Cómo citar este documento

ABSTRACT

This paper discusses the relationship between stabilization and fiscal reform in the context of the Convertibility Plan presently being implemented in Argentina, mainly focusing on tax reform and privatizations.

The rapid increase in tax receipts has been a decisive factor in the credibility and sustainability of the stabilization program. Two were the components of the reform process which were responsible for that outcome: the progressive establishment of VAT as the core of the tax system, and a radical transformation in tax administration, including much more rigorous penalties for tax evasion. Nevertheless, according to the analysis presented in the paper, there is still ample room to further improve the tax system's performance, not only from the point of view of tax collection, but also from that of equity and efficiency.

With regard to privatizations, it is still too early to fully appraise their long-run impact on the Argentine economy. However, despite the standard justification on long-term efficiency grounds, the paper argues that so far they have basically served short-term macroeconomic purposes.

Introduction[1]

The two hyperinflationary episodes of 1989 and 1990 left the Argentine economy in a critical situation. In a context of huge fiscal imbalances, massive capital flights and swift demonetization, prices rose almost fifty times in 1989 and another thirteen in 1990, while real investment collapsed, and output dropped 6.6% in 1989 to remain constant the following year.

At the end of March 1991 the government announced a new stabilization policy based on the full convertibility of the domestic currency at a fixed exchange rate established by law, and a 100% backing of the money base with foreign reserves. A sharp fiscal adjustment was the other key component of the program. At the same time, the government hastened the rhythm of privatizations, trade liberalization, and other market-oriented structural reforms already underway since the beginning of Mr. Menem's administration.

The Convertibility Plan was considerably effective in stabilizing the economy. The fiscal balance showed a remarkable improvement, whereas optimistic expectations and a favourable international background triggered an impressive inflow of private capital and a rapid process of remonetization. As a result, inflation dropped dramatically and economic activity began a fast recovery.

However, given the introduction of a fixed nominal exchange-rate regime, the convergence of domestic inflation towards international rates was not fast enough to prevent a substantial appreciation of the domestic currency in real terms.

Furthermore, against the background of a falling real exchange rate, trade liberalization caused an abrupt rise in imports. As exports remained stagnant, the trade balance sharply deteriorated and the current account, which run a surplus in 1990, has turned into a growing deficit since 1991.

The other source of potential trouble for the prospects of stabilization is the fragility still displayed by the public sector. No matter how important the achievements in tax revenues have been so far, nonfinancial current expenditures have also shown a significant growing trend, basically because of both the heavy burden of the social security system and the lack of fiscal adjustment in the provinces. In such a context, investment cuts, the decline in international interest rates, the real appreciation of the exchange rate, and once-and-for-all privatization revenues have played a key role in keeping the public sector borrowing requirements under control. Nevertheless, this is clearly not a permanent solution.

Taking all this into account, the paper discusses the relationship between stabilization and fiscal reform in the context of the Convertibility Plan presently being implemented in Argentina, mainly focusing on tax reform and privatizations. The paper is organized into seven sections. The first two sections summarize the recent performance of the Argentine economy and the fiscal accounts respectively. The third one traces the evolution of the tax system throughout the eighties, while the fourth examines the main components of the current process of tax reform, as well as its first results. The chief reasons underlying the improvement in tax performance are analyzed in the fifth section. The sixth one, in turn, discusses the role played by privatizations. Finally, the seventh section presents a preliminary assessment of the current process of fiscal reform, and points out the still unsettled issues of this process.

1. Recent macroeconomic trends

The Convertibility Plan announced in March 1991 achieved considerable success in overcoming hyperinflation and stabilizing the Argentine economy. As a result, the annual inflation (measured in terms of the CPI) was cut down to only 7.4% in 1993, an exceptionally low rate according to Argentine historic standards, and it is expected to be even lower in 1994 (see Table 1). A fixed rate of exchange, a swift trade liberalization and a sharp fiscal adjustment were the key factors explaining this outcome, but legal deindexation of nominal contracts, price agreements in some strategic industries, and other incomes policy measures were also important at the initial stages of stabilization.

Regarding the fiscal stance, in particular, the improvement has been remarkable, even allowing the government to begin producing surpluses in the overall fiscal balance since 1992: 0.7% that year and 1.4% in 1993 (see Table 1).

The success of stabilization has also brought about a large increase in all monetary aggregates. The sharp reduction in macroeconomic uncertainty, the abrupt fall in inflation and economic recovery have caused the demand for money to rise steadily. The expanding money needs of the public were matched by massive capital inflows. Privatization of public concerns and some other structural reforms launched by the government were major determinants underlying this trend. The repatriation of flight capital was equally important. However, the significance of the sudden reversal experienced by international financial conditions since the early nineties, which also benefitted most of the other Latin American countries, can hardly be downplayed.[2] As a result, private capital inflows amounted to nearly 26 billion dollars in 1991-1993, while total monetary resources (M2 in pesos and dollars) rose from 5% of GDP in 1990 to 18.2% three years later (see Table 1).

Economic activity also underwent an outstanding expansionary phase from the beginning of the program. Indeed, the GDP growth rate stood at 8.9% in 1991, 8.7% in 1992, and 6.0% in 1993. The performance of industrial output was also impressive: its average growth rate was 7.9% for those same years (Table 1). The huge capital inflow induced by both domestic stabilization and a positive external environment was essential in determining this trend.

Economic recovery was led by consumption, which experienced a boom as a consequence of the repatriation of capital from abroad and of the increased availability of private sector credit. The consumption boom was so strong that, somewhat paradoxically, it deepened the decline in the national savings rate brought about during the past decade by the debt crisis first and by the hyperinflation episodes afterwards. As a result, the level of national savings dropped to only 13.9% of GDP in 1992 and while it recovered to 15.8% of GDP in 1993, it was still far below the 1988 rate of 19.5%.

Aggregate investment also showed a significant improvement. According to official data gross fixed capital formation grew from 14% of GDP in 1990 to 17.9% of GDP in 1993 (Table 1). So far, this improvement has only enabled the economy to recover the investment rates prevailing in the mid-eighties, before the hyperinflationary outburst but after the debt crisis. Nevertheless, these rates are substantially lower than the ones observed in the seventies. Meanwhile, the substantial growth in private-sector money demand has allowed the Central Bank to accumulate a considerable sum of foreign currency over the last three years. As a result, the stock of international reserves (excluding gold) added up to over 15 billion dollars at the end of 1993. This amount is equivalent to almost one year of imports, but represents about one third of total monetary resources.

Other foreign sector indicators also evolved favorably. For one thing, privatizations and debt-equity swaps contributed to cutting down the stock of foreign debt by more than 8% between 1989 and 1992. Moreover, as a proportion of GDP, foreign debt fell from 38% in 1989 to 27% in 1993, even though the appreciation of the domestic currency was largely responsible for this result. Interest due payments, in nominal terms and also as a percentage of both total exports and GDP, diminished as well because of both the reduction in foreign debt and the fall in international interest rates.

Additionally, after complex negotiations, a final agreement with the creditor banks was signed in December 1992 under the auspices of the Brady Program, thereby normalizing Argentina's relations with the international financial community.

The Convertibility Plan, however, despite all its achievements, did not perform that well regarding other indicators.

In the first place, and in spite of the impressive growth rates of the last three years in aggregate output, economic activity during that period was extremely uneven. As a result of a sharp trade liberalization and a negative evolution of relative prices, the level of activity in the economic sectors and regions more exposed to foreign competition notoriously lagged behind the average trend, and in some cases contracted in absolute terms. Even more striking, output expansion was accompanied by a steady rise in unemployment rates, which jumped from 6.9% in May 1991 to 10.8% in May 1994.

