Chapter

16 Argentina

Editor(s):
Teresa Ter-Minassian
Published Date:
September 1997
Share
  • ShareShare
Show Summary Details
Author(s)
Gerd Schwartz and Claire Liuksila 

Argentina is a large and decentralized country with significant constitutional powers bestowed to the provincial level of government. As defined in Article 1 of its National Constitution, there are three main levels of administration: federal, provincial, and municipal. The general constitutional organizing principle is that the provinces hold all powers that they have not delegated to the federal level.

At the same time, Argentina has large regional concentrations of population, and strong regional disparities in levels of economic development. Given such disparities, the ability of subnational governments to provide goods and services to the population and finance these out of own resources differ considerably. As a result, the federal government has faced, and continues to face, very different demands from the different provinces.

Until the mid-1970s, Argentina was characterized by extensive centralization of various aspects of public policymaking, particularly in the areas of taxation and the exercise of regulatory functions. Since then, substantial decentralization of spending responsibilities has taken place; revenue-raising powers, however, remain largely concentrated at the federal level, with attendant large vertical imbalances and heavy dependence of the provincial and local governments on transfers from the central government, including revenue sharing. A reform of the current system of intergovernmental fiscal relations is clearly a high priority for Argentina, and one that the authorities will need to face in the near term.

This chapter describes the main features and characteristics of Argentina’s system of fiscal federalism, analyzes key policy issues and problems, and discusses various reform options, focusing on revenue and expenditure assignments and responsibilities, the system of transfers, borrowing, and the federal-provincial fiscal pacts. Some aspects of tax administration and public expenditure management are also discussed. The concluding section analyzes major implications of the existing system for macroeconomic management and structural reform.

Structure of Argentina’s Public Sector

Notwithstanding substantial progress with privatization, deregulation, and other structural reforms over the last several years, Argentina’s public sector remains fairly extensive. Its principal public sector entities are three levels of public administration (that is, the federal administration, 23 provincial administrations plus the administration of the municipality of Buenos Aires, and 1,110 municipal administrations), the quasi-public health funds, and the financial public sector. The federal administration includes the central government, various decentralized agencies, federally owned nonfinancial public enterprises, and the national social security system. Provincial governments, including the municipality of Buenos Aires, have their own nonfinancial public enterprises and social security systems for provincial government workers.1 In March 1996, the financial public sector included the Central Bank, 6 national official banks, and 21 provincial and municipal banks.

The National Constitution stipulates that provinces must organize and assure in their territories an effective municipal system. The concrete design of this system is left to each province to decide, but it has to be outlined in the respective provincial constitutions. In practice, provinces have granted varying degrees of independence to their municipalities; most, but not all, have granted autonomy to their municipalities, implying that municipalities are free to establish their own procedures for government, including, at least in theory, the right to establish and administer their own taxes. A significant development in this regard is the direct election of mayors, a process that was initiated in the municipality of Buenos Aires in June 1996.

Fiscal Federalism: Main Aspects and Recent Reforms

The major policy shaping Argentina’s system of fiscal federalism since the mid-1970s has been “re-decentralization,” which has been intensified over the last several years and is expected to continue over the short to medium term. While, according to the National Constitution, provincial governments hold significant powers, historically, more and more decision-making authority had gradually been transferred to the federal level. For example, according to the National Constitution, the provinces are to guarantee basic education, while other levels of education are understood to be the joint responsibility of the federal and provincial administrations. Over time, however, federal educational policy had gradually come to encompass all levels of schooling and largely absorbed provincial powers in this area (Artana and others, 1995). Since the mid-1970s, this process is being reversed, and more and more responsibilities in the education sector have been transferred back to the provinces. Similar processes of centralization and subsequent decentralization have also taken place in other sectors, particularly in health care.

One of the more notable features of the current system of fiscal federalism in Argentina is the imbalance between own resources and spending at the different levels of government (Artana and López Murphy, 1994; Artana and others, 1995; and Fundación de Investigaciones Economicas Latinoamericanas (FIEL), 1993). Table 1 shows that, in 1995, for example, provinces carried out 54 percent of all national and provincial expenditures, while own revenues of provincial governments accounted only for about 20 percent of total provincial and national revenue. Overall, provincial government own revenues only covered about 40 percent of provincial government expenditure in 1995 (Table 2).

Table 1.Argentina: Revenue and Expenditure Shares of Federal and Provincial Levels of Government(In percent of total)
1991199219931994Est. 1995
Revenue shares
Before transfers100100100100100
Federal level18482808080
Provincial level1618202020
After transfers100100100100100
Federal level15654565655
Provincial level4446444445
Expenditure shares100100100100100
Federal level14842424446
Provincial level5258585654
Memorandum items:
Provincial own revenues in percent of provincial expenditure3439424141
Transfers in percent of total federal and provincial revenue22828242425
Sources: Government of Argentina; and authors’ calculations.

The federal level comprises the federal government, decentralized agencies, the national social security system, and the operating surplus of national public enterprises. Operations to clear expenditure arrears to suppliers, provinces, and pensioners are excluded from expenditures.

Based on the consolidated accounts of the 23 provinces plus the municipality of Buenos Aires.

Sources: Government of Argentina; and authors’ calculations.

The federal level comprises the federal government, decentralized agencies, the national social security system, and the operating surplus of national public enterprises. Operations to clear expenditure arrears to suppliers, provinces, and pensioners are excluded from expenditures.

Based on the consolidated accounts of the 23 provinces plus the municipality of Buenos Aires.

Table 2.Argentina: Provincial Government Finances1(In percent of GDP)
198919901991199219931994Est. 1995
Total revenues7.97.18.29.79.79.69.3
From own sources22.62.43.03.84.44.34 1
Federal transfers5.24.75.25.95.35.25.2
Coparticipation3.22.93.84.44.03.83.7
Housing Fund (FONAVI)0.40.50.40.30.30.30.3
Tax advances0.50.20.1
Other1.20.80.91.11.01.21.1
Total expenditures8.48.59.09.910.510.510.2
Current expenditures7.17.27.88.89.29.19.0
Wages4.14.44.75.35.55.35.4
Goods and services0.90.91.01.01.11.11.0
Interest0.30.10.10.20.20.20 2
Transfers to municipalities0.80.81.01.11.21.11.0
Other0.91.01.01.31.31.31 3
Capital expenditures1.31.31.21.11.31.41.2
Overall balance−0.5−1.4−0.7−0.2−0.8−0.9−0.9
Memorandum item:
Overall balance excluding federal transfers−5.7−6.1−5.9−6.1−6.1−6.1−6.1
Sources: Government of Argentina; and authors’ calculations.

Consolidated accounts of 23 provinces and the municipality of Buenos Aires.

In addition to own revenue (current and capital), contains nonclassified revenue. Privatization receipts are excluded.

Sources: Government of Argentina; and authors’ calculations.

Consolidated accounts of 23 provinces and the municipality of Buenos Aires.

In addition to own revenue (current and capital), contains nonclassified revenue. Privatization receipts are excluded.

These own revenue and expenditure imbalances have meant that revenue-sharing mechanisms have traditionally been a central element of fiscal federalism in Argentina. Decentralization initiatives on the expenditure side increased even further the importance of various revenue-sharing mechanisms, and they featured prominently in two recent federal-provincial fiscal pacts. In 1995, federal transfers covered about 50 percent of provincial government expenditure, and provincial transfers covered roughly 55 percent of municipal government spending (Artana and others, 1995). The 1994 Constitution stipulated that a new revenue-sharing agreement was to go into effect on January 1, 1997, but that date has been postponed.

Argentina’s revenue-sharing mechanisms have a redistributive character, that is, they are meant to benefit poorer provinces more than richer provinces. Table 3 shows per capita revenue and expenditure for various groups of provinces relative to the average for all provinces. In per capita terms, advanced (that is, richer) provinces had significantly lower total revenue and expenditure than less-developed provinces, received less funds through revenue sharing mechanisms, but raised more revenue from own sources.

Table 3.Argentina: Per Capita Revenue and Expenditure in Different Groups of Provinces(Average for all provinces = 100)
197219811991
Total revenue
Advanced provinces788182
Low-density provinces217216208
Intermediate provinces123113121
Less-developed provinces127149143
Own revenues
Advanced provinces111113121
Low-density provinces15193146
Intermediate provinces706870
Less-developed provinces486042
Coparticipation transfer revenue
Advanced provinces837466
Low-density provinces216200169
Intermediate provinces128143157
Less-developed provinces140166192
Other transfer revenue
Advanced provinces214038
Low-density provinces319422499
Intermediate provinces198151130
Less-developed provinces236271228
Total expenditure
Advanced provinces868478
Low-density provinces235205230
Intermediate provinces121110116
Less-developed provinces121143145
Sources: Fundación de Investigaciones Economic as Latinoamericanas (1993); and authors’ calculations.Note: The advanced provinces are the Province of Buenos Aires, the Municipality of Buenos Aires, Santa Fé Córdoba, and Mendoza; the low-density provinces are Chubut, Santa Cruz, La Pampa, Río Negro, Neuquén, and Tierra del Fuego; the intermediate provinces are San Juan, San Luis, Entre Ríos, Tucumän, and Salta: and the less-developed provinces are Catamarca, Chaco, Corrientes, Formosa, Jujuy, La Rioja, Misiones, and Santiago del Estero.
Sources: Fundación de Investigaciones Economic as Latinoamericanas (1993); and authors’ calculations.Note: The advanced provinces are the Province of Buenos Aires, the Municipality of Buenos Aires, Santa Fé Córdoba, and Mendoza; the low-density provinces are Chubut, Santa Cruz, La Pampa, Río Negro, Neuquén, and Tierra del Fuego; the intermediate provinces are San Juan, San Luis, Entre Ríos, Tucumän, and Salta: and the less-developed provinces are Catamarca, Chaco, Corrientes, Formosa, Jujuy, La Rioja, Misiones, and Santiago del Estero.

