Fiscal Policy, Growth, and Design of Stabilization Programs
- Ana María Martirena-Mantel
- Published Date:
- January 1987
The objectives of Fund-supported stabilization programs include a balance of payments viable over the medium run, the promotion of growth under a stable economic environment, price stability, and the prevention of excessive growth in external debt. These objectives do not have the same weight, but each is important in stabilization programs. A narrow interpretation of the Fund’s role would emphasize the balance of payments objective and de-emphasize the others.
This paper deals with the role of fiscal policy in stabilization programs, emphasizing the structural aspects of fiscal policies since, over the years, these aspects have attracted less attention than has demand management. The Baker initiative of October 1985 called attention to the importance of these structural aspects. The paper does not discuss other elements of program design, such as incentive measures implemented through the exchange rate, through import liberalization, through financial deregulation, or through pricing policy, even though these structural elements are obviously important. In countries where institutions necessary for the effective use of other policies are not adequately developed, fiscal policy may be the main avenue to economic development and stability, although, unfortunately, political pressures, external shocks, and administrative shortcomings have frequently weakened government control over this instrument. Tax evasion, inflation, and the proliferation of exonerations have reduced the government’s ability to control tax revenues, while political pressures, fragmentation of the public sector, and inadequate monitoring systems have undermined its ability to keep public: expenditure in cheek. Far from being the stabilizing factor in the economy that it should be, fiscal policy has itself, in too many instances, become a major destabilizing force contributing to disequilibrium in the external sector.1
In recent years the connection between fiscal developments and external sector developments has been increasingly recognized. Some have gone as far as to suggest a “fiscal approach to the balance of payments” that considers fiscal disequilibrium as the main cause of external imbalances.2
Although growth was always a primary objective of economic policy, the sustained rates of growth experienced by most countries until the mid-1970s (except for occasional and transitory periods of balance of payments difficulties) made it possible for the Fund, in negotiating stabilization programs, to concentrate on the objective of stabilization in which it had more expertise and an accepted mandate. The increase in oil prices during the 1970s and especially the more recent debt crisis accompanied by the sharp fall in commodity prices brought about a new environment in which external sector disequilibrium could not be easily financed. This forced many countries to pursue (over longer periods than had earlier been the case) stabilization policies aimed at reducing external imbalances or the rate of inflation, policies that some critics considered as inimical to growth.
In the face of external shocks, some countries (for example, the Republic of Korea) succeeded in stabilizing their economies and in advancing once again along the road of economic development. Others were less successful. When the need to pursue stabilization policies extended over several years, the short-run political costs of these policies began to loom larger than the longer-run economic benefits; political fatigue set in and some countries became restive under the harness of traditional stabilization programs. The cries of critics that stabilization policies were inhibiting growth became louder and attracted a larger following. Critics advised policymakers to abandon stabilization policies recommended by the Fund and to concentrate on growth, regardless of the consequences for the balance of payments and the rate of inflation. They espoused the position that inflation is a lesser evil than stagnation and that the external sector can be kept in equilibrium through quantitative restrictions and export subsidies, or by repudiating external debt obligations.
As already mentioned, stabilization and growth have always been legitimate policy objectives. Although in the past it was thought that at any given moment a country could focus on policies aimed specifically at one or the other of these objectives, the view that it is unwise to separate these objectives currently predominates. Stabilization programs must pay attention to growth to ensure that stability is not won at the price of stagnation.3 Growth policy must pay attention to stability to ensure that the pursuit of growth is not aborted by excessive inflation or by pressures on the external sector, as has happened in several cases in recent years. Growth without stability may be technically impossible over the longer run; stability without growth may be politically impossible except in the short run. This paper attempts to reformulate the fiscal design of stabilization programs in order to emphasize the growth objective.
If stabilization were the only objective of economic policy, stabilization programs could rely mostly on traditional demand-management policies.4 Stabilization with growth, however, requires that demand-management policies be complemented by policies aimed at increasing potential output. Misguided structural policies have reduced potential output by misallocating resources and by reducing the rate of growth of the factors of production. They have thus been the main cause of stagnation and a contributor to economic instability. The design of adjustment programs should integrate stabilization with growth, or demand-management policies with structural, supply-side policies.
Fiscal Policy and the Design of Fund Programs
Stabilization programs can, in theory, emphasize either specific or general fiscal policies. For example, the country and the Fund could agree on a whole range of specific fiscal measures, such as changes in various taxes and tax rates and changes in specific public expenditures, subsidies, and public utility rates. These measures, however, would have to add up to the required adjustment in aggregate demand and supply. They must reduce the balance of payments disequilibrium and the rate of inflation to the desired level by reducing aggregate demand and by increasing aggregate supply. For identification I shall call this the microeconomic approach to stabilization programs, an approach that explicitly recognizes both the demand-management and the supply-management aspects of fiscal policy. It recognizes that fiscal policy changes usually affect not only aggregate demand but also aggregate supply.5
Alternatively, the country and the Fund could limit their agreement on a program to general, macroeconomic variables. In the extreme version of this alternative, the Fund and the country might not even discuss specific fiscal policies, but would limit not only their agreement but also their discussions to the size of the fiscal deficit and to the expansion of bank credit associated with that deficit. If specific policies are discussed, it would be to assess their immediate impact on the size of the fiscal deficit and on aggregate demand.
In this approach, supply-side aspects of fiscal policy (what I have called the supply-management aspects) would be largely ignored. I shall call this the macroeconomic approach to stabilization policy. This approach implies that once the size of the deficit has been determined, the balance of payments consequences of that deficit have also been determined regardless of the specific measures that the country will employ to achieve the stipulated level of fiscal deficit.6 Whether the deficit is reduced by raising taxes or by cutting spending, and regardless of the specific tax and spending measures used to achieve such a reduction, the balance of payments consequences are assumed to be the same.7
Although these alternative versions of the design of stabilization programs have probably never been pursued in their pure form, over the years the formulation of stabilization programs has been much closer to the macroeconomic than the microeconomic alternative,8 in conformity with the common interpretation of the guidelines on conditionality.9 Until recent years, stabilization programs established fiscal ceilings on the basis of an implicit model that connected monetary expansion associated with the fiscal deficit to developments in the balance of payments. The countries themselves would then choose the specific ways in which the fiscal ceilings would be observed. Which tax rates should be changed, which new revenue measures should be adopted, and which expenditures should be reduced (or expanded) were left to the authorities to determine, although Fund missions did provide some advice based, where possible, on technical assistance reports. As Sir Joseph Gold put it” … performance criteria… must be confined to macroeconomic variables… . The concept of ‘macroeconomic’ variables involves the idea of aggregation… . [and] includes the broadest possible aggregate in an economic category… .” Gold goes on to state that “… the Fund should not become involved in the detailed decisions by which general policies are put into operation…” He concludes that “specific prices of commodities or services, specific taxes, or other detailed measures to increase revenues or to reduce expenditures would not be considered macroeconomic variables” (Gold, 1979, pp. 32–33).
Specific measures (such as the elimination of subsidies) were on rare occasions made performance criteria in Fund programs, but the main reason for doing so was often deficit reduction and thus demand management.10 Fiscal changes without direct and immediate bearing on the size of the fiscal deficit (say, revenue-neutral tax reforms) did not receive explicit attention in formal agreements, even though they might have a bearing on the efficiency of the economy. Changes that would increase the fiscal deficit in the short run but would have desirable supply-side effects on the economy over the medium run were not encouraged. The observance of the fiscal ceilings was the most essential fiscal element of a program.
If the country wanted advice on its tax structure, on the structure of its public spending, or on their respective administration, it could request technical assistance from the Fund. No conditionality was attached to the provision or the use of this advice, although Fund missions occasionally used technical assistance reports to provide advice to the countries, especially on how to raise revenues.11 Technical assistance has been the major channel through which the Fund has directly influenced the structure of tax systems and their administration and, to a lesser extent, the structure of public spending.
