Information about Europe Europa
Journal Issue
Share
Article

Italy: Selected Issues

Author(s):
International Monetary Fund
Published Date:
February 2005
Share
  • ShareShare
Information about Europe Europa
Show Summary Details

V. Fiscal Decentralization in Italy: Issues and Prospects130

Core Questions, Issues, and Findings

  • Italy has been reforming its intergovernmental fiscal relations over the last three decades, to give local governments more autonomy over taxation and spending. However, the reform process has been slow, reflecting political difficulties and concerns about the possible impact of a reform on overall fiscal discipline.
  • While recognizing that the design of intergovernmental fiscal relations must inevitably reflect political choices, this paper focuses on aspects that are key to the successful implementation of any federal fiscal framework: stability of norms and regulations, monitoring mechanisms, and the credibility of enforcement. To put these issues in perspective, the paper reviews experiences in other advanced countries.
  • The main recommendations call for strengthening mechanisms for coordinating budgetary policies and management across levels of government; ensuring appropriate recording and reporting of all subnational government operations; improving transparency and stability of rules and regulations; and backing rules with clear and effective sanctions for non-compliance.

A. Introduction

179. Over the last three decades, Italy has gradually implemented a process of fiscal decentralization. This process appears to be part of a general trend experienced in many countries as a result of a desire for greater government accountability (Ter-Minassian, 1997, Reforme, 2004). Therefore, the Italian experience with fiscal decentralization is not unique.

180. In 2001, a major constitutional reform dramatically changed the design of intergovernmental fiscal relations.131 The main guiding principles of the reform are codified in the new Title V of the Constitution, which delegates a number of functions to regions; it also sets the legal basis for widening regions’ fiscal autonomy by assigning tax and spending powers to them—a clear break from the centralist experience of the past.

181. Three years into the constitutional reform, however, the legislative framework for its implementation is not complete. Comprehensive implementing regulations have yet to be formulated—most importantly regarding the financing mechanisms of local governments, whose spending responsibilities will be significantly expanded. Meanwhile, proposals for another round of constitutional amendments that would further deepen the role of local governments are on the table.

182. Progress toward establishing the fiscal autonomy of local governments has been slow. This reflects in part the legislative delays noted above, but also concerns over widening of regional disparities and loss of fiscal discipline possibly induced by fiscal decentralization. Previous experience with decentralization in Italy, especially with the devolution of health expenditure in the 1970s, resulted in spending systematically and significantly exceeding budgetary allocations; ex post, the central government had to bail out local governments.

183. The possible costs associated with implementing fiscal decentralization have also been a source of concern. Over the medium term, Italy already faces daunting spending needs associated with an ageing population. Initiatives that may exacerbate spending pressures, threaten adherence to the commitments under the Stability and Growth Pact (SGP), and contribute to further increases in public debt (which already exceeds 100 percent of GDP) therefore need to be carefully assessed.

184. This paper reviews key issues emerging from Italy’s experience with fiscal decentralization. There is a vast literature on the normative approach to fiscal decentralization in Italy, focusing in particular on the appropriate design of tax assignments, expenditure mandates, and equalization systems.132 As this is well documented, the paper will discuss the framework of fiscal decentralization, focusing on aspects that have received less attention—but appear crucial for success, namely the stability of norms and regulations, the monitoring mechanism, and the credibility of enforcement. To put these in perspective, experiences in other advanced countries in these areas will also be reviewed.

185. The paper is organized as follows. Section B presents some main stylized facts about fiscal decentralization in Italy, and reviews the main legislative steps. Section C focuses on the key challenges highlighted above. Section D concludes.

B. Key Stylized Facts about Fiscal Federalism in Italy

186. Initial steps toward fiscal federalism were taken in the 1970s. Although “ordinary statute” regions were established in the 1948 Constitution, they began operating only during the 1970s.133 The transfer of limited administrative functions to regions started in 1972. By 1978, regions carried out most functions related to health care administration; however, the central government retained effective control over the design of health care policies. Over the same period, a tax reform vastly centralized revenue (see Box 1).

187. Reflecting these developments, expenditure by regions increased while their own revenue remained flat. Over the two decades starting in 1970, regional expenditure increased from 11 to 14.8 percent of GDP; own revenue remained relatively unchanged at under 3 percent of GDP. The deficit was covered by central transfers, mostly earmarked for health spending undertaken by regions.