The evolution of relative prices, in turn, had much to do with the dynamics of inflation. In fact, the rapid growth in nominal aggregate demand, and the persistence of implicit defensive pricing rules inherited from the high-inflation period, especially in the sectors supplying services, made the convergence of domestic inflation towards international rates a rather slow process. Under convertibility rules, and with an absolutely passive policy regarding capital inflows, this caused a sizeable appreciation of the domestic currency, impairing the competitiveness of the tradable sectors. Indeed, the real exchange rate, which at the beginning of the plan was almost 34% below its average level in the second quarter of 1990 (generally accepted, even in the government's view, as a reasonable reference for this variable) fell another 15% from April 1991 to the end of 1992.

Real appreciation, coupled with trade liberalization and a booming domestic absorption induced a sudden worsening of the trade balance. On the one hand, total imports grew rapidly in both volume and value during the last two years: they rose from an extremely low level of 4.1 billion dollars in 1990 to 8 billion dollars in 1991, and on to almost 16.8 billion dollars in 1993. Total exports, in turn, remained stagnant around the peak attained in 1990 in the aftermath of hyperinflation, when they reached 12.4 billion dollars. As a result, the huge 8 billion-dollar trade surplus achieved in 1990 was more than halved in 1990, and it has turned into a growing deficit since 1992: 2.6 billion dollars that year, and 3.7 billion dollars in 1993 (Table 1). Meanwhile, the deficit in real services sharply worsened as well, tourism expenditures abroad largely accounting for this outcome.

In spite of a sizeable reduction in financial services, which went down from 6.1 billion dollars in 1990 to less than 4 billion dollars in 1993, the deterioration in foreign trade and real services caused the current-account balance to revert from a 1.9 billion-dollar surplus in 1990 to a 2.8 billion-dollar deficit in 1991, which soared to almost 9 billion dollars last year (Table 1).

The rapidly growing deficit in the current account since 1991 has been totally financed through the capital account. In fact, the impressive amount of capital inflows even made a considerable increase in foreign reserves possible. As has already been said, private sector capital movements were mainly responsible for this outcome, accounting for a 26 billion-dollar inflow throughout the same period.

As can be concluded from this analysis, future economic prospects are simultaneously influenced by conflicting factors. On the positive side, inflation has roughly converged to international rates, foreign reserves look high enough to sustain the credibility of the economic policy, and the agreement reached with foreign creditors under the Brady initiative has eased Argentina's access to voluntary lending in the world financial markets.

Nevertheless, many difficulties still lie ahead. From a long-term perspective, the rate of investment is still low, and its current growth is being geared primarily toward production for the domestic markets and nontradable services, the recently privatized public utilities leading the process. Moreover, practically all the recent expansion in aggregate investment has been financed by foreign savings. National savings, on the contrary, has not yet reacted to ongoing structural reforms, still remaining at quite low levels.

From a shorter-term perspective, the rise in unemployment and the deep crisis underway in many regional economies are having obvious negative repercussions on income distribution and social cohesion, thereby threatening the sustainability of the program on political economy grounds. Additionally, the rigidity imposed on the exchange-rate policy by the Convertibility Law will make it hard to reverse the real appreciation of the domestic currency and will probably aggravate the problems of the tradable sectors, as well as the current-account deficit.

As a result, the external balance will remain highly dependent on the continuity of substantial capital inflows at a time when the international financial markets begin to show clear symptoms that their future evolution will not be as favourable to the country as it has been in the recent past. This trend, coupled with the growing disparity between the monetary liabilities of the domestic financial system and Central Banks's foreign reserves, is making the overall economic environment increasingly fragile before potentially destabilizing shocks, either external or domestic.

The other source of potential trouble for the prospects of stabilization is the fragility still displayed by the public sector. We will consider fiscal performance more carefully in the following section.

2. Fiscal performance after Convertibility

The "debt crisis" that emerged at the beginning of the eighties dramatically worsened the already chronic budgetary imbalances of the Argentine public sector.

In fact, the credit rationing which came up in the world financial markets as a consequence of the crisis made it necessary for the country to produce a permanent trade surplus to meet at least partially the large interest payments due on the previously accumulated foreign debt. The required adjustment had long-lasting negative implications on the overall functioning of the economy, constituting the well-known "external transfer" problem.

But that was not all. Given that almost all the foreign debt had become public by the time of the crisis, the transfer problem simultaneously had a "fiscal dimension": on the one hand, the only way the government could obtain the necessary foreign currency to face its enlarged foreign payments, since Argentina's foreign trade business is mostly in private hands, was by "buying" the trade surplus from the private sector; but on the other, the magnitude of the trade surplus to be transferred from the private to the public sector was extremely large when compared with either the revenue generating capacity of the tax system or the availability of domestic credit[3].

This "internal transfer" problem also had serious macroeconomic repercussions throughout the eighties. Because of the government's failure to induce a permanent increase in the tax burden, the repeated attempts to adjust the fiscal accounts so as to create the primary surplus required by interest payments were concentrated primarily on the expenditure side, and mostly in investment, with obvious consequences on the economy's growth performance.

Nevertheless, expenditure cuts were not enough either, so the government had no choice but to rely heavily on Central Bank financing and domestic indebtedness. Given the prevailing low levels of monetization, this move exerted enormous pressure on both inflation and real interest rates, exacerbating the adverse economic effects brought about by the "external transfer" problem.

By the end of the eighties, the failure to close both the external and fiscal gaps simultaneously led the economy to hyperinflation. Notwithstanding, this traumatic experience had one positive outcome: it created a widespread demand for economic stability and made a sustainable fiscal adjustment feasible.

The current Administration, in office since 1989, has taken advantage of this opportunity by promoting several ambitious reforms in the sphere of the State, which at a first stage have focused primarily on the tax system and the privatization of public enterprises. These reforms, together with the implementation of the Convertibility Plan and the positive change experienced by the external environment, have produced a remarkable fiscal improvement.

In the first place, tax receipts have experienced a dramatic increase over the last three years. The impressive recovery in tax performance, which represented about five points of GDP including both federal and provincial taxes, was due to several reasons. First, the success of stabilization in sharply reducing inflation while rapidly expanding output. Second, the impact of stabilization was greatly amplified by the decision to concentrate the tax effort on a broadened VAT and a few other taxes, and by the outstanding improvement achieved in tax administration, particularly regarding collection and control procedures. Last but not least, the private sector showed an increasing disposition to pay conventional taxes.

As a result, tax pressure at the federal level jumped to an unprecedented 16.6% of GDP in 1992, four points higher than in 1990, remaining roughly constant the following year. At the same time, current income (which also includes the public firms' surpluses and other nontax revenues) reached 17.5% of GDP in 1992, and stayed practically at the same level in 1993 (see Table 2).

However, nonfinancial current expenditures also displayed a growing trend, climbing to about 15% of GDP in 1992-93, mainly as a consequence of an adverse evolution of transfers to both the social security system and the provinces.

The resulting primary savings were about 2.5% of GDP in the same period. This level, though much higher than the one attained in 1991, was roughly similar to that of 1989, and even lower than that of 1990 (Table 2).

The evolution of current savings, in contrast, proved to be much more favourable, reverting from a negative 2.7% of GDP in 1989 to a positive 1.4% in 1993. An unexpectedly swift decrease in foreign debt interest payments, which went down from 4.4% of GDP in 1989 to 1.0% in 1993, helped explain this outcome (Table 2). Indeed, the lower interest payments were the consequence of three different factors. First, through privatizations and debt-equity swaps, the government achieved a modest reduction in its overall foreign indebtedness. Second, and more important, the government benefitted from the exogenous change in world financial markets, which brought about a sharp decline in international interest rates. Finally, the real appreciation of the exchange rate induced by stabilization had, from a fiscal perspective, the positive side effect of reducing the burden of foreign interests when measured as a proportion of GDP.