Interestingly, the wedge between advanced and less-developed provinces may have widened over the period 1972–91. For example, data in Table 3 show that, during 1972–91, per capita own revenues of advanced provinces increased relative to the average for all provinces from 111 percent to 121 percent, but fell in less-developed provinces from 48 percent to 42 percent. Over the same period, however, per capita expenditure in advanced provinces fell relative to the average of all provinces from 86 percent to 78 percent, but increased in less-developed provinces from 121 percent to 145 percent; this may point to a widening gap in expenditure efficiency and productivity across provinces.2

Expenditure Assignments

Expenditure assignments in Argentina are not dissimilar to those in other federal countries. The central government is exclusively responsible for matters concerning defense, foreign affairs, international trade, the regulation of interstate trade, monetary policy, immigration policy, and provision of unemployment insurance3 (Table 4). Responsibility for health care and primary and secondary education rests jointly with federal, provincial, and municipal governments; responsibilities for social welfare services, police, and highways are shared between federal and provincial levels of government.

Table 4.Argentina: Selected Expenditure Assignments of Federal, Provincial, and Municipal Governments
ExpenditureDelivery of ServiceFinancingRegulatory Powers
DefenseFFF
Environmental policyF, P, MF, P, MF, P, M
Education
PrimaryP, MP, MF, P, M
SecondaryPPF, P, M
UniversityFFF
Foreign affairsFFF
HealthF, P, MF, P, MF, P
Health insuranceF. PF, P, MF, P
Immigration policyFFF
International tradeFFF
Interstate trade regulationFFF
JusticeF, P, MF, P, MF, P, M
Monetary policyFFF
Public safety
PrisonsF, PF, PF, P
PoliceF, PF, PF, P
RoadsF, P, MF, P, MF, P
Social housingPF, P
Social welfareF, PF, PF, P
Transport
SeaFFF
Rail (passengers)PFF, P
AirPF, P
Unemployment insuranceFFF
Sources: Shah (1994); and Artana and others (1995).Note: F = federal level of government; P = provincial level of government and M = municipal level of government. In addition, there is private sector provision for different expenditure categories alongside public sector provision, as in the case of unemployment insurance.
Sources: Shah (1994); and Artana and others (1995).Note: F = federal level of government; P = provincial level of government and M = municipal level of government. In addition, there is private sector provision for different expenditure categories alongside public sector provision, as in the case of unemployment insurance.

Consolidated nonfinancial public sector expenditure increased by over 3 percentage points of GDP during 1989–95 and amounted to 23 percent of GDP in 1995 (Table 5). The federal level accounted for about 55 percent of this total.4 However, after subtracting interest payments, mandated social security transfers, and other entitlement programs, the federal level had discretion over only about 12 percent of total spending of the nonfinancial public sector. The increase in expenditures that occurred during 1989–95 was about evenly distributed between the federal level and the provincial level. While, for the provincial level, the increase in expenditures was largely due to the transfer of responsibilities from the federal level, the increase in expenditures of the federal level was mostly because of growing pension spending, and occurred notwithstanding the ongoing transfer of responsibilities—in health and education—to the provinces.

Table 5.Argentina: Consolidated Nonfinancial Public Sector1(In percent of GDP)
198919901991199219931994Est. 1995
Consolidated nonfinancial public sector
Revenue16.416.218.821.022.221.620.7
Expenditure19.720.322.121.422.123.023.0
Balance−3.3−4.1−3.2−0.40.1−1.4−2.2
National nonfinancial public sector
Revenue13.713.815.817.217.817.316.6
Tax revenue11.211.914.316.116.516.215.5
Non-tax revenue1.71.01.20.91.01.11.1
Operating surplus of state enterprises0.80.90.30.20.3
Expenditure16.516.518.317.416.917.818.0
Transfers to provinces25.24.75.25.95.35.25.2
Own spending11.312.113.111.511.612.612.8
Balance−2.8−2.7−2.5−0.20.9−0.5−1.4
Provincial nonfinancial public sector
Total revenue7.97.18.29.79.79.69.3
Own revenue2.72.43.03.84.44.34.1
Transfers to provinces25.24.75.25.95.35.25.2
Expenditure8.48.59.09.910.510.510.2
Balance−0.5−1.4−0.7−0.2−0.8−0.9−0.9
Sources: Argentine authorities; and authors’ calculations.

As defined in this table, the consolidated nonfinancial public sector comprises the national and the provincial nonfinancial public sector; it excludes the municipal nonfinancial public sector. The national nonfinancial public sector comprises the federal government, decentralized agencies, the national social security system, and the operating surplus of national nonfinancial state enterprises. The provincial nonfinancial public sector reported here only comprises the consolidated operations of the provincial governments. Federal government operations to clear expenditure arrears to suppliers, provinces, and pensioners, financed by bond issues (BOCONs) are excluded from expenditures.

Based on the data reported in the consolidated accounts of the provincial nonfinancial public sector.

Sources: Argentine authorities; and authors’ calculations.

As defined in this table, the consolidated nonfinancial public sector comprises the national and the provincial nonfinancial public sector; it excludes the municipal nonfinancial public sector. The national nonfinancial public sector comprises the federal government, decentralized agencies, the national social security system, and the operating surplus of national nonfinancial state enterprises. The provincial nonfinancial public sector reported here only comprises the consolidated operations of the provincial governments. Federal government operations to clear expenditure arrears to suppliers, provinces, and pensioners, financed by bond issues (BOCONs) are excluded from expenditures.

Based on the data reported in the consolidated accounts of the provincial nonfinancial public sector.

Since the mid-1970s, and reflecting efforts to improve public sector efficiency, the policy of devolving spending from the federal to subnational levels of government and to the private sector has resulted in important changes in the structure of public expenditures. Examples of these changes are provided in Table 6. For instance, while in 1983 the federal level carried out 44 percent of all education spending, in 1992 its share was reduced to 22 percent, reflecting the ongoing devolution of spending responsibilities for primary and secondary education to the provinces; the share of provincial expenditures increased correspondingly. A similar pattern is found in health care spending, with the difference that municipal rather than provincial government spending was increased, at the expense of federal level spending. Similar reductions of the share of federal level spending were also observed in other expenditure categories, such as social assistance, public housing, and spending on infrastructure and general services. In all these cases, the major increase in expenditure responsibilities occurred at the provincial level of government.

Table 6.Argentina: Use and Provision of Public Funds for Selected Expenditure Items
19831992Change in Shares 1983–92Real Growth Rate During 1983–92
(In percent of total)(In percent)
Education
Spending of public funds100100+28
Federal level4422−22−36
Provincial level4970+21+82
Municipal level22+76
Private sector56+1+56
Provision of financing100100+28
Federal level4927−22−30
Provincial level4971+22+84
Municipal level22+76
Health care
Spending of public funds100100+24
Federal level1711−6−19
Provincial level7474+24
Municipal level915+6+114
Private secror
Provision of financing100100+24
Federal level2414−10−30
Provincial level6772+5+32
Municipal level915+6+114
Social assistance
Spending of public funds100100+38
Federal level192−17−85
Provincial level1837+19+188
Municipal level79+2+82
Private sector5652−4+28
Provision of financing100100+38
Federal level8165−16+10
Provincial level1226+14+203
Municipal level79+2+82
Public housing
Spending of public funds100100−24
Federal level187−11−70
Provincial level8293+11−14
Municipal level
Private sector
Provision of financing100100−24
Federal level8479−5−28
Provincial level1621+5−5
Municipal level
Infrastructure and services
Spending of public funds100100−50
Federal level6839−29−72
Provincial level3057+27+6
Municipal level24+2+32
Private sector
Provision of financing100100−50
Federal level7258−14−60
Provincial level2638+12−27
Municipal level24+2+32
Source: Authors’ calculations based on data in Artana and others (1995).
Source: Authors’ calculations based on data in Artana and others (1995).

The devolution of expenditure responsibilities in areas such as primary education, hospitals, water, electricity, and adult education, need not affect the financing of these expenditures, that is, while spending responsibilities are transferred to subnational levels, the federal level could continue to provide financing via transfers. In Argentina, this only held true in some cases. For example, for social assistance, which grew by 38 percent in real terms during 1983–92, the federal level reduced dramatically its own spending, but increased its overall provision of financing, although by much less than the overall increase in spending (see Table 6).

In many cases, however, the devolution of expenditure responsibilities to subnational levels of government took place without a comparable transfer of funds to finance these activities, even though some compensating transfer payments were and are being made. For example, in the case of health care spending, which grew by 24 percent in real terms during 1983–92, the federal level decreased its own spending and its provision of financing, suggesting that expenditure responsibilities were devolved without providing full financing for the transfer of responsibilities. Similar patterns existed for education spending, public housing, and infrastructure and services (see Table 6).

To face the increased expenditure responsibilities, subnational governments resorted, among other things, to financing from provincial banks or commercial banks (often collateralized by future receipts from revenue-sharing agreements), or issued IOUs in lieu of paying wages; provinces also received discretionary aid from the central government in the form of loans or grants.