188. During the 1990s, considerable progress was made in the decentralization of expenditure functions. Responsibilities in crucial sectors like health were assigned to regions, according to the so-called Bassanini-laws (for more detail, see IMF, 2000). As regions continued to be financed overwhelmingly by central government transfers, frequent spending overruns continued to be covered by the center (thus creating a perception of soft budget constraints).

Box 1:Italy: Main steps to regional fiscal decentralization

1948: The Constitution foresees the creation of regions; special powers are granted to five special statute regions.

1972-1977: “Ordinary statute” regions are set up. The tax reform of the early 1970s heavily centralizes revenue against the significant decentralization of expenditure.

1978: Health expenditure is largely decentralized to regions.

1992: Health service contributions and automobile taxes are attributed to regions.

1995: Specific state transfers are abolished and replaced by a share of the excise on gasoline; a new equalization fund is set up.

1997: A new tax on productive activities (IRAP) is introduced and assigned to regions; its yield more than replaces the abolished health social contributions.

1998: A surcharge on the personal income tax (IRPEF), by 0.5 percentage points, is introduced (offsetting an equal reduction in the national IRPEF).

2000: Legislative decree 56 replaces central transfers with tax revenue sharing (the most significant share is based on a time-varying formula for VAT to be assigned to regions).

2001: Law 3/2001 reforms Title V of the Constitution; a national vote ratifies it in October.

2002-to date: A number of legislative initiatives are put forward. In 2001, the “Bossi proposal” aims to significantly strengthen regional autonomy—procedures for its ratification are not followed up after the 2001 administrative elections. In 2003, the “La Loggia Law” (131/2003) delegates the government to define the “fundamental principles” delimiting regional legislative jurisdiction. In the spring of 2004, the Senate passes a new proposal for constitutional reform that would, among other things, radically change the reformed Title V.

189. A few local taxes were also introduced, although with limited regional authority to set tax bases and rates. The ability to impose taxes marked a shift from “derived” financing, with all financing provided via earmarked transfers from the center, to more “autonomous” financing, with block (not earmarked) grants, some regional sharing of central government taxes (such as the VAT) and own taxes; allocation of these funds was left to regional decision. Hence, regions’ own revenue significantly increased, from 2.8 percent in 1990 to 7 percent of GDP in 2000, in part replacing lower central transfers.

190. As a result, vertical fiscal gaps (defined as the ratio of central transfers to total regional revenue) declined (Figure 1). However, lower vertical fiscal gaps may not be a sign of deeper fiscal autonomy of regions, as these have continued to rely on the center for financing (Giarda, 2004). Despite higher own revenue, regions could not match spending increases, particularly for health care. Moreover, as most equalization mechanisms were defined in terms of spending (where spending limits were set by the central government), revenue from own taxes were deducted from “admissible” spending to compute the size of the equalization transfers—thus effectively curtailing the regions’ already limited financial autonomy and possibly creating a disincentive to raise own revenue.

Figure 1.Italy: Regions’ Vertical Fiscal Gaps, 1980-2003 1/

Source: Staff calculations, based on data from Franco, Messina and Zotteri (2004).

1/ Computed as the share of transfers received by regions over regions’ total revenue.

191. The need to enhance regions’ fiscal discipline induced a rethinking of the system of intergovernmental fiscal relations. Persistent financial difficulties called for increasing accountability of local administrators by assigning wider responsibilities and strengthening the financial means to undertake them. Backed by a reform in the electoral system—with direct election of regional governors and municipal mayors bringing local authorities closer to their electorate—time was ripe for a reform of intergovernmental relations. A first major step was the passage of a legislative decree in February 2000 (Law 56/2000) which replaced most of the previous central transfers with sharing of central taxes, most notably the VAT.

192. The process culminated in the reform of Title V of the Constitution in 2001. This reform moved further toward a federal structure based on the subsidiarity principle—functions shall be allocated to the lowest level of government with the exception of services that cannot be performed at the local level. Only the functions explicitly stated in the constitutional law remain exclusive jurisdiction of the central government—according to Article 117, these cover some strategic areas, such as foreign affairs, justice, defense, monetary policy, electoral rules, social security, and other sectors that have to be dealt with nationwide. In a number of areas, the central government and regions have “shared” jurisdiction—where the regions can legislate within the “fundamental principles established by the law of the state.” All other functions not explicitly mentioned in the Constitution were transferred under regional jurisdiction.134

193. The new Article 119 establishes revenue raising and spending autonomy for regions within the framework of general principles set by the state.135 It also allows local governments to borrow—without state guarantee—to finance investment. It mandates both “an equalization fund, without allocation commitments, for territories having lesser fiscal capacities per inhabitant” as well as the possibility for the government to allocate “additional resources” to specific areas.