Cash revenues arising from privatizations, which were especially necessary at the beginning of the stabilization process, extremely low levels of public investment, and an almost balanced consolidated budget of public firms also contributed to the impressive fiscal adjustment of the early nineties.

Considering the evolution of the national government's overall balance, the turnaround in the fiscal stance looks indeed so impressive that it is difficult to think that a permanent solution to the "domestic transfer" problem of the eighties has not been finally reached. A closer scrutiny reveals, however, that the current fiscal position is still quite fragile.

There are several reasons which support this assessment. First, there is the role played by privatizations in the last three years. As was already mentioned, their cash proceeds, which averaged over one point of GDP in 1991-93, helped close the fiscal gap, especially in the early stages of stabilization. But their importance in strengthening the financial position of the government was much greater because they also contributed to reducing the public debt through the implementation of debt-equity-swap schemes.

Privatizations are, nonetheless, a temporary instrument. In order for the current fiscal equilibrium to be sustainable, they will eventually have to be replaced by revenues arising from other sources or by expenditure cuts.

Second, current trends in world financial markets suggest that Latin American countries as a whole are likely to face higher interest rates and less capital inflows in the near future. Besides the direct impact of higher rates on interest due payments, the fiscal balance could also be affected by the negative effects of a tightened external constraint on domestic economic performance. When it finally becomes necessary, external adjustment will inevitably mean a slowing down of economic activity and, consequently, lower tax revenues. Likewise, if the adjustment also requires an exchange rate devaluation, the real burden of interest payments will additionally be heightened through this channel.[4]

The fragility still displayed by fiscal accounts is also explained by domestic factors. For one thing, economic policy will have to face enormous pressure to raise nonfinancial current expenditures. In fact, overall current public expenditure at both the federal and provincial levels have already increased from 25.3% in 1990 to 27.5% in 1993. In the immediate future, the challenge will mainly come from the political front: general elections will be held in mid-1995 and Mr. Menem is expected to run for his second term.

But such a challenge will hardly disappear after next year's elections for at least two reasons. One is the ongoing crisis in the regional economies. So far, regional difficulties have prevented the provinces from replicating the fiscal adjustment attained at the federal level despite the national government's many attempts to persuade provincial administrations to do so, through either financial benefits or threats with severe penalties in the distribution of federal tax revenues.

The second reason is that public investment has almost vanished and cannot go anywhere but up. As a matter of fact, and leaving aside investments made by public firms (because they are obviously much lower than in the past after privatizations), overall capital expenditures in the public sector have experienced a dramatic decrease, even when compared to the already meager levels in the late eighties. Consequently, if a collapse in the economic and social infrastructure is to be avoided, they will no doubt have to be raised substantially.

There is still the impact of the recently implemented social security reform to be considered. No matter how relevant this structural reform is to enhancing national savings and fostering the development of domestic financial markets, the transition from the old regime to the new will certainly have significant and enduring fiscal costs.

Indeed, social security reform has established that a restructured public pay-as-you-go system will coexist with a new capitalization scheme to be operated by private pension funds. The problem is that the transition to the new regime will inevitably produce a financial imbalance in the remaining public system: while outlays will stay basically unaffected by the reform, as long as they depend on the number of passive workers retired under the old regime, revenues will suffer a sudden fall as the active workers who now choose the capitalization scheme shift their payments to the private pension funds. First estimates suggest that the initial loss in revenues will be over three quarters of a percentage point of GDP.

We have already advanced a number of reasons why the sustainability of the present fiscal equilibrium cannot be taken for granted. If anything, they all stress the vital importance that future tax performance will have in preserving the fiscal balance.

However, some symptoms have already appeared suggesting that at some point during the first half of 1994 tax revenues might have reached a plateau, which will not be easy to surpass in the near future. In fact, for the first time since Convertibility, the June 1994 federal tax collection was lower than it had been a year ago, and the same happened in July. A gradual cooling down of economic activity and an incipient fatigue in tax compliance on the part of the private sector might explain this trend.

Moreover, fiscal revenue prospects are further endangered by a complex policy dilemma regarding the targets of tax policy. In fact, while convertibility rules have extremely reduced the feasibility of active monetary policy, and the government's commitment with trade liberalization precludes the systematic use of import restrictions (though there have been some exceptions), fiscal policy is practically the only remaining instrument for macroeconomic management.

But fiscal policy, which given the prevailing constraints on public expenditure management basically means tax policy, is subject to conflicting demands. On the one hand, as we have already emphasized, it is of critical importance to prevent the emergence of fiscal imbalances from jeopardizing economic stability. On the other hand, it is the only instrument available to counteract the most negative side effect of the ongoing stabilization process, namely the impact of exchange-rate appreciation on competitiveness and unemployment.

In fact, the need to deal with an overvalued domestic currency explains the recurrent attempts to produce a so-called "fiscal devaluation", in other words, to improve effective exchange rates by resorting to tax policy. Along this line, a first policy package was announced in October 1992, including export rebates, fuel tax cuts, and the elimination of tax stamps and other minor taxes. Since then, other comparable measures have followed, the most important one being a selective reduction of employers' contributions to the social security system in tradable sectors. Furthermore, out of concern over the recent trend in unemployment, the government has lately announced that this last measure will be extended to nontradable sectors as of 1995.

Preliminary estimates suggest that, all measures included, the improvement in nominal effective rates brought about by this "fiscal devaluation" does not exceed 5%, quite a modest figure considering the actual level of real appreciation. However, preliminary calculations also estimate the fiscal cost of these measures to be substantial, ranging between 1.0% and 1.5% of GDP.[5]

In sum, the sustainability of the current fiscal equilibrium will largely depend on how successfully the government handles the tax policy dilemma we have just discussed. With this in mind, we will now examine the tax reform currently underway in greater detail. Let us start considering its initial conditions. 3. The evolution of the tax system throughout the eighties

During the eighties the tax structure was hit by an extremely unstable macroeconomic scenario. Sharp fluctuations in tax revenues were much more the result of macroeconomic factors, such as high inflation, relative price volatility and wide output swings, than the consequence of changes within the tax system.

As we have already seen in the previous section, the external adjustment imposed by the debt crisis had strong negative repercussions on the fiscal accounts. Since the foreign debt was mainly concentrated in the public sector and the foreign exchange generated by the trade surplus was in private hands, fiscal imbalances worsened sharply and became one of the dominant features of the economy.

Recurrent attempts to cut the fiscal deficit were made throughout the period. Each of those attempts included a tax package aimed at sharply increasing the tax effort. Their design, primarily guided by short-run revenue considerations, fell too short of what a comprehensive tax reform should have been. In fact, they heavily relied on temporary "emergency" taxes, and could not properly take into account either efficiency or equity matters. As a result, the fiscal accounts usually showed an ephemeral recovery after every adjustment, but the achievements did not last long.

The chronic fiscal gap and the volatility of the macroeconomic environment submitted the tax system to permanent pressure, and prevented it from consolidating a definite pattern. Consequently, regarding its tax structure, Argentina's performance during the eighties seems to have followed a hybrid path.[6]

The behaviour of both direct taxes and VAT, in particular, was rather unsatisfactory. On the one hand, revenues arising from income and wealth taxes were too low: 1.6% of GDP on average over the whole decade, representing about 13% of federal tax revenues.[7] It should be mentioned, though, that Argentina had already shown a poor record on direct taxation in the seventies when economic conditions were much better.