In general, though, the devolution of expenditure responsibilities forced subnational governments to prioritize spending. As a result, the structure of subnational government expenditures underwent significant changes. The observed real growth of health and education spending, for example, reflected a redirection of spending by the municipal and provincial administrations that more than compensated for the decision by the federal government to reduce its own funding of these functions. The increased spending on health and education by provincial and municipal governments was matched by substantial expenditure cuts elsewhere, such as public housing and public infrastructure and services. This view is corroborated by evidence of neglect of public infrastructures in various Argentine provinces (see Schuler, 1995).

Tax Assignments

The National Constitution establishes the principles of separation of tax sources between the federal government and the provinces. However, ever since the federal government experienced a fiscal crisis in the late nineteenth century, a number of tax sources, notably consumption, have been shared between the federal and provincial governments. In theory, the current system is rather simple: the federal government has exclusive powers to set import and export duties, and, for a limited period of time, may also establish direct taxes; indirect taxes are concurrent taxes that can be imposed by the federal and/or provincial governments; the provinces have the exclusive right to impose permanent direct taxes.5

In practice, the provinces have delegated much of their responsibility for legislating, administrating, and collecting taxes to the central government, but receive a share from these taxes. Two key reasons for this delegation of power are that the central government can collect most taxes more efficiently than provincial governments, and that the potential tax bases differ significantly from province to province. Taxes collected by the federal government include the income tax, the value-added tax (VAT), excise taxes, foreign trade taxes, liquid fuel and energy taxes, the gross assets tax (which was levied on companies but has recently been eliminated), the personal assets tax (levied on individuals), social security taxes, and a number of minor levies. Taxes collected by provincial and municipal governments are the real estate tax, the automobile tax, road taxes, and the provincial turnover tax, which, in 1993, accounted for over 50 percent of all own revenue of the provinces and is levied in addition to the federal VAT.

Total revenue of the consolidated nonfinancial public sector increased by 4.3 percent of GDP during 1989–95 to a level of 20.7 percent of GDP in 1995 (see Table 5). About two-thirds (2.9 percent of GDP) of the increase in overall revenue occurred at the federal level, which experienced a rapid growth of tax revenue (by 4.3 percent of GDP), but reductions in nontax revenue (by 0.6 percent of GDP) and in operating surpluses of state enterprises (by 0.8 percent of GDP), the latter reflecting extensive privatization. In 1995, about 80 percent of all revenue was collected at the federal level, slightly less than in 1989.

As already suggested, the centralization of revenue-raising functions at the federal level reflects, to some extent, the different capacities of different provinces to raise their own revenue, which in turn is mainly due to differences in provincial tax bases, and, to a lesser extent, to differences in the strength of provincial tax administrations. Overall, provincial own revenues represented about 45 percent of total provincial government revenue in 1993; the remainder came from various federal transfer mechanisms. In 1992, Argentina’s most advanced subnational entity, the municipality of Buenos Aires, with 9 percent of the country’s population (according to the 1991 census), accounted for 27 percent of all tax revenue collected by the provincial administrations. In contrast, 6 poorer provinces,6 that together also accounted for 9 percent of the country’s population, raised only 3 percent of all tax revenue collected by the provincial administrations.7 Other evidence also points to the different taxing abilities of different provinces according to their level of economic development (see Table 3): in 1991, for example, per capita own revenue of the economically advanced provinces exceeded the average for all provinces by about 20 percent; in contrast, the less-developed provinces only collected 42 percent of the average per capita own revenue for all provinces.

If own tax collections are taken as an indicator, economic differences between the advanced and the less-developed provinces have been increasing (see Table 3): in 1991, advanced provinces exceeded the national average for per capita revenue from own sources by more than they did 20 years earlier, whereas the less-developed provinces fell short of the national per capita average by more than they did 20 years earlier. The mirror image is that, in per capita terms, the less developed provinces had to rely more and the advanced provinces less on transfers from the federal level in 1991 than they did in 1972.

Intergovernmental Transfers

Argentina has elaborate federal-provincial transfer mechanisms, which are necessary given that revenue-raising functions continue to be fairly centralized while expenditure functions are being gradually decentralized. There are similarly elaborate provincial-municipal transfer mechanisms, but these differ from province to province. As shown in Table 1, in 1995, transfers from the federal level to the provinces amounted to 25 percent of the total national and provincial revenue collected. However, the quantitative importance of federal-provincial revenue transfers differs significantly across provinces. While, for example, in the province of Buenos Aires transfers represented only 51 percent of provincial expenditures in 1993, in four provinces (Catamarca, Formosa, La Rioja, and Santiago del Estero) they accounted for more than 90 percent.

Improving the structure of the federal-provincial transfer mechanisms is of key importance for improving Argentina’s public sector efficiency: given the wide regional economic disparities, a transfer system will always remain a central element of Argentina’s system of fiscal federalism, no matter by how much provincial tax bases and tax administrations can be improved.

It has sometimes been argued that the redistributive effect of transfers falls short of what would be needed to address Argentina’s significant regional differences. For example, the World Bank (1996), based on data presented by Porto and Sanguinetti (1993), argues that per capita transfers to the poorest provinces (Chaco, Formosa, and Santiago del Estero), where close to 40 percent of the population lived below the poverty line in 1991, are only slightly higher than the average per capita transfer to all provinces, whereas some wealthier provinces received almost double the average per capita transfer. While the data presented in Table 3 and other available evidence do not support this statement, they do suggest that interregional redistribution may be insufficient. For example, the correlation coefficient between a poverty headcount index and per capita shared tax revenue, calculated on the basis of data presented in Rezk and others (1996), was only about 0.31 in 1994; this is, however, better than the 0.15 that prevailed in 1984. Similarly, during 1981–90, the share of less-developed provinces in total federal revenue transfers to the provinces rose from 19 percent to 24 percent. This improvement came mainly at the expense of the advanced provinces which saw their share in the total decline from 50 percent to 42 percent over this period.

Currently, there are three basic mechanisms for revenue sharing between the national and the provincial level: (1) the “coparticipation” scheme, which provides automatic, nonearmarked transfers; (2) other automatic transfers, all of which are earmarked for specific purposes; and (3) discretionary, that is, nonautomatic, transfers and grants that may be either earmarked or nonearmarked. The detailed structure of the system is shown in Figure 1.8 Accordingly, excise taxes, income taxes, and the VAT (and, before being phased out in 1995, also the gross assets tax) are subject to revenue sharing under the coparticipation scheme. However, the provinces receive a certain share of the income tax and the gross assets tax under separate arrangements prior to the distribution of tax revenue under the coparticipation scheme, just as 11 percent of VAT revenue is given to the social security system prior to entering into the coparticipation scheme. Other taxes, such as the fuel tax, the energy tax, the personal assets tax, wage taxes, the import surcharge (called the “statistical” tax), and charges on insurance premiums are shared according to separate earmarking rules outside of the coparticipation scheme. In principle, all discretionary transfers and grants are provided in the form of National Treasury Contributions (ATN), which is a mechanism to allocate in discretionary fashion to the provinces 2 percent of income tax revenue and 1 percent of “net” coparticipation revenue.9 The various transfer arrangements have undergone numerous changes, partly reflecting changes in the tax system. For example, the introduction in 1995 of the personal assets tax (for individuals) and the phasing out of the gross assets tax (for companies) also had implications for the revenue-sharing arrangements (see Figure 1).

Figure 1.Argentina: Assignment of Federal Revenue Collections in 1996

1/ Until end-1998, 21 percent of liquid fuel taxes (with some exceptions) are assigned to the National Social Security system; the rest is distributed in the proportions shown.

The coparticipation scheme is the centerpiece of the transfer system, and currently responsible for over two-thirds of all federal-provincial transfers. The basic structure of the scheme, under which the provinces entrust the federal government with the task of administering a number of taxes, the revenues of which are then shared, has existed since 1935. Since its inception it has seen frequent changes, reflecting both the continuing devolution of expenditures and the distributional issues that have arisen in this context. Since 1980, coparticipated revenue is not only shared between the federal and the provincial governments but also with the social security system.

During 1980–96, the primary distribution of coparticipated revenue has been governed by three major regimes, successively. In the early 1980s, and based on a 1973 legal regime, the provinces received 48.5 percent of coparticipated revenue directly and a further 3 percent through the Regional Development Fund. The remaining 48.5 percent was retained by the federal government. However, by 1983, various “pre-coparticipation” arrangements had reduced the effective share of the coparticipated revenue going to the provinces to 29 percent.

When the 1973 law expired at the end of 1984, it took until 1988 to approve a new coparticipation law, and the three-year period 1985–87 was characterized by the absence of a legal regime for “coparticipating” tax revenue between the federal and provincial levels. In practice, each year the various provinces negotiated bilateral agreements with the federal government. The overall revenue share of the provinces grew during 1985–87, and this higher share was by and large validated by a new “transitory” coparticipation law in 1988, which stipulated that 56.66 percent of coparticipated revenue accrue directly to the provinces, with an additional 1.0 percent allocated for discretionary distribution to the provinces in the form of ATNs, while the remaining 42.34 percent are retained by the central government.

The main features of the 1988 coparticipation scheme essentially prevail until today, even though there have been numerous changes and adjustments. The two major changes have been to establish “pre-coparticipations,” that is, to redirect parts of the tax revenue originally destined toward the coparticipation scheme toward other purposes, and to provide some fixed-sum transfers and a minimum transfer guarantee to the provinces.