With important qualifications, this macroeconomic approach to stabilization programs predominated until a few years ago. Starting with extended Fund facility programs, however, Fund missions began paying more attention to structural aspects in general and specific fiscal aspects in particular,12 and today much more attention is paid to structural (supply-side) elements in stabilization programs. The transition from the macroeconomic to the microeconomic approach is, however, far from complete. The approach followed in negotiating stabilization programs begins with an estimation of the required reduction in a country’s fiscal deficit, given its balance of payments position and the foreign financing presumed to be available, and proceeds, separately and often ex post, to a discussion of specific policies.13 The connection that is likely to exist, especially over the medium run, between the “required” deficit reduction and the specific measures adopted to make that reduction possible is not accounted for in setting program ceilings. For example, the removal of growth-retarding taxes is not encouraged if alternative revenue sources are not immediately available, as such a removal will immediately increase the fiscal deficit and, given the underlying model used, will presumably lead to a deterioration in the country’s external position. Thus, the approach still goes from the macroeconomic to the microeconomic and much attention is focused on the size of the deficit and on its financing.
Nevertheless, recent Fund programs have increasingly recognized that the specific measures through which fiscal deficits are reduced may determine, especially over the medium and longer run, whether a stabilization program will have durable, beneficial effects on the balance of payments and on growth, or whether these effects will vanish as soon as the program is over. An adequate macroeconomic framework (consistent with a viable balance of payments and with price stability in the short run) is a necessary, but not a sufficient, condition for growth and for stability over the longer run. Stability requires in addition efficient structural policies.
Should the Fund and the authorities focus mainly on macro-economic fiscal variables, as has traditionally been the case? Or should they make specific fiscal policies of equal importance in a program? Putting it more starkly, should the Fund be prepared to walk away from an arrangement with a country in which resources have been badly misallocated, thus reducing its growth potential, if an acceptable core of structural policies is missing even though the traditional macroeconomic framework appears adequate? Should Fund missions start the analysis of a program by identifying such a structural core of required policies—that is, a set of specific supply-side measures—that must be implemented over the course of the program before the macroeconomic ceilings are set? 14 The answers to these questions are not as obvious as they might appear at first, as convincing arguments can be presented on both sides.
A first argument in favor of continuing with the traditional, macro-economic approach is that, at least in theory, this approach is objective. Whether or not performance criteria are satisfied is an issue subject, in most cases, to quantification and verification and thus beyond dispute.15 As such, this approach reduces the uncertainty faced by authorities. They know that if the country satisfies the performance criteria it will obtain from the Fund the agreed financial support. And, once again, those performance criteria normally relate to macroeconomic variables.
A second, and perhaps more important, argument is that performance criteria based on ceilings imply less political interference by the Fund in the internal affairs of countries than do criteria related to specific measures. Authorities are likely to object to having to agree to modify a tax in a given way or to modify the level or pattern of public spending.16 Critics who find present Fund conditionality too rigid are likely to object even more to what might be seen as an extension of that conditionality. Examples of this reaction exist in connection with Fund recommendations to eliminate or reduce subsidies. Many observers feel that these are political decisions that should be left to the authorities and that the Fund should at best offer only an opinion on them.
A third argument in favor of the traditional approach is that discussions about fiscal ceilings, as well as the review of the outcome of these discussions at headquarters, require fewer and less specialized staff resources than do discussions of specific measures. For an institution concerned about its own budget, this is an important consideration. The design of a program can be based on a relatively straight-forward view of the relationship between fiscal deficits and balance of payments. Once some assumptions are made, it is far easier to decide what the size of a fiscal ceiling should be than to decide the details of specific policy changes and how these changes influence program objectives.
A fourth argument, closely related to the preceding one, is that, at least in the fiscal area, it is far easier to write a letter of intent in which a country’s formal commitments are couched in the form of general ceilings than to write documents that spell out formal commitments in terms of many specific policy changes. It is always difficult, for example, to specify the precise requirements of a tax reform.
There are, however, arguments that caution against exclusive or excessive emphasis on traditional performance criteria that emphasize fiscal ceilings. They favor paying close attention to the microeconomic aspects of fiscal policy, such as the structure of individual taxes, the structure of expenditure, the allocation of investment, the prices charged by public utilities, and public employment. To avoid any misunderstanding on this issue I should emphasize here that the questions raised below about fiscal ceilings should not be interpreted as supporting Fund critics of conditionality. They simply call attention to the arguments (a) that a good stabilization program must not rely exclusively on demand management and (b) that the ceilings used to serve demand management should not be set independently from the structural changes that the country is willing to make. The main justification for this change of emphasis is that, provided the supply-response is not insignificant and occurs fairly rapidly, the more far-reaching the structural reform agreed to by the country, the greater will be that supply response (in terms of output, exports, capital repatriation, and the like). Such a supply response may imply that a less stringent demand-management policy may be necessary.
Problems have at times been encountered when ceilings have been imposed on macroeconomic variables. These problems are mentioned to indicate that a program that relies exclusively on performance criteria related to macroeconomic variables may not be capable of providing the hoped-for results. First, the longer ceilings on macro-economic variables are in use, the more ways countries learn to get around them. Ceilings are most useful when a country complies not just with the letter of an agreement but also with its spirit. Unfortunately, there have been instances in which countries have complied with the letter and defied the spirit of an agreement. They have engaged in operations aimed at circumventing the ceilings in order to draw resources from the Fund without making genuine adjustments. To deal with this problem, the Fund has been compelled, in some programs, to increase the number of performance clauses related to the fiscal deficit. This has created a perception of excessive conditionality.
Second, the usual formulation of a stabilization program may give the impression that the relationship between fiscal deficits and program objectives, and especially their relationship with the balance of payments, is clearcut and unambiguous. In other words, it may give the impression of a single-valued functional relationship—that is, so much fiscal deficit implies so much deficit in the current account of the balance of payments. Unfortunately, our knowledge about important economic relationships (such as that between changes in the money supply and changes in prices, and that between changes in prices, changes in nominal exchange rates, and their effects on the balance of payments) is too limited to inspire excessive confidence about the precise level of the fiscal deficit required to achieve a given change in the current account of the balance of payments or in other economic objectives. The truth is that a given fiscal deficit may be associated with a range of balance of payments outcomes.17
Third, the ceilings may, in some cases, divert attention away from the basic objectives of economic policy. Meeting the ceilings may come to be seen, within the program period, as an end in itself. During this period, programs may be judged successful or not depending on whether ceilings are being met rather than on whether the ultimate objectives of the program (durable improvement in the balance of payments, growth, price stability, and so forth) are being achieved.
Finally, and most important, excessive reliance on macroeconomic ceilings may divert attention away from the quality as well as the durability of the specific measures used by a country to comply with its performance clauses. Let me give some examples, starting with the question of the durability of the fiscal measure. The question to be raised is: Will a fiscal measure have a permanent impact on the fiscal deficit? Is, for example, a revenue increase or an expenditure cut of such a nature as to affect the deficit for years to come, or is it of a once-and-for-all type? This is an important question if the program’s objective is, as it should be, a permanent improvement in the economy.
Sometimes tax payments by enterprises have been advanced at the request of the government, 18 or public expenditures have been postponed (through the building up of arrears or through the postponement of inevitable expenditures)19 so that the country can meet the fiscal ceilings and can, thus, make the next drawing. At other times, temporary sources of revenue (once-and-for-all taxes, temporary surtaxes, tax amnesties, sales of public assets, and so forth) have allowed the country to stay within the agreed ceiling without doing anything to reduce its underlying or core fiscal deficit.20 At times governments have used up so much of their political capital in introducing these temporary measures that they no longer have the stamina to make the permanent and growth-promoting policy changes required to achieve durable adjustment with growth.
In addition to the question of the durability of the fiscal measures (will their effects survive the program?), there is the important question of the quality (or, if one wishes, of the economic efficiency) of those measures. As far as short-term demand-management policy is concerned, whether a country reduces the fiscal deficit by raising revenue or by cutting expenditure is inconsequential.21 It is also inconsequential whether it does it through the use of measures that have disincentive effects or through measures that do not have such effects. The stabilization program will fail if the ceiling is not observed; it will not fail if it is observed through growth-retarding measures.