On the other hand, the country adopted the value added tax in the mid-seventies, earlier than most Latin American economies. Until very recently, however, experience with this tax has also been disappointing. From the beginning, its base was not comprehensive, and a multiple rate scheme was implemented. By 1980 the tax base was broadened, and in 1981 the tax rate was increased to 20%. This enabled VAT collection to reach 3.8% of GDP that year, but the attempt proved to be short lived. Between 1982 and 1984 new exemptions were granted for public services, foodstuffs and medicines. Moreover, the tax rate was progressively reduced until it marked only 14% in 1990. These changes, coupled with a deteriorating economic activity and declining imports, led to a fall in VAT revenues, which on average amounted to slightly more than 2% of GDP (16.5% of total federal collection) from 1984 to 1990.

Due to the meager performance of both direct and value added taxes, the share of excise taxes in total collection gained enormous relevance during the past decade. In fact, excise taxes represented on average one-third of the federal tax burden over that period. Fuel tax revenues, in particular, were roughly equivalent to those stemming from VAT.[8]

Wage income was the other significant source of federal taxation. The funds coming from this source, representing almost 25% of the total, are earmarked for the social security system. Nevertheless, tough wage taxes are quite high, the structural crisis of the system demanded additional funds, and even so pensions were kept well below legal standards.

Finally, federal taxation was completed by trade taxes. On average, they comprised around 14% of the total but were subjected to sharp fluctuations owing to the instability of trade policies throughout the period, particularly regarding import tariffs and export tax rates.

As has already been said, neither efficiency nor equity implications of the tax system were a major issue during the eighties. The inflationary environment and the volatility in tax revenues help explain this fact. However, during the first half of the eighties, while the economy was plunging into deep stagflation, the government authorized massive tax expenditures. Though phrased in terms of regional and sectoral targets of industrial policy, the social costs and benefits of the decision were not duly taken into account. Indeed, tax expenditures were appropriated by a small number of firms, while there was hardly any impact on aggregate employment or investment.

Anyhow, despite temporary improvements, the evolution of the overall tax burden (considering both federal and provincial taxes) showed a declining trend over the decade. Starting from over 17.5% of GDP in 1980, total collection dropped steadily until 1983, reaching 11.7% of GDP that year and remaining at about the same level in 1984. Partly as a consequence of the tax package implemented during 1984 and 1985, but largely as a result of the temporary stabilization achieved by the Austral Plan, the tax burden recovered its 1980 level in 1985-86.

This significant improvement, however, was basically brought about by increases in both export taxes, which later proved politically unsustainable, and wage taxes. To a lesser extent, other equally distortive taxes, such as the fuel tax and that imposed on bank drafts (which reached 0.7% of GDP in 1988), also contributed to the same end. Additionally, a compulsory savings regime was established from 1985 to 1989, which in practice worked as an "emergency" tax.

Instead, other measures aimed at changing the tax structure in a deeper and more permanent way did not produce the expected outcome. That was the case, in particular, of the measures taken to improve the performance of direct taxes and the VAT, which failed to induce any relevant rise in revenues. To a great extent, that was because they were heavily contingent on administrative resources which were simply unavailable.

Be that as it may, the recovery did not last long. In 1987, the Austral Plan began to weaken and inflation accelerated, leading to a renewed fall in tax revenues. As a result, the average total tax burden for 1988-1990 stayed below 15% of GDP.

Moreover, the deterioration in tax collection attained unprecedented levels during the two hyperinflationary outbursts, in the second quarter of 1989 and the first quarter of 1990. In both periods the federal tax burden fell beneath 10% of GDP, reaching its lowest levels in the decade.

4. The current process of tax reform

After the failure of a first stabilization attempt, which gave rise to the second hyperinflationary outburst in late 1989 and early 1990, President Menem's administration firmly adopted the market-friendly approach to structural reforms advocated by the World Bank and the Washington Consensus. The public sector thus became one of the chief reform targets. As a consequence, both tax reform and privatizations were placed at the top of the government's agenda.

Regarding tax reform, it is possible to distinguish two different stages. First, the priority was given to increasing revenues by any means, in order to overcome the effects of hyperinflation on the fiscal accounts. Second, still in progress, more attention is being paid to allocative and distributive issues. As a result, the tax structure is acquiring a new profile, quite different from the one it had in the eighties.

In the midst of hyperinflation, as was already mentioned, tax collection reached extremely low levels. In the first half of 1989, total tax revenues in real terms were 22% lower than those in the same period of 1988. The Olivera-Tanzi effect was the basic underlying factor: according to official estimates, during the second quarter of 1989 the loss due to this effect amounted to 31.5% of earned revenues (equivalent to 3.2% of GDP) at the federal level.[9] Moreover, the lack of private credit and high real interest rates motivated long delays in tax payments.

The performance of taxes linked to domestic activity was also affected by a deep recession and the virtual interruption of economic transactions: revenues arising from VAT, excises and income tax were reduced to about one-half of their level in the second quarter of 1988. The extreme instability of the exchange rate regime, in turn, impaired the performance of trade taxes.

The sudden fall in inflation brought about by the first policy announcements of President Menem's administration fueled a rapid recovery in tax collection from July to November 1989. The reverse Olivera-Tanzi effect was responsible for about 40% of the recovery, but several emergency measures were also adopted -a major contribution being higher taxation on exports, which tried to capture at least part of the windfall gains arising from the steep devaluation of the domestic currency.

Despite the above-mentioned second hyperinflationary episode in late 1989 and early 1990, a second stage in tax policy has been developing over the last three years, particularly since the 1991 stabilization program. In this new stage, a series of successive reforms seems to be attaining a permanent improvement in tax collection, while also shaping a new tax structure. The core of the reform strategy seems to be a simpler system, eliminating all minor distortive taxes, and placing VAT at the center of the new scheme.

Indeed, the VAT base has been broadened to unprecedented levels (including agriculture and all major service activities), while multiple rates have been eliminated and the remaining general rate has been strongly raised from 14% in 1990 to 18% as of 1992. At the same time, VAT compliance has been enhanced through several administrative mechanisms, such as widespread withholding and standardized billing, which are diminishing evasion, particularly by small businesses and informal trading.

More recently, important steps have also been taken concerning the reform of the income tax. First, Congress failed to enact the administration's extremely innovative proposal of a tax on the net surpluses of firms. This tax was intended to replace both the traditional income tax and employer's contributions to social security, altering tax incentives in the right direction, while radically simplifying tax administration, because of the close connection existing between its base and that of VAT.

In the end, a more conventional reform was passed. With regard to the corporate income tax, besides increasing the tax rate from 20% to 30% in April 1992, the restrictions applied to deductions of past losses arising from business declarations were the most important aspect of the reform.

Deductions of past losses were a major drag on corporate income-tax collection throughout the eighties, due to abuses with tax-expenditure schemes perpetrated by discretionary industrial policies, and also because of the negative impact of inflation-adjustment accounting mechanisms on taxable profits. Owing to the sharp reduction in inflation and the drastic cut in tax expenditures, it is unlikely that this issue will represent a serious problem in the future. However, something had to be done with the declared losses accumulated in previous years but not yet deducted in subsequent declarations. The solution adopted in 1993 was to transform those losses into fiscal credits at a uniform rate of 20%, and to pay the credits with long-term public bonds (Bonos de Consolidacin de Deudas).

Regarding personal income tax, the marginal rate was cut down from 45% to 30% and the number of income brackets was reduced to six. The major innovation, however, was the suppression of the nontaxable minimum income. Instead, taxpayers incorporated into the social security system will be allowed to partially compute their monthly contributions to the system as payments in advance in their annual income tax declarations. It is expected that this measure will help diminish evasion in the social security system, especially among self-employed workers and professionals.

Although some minor improvement in revenues has already been achieved, the full effect of income tax reform is likely to be perceived from 1994 onwards.