The secondary distribution of coparticipated revenue, which has the explicit objective of redistributing income among the provinces, has remained by and large unchanged since 1984. The allocation among the provinces is based on the following basic criteria: 65 percent is distributed according to the number of inhabitants, 10 percent according to population density (inhabitants per square kilometer), and 25 percent according to the inverse of the number of houses and automobiles per inhabitant and the inverse of the average level of education per inhabitant. As shown in Table 3, in 1991, per capita coparticipation transfers to less-developed provinces exceeded the average for all provinces by 92 percent, whereas in advanced provinces they amounted only to 66 percent of the average for all provinces.

There are various other automatic transfers apart from the coparticipation scheme, as shown in Figure 1. These include transfers earmarked for different funds that are under either the spending responsibility of the provincial governments (for instance, the National Housing Fund (FONAVI), the Road Fund, and the Rural Electrification Fund), or the federal administration (for example, the Agricultural Technology Institute and the Reinsurance Institute). They also include direct transfers to the provinces (from part of income tax revenue) and the social security system (from part of the VAT and income tax revenue). For example, provinces (excluding the municipality of Buenos Aires) receive 4 percent of all income tax revenue directly, but they also receive part of the 64 percent of income tax revenue that enters into the coparticipation scheme.

There are also discretionary transfers, mostly in the form of (nonreimbursable) grants. The main purpose of these transfers is to fill resource gaps at the provincial level. During the first half of the 1980s, grants rose sharply and peaked at 3 percent of GDP in 1983, representing one-third of total provincial revenue. During this period, grants were not conditioned on improvements in the financial performance of provincial governments, but they were allocated in direct proportion to the size of the provincial deficit, thus penalizing provinces with prudent fiscal policies and rewarding those with weak fiscal discipline.

Finally, there are discretionary transfers that are reimbursable, at least in principle. These include those made through FONAVI, and Treasury advances against future tax revenues. FONAVI obtains earmarked resources in the form of 42 percent of fuel tax revenue (around 1 percent of GDP). These funds are lent to the provinces, which then onlend the money through provincial housing organizations to individuals to finance housing construction. FONAVI transfers, though nominally reimbursable, are effectively nonreimbursable because of low repayment levels—the loan recovery rate is less than 10 percent. In addition, FONAVI does not reach the very poor, and, in fact, it has evolved into a mechanism for subsidizing middle-class housing.

Historically, the existence of the transfer system has been a source of instability. The many changes in the system made it difficult for the provinces to forecast the funds available to them each year, which did not, however, prevent individual provinces from embarking on expansionary expenditure policies. Particularly in the early 1980s, when the funds channeled through the coparticipation arrangement were reduced via pre-coparticipation arrangements, the provinces began to rely more heavily on discretionary ATN transfers to fill financing gaps. In 1983, for example, ATN transfers exceeded coparticipation transfers by 75 percent. As originally designed, ATN funds had the objective of providing reimbursable grants to the provinces in the case of “unusual events”; it was an emergency “gap-filling” transfer mechanism that provided advances against future tax revenue in exceptional circumstances, and advances were supposed to be repaid by the provinces at the time they received funds through the regular revenue-sharing arrangements. In practice, ATN funds became the major mechanism for providing discretionary funds to provinces, largely reflecting their respective relative political leverage.

During 1985–87, after the 1973 coparticipation law had expired and a new accord could not be reached, all transfers to the provinces were channeled through the ATN mechanism. Sanguinetti (1993) argues that, during this period, political factors played the dominant role in allocating transfers across provinces. He found that provinces ruled by the federal opposition party received higher per capita transfers from the federal government than those ruled by the same party governing the federal level. He attributed this to “coordination failure,” where the federal government was generally unable to ensure fiscal discipline (which is reflected in higher overall transfers to the provinces during this period), but was particularly unable to do so in provinces ruled by the opposition party, and therefore was forced to cover the higher deficits of these provinces via higher discretionary transfers.

The new coparticipation agreement that came into effect in 1988, capped discretionary ATN funds at 1 percent of coparticipated revenue (in addition to 2 percent of income tax revenue going toward ATN). Thus, it greatly enhanced the importance of fixed rules in the federal-provincial revenue-sharing arrangements, even though, during 1989 and 1990, when provincial finances came under severe pressure, tax advances were again increased. These tax advances reached 6.2 percent of provincial revenues in 1989, and were repaid in nominal terms during a period when the annual inflation rate was 3,000 percent. Even until today, and within the general environment of relative economic stability that has been prevailing during the last few years, the federal government has been forced sometimes to provide extraordinary discretionary transfers to provinces, outside of the normal transfer framework, for instance in the form of advances against future tax revenue.

Subnational Borrowing and Debt

Within Argentina’s federal structure, all levels of government are generally permitted to borrow both domestically and abroad to cover their deficits. During the 1980s, both federal and local governments borrowed extensively, reflecting the weak fiscal management during the period. In addition to direct borrowing, both federal and local governments accumulated sizable arrears on payments for wages and pensions, to suppliers, and for debt service. During 1991–95, the federal government tried to clear its accumulated expenditure arrears to pensioners, suppliers, and to the provinces by issuing “consolidation bonds” (BOCONs) and other government debt instruments. These federal arrears clearance operations added up to a total of about 9 percent of Argentina’s 1995 GDP.10

Lack of financial control prevailed in particular at the provincial level and was an important source of financial and macroeconomic instability throughout the 1980s. In the late 1980s, and before discretionary ATN transfers, the provinces accounted for roughly 40 percent of the deficit of the consolidated nonfinancial public sector. These deficits were financed by discretionary ATN transfers and loans from the federal government, but also by loans from the provincial banks and other parts of the financial system, arrears to suppliers, and delays in wage payments to provincial government employees (Fundación de Investigaciones Economicas Latinoamericanas, 1993).

Provincial banks, in particular, acted as captive sources of financing. The provincial government banks were considered to be akin to the central bank of each province: they provided funds to the provincial governments upon demand, and in turn, received rediscounts from the Central Bank of Argentina. During 1983–90, for example, these central bank rediscounts amounted to over 2 percent of annual provincial spending (Fundación de Investigaciones Económicas Latinoamericanas, 1993). Given their portfolio of bad assets and assets of doubtful quality, resulting to a significant extent from lending to provincial governments, provincial banks were among the prime candidates for the bank restructuring and consolidation exercises that started in earnest in 1995.

By establishing a currency board arrangement for monetary management, the Convertibility Law of March 1991 ended inflationary central bank financing of public sector deficits at all levels. This also meant that new rediscounts from the central bank to the provincial banks were eliminated, and the central bank began to recover outstanding rediscounts. To strengthen further control over provincial bank operations, in 1993 the reserve requirement on deposits of provincial entities held at provincial banks was gradually raised to equal that for ordinary bank deposits. In cooperation with provincial governments, the federal government has begun to restructure, privatize, or liquidate provincial banks. As of March 1996, six provincial banks had been privatized, and seven other provincial banks were in the process of being privatized. Also, since the beginning of 1994, provincial governments have been required to obtain congressional approval for borrowing in foreign currencies. So far, only 0.2 percent of GDP of foreign borrowing by the provinces has been authorized.

Still, while the currency board arrangement reduced, among other things, the degrees of freedom for carrying out extraordinary financing operations, it has not been sufficient to bring about financial discipline at the provincial level. There are no legal limits on domestic currency borrowing operations of the provinces, and provincial governments have continued the practice of pledging future transfer receipts from coparticipation as a collateral for borrowing from commercial banks. In addition, provincial governments have sometimes developed alternative sources of financing. For example, when faced with a cash crisis in 1995, several provinces issued “coupons” in lieu of wage payments.11

Given these continued problems with financial discipline, many provinces remain overindebted. Provincial debt per capita averaged Arg$430 in 1995 (which compared to an average of US$400 for the U.S. states),12 ranging from Arg$50 in the province of San Luis to Arg$2,200 in La Rioja. Total provincial debt amounted to 58 percent of total provincial revenue in 1995, and while in six provinces (La Pampa, Neuquén, San Luis, Santa Cruz, Santa Fé, and Tierra del Fuego) provincial debt was less than 20 percent of total revenue, in nine provinces (municipality of Buenos Aires, Chaco, Córdoba, Corrientes, Formosa, La Rioja, Mendoza, Misiones, and Rio Negro) it exceeded 80 percent.

The fact that many provincial governments had to resort to extraordinary financing operations and remain heavily indebted suggests that further adjustment and reform are needed. This would be supported by the disaggregated provincial fiscal accounts, as shown in the World Bank report (1996), which suggest that only two provinces (Mendoza and the municipality of Buenos Aires) had a primary fiscal surplus in 1995, whereas nine provinces had a primary fiscal deficit between 20–36 percent of their total revenue, and eight other provinces had a primary deficit between 10–19 percent of their total revenue. Such fiscal positions are clearly unsustainable.

Main Aspects of the Federal-Provincial Fiscal Pacts

By the early 1990s it had become clear that economic reform efforts by one level of government alone, regardless of how well designed, would not be sufficient to yield the desired fiscal and macroeconomic adjustment. To coordinate their reform efforts, two successive fiscal pacts were negotiated between the provincial and federal authorities in the 1990s. While these two pacts provided a starting point, they will need to be succeeded by further adjustment measures by the provinces.

