation,


Substituting equations (5) and (6) and (7) into (4) and rearranging yields:

APPENDIX II Behavioral Equations
  • 1. Money Demand:
    • a. Currency:log (CU/P)= α + βlog(Y/P) + ψlog(1+itd) + δlog(1+P) + θlog(CU/P)−1
    • b. Demand Deposits:log(DD/P) = α + βlog(Y/P) + ψlog(1+itd) + δlog(1+P) + θlog(DD/P)−1
    • c. Time Deposits:log(TD/P) = α + βlog(Y/P) + ψlog(1+itd) + δlog(1+P) + θlog(TD/P)−1
  • 2. Private Consumption:log(C/P) = α + βlog(1+rdep) + ψ(1+P) + δlog(Yp) + Φ(Y/Yp)+Dum1
  • 3. Private Investment:log(V/P) = α + βlog(1+r1nd) + ψ(1+P) + δlog(DCp/Y) + Φlog(V/P)−1
  • 4. Real Interest Rates:log(1+rgov) = α + βlog(1+r∼) + ψlog((B+(B*−NFA*)))/Y + δ(Δe/e) + Dum2

Note: Some of the equations contain parameter values which are zero.

  • CU = currency in hands of the private sector
  • DD = demand deposits in the banking system
  • TD = time deposits in the banking system
  • C = private sector consumption
  • V = private sector investment
  • DCp = domestic credit to the private sector
  • P = expected inflation
  • itd = nominal interest rate on time deposits
  • rdep = average real deposit rate (including govt. bonds)
  • rind = average real lending rate
  • rgov = real interest rate on government bonds
  • Y = Nominal GDP
  • Yp = Permanent nominal output
  • B = domestic debt of public sector
  • B* = foreign debt of public sector
  • NFA* = net foreign assets of central bank
  • Δe/e = expected real currency depreciation

Ms. Parker is an Economist in the Central Asia Department. The paper was prepared while Mr. Kastner was an intern in the Policy Development and Review Department. He is a doctoral candidate at the University of Mainz, Germany. The authors gratefully acknowledge helpful comments from Jack Boorman, Maria Carkovic, Charles Collyns, Mohamed El-Erian, David Goldsbrough, Mohsin Khan, Paulo Leme, Ranjit Teja, Sweder van Wijnbergen, and Roberto Zagha.


The experience of Mexico during the mid-1980s illustrates this point. Viewed in isolation, Mexico’s balance of payments appeared viable. The country was generating current account surpluses on the order of two percent of GDP, despite a large terms of trade shock. By 1987 gross reserves had reached 12 months of merchandise imports. Mexico was expected to graduate from rescheduling and concerted lending. Yet the magnitude of the current account surplus needed to meet the country’s external obligations was not compatible with low inflation and renewed growth. An unsustainable fiscal situation gave rise to increasing inflation, higher real interest rates, and declining secondary market prices of foreign debt. The Mexican example shows that the apparent absence of financing gaps is insufficient to ensure external viability.


An analogous condition for the infinite-horizon case is that the growth rate of public debt not exceed the real interest rate. See Buiter and Patel (1992) for an elaboration of this point.


Buiter cites an attempt by Hills (1984) to develop an intertemporal budget identity for the U.K. Guidotti and Kumar (1991) made illustrative calculations of net worth for four hypothetical debtor countries.


For example, the government could pursue an expansionary fiscal policy financed with foreign borrowing. At some point it may no longer be possible to extract sufficient resources from the private sector to service the foreign debt. The result could be a decline in secondary market prices followed eventually by debt reduction or repudiation. Either way, the prospective liabilities of the public sector would be brought into line with its assets, thus ensuring that the intertemporal budget identity is satisfied. However, this would not be a sustainable fiscal strategy.


Even if such a critical value could be computed, it would not yield the preferred debt strategy; to derive an optimal public debt path, one would have to maximize some intertemporal social welfare function subject to the solvency constraint.


However, higher real interest rates on government debt also increase the income of private bondholders, raising additional funds for investment.


Solvency is a weaker criterion since it requires only that the debtor be able, not willing, to repay its debt in full.


Anand and van Wijnbergen (1988) apply the Cohen index to Turkey.


Revenues should exclude dividends from the central bank as well as temporary revenue from asset sales. See Appendix 1 for a derivation of equations 2 and 3.