The above discussion should not be interpreted as arguing that stabilization programs should no longer rely on demand management based on a macroeconomic framework that sets ceilings on relevant macroeconomic variables. In my view, the need for such a framework is too obvious to require justification. The discussion simply argues that this framework needs to be supplemented by measures aimed at ensuring that stabilization programs are, first, durable and, second, as growth-promoting as possible. Under present guidelines on conditionality, under which the Fund staff operates, the change advocated in this paper might not be possible. A decision by the Executive Board of the Fund states that “Performance criteria will normally be confined to (i) macroeconomic variables, and (ii) those necessary to implement specific provisions of the Articles or policies adopted under them. Performance criteria may relate to other variables only in exceptional cases…” (International Monetary Fund, 1986, pp. 27–28).
Stabilization Policy and Economic Growth
Growth-promoting stabilization policy requires that the reduction in the fiscal deficit be carried out through fiscal measures that are (a) durable in their effects and (b) efficient in their impact. In other words, the policies chosen must not self-destruct once the program is over and must achieve their deficit-reducing objective with the least possible inhibition of economic growth.
The efficiency of fiscal instruments is important for growth, as much recent work on this issue has demonstrated. Work effort, exports, productive investment, saving, capital flight, foreign investment, and so on, can be affected by the choice of specific fiscal instruments.22 These choices may play a large role in determining the amount of foreign resources a country will have available during and after the program period. Thus, the relationship between changes in the size of fiscal deficits and changes in the ultimate objectives of economic policy, such as growth and stability, is inevitably influenced by the fiscal policy measures utilized. It can make a substantial difference to the growth prospects of a country if the fiscal deficit is reduced by eliminating a totally unproductive expenditure or by raising a tax that has strong disincentive effects, even though in terms of traditional stabilization policy (in terms of short-run fiscal deficit reduction) the result would appear to be the same. The more efficient the measures used to achieve a given deficit reduction, the greater will be the rate of growth, and, assuming an unchanged monetary policy, the lower will be the rate of inflation.
The implication of the above conclusion for stabilization programs is obvious: provided that a country is willing to implement a considerable number of structural measures early enough in a program so that the positive effects of these measures can be fell relatively soon, the Fund should be prepared to require less reduction in the overall fiscal deficit (i.e., to require less austerity) than it would if the structural package were less far reaching or if the country delayed its introduction. Thus, the Fund should explicitly recognize, at the time it enters into an agreement with a country, a trade-off between quantity and quality of fiscal adjustment, a trade-off that would also be influenced by the timing of the introduction of the structural measures. This trade-off should be recognized and, possibly, formalized in program design and negotiations.23
This is not the place to discuss in detail the quality of the fiscal measures that could form the structural core of a stabilization program, but a few examples may help convey the importance of this issue. Suppose that an agricultural commodity of wide consumption (say, wheat, corn, or rice) has been subject to an export tax in a country negotiating a Fund program. The elimination of this tax would reduce tax revenue and thus raise the fiscal deficit. This, in turn, would have monetary and consequently balance of payments implications, which the macroeconomic framework of Fund programs would assess. But let us consider whether there are countervailing supply-side effects. The removal of the tax would raise the domestic price of the commodity and lead to a reduction in domestic consumption, thus making some additional supply available for exports.24 In addition, the removal of the export tax would encourage producers to produce more of that product. When this additional production becomes available, exports will increase further. Since the availability of foreign exchange is always a key factor in a stabilization program, focusing only on the demand effect (through the increase of the fiscal deficit) that the elimination of the tax will have, and ignoring the supply effect (through the incentive to produce and export more), is likely to introduce a bias against the elimination of that tax. It may thus possibly lead to programs that require greater demand reduction than might have been necessary.25
Or suppose that some additional spending is carried out by the government to repair a road that facilitates the shipping of agricultural products out of the country. Here again the short-run negative effect on the balance of payments associated with the larger fiscal deficit is partly or fully neutralized by the positive effect associated with larger exports. These examples may be extreme but are far from rare. It would be easy to provide additional illustrations of the link between quantity and quality of fiscal adjustment. A perusal of stabilization programs indicates that, despite an increasing awareness of these issues, political difficulties, guidelines on conditionality, and timing concerns have prevented their formal inclusion in Fund programs.
In negotiating programs, the Fund has attempted, with increasing frequency, to ensure that cutbacks in government expenditure are focused on less productive activities. World Bank guidance is sought in this connection. Nevertheless, obvious political sensitivities have limited the degree of Fund involvement in decisions on expenditure policy. As a result, the expenditure policies pursued have in several instances not been as supportive of the growth objective as they could have been.26
An examination of actual cutbacks in capital expenditure in various countries indicates that they have at times been borne by some of the more productive projects. To reduce the budget deficit, cutbacks have sometimes affected productive, externally financed projects despite the fact that loans for part of the total cost of the projects were highly concessionary. At other times, cutbacks have focused on productive, domestically financed, small-scale projects, while externally Financed, highly visible, but less productive projects backed by important donors have been protected. Even where a core investment program has been agreed between the country and the World Bank, higher implementation rates for lower-priority projects have often occurred.
A common feature of such policies has been the disproportionate cutback in expenditure on materials, supplies, and maintenance relative to other types of expenditure. As a result, roads, bridges, public buildings, irrigation projects, airports, and other public sector infrastructure have deteriorated by more than would have been necessary, notwithstanding the inevitability of certain adjustments necessitated by the debt crisis.27 Inadequate maintenance eventually requires expensive projects for reconstruction of deteriorated plants and equipment.28 In agricultural regions, impassable roads have drastically limited the impact of market-oriented policies aimed at encouraging increased agricultural production. Shortages of materials and supplies have also dramatically limited the productivity of public sector employees, whether in education, medical care, agricultural extension, or tax administration. Across-the-board cutbacks in expenditure have been common. Such an approach fails to address the enormous waste of expenditure in many politically sensitive but unproductive sectors, including defense spending. Significant cutbacks in public sector employment remain the exception. As a result, efforts to cut the public sector wage bill have typically resulted in a deterioration in real wages, often greatest among the higher-paid civil servants. The factors encouraging corruption, low productivity, and multiple jobs of civil servants have therefore been intensified.
Tax increases have in some instances included measures that can be expected to have detrimental effects on growth. This has at times occurred in countries that already have very high tax ratios. For example, on many occasions the rates of export duty have been raised (or an export duty has been imposed) following devaluation, on grounds that the exporters would enjoy some sort of “windfall” profit. However, devaluations often simply offset past cost increases. Import surcharges have been levied, or the rates of import duties have been raised, for balance of payments and revenue reasons. As these surcharges have been imposed on products already highly taxed, they have, by increasing the differences between taxed and untaxed imports, increased distortions and reduced growth prospects.29 Surcharges on the income taxes of individuals and corporations have often been used. Sometimes countries have raised payroll taxes or taxes on interest incomes with undesirable repercussion on employment, saving, and capital flight. In a few cases countries have levied taxes on expatriate employment, or have raised the rates of mining taxes, or have levied taxes on foreign exchange transactions, thus discouraging foreign participation in economic development.
The main point of this discussion is worth repeating. The impact of changes in fiscal deficits on economic objectives depends to a considerable extent on the quality of the specific measures employed. A change in the quality of those measures will change the relationship between the fiscal deficit and the balance of payments, especially over the medium and longer run. The required reduction in the fiscal deficit (the required austerity) needed to achieve a given effect on the basic objectives of economic policy will be more severe as less efficient measures are chosen. For this reason, stabilization programs should systematically deal with microeconomic issues of public finance in addition to other structural policies. Programs must include needed structural changes and must integrate them with the macroeconomic framework.
Several problems arise in connection with the implementation of the approach suggested in this paper. They relate to (a) our knowledge of incentive effects, (b) timing considerations, and (c) political implications.