The first federal-provincial fiscal pact, which was agreed in August 1992, aimed at (1) further decentralization by transferring expenditure responsibilities for public health, education, and housing from the federal to the provincial governments; (2) getting the provincial governments to shoulder part of the burden of improving the financial situation of the national social security system that had accumulated enormous payment arrears to pensioners; and (3) promoting tax reform at the provincial level. In this context, the pact also sought to ensure that the transfer windfall for the provinces from the strengthened revenue performance would be used efficiently. A key change under this first pact was to divert to the national social security system a portion of the revenue transferred to the provinces under the coparticipation scheme. It was agreed that 15 percent of all coparticipated tax revenue would be transferred to the social security system in a pre-coparticipation arrangement (see Figure 1). This change effectively reduced the share of coparticipated revenues allocated to the provinces from 57.7 percent to 49.0 percent, but it also cut the coparticipated revenue retained by the federal government from 42.3 percent to 36 percent.13

At the same time, the pact’s fiscal decentralization policies were implemented. In 1992, spending responsibilities in education, health, and social expenditure equivalent to roughly 1 percent of GDP were transferred from the federal government to the provincial administrations. In the area of health care, all hospitals and clinics operated by the federal government were transferred to the provincial governments. While most social welfare programs were transferred to the provinces, about two-thirds of the remainder of the budget of the Ministry of Health and Social Welfare (about 1 percent of GDP) was channeled to the provinces through FONAVI, which was transferred to the provinces in 1992. While nominal responsibility for secondary education was transferred to the provincial administrations in 1992, the actual transfer was arranged separately with each province over a two-year transition period. The provinces already had responsibility for primary education before the pact went into effect, so that with these additional changes, postsecondary education is now the only level of education remaining under the responsibility of the federal government.

Notwithstanding the transfer of spending responsibilities to the provinces, concerns about the need to rein in provincial spending resulted in an agreement that current expenditures that were financed by coparticipated revenues, including outlays for public services transferred to the provinces, could only grow by 10 percent in nominal terms in 1993, the same rate as the projected growth of nominal GDP at the time. Any coparticipated revenue in excess of the projected amount would have to be used to cancel provincial government debt or to finance capital expenditure. In addition, provinces would be responsible for the debt service on loans contracted with international organizations for infrastructure projects executed in their territories.

The second federal-provincial fiscal pact, which was proposed in August 1993 and went into effect in 1994, carried the reforms a step further by (1) encouraging the provinces to carry out deregulation and privatization; (2) offering provinces an option to shift their troubled social security systems for provincial government employees to the national system; and (3) supporting the reduction or elimination of provincial taxes that affect enterprise costs or may impede the development of financial markets.

Tax reform was clearly the centerpiece of the second fiscal pact, as shown in Tables 7, 8, and 9. Provinces adhering to the pact committed themselves to eliminating stamp taxes on checking accounts, taxes on the transfer of fuel, gas, and electricity, and, most important, phasing out the provincial turnover tax. While the provincial turnover tax constitutes the largest source of provincial own tax revenue, it is a cascading tax, constitutes a drag on enterprise costs, benefits imports over domestic products and increases the cost of exports, has a tax base that overlaps with the federal VAT, and makes it difficult to audit inter-provincial transactions.

Table 7.Argentina: The Second Fiscal Pact—An Overview
(1) Provinces that are signatories receive:
The minimum amount of coparticipated revenues is increased to Arg$740 million a month, compared with a minimum of Arg$725 million a month that was established under the previous federal pact that was in force during 1993.
To the extent that they eliminate stamp duty and taxes on gross income for productive activities, provinces are not required to refund advances made by the Treasury since August 1992 to comply with the guaranteed coparticipation minimum (during September 1992 to June 1993; advances from the federal government amounted to Arg$0.9 billion).
The option to transfer their pension funds to the national system, including the deficits they generate, that is, approximately Arg$1.2 billion a year.
Political guarantees to negotiate the offsetting of claims and debts between the provinces and the federal government.
(2) Productive sectors1receive:
Exemptions from the provincial turnover tax—the process must be completed by June 30, 1995, the provinces having the option to apply it partially and gradually—or primary production; industry; mining; tourism; financial services; savings and investment companies and mortgage security companies; the provision of electricity, water, and gas services, except for those generated in homes for domestic use.
Exemptions from stamp duties for financial or insurance operations for the agricultural, industrial, mining, or construction sectors, with the commitment that this will fully apply to the remaining operations and sectors by June 30, 1995.
Exemptions from specific provincial taxes levied on transfers of fuel, gas, electricity, including taxes on self-generated energy, and domestic services.
The elimination of rates or taxes that are levied directly or indirectly on the flow of goods among jurisdictions or the use of physical space, including airspace, for services.
Waiver for the taxes on interest earned on fixed term and savings bank deposits, banking debts, and gradually all taxes levied on payroll, with completion of this waiver process by June 30, 1995.
Relief from the assets tax, to the extent that they are affected by the repeals and exemptions arranged by each province in connection with stamp tax.
Reductions of 30–80 percent in rates of employers’ social security contributions, applying only to those sectors that are exempt from stamp duties and turnover taxes.
(3) Provinces are also required to:
As of January 1, 1994, revise taxes on real estate property, so that in no case they exceed, for rural real estate, 1.2 percent; suburban real estate, 1.35 percent; and urban real estate, 1.5 percent of the taxable base. The taxable base may not exceed 80 percent of the market value of the real estate.
As far as possible, strengthen the tasks of auditing and supervising compliance with tax obligations, implementing standard systems that give precedence to regimes of at-source withholding and collection or payment on account.
Within three years, replace provincial turnover taxes with a general consumption tax, with a view to ensuring tax neutrality and an improved competitiveness of the economy.
Move toward the full or partial privatization, or leasing/concessions to the private sector of provincial public enterprises.
Undertake deregulation, removing the restrictions on the supply of goods and services and on interventions in the various markets.
Adopt rules consistent with national legislation on occupational accidents.
Source: Ambito Financiero of January 21, 1994.

Agricultural production, industry, construction, mining, tourism, and scientific and technological research.

Source: Ambito Financiero of January 21, 1994.

Agricultural production, industry, construction, mining, tourism, and scientific and technological research.

Table 8.Argentina: Tax Stipulations of the Second Fiscal Pact of August 1993—Obligations of the Provinces1,2
TaxMeasures AdoptedSector or Activity
Stamp taxRepealAny institutionalized financial or insurance operation for the agricultural, industrial, mining, or construction sectors.
Specific provincial taxes and municipal taxesRepealTransfers of fuel, gas, electricity, including taxes on self-generated energy, and similar services.3
Taxes on interesr on fixed term and savings bank deposits and on banking debtsRepeal
Provincial turnover taxesEliminate
  • Primary production, financial services provided by financial institutions.

  • Savings and investment companies, mortgage security issuing companies, private pension funds (AFJPs).

  • Mutual fund management companies.

  • Insurance companies, with respect to income from their specific activity.

  • Currency transactions, with respect to income from such activity.

  • Production of goods, not including income from sales to consumers.4

  • Provision of electricity, water, and gas services for commercial and/or industrial purposes.

  • Real estate construction.

  • The elimination process should be completed by June 30, 1995.

Real estate property taxesReviseAs from January 1, 1994:
  • Average tax rates should not exceed, for rural real estate, 1.2 percent; suburban real estate, 1.35 percent; and urban real estate, 1.5 percent.

  • The taxable base should not exceed 80 percent of the market value of urban and suburban real estate or the value of undeveloped land in the case of rural real estate.

Road taxes and road maintenance taxesReviseIt is recommended to municipal governments that these or similar taxes not exceed 0.40 percent of the value of the provincial taxable base and be adjusted to reflect the cost generated by the actual provision of the service.
Taxes on driver’s licensesRevise and coordinateThe obligation will be to ensure the uniformity of valuations or applicable tax rates among all jurisdictions as of 1994. Valuations published by the Directorate General of Taxation (DGI) are to serve as reference.5
Sources: Based on information provided by Instituto de Estudios Económicos sobre la Realidad Argentina y Latinoamericana (IEERAL); and Fundación Mediterránea.

It was also agreed to work toward a repeal of those municipal taxes that are applicable to the same tax bases as provincial taxes.

In addition to the measures mentioned in the table, the provinces also accept the obligations to (1) move toward the partial or complete privatization of services, works, or projects that are the responsibility of provincial or municipal governments; (2) remove the restrictions on the supply of goods and services and on interventions in the various markets; and (3) adopt the rule established in Article 16 of Law No. 24,028, for determining jurisdiction in occupational accident cases.

The inclusion of the remaining operations and sectors should be gradual, and to be completed by June 30, 1995, in accordance with the schedule to be determined by each province.

These will be treated in the same way as the retail sector.

If this tax is wholly or partly the responsibility of municipal governments, then it will be proposed to these governments that they align themselves with this system.

Sources: Based on information provided by Instituto de Estudios Económicos sobre la Realidad Argentina y Latinoamericana (IEERAL); and Fundación Mediterránea.

It was also agreed to work toward a repeal of those municipal taxes that are applicable to the same tax bases as provincial taxes.

In addition to the measures mentioned in the table, the provinces also accept the obligations to (1) move toward the partial or complete privatization of services, works, or projects that are the responsibility of provincial or municipal governments; (2) remove the restrictions on the supply of goods and services and on interventions in the various markets; and (3) adopt the rule established in Article 16 of Law No. 24,028, for determining jurisdiction in occupational accident cases.

The inclusion of the remaining operations and sectors should be gradual, and to be completed by June 30, 1995, in accordance with the schedule to be determined by each province.