194. In 2002, a High Commission was mandated to design mechanisms for the implementation of Article 119. While it was initially envisaged that the Commission would complete its work by end-March 2003, due to procedural delays it did not start working until the summer of 2003. The 2004 Budget Law extended the mandate of the High Commission to end-September 2004, in October 2004, the mandate was further extended through end-September 2005. In the meantime, the government would not propose any reform to implement Article 119; and the freezing of all subnational powers over their own taxes, initiated in 2003, was extended to 2004.

195. After two years and three extensions of its mandate, the High Commission has yet to publish its work. Although the Commission stands ready to do so, the key challenge is the political will to come to a firm conclusion on these issues—exacerbated by the current political spectrum and the differing views across parties.136 The possible costs of implementing fiscal decentralization have also been a main concern (Box 2).

196. A number of new legislative initiatives have further complicated the legal framework. In June 2003, the “La Loggia law” (131/2003) authorized the government to define the “fundamental principles” delimiting the regional legislative jurisdiction. However, in the spring of 2004 a new proposal was passed by the Senate for constitutional reform (Senate Act 2544) radically changing once again the structure of intergovernmental relations (among other things, regions would have been assigned exclusive jurisdiction over health, education, and local police). The proposal was amended by the Chamber of Deputies in the fall of 2004, thereby further lengthening the reform process.137

Box 2:The costs of fiscal decentralization

According to the 2001 reformed Constitution, the reform of the intergovernmental fiscal relations would increase local governments’ spending, in principle to be matched by central government’s savings. As additional local spending responsibilities are to be devolved from the center, there would be no additional expenditure at the general government level. Existing estimates put the additional spending for local governments within a range that could be as high as € 91 billion (6.7 percent of 2004 GDP) on a gross basis.

There are, however, significant risks that fiscal decentralization will translate into additional net costs at the general government level. Additional costs could emerge from duplication of functions; overall higher wages, as the wage structure and career progression are generally higher/faster at the regional than the central level; and additional new staff, to the extent that economies of scale are broken up (ISAE, 2004).

Past experience with decentralization also suggests that the transfer of functions may imply higher overall spending. When functions were transferred from the center to the regions during the 1990s, the latter created strong pressures to augment resource envelopes relatively to what they viewed as insufficient spending margins; at the same time, there was a need to placate central agencies that would have lost from the transfer of functions. This resulted in additional funds from the national budget and additional net spending (Giarda, 2004).

197. Absent a clear legal framework, the Constitutional Court has been called to deliberate on the “directions” of fiscal federalism. The Court has so far been very cautious, by providing, in its deliberations, a “centralistic” interpretation of various legislative provisions. For example, it has recently ruled that IRAP—a regional tax on business activities that is seen as the main own regional tax—is a “state” tax, since the state established it (while the regions, to which the proceeds from this tax are transferred in their totality, manage it).

C. The Current Challenges

198. Successful fiscal decentralization requires a sound and stable legislative and institutional setup. The “rules of the game” need to be transparent and unchanging, to ensure buy-in into the reform from all interested parties. In addition, mechanisms are needed to coordinate fiscal policy across government tiers and detect and correct possible noncompliant behaviors. These issues are covered below.

Stability of norms and regulations: the Domestic Stability Pact

199. Since 1999, the Domestic Stability Pact (DSP) has set fiscal constraints on local governments. The DSP was originally intended to engender commitment to fiscal discipline at all levels of government (reduction of both budget deficits and debt levels) within the SGP limits.138 By setting common objectives, the DSP aimed to establish common standards and transparent monitoring mechanisms to asses local governments’ fiscal behavior in a cooperative manner. By virtue of these features (some more notional than real, see below), the DSP has become an integral part of annual budget laws and serves as an instrument to guide and influence the dialogue among levels of government on fiscal matters (Goretti and Mercuri, 2004). However, the DSP has lost some of its effectiveness; for regions, it has been replaced by discretionary negotiating mechanisms for health spending.