As to the first point, one could argue that not enough is known about the incentive effect of particular policies to place precise quantitative values on them. This is apparent, but irrelevant. Stabilization programs often rely on exchange rate devaluation even though precise estimates of these responses are not available. They also rely on changes in real interest rates even though, again, the size of the response of financial (and real) saving to changes in real rates cannot be known with precision. The important point is to have a sense of the direction of the effects and some “feel” for their size. If one waits for precise and objective quantifications of these effects, no formal agreement on a stabilization program would ever be included.
As to the timing issue, one could agree that the choice of better policies would in time bring about a more efficient economy and higher rates of growth. But what about the present? Wouldn’t, for example, the elimination or the reduction of an efficient tax or an increase in a highly productive government expenditure raise the deficit in the short run, thus necessitating more external or inflationary financing? A simple answer to that question is that important structural changes often bring with them immediate changes in expectations that can influence individuals and corporations to make further changes reinforcing their initial effects.30 For example, changes that create an environment more favorable to the private sector may encourage individuals to repatriate capital, encourage foreign enterprises to invest in that country, and facilitate foreign borrowing. More foreign money is likely to be made available to countries pursuing structural reforms.31 Still part of the answer is the fact that, as shown in the example of the export tax, some real effects will often occur early. If structural changes are made early in a program, or even before its formal approval by the Fund, their supply-side effects would probably also occur within the program’s duration, so that the initial negative effect on the size of the fiscal deficit could be balanced by a positive effect in the latter phase of the program. Reluctance to allow some initial expansion in the deficit through, say, the removal of inefficient taxes may contribute to the postponement of essential structural adjustment.32 Finally, this timing question is not limited to these policies. For example, the existence of J-curve effects indicates that the same problem exists with exchange rate devaluation. Also, so-called ratchet effects may postpone the time when the impact of demand-management policies is felt on effective demand.
The proposed departure is not without political implications. The conditionally guidelines may have to be amended to make it possible for the Fund to include formally in a stabilization program understandings about tax or expenditure reforms in the countries that approach the Fund for programs and where there are significant structural distortions.33 In some ways this would be a change more of form than of substance, because the Fund has already in recent programs been involved in structural aspects and has tried to persuade some countries to implement particular policy changes. The countries’ authorities may object to the proposed change, especially if they perceive it as additional conditionally without their receiving anything in return. Nevertheless, if they became aware that, at the time a program is negotiated, there might be some trade-off between the size of the required macroeconomic adjustment on the one hand (the required austerity) and structural changes on the other, their possible objection to the proposed change might in some cases be less than one would assume a priori.
The above discussion indicates that, if at all possible, a more inductive approach to determining the particulars of the fiscal policy required in stabilization programs would be desirable. In this approach, in addition to identifying the range of adjustment needed at the macroeconomic level, the Fund, in cooperation with the country’s experts, would make an inventory of the various changes in both the level and structure of taxes and of public expenditure that would be required to promote the country’s growth objective.34 It would have to take into account the importance that the country’s authorities attach to such objectives as equity and the provision of basic needs. The task would then be to determine whether the proposed changes add up to a macroeconomic adjustment package that is consistent with the balance of payments objective. The structural adjustment would be made up of a basic structural core of fiscal measures representing a sine quanon for a program. If this structural core did not add up to the macroeconomic adjustment assumed to be needed, the Fund and the local experts would look for progressively less efficient ways to add to revenues or to reduce expenditures. Should the country’s economic difficulties be assumed to originate exclusively from excess demand (that is, if no major structural problems are identified), the negotiations would proceed along more traditional lines.
The country’s authorities would be aware that there is a trade-off between the size of the needed demand constraint and the extent of the structural changes. They would know that the more daring and timely they are in introducing structural changes, the more flexibility they would have in demand-management. In essence, the program would be made up of three elements, possibly all of major importance: (a) the traditional macroeconomic framework with ceilings and targets; (b) the structural core; and (c) the investment core, which presumably would indicate, on the basis of World Bank recommendations, the minimum investment, as well as the allocation of that investment, consistent with both growth and balance of payments objectives.
One should not underestimate the difficulties, both technical and political, that a formal pursuit of this alternative would present; and one should recognize that this alternative would be considerably more labor intensive for both the Fund and the countries’ experts and policymakers. It is an alternative that requires further thinking before it can be fully implemented.35 Initial experimentation in well-chosen and willing countries would be indispensable to a full assessment of its general feasibility and to an outline of the procedural steps to be followed.
In this year when the Nobel Prize in economics has been given to James Buchanan for his contributions to public choice theory, it may be appropriate to conclude this paper with a few highly personal thoughts on the political implications of the suggestions it contains.
While aggregate demand may grow independently of structural policies, so that a traditional stabilization program would be sufficient in itself to bring about the needed reduction in that demand and thus the needed adjustment to the economy, it is more often the case that excess demand exists not (or not only) because demand has grown more than it should have, but because supply (including that of foreign exchange) has been constrained by misguided structural policies. For example, financial savings may have been reduced by constraints on nominal interest rates or by excessive taxation of interest income; this reduction may have constricted the supply of domestic financial savings available to finance the deficit and private investment in noninflationary ways, and, because of capital flight, it may have reduced the availability of foreign exchange.36 Agricultural output may have been reduced by low producer prices that necessitate the import of food. Agricultural exports may have been limited by excessive export taxes, by overvalued exchange rates, and by low prices paid to producers. Food supplies may have been limited by deteriorating transportation systems because of misallocation of public expenditures. In all these examples it is the supply that has been reduced, thus creating imbalances that, in time, manifest themselves as excessive demand. In these cases, demand-management policies alone would reduce the symptoms of these imbalances but would not eliminate the causes. Thus, stabilization programs might succeed stabilization programs without bringing about a durable adjustment unless the basic causes of imbalances are addressed.
One major difficulty in dealing with these basic issues is that the policies that I have called “misguided” may be misguided only in an economic and not in a political sense. Public choice theorists would emphasize the fact that these policies may be quite rational, at least in the short run, if assessed from a purely political viewpoint.37 They would argue that structural problems exist not necessarily because policymakers made technical mistakes in their policymaking, perhaps because of poor economic understanding. Rather, public choice theorists would argue that through these policies policymakers have tried to promote their own political objectives. Furthermore, the time horizon of policymakers is generally so short that they do not take into full account the long-run implications of their policies on the economy. These policies create “rents” for groups whose support the government needs in order to stay in power, even though they may in time reduce the level of income for the majority of citizens.38
If this public choice interpretation of economic policy is at least partly valid, and I do not know to what extent it is, it implies that policies aimed at structural reforms will often be resisted more than macroeconomic stabilization policies. They would be resisted because they would remove these rents from precisely those whose support the government needs and would thus reduce the leverage that the policymakers have for staying in power. In part, structural reforms would reduce the raisond’être for the government in power. As a consequence, it would seem to follow from these theories that major structural reforms have the best chance of being carried out when there is a major political change—that is, when a government that has long been in power is replaced by a totally different one—so that the political interests of the new policymakers are not tied to existing structural policies. This public-choice-inspired hypothesis should be amenable to testing. It seems to have some plausibility, but only a careful analysis of actual situations can assess its validity as a useful tool to explain changes that occur in economic policy.
In this paper the impact of fiscal developments on the balance of payments is emphasized. But, of course, the relationship is not unidirectional. In some cases fiscal disequilibrium may initially be created by developments in the balance of payments (say, a fall in export prices). In those cases the important question is whether the government should finance the shortcoming, or whether it should immediately or progressively lower domestic spending to reflect the lower real income of the country. On this, see Tanzi (1986), pp. 88–91; Tabellini (1985); and Chu (1987).
This is particularly important in order to reduce over time the burden of the foreign debt of the countries.
But, of course, changes in the exchange rate, which have often been part of traditional stabilization programs, have incentive effects in addition to their demand-management effect.