These will be treated in the same way as the retail sector.

If this tax is wholly or partly the responsibility of municipal governments, then it will be proposed to these governments that they align themselves with this system.

Table 9.Argentina: Tax Stipulations of the Second Fiscal Pact of August 1993—Obligations of the Federal Government1
TaxMeasures AdoptedContent
Gross assets taxRepealAssets earmarked for productive processes in the specific priority sectors affected by the repeal and exemptions of provincial stamp taxes.
Payroll taxesRepealReduction in employers’ social security contributions for specific sectors. This will be effected in keeping with the sectoral priorities established for the elimination of provincial turnover taxes.
VATReviseAdjustment of the rules relating to VAT withholdings and payments on account, so that in no case shall taxpayers be liable for an effective tax rate exceeding 18 percent.
Other commitments
  • To accept the transfer of provincial pension funds to the National Social Security System.2,3

  • To organize and implement the system of rural mortgage bonds and negotiable bonds backed by urban mortgages.

  • 60-day suspension of the withholding of coparticipation surpluses in excess of Arg$725 million.4

  • The guaranteed minimum monthly revenue transfer to the provinces is increased from Arg$725 million to Arg$740 million as of January 1994.

Sources: Based on information provided by Instituto de Estudios Económicos sobre la Realidad Argentina y Latinoamericana (IEERAL); and Fundación Mediterránea.

The federal authorities agree to revise the taxes collected by the Municipality of Buenos Aires (MCBA) along the same lines and on the same terms as those undertaken by the provinces.

Excluding the professional funds stipulated by Article 56 of Law No. 10,038.

In cases where provincial social security systems are included in the new social security arrangements approved by the federal authorities.

This guarantee amount was established under the fiscal pact concluded on August 12, 1992 and ratified by Law No. 24,130. This suspension was to be valid provisionally for 60 days and permanently once the provinces have met the obligations for immediate application undertaken pursuant to this agreement.

Sources: Based on information provided by Instituto de Estudios Económicos sobre la Realidad Argentina y Latinoamericana (IEERAL); and Fundación Mediterránea.

The federal authorities agree to revise the taxes collected by the Municipality of Buenos Aires (MCBA) along the same lines and on the same terms as those undertaken by the provinces.

Excluding the professional funds stipulated by Article 56 of Law No. 10,038.

In cases where provincial social security systems are included in the new social security arrangements approved by the federal authorities.

This guarantee amount was established under the fiscal pact concluded on August 12, 1992 and ratified by Law No. 24,130. This suspension was to be valid provisionally for 60 days and permanently once the provinces have met the obligations for immediate application undertaken pursuant to this agreement.

Initially, the provinces were slow to join this second pact, largely because of the revenue implications of the tax reforms, particularly the initial stipulation to abolish the provincial turnover tax before June 1995. While the provinces were free to replace the turnover tax with other taxes, many have not yet done so. Some provinces have begun to replace the turnover tax with a provincial tax on final retail sales. While this would preserve revenue, it is not most efficient as it exploits the same tax base twice, by the federal level (VAT) and the provincial level (turnover/sales tax), and with separate administrative structures. Also, a retail sales tax is notoriously difficult to administer, especially in a country with a very fragmented retail sector.

Overall, there is no easy short-term alternative for replacing the provincial Turnover tax. Various possible alternatives (such as a provincial level VAT (PVAT), a provincial final retail sales tax, a provincial wholesale and retail sales tax, a provincial income tax, or increased reliance on real estate taxes) have their own specific drawbacks.14

Other alternatives for improving provincial revenue would be beneficial in the long run, but would not yield short-term results. For example, as Crotty and dos Santos (1996) point out, broadening federal VAT and income tax bases, and tackling evasion through a soundly based and well-executed program of audit and enforcement measures at the federal level have the potential in the long term to produce more revenue for the provinces through revenue sharing or through a provincial surcharge on these taxes, but would not yield short-term results. Similarly, improving real estate taxation would require substantial initial efforts, including, for example, improving property mapping and property registries; providing better and more consistent application of valuation techniques; improving the exchange of information between local tax offices, property registries, local building license authorities, and public utilities; and improving collection and enforcement procedures.15

The announcement in December 1993 that federal payroll taxes levied on employers would be reduced, depending on region and sector, in those provinces participating in the second pact, increased pressure on provincial governments to join. By May 1994, all but one provincial legislature had ratified the second fiscal pact, and most had taken at least some initial steps toward implementation. Also, the provinces were given a minimum revenue guarantee and some other guaranteed fixed payments that provided a floor of federal transfers equivalent to about 4.5 percent of GDP annually.

The second fiscal pact clearly shows the “horse-trading” that is involved in implementing structural reforms of the system of fiscal federalism, that is reminiscent of the old adage “two steps forward, one step back.” An example is the reduction in federal employer payroll taxes, which reduced enterprise costs (“the two steps forward”), but came at the expense of making payroll taxes an explicit instrument of regional and sectoral policies, and contributed further to the growing social security deficit (“the one step back”).

The reduction in the federal employer payroll tax rate became a centerpiece of the second fiscal pact, even though initially it was somewhat of an afterthought largely designed to induce provinces to participate.

Previously, the employer payroll tax had amounted to 33 percent, with the exception of three provinces, where it was 28.5 percent. The new rates, which began to take effect in most provinces in early 1994, implied reductions from anywhere between 0 to 80 percent, depending on region and sector of production. Economically weak regions received larger reductions, and these reductions were restricted to the agriculture, industry, construction, mining, and scientific and technological research sectors. Tourism and the service sectors (basically all nontradables) received no reductions. About 50 percent of the labor force was affected by these reductions in employer payroll taxes.

The new system of employer payroll tax rates was excessively complicated and distortive, as it meant that different industries in the same province could have different employer payroll tax rates, and the same industry in different provinces could have different tax rates. Even within the same province, employer payroll contributions were differentiated according to rural or urban location. For example, in the province of Entre Ríos, the new employer payroll tax rate was 18.2 percent in the capital city, 13.2 percent in the district of Feliciano y Federación, and 16.5 percent in the rest of the province.

In March 1995, the new system of employer payroll tax rates was simplified, and tax rates were generally unified across sectors at a level of 16.5 percent, half the previous rate, with some differentiation according to provinces still being maintained. This involved rate reductions in the services and tourism sectors, and a partial rollback of previous reductions in most other sectors.

The second fiscal pact gave to provincial governments the option to transfer to the federal social security system the public pensions systems for provincial government employees. These systems all had actuarial deficits.16 This option implied that the federal government had to be ready to absorb additional social security deficits of about 0.5 percent of GDP initially, but, given the large actuarial deficits in some provinces, over time, the cash deficits would become even more significant. Another example was the agreement under the second pact that the federal government would forgo provincial government repayment obligations equivalent to 0.4 percent of GDP, which were related to higher-than-mandated coparticipation transfers,17 and to increase, albeit only by a small amount (Arg$17 million a month), the fixed minimum guaranteed coparticipation transfer.

Tax Administration and Expenditure Management

Weak tax administration and expenditure management systems contributed to large fiscal imbalances at all levels of government during the 1980s. Since then, major improvements have been made at the federal level; it remains a major challenge to translate these into improvements at subnational levels of government.

Since 1989, major efforts have been made to strengthen tax administration, particularly at the federal level, and these helped to boost tax revenue collections, which rose to 15.5–16.5 percent of GDP during 1993–95, from about 11–12 percent of GDP at the beginning of the decade.18 VAT compliance, for example, increased from an estimated 34 percent in 1989 to 64 percent in 1992, but dropped back to 60 percent in 1993, and to 55 percent in 1994 (Crotty and dos Santos, 1996). Own revenues of the provinces also improved during this period, by about 1.5 percent of GDP. In both cases, however, some of the improvement may not have been due to better tax administration, but to the reversed Tanzi effect that set in when inflation was drastically reduced under the convertibility scheme. In addition, Argentina still falls short of the tax compliance levels that prevail in neighboring countries, such as Chile (where VAT compliance, for example, is over 80 percent) or Uruguay (where it is over 70 percent) (Crotty and dos Santos, 1996).19 Similarly, several revenue sources still appear underexploited in Argentina, like, for example, income taxes, which yielded 2.2 percent of GDP in Argentina in 1995, but yielded over 4 percent of GDP in neighboring Brazil and Chile.

Still, the growth in national tax revenue was remarkable, also in light of the fact that over 20 taxes, mainly inefficient and distortive tax handles, were abolished during that period. Among the administrative measures that contributed to this growth were the strengthening of penalties for evasion, computerization of tax returns, and improved training of auditors. In April 1993, responsibility for collecting social security contributions was transferred from the social security system to the national tax service, which helped significantly to improve compliance. However, in contrast with all these improvements at the federal level, most provincial and municipal tax administrations remain weak.

A further current problem is reporting systems that make it difficult to consolidate the accounts of different levels of government. Some of the fundamentals for improving the flow of information are being put in place, though. In particular, the integrated financial information system (SIDIF) of the national administration has developed into a powerful tool for controlling the execution of the federal budget. Unfortunately, it only covers the federal level: provincial and municipal governments formulate their budgets independently from the federal government and are not required to consult with the federal government or even to provide data on budget execution.

To be sure, coordination between different levels of government, particularly between the federal and provincial levels, has been intensified over the last couple of years (also in the context of the two federal-provincial fiscal pacts). However, because of the statutory independence of provinces and municipalities, federal government attempts to coordinate the operations of subnational governments in line with national macroeconomic stabilization and adjustment objectives usually required lengthy negotiations and many compromises.