Other important reforms have comprised the elimination of virtually all tax-expenditure schemes (measure included in the Economic Emergency Law of 1990), the suppression of export tax and over twenty other distortive taxes (i.e. tax stamps, that on bank drafts, etc.), and a simplified collection and distribution mechanism for the fuel tax.

It is also worth mentioning the 1993 administrative decision to assign the Federal Tax Bureau (Direccin General Impositiva) the responsibility for collecting social security contributions, previously under the Ministry of Labor's competence, in order to enhance tax compliance in this area.

In contrast, the institution of a corporate gross asset tax, similar to the one applied in Mexico and other Latin American countries in 1990, did not produce the expected outcome. Despite increasing the tax rate from the initial level of 1% to 2%, the proceeds remained negligible (less than one-quarter of a percentage point of GDP), so in 1993 the government finally decided to eliminate it. The process will be gradual, incorporating different sectors and regions progressively, and is expected to be concluded by mid-1995.

Additionally, a much more rigorous legal framework was established in 1990, when the Ley Penal Tributaria (No. 23771) was passed by Congress, intending to foster tax compliance by imposing severe penalties (including imprisonment) on evasion. By the end of 1993 over 1,700 cases had been opened under this new law.

Nevertheless, it is highly probable that the impact of all these reforms in actual tax collection would not have been so positive had it not been for the impressive improvement observed in the design and functioning of the Direccin General Impositiva (DGI). Indeed, the betterment in tax administration and the consequent reduction in evasion were especially relevant for two reasons: 1) they enabled the government to increase revenues without resorting to emergency taxes, and b) they improved horizontal equity, in particular regarding VAT collection.



As can be seen in Table 3, the restructuring of the Tax Bureau involved higher wages and a swift increase in personnel. In fact, the number of employees jumped from 10,500 to 16,000 in just three years (from November 1989 to November 1992) and according to the 1994 Federal Budget it is expected to reach 26,500 this year, with average wages rising from 1,100 to 1,800 dollars. The restructuring also involved major changes in working routines, the adoption of wage incentives associated with performance, and a new wage scale encompassing greater pay differentials. Administrative efficiency was further pursued through technological upgrading: personal computers in use numbered only 20 in 1989 but rose to 1,164 in 1992; more powerful equipment was incorporated as well.

Meanwhile, a much larger share of the Tax Bureau staff was assigned to monitoring and controlling tax compliance (6,500 employees in 1992 as compared to 1,400 in 1989). As a result, the number of monthly inspections rose from 360 to 1,800 during the period considered. Likewise, temporary closing down of establishments responsible for tax offenses jumped from 751 in 1990 to over 16,000 in 1993.

The reform in tax administration was truly effective. While nominal revenues doubled from November 1989 to November 1992 (in fact they grew another 75% from then to mid-1994), collection costs lagged far behind: they represented 3.2% of total revenues in November 1989 but dropped to only 2.7% in November 1992.[10]

Altogether, the reform process has been quite successful so far. As a matter of fact, tax collection at the federal level increased by more than 50% in real terms from its yearly average in 1988-90 to 1992. As a consequence, the federal tax burden climbed from an average 12.4% of GDP in the former period to a historic peak of 16.6% in the latter. Adding to that figure the 3.2% of GDP collected at the provincial level, the total tax burden amounted to almost 20% of GDP in 1992, about 36% higher than in 1988-90, and it remained around that level in 1993.

As can be inferred from Table 4, the federal tax structure also changed dramatically, and basically in the right direction regarding economic incentives. The most relevant feature of this change was the central role won by the VAT. In 1989 the value added tax collected 1.6% of GDP, representing 12.4% of the federal tax burden. In 1993, it collected 6.1% of GDP, or almost 38% of that burden.

At the same time, revenues arising from social security contributions (wage taxes) increased sharply as well. They climbed from 2.7% of GDP in 1989 to 4.9% in 1993, while their share in the federal burden rose from nearly 21% to over 30% during the same period. Nonetheless, as was already explained, the initial impact of the ongoing social security reform will be the reduction of the funds going to the public segment of the system.

Income and wealth taxes, instead, did not perform that well. Indeed, their joint revenues dropped slightly from 2.4% of GDP in 1989 to 2% in 1993. As a result, their share in the federal burden also diminished, falling from 18% to 12.4% of the total. As was already mentioned, however, an improvement in income tax collection is expected as from 1994.

Nevertheless, the fact is that both VAT and social security contributions produced a huge increase in fiscal revenues: contrasting 1993 to 1989, this increase represented 6.7% of GDP. About half that total explained the overall growth experienced by federal tax collection between those years, while the balance made it possible to sharply reduce the burden of trade and other highly distortive minor taxes in GDP.

As a consequence of the virtual elimination of export taxes and the significant reduction in import tariffs, foreign trade taxes dropped from 2.8% of GDP (almost 22% of the federal burden) in 1989 to 0.9% of GDP (5.5% of federal collection) in 1993. In turn, minor taxes which jointly totalled 1.7% of GDP (13% of the federal total) in 1989, fell to 0.7% of GDP (4.% of that total) in 1993.

Finally, excise taxes (including fuel) grew only slightly from 1989 to 1992 but dropped again in 1993, reaching only 1.5% of GDP and a share of less than 10% in the total tax burden that year.

5. Reasons behind the improvement in tax performance[11]

The previous section showed the significant increase in fiscal revenues which took place over the last three years. In fact, the federal tax effort reached 16.1% of GDP in 1993, almost 1.5 points higher than the preceding historic maximum attained in 1986, and about 3.5 points higher than the 1980-1991 average (Table 5). The purpose of this section is to identify and quantify the main factors explaining that outcome. Of course, this is just a preliminary analysis intending to establish approximate orders of magnitude among those factors. A deeper appraisal would demand a much more disaggregated analysis and a data base which is currently unavailable in the Argentine case.

For our purpose, we have compared the performance of tax receipts in 1993 with their performance in some selected years of the past decade: 1981, 1986 and 1989. This choice was not arbitrary. Indeed, given the extreme economic instability which characterized the eighties, it enabled us to make a coverage of different macroeconomic and fiscal contexts.

1981 marked the commencement of the debt crisis, and both aggregate economic indicators and tax performance began a fast worsening process. Owing to a broad legal base and a high rate, however, VAT collection was quite important in 1981. The Austral Plan launched in mid-1985 temporarily interrupted the declining trend of the Argentine economy. As a result, tax performance reached the already-mentioned historic maximum of 1986. In 1989, instead, the country was in the midst of hyperinflation and tax revenues fell again to a low level.

From Table 5, it is immediately apparent that VAT and Social Security contributions were practically the only factors responsible for the recent improvement in tax performance. In effect, as a percentage of GDP, revenues originating in both taxes were 5.3, 4.6 and 6.7 points higher than those of 1981, 1986 and 1989 respectively. In all three cases,indeed, this rise was even greater than the change in the overall tax burden, while the contribution of income and wealth taxes stayed roughly constant and that of the remaining taxes actually declined.

Therefore, in order to explain the current positive changes in overall tax collection, it is both necessary and sufficient to identify the main factors behind the recent good performance in VAT and Social Security contributions.[12]

Let us first discuss the VAT case, which has been responsible for the largest revenue increases since the beginning of the current process of tax reforms.

This tax is levied on both domestic transactions ('VAT-national') and imports ('VAT-imports'). Revenues arising from the latter are obviously much smaller than those coming from the former. Consequently, according to the reference year, the growth in VAT-imports explains only 10% to 15% of the growth in total VAT receipts, the remaining proportion being explained by VAT-national.