Over the years, what I have called the supply-management aspect of fiscal policy has received far less attention than the more traditional demand-management aspect. To put it differently, price (or micro) theory was rarely integrated with income (or macro) theory. Fiscal policy based on the Keynesian framework normally concentrated on the effects of changes in tax levels and public spending levels on aggregate demand. Supply management is a relative newcomer to economic policy, even though it had been clearly recognized by Joseph Schumpeter in his classic book. The Theory of Economic Development, first published some seventy years ago. Supply management emphasizes that the way the factors of production are used may be more important than their amounts. It emphasizes that growth requires not only that the factors of production keep growing at a desirable pace but also that they are allocated as efficiently as possible. If, for example, investment grows but is progressively channeled into less productive projects, the country’s output may not grow.
Most of the formal models that link the fiscal deficit to the balance of payments follow this approach. In these models, it is the size of the macrovariables (the saving rate, the investment rate, the fiscal deficit, etc.) that plays the leading role. These variables are rarely disaggregated, so that the possibilities connected with better resource allocation are not explored.
It must be understood that even this macroeconomic approach will have to depend on specific measures to raise revenue or reduce spending.
The theoretical design of Fund programs as generally interpreted has been much closer to what I have called the macroeconomic approach.
See, for example. Gold (1979), pp. 30–34.
For subsidies, one additional reason was their direct effect on the current account of the balance of payments when the subsidy encouraged the consumption of an imported commodity.
Technical assistance is provided by the Fund only at the request of a country’s authorities.
The extended Fund facility was established in September 1974 to provide financial assistance in support of medium-term programs for up to three years to overcome structural balance of payments maladjustments. The first request for this arrangement, by Kenya, was approved by the Fund in July 1975.
In actual negotiations, the sequence may not appear as described but in essence it is.
This might require a change in the conditionality guidelines approved by the Executive Board. Of course, whenever there is no presumption that resources have been badly misallocated, Fund programs would continue to focus on a macroeconomic framework.
“Performance criteria are always objective in order that a member will not be taken by surprise by a decision of the Fund to impede transactions under a stand-by arrangement. The member has maximum assurance, therefore, about the circumstances in which it can engage in transactions with the Fund.”—Gold (1979), p. 32.
A few programs have made the total level of public expenditure a performance clause. This can be considered a departure from the traditional narrow interpretation of conditionality guidelines. Generally, the formal agreements have focused on the difference between public expenditure and revenue (i.e., on the deficit).
Programs recognize this problem by including (a) reviews to ensure that additional measures are taken to stay on track and (b) a commitment to take additional measures as needed.
A few years ago the government of a given country pressured a large foreign enterprise to advance tax payments for the next three years so as to allow the country to comply with the fiscal ceiling.
This is common with real wages for public employees that are at times reduced to unsustainable levels during the program but bounce back to a more normal level when the program is over. Permanent adjustment would more likely result from a reduction of the permanent public sector work force than from what is often a temporary reduction in real wages. On the issue of arrears in the payments of goods and services by the government, see Diamond and Schiller (1987).
For a definition of the concept of core fiscal deficit, see Tanzi and Blejer (1984), p. 119.
This is true regardless of the present level of taxation in the country.
The Fiscal Affairs Department of the Fund has produced a series of papers on this issue in the past few years. Some of these papers are expected to be published in a forthcoming book. For specific studies of the relationship between export taxes and exports, see Vito Tanzi (1976), pp. 66–76; Okonkwo (1978); and Sanchez-Ugarte and Modi (forth coming).
The potential output of a country is likely to grow if (a) the rate of investment grows while its average productivity and the average productivity of the other factors of production (labor, land, and so on) do not change; (b) if the average productivity of the factors of production increases owing to the removal of distortions, or to technological change, even though the supply of the production factors does not change. If the distortions have, as is often the case, reduced the country’s ability to earn foreign exchange or have led to the misuse of the foreign exchange available, their removal will over time increase the flow of foreign exchange available to the country. In other words, the removal of the distortions would have the same effect as an increase in foreign lending to the country. It would thus reduce the need to constrain demand, as this need, in a typical Fund program, is often a function of the scarcity of foreign exchange.
It should be recalled that an export tax on a commodity X can be decomposed in a production tax on X and a consumption subsidy on X. Thus, the removal of the export tax removes the subsidy to domestic consumption and removes the tax on production. Domestic consumption falls while production and, presumably, exports rise.
Of course, if the authorities propose to reduce the fiscal deficit through an increase in export taxes, then the negative supply-side effects of this policy would require even greater demand management than when these effects are ignored.
At this point it may be useful to state the obvious: government decisions are often influenced more by political considerations than by considerations of economic efficiency.
These expenditures are generally classified as “current” rather than “capital” expenditure. Therefore, the common view that stabilization programs must protect “investment” may not necessarily lead to the best policy. In some cases, the most productive expenditures are “current” ones.
There is now a growing concern among some experts that the present reductions in fiscal deficits associated with these lower expenditures for maintenance of roads and other infrastructure will necessitate much higher expenditures (and thus higher deficits) in future years, as the lack of maintenance will require expensive rebuilding. This is again an example of the shifting of the fiscal deficit from the present to the future.
Imports subjected to import duties are often less than 50 percent of total imports, so that substantial rate increases on the taxed imports are needed to generate significant tax revenue. Of course, as the rates go up so does smuggling.
This is particularly true when the attitude of the government indicates that these changes are not likely to be reversed soon.
It should be recalled that the Baker initiative is postulated on this assumption.
Many structural changes can be made to be revenue neutral by removing some taxes (adding some expenditure) while at the same time adding some other tax (reducing some other expenditure).
That is, some of the documents that reflect the formal understandings between the Fund and the country must spell out the details of the agreement between the two parties as to the tax modifications, changes in public expenditures, and so forth.
Obviously, other structural aspects would also be considered.
The full and formal introduction of structural changes in the theoretical design of Fund programs should be considered one of the main challenges to our future research effort.
For an analysis of the ways in which deficits get financed in developing countries and on the limits to those sources of financing, see Tanzi (1986), pp. 139–52.
All the literature on rent-seeking that represents an important chapter of public choice would support this view. See especially Buchanan, Tollison, and Tullock (1980); and Tollison (1982), pp. 575–602.
For example, if agricultural prices are kept low in order to subsidize the real wages of urban dwellers, the government may acquire the support of the latter, but the cost may be a low rate of growth and increasing economic difficulties over the longer run. See, for example, some of the studies in Harberger (1984).
BuchananJames M.RobertD. Tollison and GordonTullockeds.Toward a Theory of the Rent-Seeking SocietyTexas A&M University, Economic Series, No. 4 (College Station: Texas A&M Press1980).
ChuKe-Young“External Shocks and the Process of Fiscal Adjustment in a Small Open Developing Economy,”IMF Working Paper No. 87/11 (Washington: International Monetary FundMarch21987).
DiamondJack and ChristianSchiller“Government Arrears in Fiscal Adjustment Programs,”IMF Working Paper No. 87/3 (Washington: International Monetary FundFebruary101987).
GoldJosephConditionality IMF Pamphlet Series No. 31 (Washington: International Monetary Fund1979).
HarbergerArnold C.ed.World Economic Growth (San Francisco, California: Institute for Contemporary Studies1984).
International Monetary FundSelected Decisions of the International Monetary Fund and Selected DocumentsTwelfth Issue (WashingtonApril301986).
KellyMargaret R.“Fiscal Adjustment and Fund-Supported Programs, 1971–80,”Staff PapersInternational Monetary Fund (Washington) Vol. 29 (December1982) pp. 561–602.
OkonkwoUbadigbo“Export Taxes on Primary Products in Developing Countries: The Taxation of Cocoa Exports in West Africa” (unpublished; Washington: International Monetary FundNovember291978).
Sanchez-UgarteFernando and Jitendra R.Modi“Are Export Duties Optimal in Developing Countries?: Some Supply-Side Considerations” in Supply-Side Tax Policy: Its Relevance to Developing Countries (Washington: International Monetary Fundforthcoming).