Given the need for further reform and adjustment at all levels of government and given that decentralization will continue, it would seem desirable to strengthen provincial and municipal government reporting and management systems along the lines of the federal system. The key purpose of such systems would be to provide a comprehensive tool for all levels of government to help with cash management, financial planning, and public debt management. Such systems have worked well in other federal states, including neighboring Brazil, where one has been in existence for over a decade.

Implications for Macroeconomic Management and Structural Reform

In 1990, the World Bank, in analyzing provincial government finances in Argentina and their macroeconomic implications, concluded that “provincial-national financial practices have contributed to unsustainable public sector fiscal and quasi-fiscal deficits, and their continuation would undermine national efforts to attain price stability and to promote sustainable economic development.”20

Since then, the efforts of the federal government to stabilize and restructure the economy have met with considerable success. Similarly, as a result of substantial progress in restructuring the system of fiscal federalism, it is no longer true that “Argentina provides a good illustration of the ‘fiscal perversity’ of subnational governments” (Prud’homme, 1995, p. 206), as had been the case during much of the 1970s and 1980s.

Often, the provinces have been blamed for a large part of Argentina’s problems with macroeconomic management. Sanguinetti (1994), for example, has suggested that the fiscal behavior of the provinces was a major culprit in various failed stabilization attempts of the past, particularly the fiscal crises of 1975 and 1983, and the failed stabilization attempt during 1985–87 under the Austral Plan. This claim is supported by the data shown in Table 10. From 1970 to 1975, the balance of the consolidated nonfinancial public sector worsened by 13.6 percentage points of GDP, to which the provinces contributed the most. Similarly, from 1980 to 1983, when the balance of the consolidated nonfinancial public sector worsened again by about 8 percentage points of GDP, the provinces again contributed the most. During 1983–87, when the overall balance of the consolidated nonfinancial public sector improved by over 7 percentage points of GDP, the overall balance of the provinces continued to deteriorate by almost 2 percentage points of GDP. By contrast, during 1975–80, provincial finances improved significantly more than those of the rest of the nonfinancial public sector.

Table 10.Argentina: Changes in Consolidated Nonfinancial Public Balances During Four Time Periods

(In percentage points of GDP)1

1970–751975–801980–831983–871990–94
Overall change−13.67.9−8.17.32.7
Provinces−5.94.8−3.3−1.90.5
Other−7.63.1−4.99.32.2
Federal government−5.53.3−2.87.40.7
Stare enterprises−2.30.1−1.32.0−0.92
Social security0.2−0.3−0.8−0.1
Other32.4
Sources: Sanguinetti (1994); Argentine authorities; and authors’ estimates.

An improvement in the balance is denoted by a positive number: a worsening by a negative number.

Data for 1990–94 refer only to the operating surplus of nonfinancial stare enterprises.

Primarily consists of the balance of decentralized agencies and the balance of quasi-fiscal operations of the central bank.

Sources: Sanguinetti (1994); Argentine authorities; and authors’ estimates.

An improvement in the balance is denoted by a positive number: a worsening by a negative number.

Data for 1990–94 refer only to the operating surplus of nonfinancial stare enterprises.

Primarily consists of the balance of decentralized agencies and the balance of quasi-fiscal operations of the central bank.

It may be argued that the failure of provincial governments to adjust was partly a reflection of distorted incentive systems. For example, as Tanzi (1996) has pointed out, historically, four provinces used to have legal authority to grant exemptions from the national VAT. At least in the short run, using their authority was clearly in the best self-interest of these four provinces, but it resulted in a considerable erosion of VAT revenue at the federal level, and aggravated national fiscal problems.

Also, for the more recent period of 1990–94, Table 10 would suggest that both provincial governments and the federal government contributed to the adjustment of the finances of the consolidated nonfinancial public sector, but that other factors, such as ending quasi-fiscal operations of the central bank under the convertibility law, contributed the most.

Finally, without underestimating the efforts of the federal government to increase tax revenue at the federal level, and while recognizing the revenue windfall that this created for the provinces (via the automatic revenue-sharing arrangements), it should also be noted that the improved fiscal performance of the central government has reflected in part the policy of decentralizing expenditure responsibilities without a corresponding decentralization of revenue-raising powers or tax bases.

Clearly, numerous problems remain to be addressed. In particular, macroeconomic management has been complicated by the fact that adjustment initiatives usually originated at the federal level, and, up to now, have largely failed to translate into similar adjustment initiatives at the provincial level. As Tanzi (1996) has suggested, by and large, the attitude of provincial and municipal governments has been that economic stabilization is a national public good and is thus the sole responsibility of the federal government. One reason for this may be that the nature of the revenue-sharing arrangements has allowed provincial government to delay adjustment, and thereby water down the macroeconomic impact of federal initiatives. For example, faced with the need to correct large macroeconomic imbalances, the federal government introduced major tax and administrative reforms that succeeded in raising sharply the ratio of taxes to GDP. Through these efforts at the federal level, provinces received an automatic revenue windfall via the various revenue transfer mechanisms. The financial problems the provinces experienced during the 1995 recession reflected difficulties in cutting back expenditures in line with reduced transfers, particularly from coparticipation.

It has been suggested that not only the vertical revenue-expenditure assignment imbalance but also the deficient design of the transfer system itself remain fundamental weak points in the otherwise successful stabilization program now under way (López Murphy and others, 1995). The current revenue sharing arrangements rely almost exclusively on fixed shares, that is, a certain percentage of whatever is collected is transferred to the provinces (Figure 1).21 Hence, when economic activity is high or when reforms at the federal level increase tax revenue, transfers are high, thus allowing provinces to increase spending; when economic activity is low or tax evasion or avoidance is on the increase, transfers are low (but not below the minimum guarantee). This tends to exacerbate cyclical fluctuations in activity and to frustrate in part restrictive budgetary policies at the federal level.

At the same time, and given the need to raise revenue, the federal government has had an incentive to focus its efforts on those taxes that mainly accrue to the federal level. As Tanzi (1996) has suggested, this may easily lead to a situation where nonshared taxes will acquire a greater weight in the tax system, even when they are less efficient. In Argentina, the incentives to focus on nonshared taxes for increasing revenue efforts have prompted the federal government to raise import-related revenues, including through a temporary import surcharge in the form of a “statistical tax” or by increasing to the permitted maximum the list of MERCOSUR imports that still have to pay import duty. Also, there has been a tendency to implement different pre-coparticipations (most recently, in the form of a temporary 3 percentage point increase in the VAT that was initially excluded from the revenue-sharing arrangements and given directly to the federal level). As argued by Porto and Sanguinetti (1993), these pre-coparticipations are reminiscent of the early 1980s, when similar arrangements undermined the revenue-sharing mechanism that was in place and contributed to its demise at the end of 1984.

A new revenue-sharing mechanism needs to be put into place in the near future. While it is difficult to envision that a radically different mechanism could be negotiated easily, it would clearly be desirable to address the major shortcomings of the current mechanism. This would include addressing the need further to reduce discretionary elements, broaden the taxes included in the revenue-sharing arrangement (ideally, all taxes should be shared to an equal extent), and to reduce the procyclicality of transfers, maybe by establishing a stabilization fund (even though it should be made clear that this fund would not be available for discretionary transfers).

Similarly, problems remain concerning progress with structural reform, particularly at provincial and municipal levels. Notwithstanding substantial progress in restructuring Argentina’s system of fiscal federalism, most recently under the two federal-provincial fiscal pacts, structural reforms at subnational levels have been lagging; a key issue in this regard is government employment, particularly at the provincial level. During 1983–90, when the provinces enjoyed relatively easy access to extraordinary financing, provincial civil service employment expanded by about 48 percent, and made provincial governments an employer of last resort, particularly in the poorer provinces. For example, of the four provinces that in 1992 raised less than 10 percent of their total expenditure through own revenues (Catamarca, Formosa, La Rioja, and Santiago del Estero), three ranked highest with regard to the level of public employees per 100 inhabitants. Similarly, the four provinces that raised more than 30 percent of their total expenditures through own revenues in 1992 (Buenos Aires, Córdoba, Mendoza, and Santa Fé) were the only ones that had less than ten public employees per 100 inhabitants. In general, provinces with the lowest level of own revenue mobilization (relative to total expenditures) are also the ones where civil service employment (per 100 inhabitants) is highest, and vice versa: during the 1990s, the correlation coefficient between own revenue mobilization (relative to total expenditure) and civil service employment of the various provinces has remained fairly stable at about −0.70.

In conjunction with the federal-provincial fiscal pacts, various programs were developed to encourage the provinces to downsize provincial government employment through incentives payable to government employees and to private firms that might be prepared to hire them. The most innovative of these schemes, the BOCEP program, functions as a marginal employment subsidy. BOCEPs, or bonds for the creation of private employment, are bonds given to provincial public sector employees who are retrenched on a voluntary basis. The retrenched workers are to hand over their BOCEPs to their new private sector employers, who can, in turn, use these bonds to obtain a loan for investment purposes from the state-owned National Bank; BOCEPs are then amortized over five years.22

However, so far, provincial public sector employment reduction schemes have all met with only limited success. On average, provinces continued to spend over 50 percent of their total expenditure on wages and salaries (see Table 2), and, even though, according to newspaper reports, about ten provinces have been experiencing repeated problems in paying civil service wages on time, major employment adjustments remain yet to be made.