As can be seen in Table 7, regardless of the base year considered, almost the entire increase in VAT-imports comes from changes in the tax base, while changes in both the tax rate and "other" factors were of minor importance. The main determinants of the tax base are the dollar value of imports, the effective import exchange rate, and legal exemptions.

The recent growth in this variable, which reached its highest level in 1993 (Table 6), is to be fully credited to the behaviour of import volumes. As a matter of fact, a swift trade liberalization has compounded with a sizeable appreciation of the exchange rate to produce an import boom over the last three years. As a result, purchases abroad nearly tripled in dollar terms from 1989 to 1993, more than compensating for the appreciation of the domestic currency.

In the case of VAT-imports, collection procedures are simple and well established. Tax evasion, then, is not a major issue here. This is the reason why effective revenues are not very different from the theoretical ones in Table 6, and "other" determinants of VAT-imports performance are almost negligible in Table 7.

Estimating the determinants of the changes in the VAT-national is a more complex task, basically because of the difficulties in rigorously calculating its tax base. In the first place, in order to properly identify all legal exemptions, it would be necessary to work with a disaggregation of GDP of more than four digits. In the second place, given that the sectoral disaggregation of final consumption is also inaccessible, a recent estimation of the input-output matrix would be needed. Finally, a sectoral disaggregation of investment and exports would also be required.

Since all these elements are unavailable in the Argentine case, we had to choose a different approach. Based on previous calculations for 1981, 1986 and 1989, and our own estimates for 1993, we determined the taxable sectoral products at a two-digit disaggregation level. Deducting export and investment fiscal credits, we obtained provisional estimations of the net tax base and the theoretical tax collection. At the same time, based on effective revenues and legal lags, a series of earned revenues was constructed.[13]

As can be seen in Table 8, the tax base (after deductions) was over 30% in 1981, dropped to less than 20% in the late eighties, and strongly recovered in the early nineties reaching 41% in 1993. In fact, taxable GDP (before deductions) in the case of VAT-national was estimated at about 65% of total GDP for that year, which is quite high according to international standards. Meanwhile, the tax rate fell from 20% in 1981 to 18% in 1986, and to a further 15% in 1989, only to recover its 1986 level as of 1992.

As a result of the evolution of both the tax base and rate, theoretical collection reached 7.4% in 1993, a level that more than doubled the figures for 1986 and 1989 and was even higher than the previous 1981 peak. The other remarkable fact in Table 8 is the improvement evidenced by the effective/theoretical revenue ratio: after marking 50% in 1981 and 1986, it dropped to less than 40% in 1989, and then recovered to almost 70% in 1993.

From the evolution of theoretical collection it is rather simple to calculate the contribution of changes in the tax base and rate to the evolution of VAT-national revenues. The difference between theoretical and earned receipts, in turn, was attributed to "other" determinants, the effects of changes in tax administration being included in this category. Finally, a "lag" effect, encompassing but not identical to the Olivera-Tanzi effect, was included to account for the difference between earned and effective revenues.

Results are reported in Table 9. Changes in the tax base, mainly due to legal revisions, are the most important factor explaining the 1993 improvement in VAT-national revenues, particularly when compared with 1986 and 1989. This factor is somewhat less relevant when compared to 1981 because the VAT-national also had quite a broad base that year.

The significance of "other" determinants, especially when the reference years are 1981 and 1989, at least partially reflects recent improvements in tax compliance.[14] Instead, tax rate changes and the "lag" effect seem to have played a less important role as determinants of the VAT-national growth.

With regard to the performance of social security contributions revenues arising from salaried workers represent more than 80% of the system's total receipts. Therefore, the evolution of aggregate wages was taken as a proxy for the evolution of the system's tax base.

As can be seen in Table 10, changes in the tax base have played a negative role in the behaviour of social security contributions in 1993, particularly when compared with their performance in 1981 and 1986. This can basically be attributed to the fall experienced by real wages over the last decade. To some extent, the recent increase in output per worker has also strengthened the negative trend in the tax base. Therefore, the significant improvement in social security contributions is essentially explained by two factors: tax rate hikes, most relevant in the comparisons with 1981 and 1986, and a positive "lag" effect, crucially important in the comparison with 1989. Unlike the importance shown by "other" determinants in the case of VAT-national, results regarding their contribution are not conclusive in this case.

6. The role played by privatizations[15]

As was already mentioned, the privatizing of public firms was one of the major goals of President Menem's administration.

Beginning in 1990, this process evolved at an extremely fast pace. The calls for bids attracted offers from a large number of local and international companies. By the end of 1992 most public utilities and oil fields had already been sold. In 1993, moreover, the government undertook the privatization of YPF, the state-owned oil company and Argentina's largest concern. As a first step in that process, 40% of the company's shares were sold to domestic and international investors in June last year. Once completed, very few companies will remain in the hands of the public sector.

The immediate beneficiary has been the national government. Including the sale of YPF shares, it has collected around 17.3 billion dollars since the beginning of privatizations: 9.5 billion dollars in cash, the rest being the market value of all other payments involved (see Table 11). So far, these funds have been an important factor in explaining the fiscal improvement, as well as the government's ability to service the public debt without resorting to inflationary financing.

In both the market-friendly literature and the current rhetoric employed by governments in many developing countries, privatizations, together with deregulation, are considered basic tools for enhancing both the productivity of existing resources and the efficiency of investment decisions. Indeed, they are expected to produce such effects through their incentives on private initiative and creativity. They are thus supposed to work mainly through microeconomic channels.

Nevertheless, in most actual cases, privatizations seem to have served other purposes as well. Argentina is a good example. With an outstanding performance regarding both the relative magnitude of the privatized assets and the speed exhibited in completing the process, Argentina's privatizations have fulfilled some crucial functions which were very different from those advocated by the standard approach. Moreover, given the way they were implemented, and even though it is still too early to assess their long-run impact on productivity, there seem to be justified doubts on their final contribution to overall efficiency and growth.

As was also the case with tax policy, two different stages can be identified regarding the privatization process. The first covered 1990 and the first quarter of 1991, while the second can be associated with the Convertibilty Plan launched in April 1991.

During the first stage, the economic situation was very bad. Even though the government had managed to curb the second hyperinflationary outburst in March 1990, inflation remained very high (fluctuating around a 10% monthly rate) and a recessionary trend persisted throughout the rest of the year.

In such a context, one of the main functions of the privatization program responded to political economy considerations. Given that the credibility of the administration was questioned for the persistency of the economic crisis, the continuity of the privatization process was meant to regain support from the business community, attesting that the government was sticking to its original commitments. At the same time, privatizations were presented to public opinion as the only way out of the economic crisis.

In order to meet the administration's political needs, then, the first major privatizations were submitted to tight deadlines. A thorough assessment of truly important economic issues, such as the pricing policy of the privatized firms, or the establishment of regulatory frameworks according to the new realities, was dismissed or postponed when it threatened to delay the process.

The second relevant role played by privatizations was related to macroeconomic stabilization. In fact, the narrowness of the domestic financial markets, aggravated by two hyperinflationary episodes and the compulsory rescheduling of short-term bank deposits and public bonds in January 1990, made it compelling to find a temporary way to finance the remaining fiscal gap so that the intended adjustment could have enough time to produce its full effects. As had previously been the case with the bridge loans provided by multilateral institutions, privatization revenues served that purpose too.

Taking into account the political and financial constraints faced by the government in 1990, the public firms chosen to start the process were those which could more easily attract the interest of the private sector: Entel (the public phone company), and Aerolineas Argentinas (the state-owned airline). In both cases the new owners were granted extremely high real tariffs, as well as the preservation of quasi-monopolistic conditions in their respective markets. At the same time, the government initiated the concession of oil fields to the private sector.