TabelliniGuido“Fiscal Policy Response to the External Shocks of 1979 in Selected Developing Countries: Theory and Facts” (unpublished; Washington: International Monetary FundDecember261985).
TanziVito“Export Taxation in Developing Countries: Taxation of Coffee in Haiti,”Social and Economic Studies (Kingston) Vol. 25 (March1976) pp. 66–76.
TanziVito“Is There a Limit to the Size of Fiscal Deficits in Developing Countries?” in Public Finance and Public Debted. by BernardP. Herber (Detroit, Michigan: Wayne State University Press1985).
TanziVito“Fiscal Policy Responses to Exogenous Shocks in Developing Countries,”American Economic Review Papers and Proceedings (Nashville, Tennessee) Vol. 76 (May1986) pp. 88–91.
TanziVito and MarioI. Blejer“Fiscal Deficits and Balance of Payments Disequilibrium in IMF Adjustment Programs” in Adjustment Conditionality and International Financinged. by JoaquinMuns (Washington: International Monetary Fund1984) pp. 117–36.
Luis Jorge Garay S.
The purpose of Vito Tanzi’s paper, “Fiscal Policy and Stabilization Programs,” is to show some methodological and operational problems that are inherent in the traditional adjustment programs of the International Monetary Fund, even within their own theoretical framework.
The paper has the merit of spelling out certain basic problems in the analytical framework, with particular attention to the fiscal area, although the naming and conceptual interpretation of the two alternative approaches contrasted with each other are unfortunate. To call the approach that is put forward as a framework of reference the “macroeconomic approach” is inappropriate, because strictly speaking the subject of study in a macroeconomic approach is the analysis of the behavioral interrelationships between those factors that play a decisive role in the evolution of a specific economic phenomenon. In point of fact, the operational approach adopted by the Fund is based on a partial study of disequilibrium situations at given moments in time, but not on a study of the behavioral relationships of the various components of factors that determine the disequilibrium itself, of the dynamics of the process leading to the disequilibrium, or of the transition to the path to equilibrium. The approach advocated by Tanzi also suffers to a large extent from this kind of fundamental shortcoming.
In this context, a “macro approach” to tackling the problem of fiscal deficit has to focus on the dynamics of the process that generates the deficit; dynamics that include not just various facets of the economy but also various instances over time and between sectors within the public sector.
The difference between a macro approach and a mere “aggregate approach” is radical and can, in general, lead to substantially different economic policy considerations. It is therefore appropriate for me to try to clarify the scope of the essential contrasts between them, making brief reference to certain basic shortcomings of the aggregate approach that could be avoided, or at least reduced, in the context of a strictly macroeconomic approach.
In the first place, a purely aggregate approach does not guarantee the ability to identify a serious partial or even quasi-generalized disequilibrium situation. Indeed, in practice it is feasible and in no way surprising to observe an “apparent” fiscal equilibrium at the aggregate level for the public sector as a whole, despite the fact that behind it there lies a significant range of partial disequilibria, which, although they may even offset one another (in terms of monetary value) between the various official entities, are nonetheless the clear expression of a typical disequilibrium situation. To use the terminology of recent economic theory on the subject to illustrate this, rather than being a symptom of a state of equilibrium, a situation of this kind is in reality merely one of a number of temporary situations that are generated in a disequilibrium process. To place special emphasis on the aggregate of public sector deficit, without having sufficient understanding of the process by which the fiscal disequilibrium is generated, of the instances of actions, or of the timing of the factors acting upon it, quite frequently leads to fiscal aggregate adjustment programs with a short-term horizon, involving the adoption of policies and corrective measures that are certainly effective in swiftly controlling the deficit level, but which may actually generate conditions that foster disequilibrium in the medium term. This runs counter to the whole purpose of the adjustment program, which is to facilitate the transition to equilibrium.
Second, the fact that most adjustment programs are short-term rather than medium-term tends to encourage the authorities to apply discretionary measures that have a visible impact soon after their introduction, which may not be the most appropriate way of achieving all their objectives, of encouraging continuity in a recovery program of the kind referred to in the adjustment programs, or of guaranteeing the consolidation of some of the vital conditions for economic growth with stability in the medium term.
There is not necessarily a one-to-one relationship between a program’s targets and the specific policy measures that will guarantee attainment of those targets. In practice, to the extent that some leeway is given in the choice of measures, the possibility increases of medium-term criteria coming into play in seeking to comply with the spirit of the adjustment, and of policy priorities additional to those established in the program being taken into consideration. Such leeway, obviously, can be correspondingly broader, ceteris paribus, the more flexible the program is toward changes in the environment (as is explained below), the longer the period prescribed for reaching the proposed targets, the less the relative intensity of the adjustment in terms of the magnitude of the initial disequilibrium, and the smaller the range and depth of the disequilibrium process that one is trying to correct.
Consequently, only when the costs and benefits over time and the trade-offs between targets have been weighed in a medium-term context and of the various modalities of adjustment will it be possible to define a “desirable” program in which the flexibility, durability, intensity, extent, and “quality” of the adjustment can be duly reconciled.
Third, the time horizon, the level of aggregation, and the selection of reference parameters used to frame the traditionally tailored adjustment programs have contributed to the imposition of qualificatory standards on the attainment of the set targets, which in principle do not adequately take into account the number of significant changes that take place in the environment (both domestic and international) in which the performance of the economic aggregates for the design and specifications of the program was simulated.
Despite this, it must be recognized that from the outset two exception clauses were established regarding compliance with adjustment programs arranged with the Fund: (i) the so-called waiver clause, under which compliance with a target or targets is waived provided the deviation in compliance is slight and reversible in the short run and that it is not detrimental to compliance with the program as a whole, and (ii) the modifiability clause applicable in cases of substantial alterations in the environment—large, irreversible, and quasi-generalized—which make it necessary to change the program in order to ensure that all those alternative measures that would be suitable in the new circumstances are adopted, as well as to ensure that the spirit and the quality of the adjustment “desired” at the outset are complied with. This latter modality has been applied only on very few occasions in very special situations, for which reason, at least until now, it has not been possible to attain a reasonable degree of flexibility in the traditional criteria for certifying compliance with a program.
Fortunately, about last October the way began to be opened in the desired direction by another form of exception, intermediate between the other two, included in the financial agreement signed by Mexico and the Fund, which provides for the introduction of adjustments in a target or targets functionally related to the degree of alteration (or change) in one or several of the economic variables that decisively affect the behavior of the aggregates of the economy, but on condition that not only the spirit of the core of policies and measures stipulated therein but also their orientation, intensity, and arrangement over time remained intact. One outstanding example of this kind of cause is the unforeseen, drastic, and lasting change (in terms of the program horizon) in the world price of a product whose exports generate a large proportion of the foreign exchange of the program country.
Fourth, because the traditional adjustment programs are based on one and the same analytical framework, regardless of the nature, depth, and scope of the disequilibrium process, the consequent therapy for rehabilitation, based on a diagnosis derived from the framework referred to, has of necessity taken a standard form. However, to avoid entering into a discussion which would not be appropriate in this commentary, suffice it to say by way of illustration that there is a near consensus on the desirability, or at least the expediency, of adopting shock tactics for those economies subject to acute, lasting, and quasi-generalized disequilibrium. Obviously, such therapy contrasts fundamentally with the one prescribed by the orthodox approach to adjustment.
In these circumstances, it seems clear that the more severe the maladjustment the more the modalities, intensity, and duration of the recovery process will need to take into account the inherent conditions of the dynamics of disequilibrium. It should be remembered in this respect that regardless of the analytical framework adopted no theory has yet been developed (even for simplified systems) to study the “optimum” level of maladjustment or disequilibrium, let alone one on a “desirable” framework for transition to equilibrium. The absence of such a theory makes the application of a single set of universal “hard-and-fast” rules regarding the “desirable” depth and timing of the adjustment even more vulnerable.