To be sure, federal government efforts to improve macroeconomic management and introduce various structural reforms have forced adjustment upon subnational levels of government, as, for example, evidenced in the fact that own revenues of the provinces grew by over 50 percent in real terms during 1989–95. At least for some provinces, the need to adjust may have arrived faster than anticipated, as suggested, for example, by recent strikes of public servants in several provinces. Usually, the less-developed provinces were the ones that had the most trouble adjusting, even though, as shown, for example, by the salary arrears public workers in Córdoba (Argentina’s second largest city) and other parts of the country had to endure mid-1995,23 adjustment has been difficult pretty much everywhere.

Fundamental reforms of the system of fiscal federalism in Argentina are inevitable. These reforms need to take into account the complex interests and needs of the various participants: the central government, the richer provinces, and the poorer provinces. In a first step, a possible reform package could comprise three main elements: (1) strengthening the own tax base of the provinces; this could include passing on to the provinces the federal income tax for natural persons (personal income tax) or creating a “piggyback” provincial income tax, and improving real estate taxation, maybe with guidance provided by the federal level; (2) reducing the overall degree of revenue sharing, and, in this context, replacing the complicated revenue-sharing arrangements that currently exist for each specific tax by a more simple system where the provinces receive a certain fixed (but lower) share of all revenue collected; and (3) establishing an explicit equalization scheme among the provinces that supports the poorer provinces.

We would like to thank Ehtisham Ahmad, Jerald Schiff, Ernesto Stein, Teresa Ter-Minassian, and the participants of a Fiscal Affairs Department seminar on “Argentina: Fiscal Federalism and Provincial Tax Issues” for helpful comments and suggestions, and Lina Annab and Derek Bills for excellent research support.

Many of the social security systems for provincial government employees are being incorporated into the national social security system, a process that started in 1994.

Low -density provinces have been separated out because they may be considered as a special case. For example, many low-density provinces have relatively high per capita own revenue due to natural resource rents. They also have high per capita expenditure due to indivisibilities; that is, basic infrastructure and services have to be provided independent of population density.

Unemployment insurance is also provided by the private sector.

As defined here, the federal level includes the federal government, decentralized agencies, the national social security system, and the operating surplus of national public enterprises.

Sales taxes, such as the provincial turnover tax, are defined as “direct taxes” in Argentina.

Catamarca, Corrientes, Formosa, Jujuy, La Rioja, and Santiago del Estero.

Fur details, see Gómez Sabaini (1993); data for earlier periods are shown in Fundación de Investigaciones Económicas Latinoamericanas (1993).

Figure 1 is a revised and updated version of a similar chart in Cetrangolo (1993); also see Ministry of the Interior (1996).

In some instances, provinces received other discretionary transfers in the form of reimbursable advances against future coparticipation revenue.

See, for example, Vernango (1995).

Arg$l equals US$1.

At the federal level, however, there was a clear overall gain as the national social security system is part of the federal public sector.

See Crotty and dos Santos (1996) for an elaboration of advantages and disadvantages of different alternatives.

Still, there is considerable room for improving the effective tax base of real estate taxes, which in 1995 accounted for only 18 percent of all own revenue of the provinces. For example, in the city of Santa Fe, a recent survey showed that two-thirds of the lots that were vacant according to the tax roll actually had buildings on them, and 56 percent of all properties were underrecorded in the sense that the extent of the construction that was recorded was less than what was actually on the lot. A similar survey in the municipality of Santo Tomé found 52 percent of all properties to be underrecorded. Also, a 1994 tax amnesty in the province of Buenos Aires led 400,000 taxpayers, including 170,000 whose land was vacant according to the tax roll, to report 30 million square meters of construction previously unknown to the tax authorities (Provincia de Buenos Aires, 1994).

Many provincial social security systems had more generous benefits than the federal system. Provincial social security systems that would be transferred to the federal system would operate henceforth under the same regulations as the federal system.

Coparticipated revenue transfers in excess of the monthly minimum contained contingent repayment obligations by the provinces. The federal government elected to waive this repayment obligation as part of the second pact.

Since inflation was drastically reduced at the same time that tax administration was being improved, it is difficult to separate out revenue increases that are due strictly to the reduction in inflation (through the reversed Tanzi effect), and revenue increases that are only due to improvements in tax administration.

The task of improving tax compliance in Argentina has been complicated by frequent tax amnesties that have been used by federal, provincial, and municipal authorities to allow people to pay previously underreported or delinquent taxes with reduced penalties. As a result, a prevailing attitude has been: “why should 1 pay my taxes if 1 can wait for the next amnesty?” (Crotty and dos Santos, 1996).

World Bank (1990). Similar arguments can also be found in World Bank (1992).

However, the provinces have a minimum guarantee under the coparticipation scheme, that is, they are legally guaranteed a minimum transfer each month, even if total revenues collected were to fall to zero.

For a detailed description of this scheme, see Montoya (1994).

See, for example, Vernango (1995).

References

    Ambito Financiero1994 issue of January21.

    Artana Daniel and Ricardo López Murphy1994“Fiscal Decentralization: Some Lessons for Latin America,”paper presented at the annual meeting of the Latin American Econometric SocietyCaracas.

    ArtanaDanielOscarLibonattiCynthiaMoskovits andMarioSalinardi1995“Argentina,” in Fiscal Decentralization in Latin Americaed. by Ricardo LópezMurphy (Washington: Inter-American Development Bank).

    CetrangoloOscar1993“El Pacto Fiscal y las Relaciones Financieras entre la Nación y las Provincias” (Buenos Aires: Fundación Union Industrial ArgentinaOctober).

    CrottyJohn andPaulo dosSantos1996“Provincial Tax Issues,”paper presented at a seminar on Fiscal Federalism and Provincial Tax Issues (Washington: Fiscal Affairs Department, Internatinal Monetary FundJuly).

    Fundación de Investigaciones Económicas Latinoamericanas (FIEL)1993Hacia una Nueva Organización del Federalismo Fiscal en la Argentina (Buenos Aires: Fundación de Investigaciones Económicas Latinoamericanas).

    GómezSabainiJuanC.1993“Coordinación de la Imposición General a los Consumos entre Nacion y Provincias (Argentina),”Serie Política FiscalNo. 47 (Santiago: United Nations Economic Commission for Latin America and the Caribbean).

    LópezMurphyRicardoOscar Libonatti andMarioSalinardi1995“Overview and Comparison of Fiscal Decentralization Experiences,” in Fiscal Decentralization in Latin Americaed. by Ricardo LópezMurphy (Washington: Inter-American Development Bank).

    Ministry of the Interior Undersecretary for Assistance to the Provinces1996Co-participación Federal de Impuestos—Un Tema de Todos y Para Todos (Buenos AiresNovember26).

    MontoyaSilvia1994“Diagnóstico y Perspectivas del Empleo Público Provincial,”Novedades EconomicasJanuary–February pp. 4555.

    PortoAlberto andPabloSanguinetti1993“Descentralización Fiscal en America Latina: El Caso Argentino,”Serie Política FiscalNo. 45 (Santiago: United Nations Economic Commission for Latin American and the Caribbean).

    Provincia de Buenos Aires1994Noticias de EconomiaSeptember/October.

    Prud’hommeRémy1995“The Dangers of Decentralization,”World Bank Research ObserverVol. 10No. 2 (August) pp. 201220.

    RezkErnestoMarceloCapello andCarlos AlbertoPonce1996“Arreglos Fiscales Intergubernamentales en Argentina,”draft presented at the International Seminar on Fiscal Federalism and Federal Finances in Córdoba, Argentina,June27–28.

    SanguinettiPablo J.1993“The Politics of Intergovernmental Transfers and Local Government Deficits: Theory and Evidence,”Estudios Economicos (Mexico) pp. 87109.

    SanguinettiPablo J.1994“Intergovernmental Transfers and Public Sector Expenditures: A Game- Theoretic Approach,”Estudios de Econamía (Chile) Vol. 21No. 2 (December) pp. 181212.

    SchulerKarl1995“Rauher Wind Über der Pampa,”MosquitoNo. 9/10 (December) pp. 3437.

    ShahAnwar1994“The Reform of Intergovernmental Fiscal Relations in Developing and Emerging Market Economies,”Policy and Research Series No. 23 (Washington: World BankJune).

    TanziVito1996“Fiscal Federalism and Decentralization: A Review of Some Efficiency and Macroeconomic Aspects,” in Annual World Bank Conference on Development Economics 1995 (Washington: World Bank).

    TeijeiroMario1996“La Política Fiscal Durante la Convertibilidad” (unpublished;Buenos Aires: Centro de Estudios PúblicosApril).

    Ter-MinassianTeresa1996“The Role of the Treasury in the Financial Management of Government Operations: The Brazilian Experience in an International Perspective,”paper presented at the First Brazil Seminar on Public Finance,BrasiliaSeptember2.

    VernangoJavier1995“Diez Provincias sin Dinero Para Sueldos,”Ambito Financiero (January24 1995).

    World Bank1990“Argentina: Provincial Government Finance Study,”World Bank Report No. 8176-AR (Washington: World Bank).

    World Bank1992“Argentina: Towards a New Federalism,”World Bank Report No. 10612-AR (Washington: World BankJune).

    World Bank1996“Argentina: Provincial Finances Study,”World Bank Report No. 15487-AR (Washington: World BankJune).

    Other Resources Citing This Publication