The Convertibility Plan marked the second stage in the privatization process. The success of stabilization helped the government recover credibility. Political considerations became somewhat less prominent in orienting that process. However, given the clear aversion still evidenced by domestic financial markets to meet government borrowing requirements, and despite a significant improvement in tax performance, privatization cash receipts, together with the sharp decline in foreign debt interest payments, continued to be relevant in closing the fiscal gap from 1991 to 1993.

In fact, in 1991 privatization cash receipts were slightly higher than primary savings and covered almost all foreign interest payments. Primary savings more than doubled in 1992-93, but privatization cash revenues still represented over 50% of those savings and almost 100% of foreign interest payments in both those years (see Table 2).

Privatizations were also critically important in the reduction of the outstanding public debt. From the beginning of the program to the end of 1993, the government had already rescued the equivalent of about 13.3 billion dollars (face value) in titles of both domestic and foreign public debt. In particular, the exchange of public foreign debt titles for equity constituted the bulk of the first two major privatizations (Entel and Aerolineas). These operations alone enabled the government to rescue 6.5 billion dollars in foreign titles (about 20% of the debt with commercial banks at that moment), therefore making it possible for Argentina to join the Brady Program.

As was already mentioned, in 1993 the government undertook the privatization of YPF, the state-owned oil company, initially selling 40% of the company's shares. The revenues of this operation amounted to slightly less than 2.5 billion dollars. Nearly half of that sum was destined to pay federal debts with the provinces, the remainder being used to diminish the government's long-standing debt with pensioners.

Besides contributing to strengthening the fiscal situation, privatizations also played a key role regarding the evolution of the economy's external accounts. In fact, capital inflows have been crucial in closing the current account's growing imbalance since the beginning of the Convertibility. Privatization cash revenues were an important component of these inflows: more than one-third of the private capital account surplus over the last three years can be directly or indirectly attributed to privatizations.

Having discussed the short-term effects of privatizations on stabilization, what can be said about their long-term impact on efficiency and growth?

It is widely acknowledged that the sustainability of the current stabilization process, and the prospects of growth resumption in the Argentine economy are critically contingent on two factors: 1) the rise in both the national savings rate and the investment rate, and 2) the improvement in the competitiveness of the tradable sectors. The long-run effects of privatizations should therefore be assessed regarding these factors.

Nevertheless, even though it is still too early to fully estimate such effects, the way privatizations were conducted casts some doubts on their final contribution to overall efficiency and growth.

Let us consider, in the first place, the different ways in which the domestic private sector can finance the purchase of public firms. Macroeconomically speaking, there are no more than four options (and all possible combinations of them): 1) a rise in private savings, 2) a fall in private investment, 3) an increase in the private-sector flow demand for credit, or 4) a decline in the private-sector flow demand for financial assets. The first two options imply that the private-sector financial surplus (deficit) is growing (diminishing); the third and fourth ones that, given the private-sector financial balance, its composition is changing.[16]

From this point of view, privatizations could have a positive effect on growth provided they induce a rise in private savings while keeping public savings constant, thereby enabling an increase in aggregate investment. Given that the additional private savings will be transferred to the public sector, this requires that the government not use privatization earnings to meet current expenses or debt payments.

However, it is pretty obvious in the Argentine case that privatizations were not financed by higher private savings. Instead, they were basically funded by a combination of higher private foreign indebtedness (option 3), and the repatriation of flight capital (option 4). Moreover, prevailing incentives since the beginning of the Convertibility have been such that the massive capital inflows helped finance not only privatizations, but also a boom in private consumption, thereby inducing a sort of crowding-out effect on private savings.

Considering the evolution of the fiscal accounts (see Table 2), it is also evident that privatization earnings did not fund higher levels of public investment, but rather replaced other financing sources unavailable to the government during the recent period. In that sense, it could be argued that the use of privatization revenues to meet current expenses and debt payments had a somewhat crowding-out effect on aggregate investment.

There is a different way in which privatizations could have a positive effect on growth. Most privatization contracts impose investment commitments on the new owners of the former public firms. These investments are expected to have a favourable effect on the productivity of those firms.

But in order to improve overall competitiveness, it is necessary for those productivity gains to spill over the rest of the economy. To make privatizations more attractive, however, the government authorized significant rises in the prices charged by former public firms, which are now at a historic peak in real terms. In fact, this event, coupled with the real appreciation of the domestic currency, is impairing the competitiveness of the economy.[17]

There is also a timing problem connected with the sustainability of stabilization because privatized firms are mainly located in nontradable sectors. Even if a proper regulatory framework could ensure that productivity gains be passed on to the rest of the economy, the positive effects of an investment spurt in nontradable sectors on the economy's international competitiveness would be far from immediate. In the meantime, the government might well have to face the familiar trade-off between a growing external imbalance and recession, there by jeopardizing the prospects of the stabilization effort.

7. Concluding remarks

The current performance of the Argentine economy provides new evidence for the ongoing debate about the proper sequencing between stabilization and fiscal reform.

If anything, the initial failure of President Menem's administration to stabilize the economy, as well as its subsequent success under the Convertibility Plan, confirmed that when inflation surpasses a certain threshold, stabilization cannot wait until the structural reforms reshape the public sector according to market-oriented principles. Some kind of shock therapy is needed to first bring inflation down to less disruptive levels, say under 20%. The Convertibility Plan did so mainly by establishing the exchange rate as the nominal anchor of the economy.

Of course, this does not at all mean that fiscal reform is irrelevant. It means rather that fiscal reform is less concerned with the generation of stabilization than its sustainability.

Nevertheless, the Convertibility regime imposed severe constraints on the government's ability to finance fiscal imbalances. Two different stages can be distinguished according to the way in which these constraints were met.

During the first stage the priority was given to increasing fiscal revenues by any means. Not only tax policy, but also privatizations were chiefly devoted to that end. During the second stage, still in progress, more attention is being paid to allocative and distributive considerations.

The outstanding recovery in tax revenues is probably one of the most significant economic changes which has taken place in the last few years. Stabilization, together with a dramatic output expansion, no doubt played a major role in the process.

It would be a mistake, however, to exclusively credit that improvement to macroeconomic reasons. In fact, immediately after the government took office, political conditions created by hyperinflation enabled it to have a wholly new legal framework passed by the Congress. Unthinkable in the past, this laid the foundation for the forthcoming structural reforms. Against this background, a comprehensive tax reform process was begun, which was to a great extent the cause of the powerful response in tax revenues to output growth during the Convertibility Plan.

So far, regardless of its long-run effects, the sharp fiscal adjustment brought about by this rapid increase in tax receipts has been decisive to the credibility and sustainability of the stabilization program.

Two were the components of the reform process which were responsible for that outcome: the progressive establishment of VAT as the core of the tax system, and a radical transformation in tax administration, including much more rigorous penalties for tax evasion.

Thanks to these measures, the government was able to begin to overcome one of the major weaknesses of the Argentine economy: the lack of an adequate tax base. As a result, the overall tax burden practically reached an unprecedented 20% of GDP in the last two years. The figure is even more impressive considering that no "emergency" taxes were applied.

As soon as the tax system managed to consolidate its revenue generating function, other measures followed, aimed at increasing efficiency. First, highly distortive export taxes were abolished. Then, other equally distortive minor taxes were eliminated, while the fuel tax was greatly reduced. More recently, income tax underwent important changes as well. The strategy is clearly to build a simpler tax system, placing VAT at its center, while gradually approaching international standards in income tax performance.