200. The definition of the DSP has varied over time. For ordinary regions, while initially the DSP aimed to cap increases in the current primary balance, targets progressively moved to exclude various spending and revenue components; as of 2002, targets have focused only on spending levels (Table 1). Health spending—accounting for about 70 percent of regions’ expenditure, mostly in the forms of transfers to health care units and hospitals—has been excluded since 2000. On the other hand, the DSP for provinces and municipalities has been based on various definitions of budget balances (for details, see Franco, Messina and Zotteri, 2004).

201. More generally, the DSP has suffered from a number of drawbacks that hamper its effectiveness:

  • Lack of consistency with the SGP targets. While the SGP caps the overall budget position of the general government, the DSP targets a subset of the overall budget position (as most spending is excluded, most notably health) of a subset of the general government (as only ordinary regions and local governments are included, see below). Thus, targeting different budget variables precludes mapping local government targets into SGP rules.
  • Frequent amendments to the DSP rules. As DSP targets have been subject to frequent changes, subsequent amendments have redefined budgetary targets and allowed higher deficits than originally mandated (Balassone, Franco, and Zotteri, 2002), thus limiting the effectiveness and credibility of the DSP rules.
  • Varying (and somewhat limited) institutional coverage. In 1999, the original DSP formulation applied the same targets to regions, provinces and municipalities; in 2000, a distinction was introduced between regions and the other entities; and since 2001, smaller municipalities (with less than 5,000 population) have been excluded from the targets. Local health units continue to be excluded.
  • Lack of coordination with debt reduction targets. The DSP included an explicit debt reduction target, as a share of GDP, only in its original 1999 formulation (covering the period 1999-2001). Although in subsequent years the DSP formulation has focused directly on reducing regions’ and local governments’ budget deficits, there was still an implicit presumption that this would allow debt reduction. However, by excluding some spending programs, compliance with the DSP targets does not necessarily cap the increase in overall debt of the general government. In other words, the DSP does not provide a robust anchor to the attainment of the (implicit) debt reduction target (Balassone and Franco, 2001).
  • Sensitivity to cyclical conditions. The DSP does not deal with the possible impact of cyclical conditions on local budgets. Increasing reliance on own taxes—as they progressively replace central transfers—may increase the exposure of local finances to varying macroeconomic conditions.
Table 1.Domestic Stability Pact for Regions, 1999-2004 1/
Definition (for regions)TargetIncentives and sanctions
1999(R-T)-(E-I-Int)Improve trend balance by 1 percent relative to 1998 primary current expenditure (E-I-Int)Sanctions: monetary (proportional to the size of the slippage) if Italy is fined at EU level.
2000(R-T-ExR-FinR-HealthR)-(E-I-Int-ExE-HealthE)

Difference from 1999

Revenue side: (ExR+FinR+HealthR)

Expenditure side: (ExE+HealthE)
As for 1999, but compensating for slippages occurred in 1999.Sanctions: monetary (proportional to the size of the slippage) if Italy is fined at EU level.

Incentives: for complying governments, reductions of costs for getting financial resources from Cassa Depositi and Prestiti (CDP).
2001(R-T-ExR-FinR-HealthR)-(E-I-Int-ExE-HealthE)

Difference from 2000: None
The budget balance cannot worsen by more than 3 percent of the 1999 outcome.Sanctions: monetary (proportional to the size of the slippage) if Italy is fined at EU level.

Incentives: for complying governments, reductions of costs for getting financial resources from CDP.
2002(E-I-Int-HealthE)

Difference from 2001: Spending limit applied
Expenditure cannot exceed 104.5 percent of its 2000 level.Sanctions: monetary (proportional to the size of the slippage) if Italy is fined at EU level.

Regions: In case of noncompliance, no additional resources for the health relative to the August 2001 agreed level. Other local governments: reductions in transfers (later removed).
2003(E-I-Int-HealthE)

Difference from 2002: None
Expenditure cannot exceed 104.5 percent of its 2000 level, adjusted for planned inflation.Sanctions: monetary (proportional to the size of the slippage) if Italy is fined at EU level.

Regions: In case of noncompliance, no additional resources for the health relative to the August 2001 agreed level. Other local governments: Limits on spending for goods and services, constraints on new borrowing, and ban on new hiring.
2004(E-I-Int-HealthE)

Difference from 2003: None
Expenditure cannot exceed 104.5 percent of its 2000 level.As in 2003.
Source: Based on Table 2 in Franco, Messina and Zotteri (2004); and staff elaborations.Note: R = revenue; T = transfers; E = expenditure; I = investment spending; Int = interest payments; ExR = exceptional revenue; FinR = financial revenue;HealthR= health-related revenue; ExE=exceptional spending; HealthE = health-related spending.