Having mentioned some of the theoretical and conceptual defects and lacunae in the aggregate approach, and the difficulties—some of them insuperable in practice—of the strict macro approach, we should also mention a number of basic institutional and operational precepts in the traditional programs which further exacerbate some of the problems just mentioned. Three of them, to refer only to those most relevant in the area of public finance, relate to the procedural and operational area of expenditure policy and to the infrastructure available for its effective execution.
In the economic analysis of public finance, according to effective operations at the cash level, the traditional approach is posited on the following basic tenets: unity of the cash area in the public Treasury, an automatic and efficient system of transfers within the State, and perfect fungibility of money in the process of public expenditure. While such premises are intimately interrelated, their relevance varies according to the stage in the process of expenditure in question. Thus, for example, the fungibility of money is fundamental in promoting adequate temporal and financial synchronization between the budget operation and the execution of the expenditure. The efficiency of the transfer system is important in promoting timely and decisive mobilization of resources between public entities (in the form of “pure” transfers or “soft” loans, among other mechanisms) as a prerequisite for the productive and (socially) profitable use of the financial surpluses that some of these generate. As is evident, the unity of the cash area ensures careful coordination between the moment the financial resource enters the cash process (for example, as a budget counterpart resource), the moment at which it generates the expenditures commitment, and the moment at which the expenditure is paid out in the cash process.
However, given that in practice such tenets are difficult to satisfy, compliance with the targets of a fiscal adjustment program whose design does not take due account of the scope and implications of failure to observe its conditions can lead to such problems as (i) accentuating the existing distortions in budget execution by not taking due account of the pressure of overhanging commitments from previous exercises (for example, by departing from a proposal to “reorganize” the accounts payable of the Treasury), (ii) obstructing proper synchrony between budget management and expenditure execution, which not only hinders the efficient and timely allocation of resources but may also even encourage such anomalies as “overfinancing” the overall fiscal deficit (with the consequent building up of the cash position, most of the time drawn from costly credit resources), and (iii) obstructing a “balanced” relationship between the priorities of fiscal, monetary, exchange (e.g., accumulation of international reserves) and debt policy targets, even to the extent of placing disproportionate emphasis on some of these policies at times when action on the others would be appropriate. (It comes as no surprise to see fiscal policy being made relatively subordinate to short-term money management.)
Thus, for reasons of an operational and institutional nature, a reduction in the level of disequilibrium, or even its stabilization at a “desirable” level (in this case, of the overall fiscal deficit for effective cash-based operations) does not necessarily imply that there has been a corresponding real adjustment (of public finances), or that the transition has been made to the path of medium-term equilibrium. On the contrary, more likely than not the adjustment will be merely “apparent, sporadic, and partial,” failing to counteract the factors determining the disequilibrium. Here, indeed, we find another of the problems that can be best dealt with by a strictly macroeconomic approach (or at any rate with better data with which to judge than in the case of a purely aggregate approach).
Before concluding, I would like to review some of the ideas with which Tanzi winds up his paper, with particular reference to the development of an approach to the design and specification of “macroeconomic structural adjustment” programs. Although sticking as closely as possible to the traditional framework as their scheme of reference, these programs would include a structural core of measures (with special importance in the fiscal area) and an investment allocation core in line with the targets of economic growth and “viability” of the balance of payments in the medium term.
As the above-mentioned approach contemplates adjustment in the medium-term and long-term horizons and as this in turn gives it a structural rather than a current or short-term character, there immediately arises the institutional concern as to what should be the proper international institution to promote the implementation of this kind of adjustment program. According to the division of roles under the Bretton Woods Agreement, the task would most properly be carried out by the competent authorities of the World Bank rather than those of the International Monetary Fund. Although it is, so to say, artificial to set forth with complete assurance a universal rule for demarcating the boundaries between short-term and medium-term policies, because a medium-term policy would in any case by its nature include a body of short-term measures, what is undeniable is that the further one moves into the framework of a macroeconomic adjustment with a medium-term horizon, with structural implications and with action required at the level of investment allocations, the more suitable the World Bank rather than the Fund will be to take charge of putting into practice programs such as those referred to.
Despite the above, there remains not just an element of contradiction, but also a lack of definition such as to the order of priority of objectives, policies, and conditionality consistent with these two kinds of adjustment programs, when we are at a transitional stage between these alternative approaches. This has been a frequent occurrence in the recent past, with program, sectoral, and structural adjustment credits granted by the World Bank simultaneously with the traditional stand-by programs agreed with the International Monetary Fund.
Consequently, in order to avoid the harmful coexistence of cross-conditionalities with mutually contradictory criteria, as has happened in some countries, it would be necessary to thoroughly revise the functions of the World Bank and the Fund with regard to the incorporation into the world monetary system of an approach to adjustment different from the one currently in use, such as a macroeconomic structural adjustment would constitute.
The paper presented by Vito Tanzi falls under the heading of the new thinking on adjustment with growth that has been gaining ground since 1985, in that it analyzes the way in which the International Monetary Fund has defined adjustment programs and proposes an alternative view which, he suggests, could make for a better connection between stabilization and growth.
The Fund’s stabilization programs have been based on the Polak model, which links domestic credit expansion with the level of international reserves, using the accounting identity developed through the monetary approach to the balance of payments. This identity stipulates that international reserves are the difference between the demand for money, considered as endogenous, and the expansion of domestic credit. This identity is formally defined as follows:
where RIN is the level of net international reserves and CDF and CDP are domestic credit to the fiscal sector and domestic credit to the private sector, respectively.
The programs operate by defining desired levels of international reserves, growth, and inflation and thereby calculate the private sector’s demand for money and credit. The accounting identity is used to calculate the public deficit that would be consistent with the above-mentioned targets. Consequently, according to this approach there is a fairly direct link between the fiscal deficit and the performance of the balance of payments.
Tanzi attacks this view by suggesting that the performance of the balance of payments is not unconnected with the particular way in which the reduction of the fiscal deficit is addressed. He illustrates this argument by referring to the adverse effects of taxes on exports and of cutting back on infrastructure maintenance. The first example increases domestic demand and reduces the supply of exportable goods, with the result that the contraction in domestic demand via a reduction in the fiscal deficit must be greater than it would be in an alternative program not using this instrument. The second example emphasizes the impact on the supply of exports and supports the same argument. Other examples relate to the durability of the proposed fiscal changes. The temporary and unsustainable reduction in the public administration wage bill, the postponement of necessary projects, and a reduction of priority investments are examples of less-than-optimum fiscal adjustments using a medium-term logic.
Tanzi consequently proposes a new way of addressing stabilization plans so that fiscal reforms can be lasting and support the needed adjustment in the balance of payments, given their effects not only on aggregate demand but also on the supply side.
The microeconomic approach he proposes would begin with an analysis of the specific revenue and expenditure reforms and then ascertain whether they provide the requisite adjustment at the macro-economic level. Tanzi also discusses the advantages and disadvantages of this system, highlighting the political interference that this new system would imply, the lack of a clear Fund mandate in these areas, and the inadequacy of the Fund’s budget to cope with this kind of program.
The need to change the way in which the Fund formulates its adjustment programs reflects a shortcoming of the Polak model. Tanzi recognized this but limits his criticisms to the failure to consider the microeconomic impact of fiscal policy on supply. This would seem to be the right time to mention other limitations of the Polak approach that should be taken into account when redefining the adjustment programs.
The Polak approach assumes that tax policies affect demand in the same way irrespective of the form they take. However, it is obvious that not every tax increase produces a similar reduction in domestic expenditure. For example, an increase in domestic taxes to offset a reduction in fiscal revenue from exports resulting from a deterioration in the terms of trade is proportionally more contractive. Likewise, a progressive income or wealth tax reduces real domestic expenditure to a lesser extent than an increase in indirect taxes on mass consumer goods.
Another assumption generally used is the stability of demand for money. This assumes a stable demand for external financial assets by resident agents. However, capital flight is frequently a countercyclical phenomenon, as economic recession reduces the return to fixed assets and increases uncertainty, thereby encouraging the outflow of capital. A reduction in economic activity causes a fall in international reserves, which makes an interest rate hike necessary, as well as further contraction of demand, leading to the need for an overadjustment to cope with the capital flight.