At the same time, it is quite evident that horizontal equity has also been greatly improved. Implementation of widespread withholding mechanisms, standardized billing and other specific administrative controls produced a clear betterment in tax compliance. Other specific measures included in the tax reform package, such as the abrupt reduction in tax expenditures and the restrictions applied for the deduction of declared past losses in the case of corporate income taxes also contributed to that end. Likewise, legal changes made tax evasion a much riskier endeavor than in the past.

Despite all positive changes achieved by tax reform in this short period, however, several questions still remain open. First, it is too early to assess whether the tax system will be able to ensure a level of revenues consistent with public expenditures on a permanent basis. On the one hand, tax performance during 1993 and the first half of 1994 seems to suggest that future growth in federal tax collection will not be as fast as in the recent past. On the other hand, political pressure and conflicting demands for better social services have kept public expenditures on a growing path.

Moreover, VAT's central role in the system has made tax revenues much more dependent on output fluctuations than in the past. In this regard, the economic authorities are currently facing a difficult dilemma: a contractionary policy is needed to reduce a dangerously growing external gap, but at the same time the fiscal balance can be seriously harmed by that very same policy. A similar dilemma applies to the exchange rate policy, because the real burden of interest payments depends on the real exchange rate.

Additionally, many of the structural reforms already in progress will also require significant public funds to finance imbalances which will inevitably arise during the transition. The social security reform is a clear case in point. In a long-run perspective, then, in order to keep the fiscal deficit under control, it will be necessary for the government to pace public sector reforms with the evolution of the revenue generating capacity of the tax system.

Regarding income distribution, there is little doubt that the government's abandonment of the inflationary tax as a way to finance its activities had positive implications on vertical equity.[18] However, the fact that the recent increase in tax revenues has been entirely due to the behaviour of VAT and Social Security contributions may have had quite the opposite effect. More research on the final incidence of these taxes is needed to reach a more definite conclusion on this matter.

Nevertheless, according to the analysis presented in this paper, there is still ample room to further improve the tax system's performance, not only from the point of view of tax collection, but also from that of equity and efficiency.

At the federal level, three issues clearly emerge. First, income tax proceeds are still extremely low: they would have to grow about two points of GDP just to reach the average level of the other major Latin American economies (Brazil, Mexico, Colombia and Chile), which in turn is only half the average level of income tax collection in OCDE countries. This fact has had quite a negative impact on both the overall tax burden and its distribution. Considering the successful experience already achieved with the VAT, it seems only reasonable to expect that better administrative procedures could also ensure much higher revenues in this case without altering current tax rates.

Second, evasion is still significant in the VAT case, despite the substantial improvements achieved in the administration of this tax. Since the estimated gap between the theoretical and effective collection is about 30% in the VAT on domestic transactions, closing one-third of this gap would imply approximately one point of GDP in higher revenues at the prevailing tax rate, or it would allow a reduction of about three points in the VAT rate without meaningfully affecting collection.

Lastly, the rate established for the employer's contributions to the social security system is excessively high, impairing the competitiveness of domestic production and affecting technological choices of investment projects. In acknowledgement of this the government is currently implementing a progressive reduction in this rate. However, a serious risk exists that this measure cause a dangerous imbalance in the social security budget, unless evasion in this area, especially in the case of self-employed workers, is greatly diminished.

On the other hand, much remains to be done concerning tax reform at the provincial level, particularly regarding efficiency. The most salient problem is that of multiple taxation generated by the current turnover tax, which applies to all (including intermediate) sales and constitutes the core of provincial tax systems. The national government is trying to negotiate its replacement by a single retail sales tax with the provinces, so far with little success. A higher VAT compliance enabling a fall in the tax rate at the federal level, a possibility suggested above, could perhaps improve the chances of acceptance of this proposal. Furthermore, other minor distortive taxes such as the provincial tax stamps and those on fuel, gas and electricity are yet to be suppressed.

In fact, the fiscal relation between the federal and provincial governments continues to be a very complex question. There is a complicated system of dividing tax collection which is legally subject to revenue sharing between the national government and the provinces. This allocation system underwent many changes in the past. Lately, as total tax collection has been increasing dramatically, the national government has been trying to prevent these increases from automatically passing on to the provinces (as they should according to current legislation). In October 1992 the so-called "Fiscal Pact" was signed whereby the national government was allowed to keep 15% of the collection of such taxes to finance the social security system. In addition, educational and health services provided by the federal government were transferred to the provinces. A second version of that arrangement, along similar lines to those of the first, was agreed upon in August 1993.

At the same time, in the context of these successive "Fiscal Pacts", the government is trying to induce the provinces to adapt to budget constraints by reducing their spending, reforming their tax systems and privatizing concerns owned by the provincial governments. In return, provinces are offered a number of benefits, including access to multilateral credit. So far, however, the provinces have been most reluctant to execute a fiscal reform such as that in progress at the federal level. The growing difficulties experienced by regional economies, and particularly the current trend in unemployment, are certainly important reasons behind this attitude.

Finally, a few remarks should be made on the several roles played by privatizations. In fact, despite the standard justification on long-term efficiency grounds, so far they have basically served short-term macroeconomic purposes.

In 1990, when the economy was still in a deep crisis, privatizations helped the government regain credibility. Privatization revenues have also been very important as a means to finance temporarily the remaining fiscal imbalances, as well as to reduce the outstanding public debt. Last but not least, they have played a key role in closing the external gap generated by the stabilization plan.

Nevertheless, the way privatizations have been conducted raises some doubts about their final contribution to efficiency and growth. On the one hand, they do not seem to have had any positive impact on aggregate savings and investment. On the other hand, due to the lack of an appropriate regulatory framework, productivity gains in privatized firms are not spilling over the rest of the economy, thereby preventing overall competitiveness from improving.

It is still too early to appraise the long-run impact of privatizations on the Argentine economy. However, there is at least one aspect of the process which seems irreversible, and that is the dramatic change experienced by the political economy environment. New private monopolies have been created, enjoying a substantial amount of economic power. At the same time, the public sector which is supposed to control these powerful new groups has been left extremely weak as a result of the long-standing crisis of the past decade. It is very difficult to foresee what the outcome of this new environment will turn out to be.

TABLE 1
Key Economic Indicators, 1988-1993.





a: Including dollar deposits in the Financial System.

Source: ECLA (several issues), DNIAF (several issues) and own estimates.

TABLE 2
National Public Sector Accounts, 1988-1993.
(Percentage of GDP)





a: Nontax Revenue includes the Public Firms' Surplus.

Source: ECLA (several issues), DNIAF (several issues) and own estimates.

TABLE 3
Tax Bureau Performance Indicators, 1989-1992.
Source: Direcccin General Impositiva and Durn and

Gmez Sabaini (1994).

TABLE 4
Tax Structure, 1988-1993.
(Percentage of GDP)




Source: ECLA (several issues), DNIAF (several issues)

and own estimates.

TABLE 5
Changes in Federal Tax Collection.
(Percentage of GDP)

TABLE 6
VAT on Imports, 1981-1993 (selected years).



Source: ECLA (several issues), DNIAF (several issues)

TABLE 7
VAT on Imports.
Determinants of Changes in Effective Performance.
(Percentage of GDP)



Source: ECLA (several issues), DNIAF (several

TABLE 8
VAT on Domestic Transactions, 1981-1993.
(Selected Years)



Source: ECLA (several issues), DNIAF (several issues)

and own estimates.

TABLE 9
VAT on Domestic Transactions.
Determinants of Changes in Effective Performance.
(Percentage of GDP)

TABLE 10
Social Security Contributions.
Determinants of Changes in Effective Performance.
(Percentage of GDP)




Source: ECLA (several issues), DNIAF (several

TABLE 11. Financial Results of the Privatization Process (1990-93)
(In millions of US dollars)




Source: Own calculations based on Secretaria de Programacin Econmica (1994).

Bibliography