Information on local (non-regional) governments is available in Franco, Messina, and Zotteri (2004).

Source: Based on Table 2 in Franco, Messina and Zotteri (2004); and staff elaborations.Note: R = revenue; T = transfers; E = expenditure; I = investment spending; Int = interest payments; ExR = exceptional revenue; FinR = financial revenue;HealthR= health-related revenue; ExE=exceptional spending; HealthE = health-related spending.

Information on local (non-regional) governments is available in Franco, Messina, and Zotteri (2004).

202. The Budget Law for 2005 extends the expenditure-based definition of the DSP (previously applied to regions only) to all levels of government. For regions, the limit will be based on a 4.8 percent increase relative to the level of spending in 2003. Health spending will however continue to be excluded; as in the past, health spending will remain subject to limits negotiated and agreed among the center and the regions. For provinces and municipalities, new limits are now set on the growth of spending (Box 3), defined to exclude wages, financial operations, and transfers to other public administrations.

203. Investment will be included in spending limits. This is not a new element for the regions, as Article 28 of the 2004 Budget Law already foresaw the “inclusion” of their investment spending as of 2005; it is a new element for provinces and municipalities. Investment spending will be allowed to exceed the specified limits if financed by asset sales.

204. Spending limits for all levels of government may be conducive to more fiscal discipline and contribute to protecting budget priorities in a multi-year framework. However, implementation of spending limits at the local level is subject to a number of risks, in particular as spending limits are not based on cost parameter changes.

Box 3:The new definition of Domestic Stability Pact in the 2005 Budget Law

The 2005 Budget Law introduces limits on the nominal increase in spending for all government levels. For provinces and municipalities, this is a major shift from the past focus was on fiscal balances. While most spending at the central government and regional level is subject to 2 percent increase relative to the 2004 outturn and a 4.8 percent increase relative to the 2003 outturn, respectively, for provinces and municipalities spending limits will be computed, for current and capital expenditure, relative to an average for 2001-03. These entities will be divided in 14 demographic “classes” depending on their population size. For those local governments that over the period 2001-2003 had an average per capita spending level lower that the average of their population class, the allowed increase in spending in 2005 is set at 11.5 percent; for the others, the increase is 10 percent.

The 2005 Budget Law also modifies limits on debt contracting. Rather than 25 percent of own revenue, the cap on borrowing is set at 12 percent of own revenue, with a medium-term adjustment path foreseen for those entities that already exceed this limit.

Monitoring

205. A system for monitoring local government finances is in place, but in some cases is not timely. Monitoring mechanisms are established within the DSP—basically unchanged over the last two years. Regions are required to submit quarterly reports to the General Accounting Office (Ragioneria Generale dello Stato), containing information on cash and commitment basis. Transmission is based on a specific reporting format and must be made within 30 days from the end of the relevant quarter. Monitoring of compliance with the DSP is undertaken by the audit court (Corte dei Conti) on data transmitted from Ragioneria. However, for some regions there have been significant and repeated delays in submitting requested information (Grilli, 2004). Finally, monitoring of health spending, which continues to be subject to separate negotiations between the central government and regions, is undertaken in the context of a computerized system (tavolo di monitoraggio), accessible by all participating entities and maintained by the Ministry of Economy and Finance (MEF).

206. While harmonized measurement of fiscal outcomes is key to effective monitoring, budgeting and accounting practices continue to vary. In 2000, decree No. 76 introduced principles of budgeting and accounting for regions, in line with those of the central government. However, as a result of the autonomy recognized by various norms, accounting differences persist, among regions and relative to the central government.139

207. Local governments’ borrowing and debt are comprehensively recorded by the Bank of Italy. The 2004 Budget Law allowed the MEF to access information from the Bank of Italy on all local governments’ operations with financial institutions; a dedicated information system has been devised to ensure the acquisition of the relevant data. As of July 2004, the MEF database covered about 80 percent of the data collected by the Bank of Italy.140 In the same vein, the 2004 Budget Law also defined what operations are included in the definition of debt; and what kind of spending programs are to be classified as investment, consistent with the SEC95. This provision will go a long way in ensuring consistency of accounting and reporting systems (Grilli, 2004).