It is important to stress that despite the fact that the Fund’s Articles of Agreement do not question capital controls, it has neither developed nor proposed any policies to limit the flight of foreign exchange by administrative means and, at the same time, it has not criticized those governments of industrial countries that have converted their financial markets into tax havens for funds coming from other countries. The impact of a recessionary adjustment on capital flight and the demand for money deserves more extensive consideration than the Polak model has suggested.
In Polak’s approach it is impossible to analyze the medium-term consequences of short-term adjustments. In particular, declining public and private investment brought on by contractionary policies in the fiscal and monetary areas delay the adjustment of the productive infrastructure and make lower growth rates in medium-term demand a necessity. The compensation over time cannot be analyzed. In this regard, adjustment programs tend to be less than perfect in that they are based on the implicit hypothesis of credit rationing. Consequently, it generates a short-term adjustment greater than that which would result from optimization over time. This results in a lower growth rate in the country and greater balance of payments difficulties, which in turn lead to a request for additional medium-term resources from the Fund and the World Bank.
The deterioration of the fiscal accounts of Argentina, Mexico, and Venezuela partly reflects the consequences of the earlier vicious circle of external public indebtedness and private capital flight. Real devaluation produces a redistribution of wealth taking the form of capital gains for external asset holders and fiscal deficits reflecting the loss of capital in the public sector.1 Polak’s answer would be to translate this loss into a contraction of domestic absorption via fiscal restraint without reflecting the specific characteristics of this situation. This points to the possibility of more carefully defining the causes of fiscal deficits in order to design policies more relevant to each situation. In particular, a tax on external asset holdings and the income they generate, combined with an international system for fiscal cooperation, would help to reduce the contractionary impact of fiscal correction.
The Polak model does not distinguish between the different effects on the balance of payments of cuts in expenditure on tradable goods and of cuts in expenditure on nontradable goods. The former strengthens the balance of payments whereas the latter only produces a contraction in domestic activity, especially when there are idle resources. The composition of public expenditure between demand for tradables and nontradable goods is, consequently, together with exchange policy, an important instrument for making changes in the structure of expenditure (expenditure switching), thus avoiding the inefficient cuts in economic activity that are characteristic of more aggregated approaches, such as Polak’s (expenditure cutting).
The thrust of these arguments is that the range of restrictions involved in the current approach used by the Fund to design the fiscal packages within its stabilization programs should be expanded. The revision proposed by Tanzi could be a propitious occasion to tackle these other issues, and expand the range of effects to include those not previously taken into account. In particular, Tanzi would like to limit criticism to the contradictions between deficit reduction and the economic distortions generated with regard to the microeconomic model of general equilibrium. It would, however, be advisable to delve further into the other macroeconomic aspects that appear to have received insufficient attention under the traditional approach.
Tanzi’s paper seeks a greater degree of integration between stabilization and growth. For this, either explicitly or implicitly, a growth theory is needed to serve as a framework for operative decisions. Nothing explicit about this is stated in the paper. However, from the examples proposed, an implicit theory that seems to support the proposition can be deduced.
This theory appears to affirm the following: some fiscal policies tend to generate distortions in relative prices, producing an inefficient allocation of resources and consequently a lower growth rate. A fiscal reform to reduce distortions will consequently permit increased growth in the medium term.
There is no doubt that this argument contains an element of truth. There are fiscal practices which act as bottlenecks to development. However, this does not necessarily mean that the allocation of resources generated by more extensive use of market mechanisms is an adequate guide for the allocation of budget resources. The dynamic comparative advantages and externalities require analysis and attention that are difficult to reduce to the argument in favor of the elimination of distortions. In the case of externalities such as public goods, the market fails by definition. The allocation of resources in this area calls for a body of criteria that are not easy to systematize and include in a generic view of the necessary changes.
The experience of successful developing countries highlights the importance of dynamic comparative advantages. This assumes government planning and direction, a feature common to the Republic of Korea, the Taiwan Province of China, and Brazil, to generate an allocation of resources that the market would not recognize as efficient in the short run. In oil exporting countries, the “Dutch disease” makes an explicit policy necessary to avoid having market signals work against industrial and agricultural activity. The decline in the terms of trade for primary commodities and energy has reduced the relevance of static comparative advantages as a resource allocation mechanism consistent with a successful development strategy. The coordinated allocation of public resources through expenditure in education, technology, financial subsidies, and trade policies and through government purchases seems to be a frequent practice in the successful countries mentioned above, but is a violation of the general equilibrium model.
A specific discussion of the fiscal reforms that the Fund would propose in the new adjustment programs could not leave out these subjects and consequently calls for some explanation of the theory and strategy of growth going beyond the mere reduction of distortions.
Given the stage that consideration of the relationship between stabilization and growth appears to have reached, it would seem a little premature to increase substantially the conditionality of Fund loans. In the first place, the tax question is still a very sensitive issue from the political point of view, as it involves questions of constitutionality. The separation of powers that is a feature of democratic republics has given the Legislature power over the budget. This means that the Executive Branch is not able to commit itself regarding aspects of conditionality such as those assumed in Tanzi’s proposal. Programs would require the conditional approval of the Executive Branch and a lengthy Legislative debate before they could be accepted. This is incompatible with the time frame and urgency generally associated with Fund programs.
The problem is exacerbated by the cartelization of the international financial market through bank advisory committees, which also require prior agreement with the Fund. The democratic principle that emerged from the American Revolution that there should be “no taxation without representation” would in practice mean that the major stockholders in the Fund are over re presented with regard to the very interests that democracy should in theory reconcile.
Furthermore, there is the question of the risk, which is unevenly distributed between countries and the Fund. As Tanzi rightly points out, our knowledge of the behavior of the various real economies is limited, despite which policy decisions still have to be taken. Put in other terms, there is a problem of our ignorance of economic dynamics when it comes to preparing policy measures. This ignorance carries with it the risk that the economy may not react in the way theory would have it. This risk, which could be called the “ignorance risk,” falls wholly on the countries. Consequently, it is to be hoped that the countries will show a stronger aversion to the ignorance risk than has the Fund. The fiscal reforms resulting from Tanzi’s microeconomic approach would reflect this asymmetry, which would make the discussions difficult. Matters are made worse by the fact that the ignorance risk increases exponentially as one attempts to move toward more specific reforms.
Consequently, if experimentation with the new approach is not coupled with a risk redistribution scheme, only the countries whose state is most critical, that is, those that have suffered the greatest erosion of their sovereignty, would be ready to participate.
One way of redistributing risk would be to tie in contingent Fund financing if the measures do not produce the expected results. For example, an export tax reduction that did not generate the expected improvements in the balance of payments would lead to an additional loan from the Fund. In this way, the countries’ reluctance to absorb the loss of fiscal revenue would be reduced because the Fund would partly cover the ignorance risk. This would not be an easy scheme to implement in many cases, but it does demonstrate the desire for change that would exist in countries if the risk were more fairly distributed.
Last, it might be advisable to change the way in which the Fund has been operating. At present, the Fund’s financing is secondary to the conditionality imposed by the agreement, as the emphasis lies in its being a requirement of the international banking community. It would perhaps be interesting for Tanzi’s ideas to be implemented, at least initially, in countries that are not renegotiating their external debts and not suffering from an acute stabilization problem. From this viewpoint, the agreement with the Fund would be seen more as Financing for a structural change than as a circumstantial stabilization measure. It would imply a redefinition of the roles of the Fund and the World Bank, as the traditional division of labor between these two institutions runs counter both to Tanzi’s proposal and to the approach to implementation that is suggested here.
On this question, see Miguel Rodríguez, “Consequences of Capital Flight for Latin American Debtor Countries,” in Capital Flight and the Third World Debt, ed. by John Williamson and Donald R. Lessard (Washington: Institute for International Economics, forthcoming).