If there is central bank lending to the private sector, an adjusted concept of the monetary base must be used.


The operational deficit is the primary deficit plus real interest payments on the public debt.


The domestic debt bg is net of private sector obligations to the government. For the sake of simplicity, we assume that the real interest rate on foreign debt equals that earned on the net foreign assets.


On the public finance approach to inflation, see the classic paper by Sargent and Wallace (1982).


The behavioral equations are given in Appendix II.


The model assumes a simple linear relationship between the default premium and the debt-to-GDP ratio. Alternatively, the risk premium could depend on the rate of growth of the debt stock or the variance of debt service obligations, as in Claessens, Oks and van Wijnbergen (1992).


Assuming the Fisher relation holds, the real interest rate differential between the home and foreign country can be decomposed into:

(rr*) = (ii* − fd) + (fdEΔE) + (EΔE − π + π*)

where fd is the forward discount, EΔE is expected depreciation and π and π* are home and foreign inflation rates. The first term on the right hand side is the covered interest differential, a measure of default risk (assuming zero transactions costs). The second is an exchange risk premium (assumed to be zero) and the third a real depreciation premium. The depreciation premium is based upon an assumed path for the real exchange rate.


There are three sets of policy variables: target and residual variables, which differ across the three scenarios, and predetermined variables, which are the same for all three scenarios.


Government foreign borrowing may be constrained by investors’ willingness to hold sovereign debt.


This assumption is appropriate for most countries whose private sectors are relatively well developed, with established export markets and overseas financial connections. In some countries that experienced debt-servicing difficulties in the 1980s, private borrowers were denied access to foreign capital markets due to the risks associated with heavy public indebtedness. However, this exclusion generally proved temporary; firms regained access when the authorities pursued credible adjustment programs and made foreign exchange available to service private debt.


A copy of the file may be obtained from Karen Parker, Central Asia Department, Room 4-528.


Yet if the reforms are fundamental, the Lucas critique still applies.


For a description of India’s financial system and its distortions, see the 1991 Recent Economic Developments report, SM/92/205.


The authorities do not publish detailed data on the domestic debt of states and public corporations; this was estimated using the Government of India’s Public Finance Statistics. 1991 and the consolidated balance sheet of public enterprises at March 31, 1992 from the Government’s Public Enterprises Survey.


For the purpose of the simulations, the adjustment of interest rates on the domestic debt is assumed to occur in a single step, As SLR-eligible securities mature and are replaced with non-SLR debt at market rates, banks holding older securities may expect to be compensated through higher interest rates or a capital injection; from a fiscal point of view the two are equivalent.


Until recently interest rates were set administratively; it is thus not possible to estimate an interest rate equation for India. For the purpose of the simulations, the real interest rate on public debt during the first year of the financial reform is assumed to equal the government’s current marginal cost of funds (about 5 percent). In subsequent years, real rates move with expected depreciation and the endogenous default premium.


These scenarios are presented for illustrative purposes only. They do not necessarily reflect the priorities of the Indian authorities or the views of the Fund.


This scenario implies an incremental capital output ratio of around 4.2, in line with the authorities’ plan projections.


The inflationary impact of fiscal deficits would be even greater were India a “monetarist economy” of the sort described by Sargent and Wallace (1981), in which real interest rates exceed the growth rate and expected, rather than actual, inflation is an argument in money demand functions.


Were there no reduction in reserve requirements, inflation would still exceed 30 percent by 1996/97 under this scenario.


The decline in real interest rates is unlikely to be maintained in a dynamic setting, as debtholders would demand higher premia to compensate them for losses on their non-indexed debt. On the other hand, the Fisher relation would probably be violated in a situation of high and variable inflation. A more complex specification of the default premium may be appropriate for high-inflation countries.


A high-inflation dummy variable is included in the consumption function.


Private investment would probably decline less than predicted, as firms would seek financing from the non-bank capital market, including inter-enterprise credit. However, the magnitude of such borrowing is difficult to forecast since, historically, inflation has been low and large firms have relied mainly on bank credit to finance their investments.


The criterion that real interest rates not exceed the growth rate of the economy is perhaps too generous; a more stringent condition is that the interest the government pays on its borrowing not be higher than the earnings on its investments.

Other Resources Citing This Publication