208. Harmonization of budgeting and accounting practices varies across countries, generally as a function of their degree of fiscal decentralization (Table 2). At one end of the spectrum, countries with a strong federalist tradition (Canada, New Zealand, and United States) allow full independence of budgeting and accounting practices and no common financial reporting; this also reflects a history of no interference of the federal government with the affairs of the states and subnational governments and a lack of strong financial support from the former to the latter. At the other end of the spectrum, countries which are now undertaking their first steps toward fiscal decentralization and where traditionally subnational governments have relied heavily on the center’s financial support, such as Italy and Spain, tend to determine centrally both budgeting and accounting rules and reporting standards. A large group of countries (Austria, Denmark, Netherlands, and Norway) lies in between, with no common budget and accounting standards among central and local governments—although the latter have to follow rules set by the former—while reporting is standardized according to rules set by the central government.

Table 2.Harmonization of budgeting, accounting, and reporting across selected OECD countries
Common budget classification and accounting rules, set by the central governmentNo common budget classification and accounting rules; rules set by the central governmentNo common budget classification and accounting rules; rules set by each government level
Common financial reporting, set by the central governmentIreland

Italy

Spain
Austria

Denmark

Netherlands

Norway
Common financial reporting, set by lower levels of governmentBelgiumUnited States

Sweden
No common financial reporting among lower levels of governmentGermanyFranceCanada

New Zealand

209. Transparent accounting rules and efficient channels of reporting are essential to timely and reliable mechanisms to monitor finances of local governments. Crosschecking information “above the line” (reported by the local governments) with “below-the-line” information (provided by the banking system) enhances the integrity of the data.

210. In this regard, the establishment of a centralized and standardized reporting system for cash operations is promising. This new system (Sistema Informativo delle Operazioni degli Enti Pubblici, SIOPE) will be piloted in 2005 in a limited number of entities, with plans to extend it gradually to the rest of the country. This would allow more timely, accurate, and harmonized reporting and recording of local governments’ treasury operations. The database, realized with cooperation of the interbank network, will be kept at the Bank of Italy and be MEF property. In addition, treasury officers would not be able to execute operations whenever the relevant coding is missing—thus adding an additional layer of control at the spending execution stage.

Credibility of enforcement, accountability and hard budget constraints

211. In its original formulation, the DSP included incentives (Table 1). This is because it was meant to serve as tool to reach shared goals and enhance cooperation across government levels, rather than act as a binding constraint imposed by the central government.

212. Progressively, though, the focus shifted on sanctions to foster compliance. While initially sanctions were monetary, based on cuts of transfers to noncompliant entities, it was soon realized that this would further exacerbate the latter’s financial difficulties at a time of distress. It is generally recognized that financial sanctions present a time-consistency problem, as fining a local government already experiencing difficulties could be politically difficult or even unconstitutional (Joumard and Kongsrud, 2003).

213. Sanctions have moved away from monetary to administrative measures—in line with the practice followed in a number of countries (Table 3).141 While administrative sanctions may be more effective than financial ones as they imply a loss of reputation and administrative autonomy, they are nonetheless difficult to design due to information asymmetries. Differences in local government sizes and historical and cultural aspects also complicate the design of sanctions that may fit all subnational governments (Joumard and Kongsrud, 2003). In some cases, penalties and sanctions may apply to public officials deemed responsible for non-compliance; in Brazil, sanctions include dismissals, fines, and even jail terms, as established under the Fiscal Crimes Law.

Table 3.Types of sanctions and enforcement mechanisms in selected countries
CountryType of sanctionsEnforcement mechanism
AustriaFinancial: Noncompliant local governments have to pay a fine proportional to the shortfall, up to a ceiling. If compliance is obtained within one year, the fine is returned; otherwise, the funds are allocated across compliant governments.Cooperative: Application of sanctions depends on the unanimous decision of a commission involving the federal and local governments.
BelgiumAdministrative: Limits on subnational borrowingCooperative: The federal government is allowed to limit regional borrowing, following a recommendation of the Supervisory Council and in consultation with regional governments.
CanadaAdministrative: In four provinces, ministries and members of the executive council are subject to significant cuts in wages for failure to achieve fiscal targets.No formal coordination. A non-binding budget coordination exists via a dialogue among ministers
GermanyNo formal sanctions.Cooperative: The Financial Planning Council (formed by the federal government, the states and representatives of the communities) is charged with monitoring fiscal developments at all government levels and making recommendations in cases of noncompliance.
IrelandAdministrative: Defaulting authorities can be removed from office and replaced by a commissioner appointed by the central government.Centralized: Subnational governments are monitored and controlled by the Department of the Environment and Local Government.
ItalyAdministrative: Limits on the purchase of goods and services; prohibition to hire new staff and to contract debt to finance investmentCooperative: The State-Local Government Conferences are involved in the monitoring process.
SpainAdministrative: Noncompliant authorities have to submit a plan for correcting any fiscal deficit.Centralized.

214. The 2005 Budget law reiterates and strengthens the system of penalties in force over the last two years. For provinces and municipalities, the law foresees limits on the purchase of goods and services (not exceeding the level of the last year when the DSP was verified to be respected; in case of continuous noncompliance, the limit would be applied to the level of spending of two years before, reduced by 10 percent); prohibition to hire new staff under any contractual terms; and ban on contracting debt to finance investment. Other penalties apply to regions for missed compliance with health spending ceilings and translate into a reduction of health financing. As an additional control mechanism, loan applications by regions, provinces, and municipalities need to be supported by proof of compliance with the DSP in the preceding year. Absent this certification, to be released by the internal audit office of the entity in question, banks and financial institutions are not allowed to grant financing.

215. Compliance with the DSP appears to have been strong, but this partly reflects the DSP limited coverage of regional fiscal operations. In 2003, 14 ordinary regions (out of 15) have complied with the DSP. However, until 2004, the share of spending monitored under the DSP was very limited, representing only 13 percent of total regional expenditure. In addition, targets are set relative to projected trends—to the extent these are set in a relatively loose way, their respect is easier to achieve.142

216.There is also a credibility problem. Bordignon and Turati (2004) stress that credibility issues are likely to emerge in sensitive areas such as health care, as it is not politically conceivable that local governments would be allowed to fail in providing essential health care services. This perception is more acute when the central government can be blamed for this failure—as may happen in countries where political and financial responsibilities across different government levels are not well defined. More generally, if the central government is involved in local governments’ affairs (for example when it is in charge of guaranteeing minimum level of services, as mandated by the Italian Constitution), then a policy of no bail-out may not be credible.

217. A cooperative approach to formulation and implementation of fiscal policies may enhance credibility and compliance. This may strengthen political commitments, and hence fiscal discipline, through a consensus-building process (Joumard and Kongsrud, 2003). All countries, especially those with a longer tradition of fiscal federalism, rely on cooperative mechanisms; this is generally achieved either by parliamentary chambers representing lower levels of government (such as in Germany and Austria) or advisory councils with representatives from all tiers (Table 3). This mechanism could be supplemented by a peer review system—whose feasibility and effectiveness would clearly be sensitive to the levels of government involved—and by the introduction of credible sanctions (Balassone, Franco, and Zotteri, 2002).

218. Eliminating the perception of soft budget constraints is difficult but not infeasible. Past experience in Italy offers an example of how this may be achieved. Substantial retrenchments in health spending were achieved in the mid-1990s—by about 1.2 percent of GDP between 1991 and 1995, when health spending reached a minimum of 5.3 percent of GDP. By that year, the regional health deficits had been eliminated. This is because the central government had managed to convince regions that it would not intervene with ex-post financing of health deficits (Bordignon and Turati, 2004). A number of developments made this course of action credible, including the need to address forcefully the causes of the 1992 financial crisis and external constraints stemming from the quest to join the final stage of EMU. However, since Italy entered into the EMU, the need to stick to non bail-out policy loosened, and with it perceptions of central government’s severity in pursuing offenders.

219. Credible enforcement is, however, impossible if the policies to be enforced are not tenable. Thus, the reduction in health spending experienced in the mid-1990s soon proved to be unsustainable—health care spending swelled back to 6.2 percent of GDP by 2001, and has hovered around that level since. This indicates that, without properly designed reforms aimed at increasing the efficiency of spending and strengthening the quality of administration and financial management at the regional level, sustainable savings are not feasible. In other words, tighter budgetary allocations may not be viable—and their enforcement would not be credible—unless regional governments are put in a position to pursue such savings in a sustainable way. Although outside the scope of this paper, this confirms that the design of expenditure mandates—and their underlying key policy parameters—will be key to ensuring fiscal discipline when fiscal decentralization is implemented.

D. Concluding Remarks and Policy Implications

220. The authorities’ cautious implementation of fiscal federalism is understandable, given the attendant risks. However, uncertainties and contradictory steps along the way to the reform of intergovernmental fiscal relations may come at a cost—not least for the credibility of government policies and their consistency with the legal framework establishing the guiding principles of these reforms.

221. The design of the intergovernmental fiscal relations system ultimately remains a political choice. But a few elements need to be in place to ensure its successful implementation:

  • Mechanisms for coordinating budgetary policies and management across levels of government should be strengthened; and the existing institutions and cooperative mechanisms, such as the Conferenza Stato Regioni and the joint state-region monitoring of health spending, could be enhanced by allowing them greater voice in defining the government’s policy intents. This would ensure an appropriate contribution of each level of government to macroeconomic objectives and fiscal sustainability consistent with the SGP objectives.
  • As a first step, it is essential to ensure appropriate recording and reporting of all subnational government operations. The current piloting of a new standardized recording system, SIOPE, is promising. Its sound functioning and prompt extension to all government entities should be encouraged. This is all the more important in cases where a “no bailout” policy may not be credible—the evidence reported in this paper would suggest that Italy is such a case. A sound and timely monitoring system is therefore essential to intervene at earlier stages of financial difficulties—rather than waiting until problems that require central government intervention materialize.
  • Transparency and stability of rules and regulations are also needed. The many changes to the DSP since its inception witness the difficulty of adapting it to changing circumstances. As the 2005 Budget Law applies the same typology of targets to both
  • central and subnational governments, this should be supported by a multi-year budgetary framework to formulate and protect budget priorities.
  • Rules and regulations should be backed by clear and effective sanctions for non-compliance; these should be non-discretionary and based on automatic mechanisms to limit political interference. Administrative measures on managers and politicians could also be applied to strengthen their accountability and respect of the “rules of the game.”
References

    BalassoneF., D.Franco and S.Zotteri, (2004), “Fiscal Rules for Subnational Governments: Lessons from EMU Countries”, in G.Kopits (ed.) Rules-Based Fiscal Policy in Emerging Markets: Background, Analysis and Prospects, Macmillan.

    • Search Google Scholar
    • Export Citation

    BalassoneF. and D.Franco (2001), “Fiscal Federalism and the Stability and Growth Pact: A Difficult Union,” Proceedings from the Bank of Italy Workshop on Fiscal Rules, Perugia, February2001, pp.601–35, available at http://www.bancaditalia.it/ricerca/statist/conv_svo;internal&action=rightFrame.action.

    • Search Google Scholar
    • Export Citation

    BordignonM., P.Manasse and G.Tabellini, (1996) “Come Fare la Perequazione Regionale in Italia?” Working Papers 107, IGIER (Innocenzo Gasparini Institute for Economic Research), Bocconi University, available at http://ideas.repec.org/p/igi/igierp/107.html

    • Search Google Scholar
    • Export Citation

    Bordignon, M. e F.Cerniglia, (2003), “I Nuovi Rapporti Finanziari fra Stato ed Autonomie Locali alla Luce della Riforma del Titolo V della Costituzione” SIEP Working Paper No. 207.

    • Search Google Scholar
    • Export Citation

    BordignonM., TuratiG., (2003), “Bailing out Expectations and Health Expenditure in Italy” CESIfo, Working Paper 1026.

    BordignonM., MapelliV. e G.Turati (2002), “Fiscal Federalism and national health service in the Italian system of governments”, in ISAE, Annual Report on Monitoring Italy, Rome.

    • Search Google Scholar
    • Export Citation

    DabanT., E.Detragiache, G.di Bella, G. M.Milesi-Ferretti, and S.Symansky (2003), “Rules-Based Fiscal Policy in France, Germany, Italy, and Spain,” IMF Occasional Paper 225.

    • Search Google Scholar
    • Export Citation

    FrancoD., G.Messina and S.Zotteri, (2004), “Fiscal Decentralisation in Italy: an Unfinished Agenda”, Wirtschaftspolitische Blätter, No. 4, 467–481, Vienna.

    • Search Google Scholar
    • Export Citation

    Giarda.P.D. (2003), “Is there a defined system of fiscal federalism in the new Italian Constitution?”, Paper presented at Convegno Federalismo fiscale: Analisi e Prospettive, Milan, Econpubblica, July 4.

    • Search Google Scholar
    • Export Citation

    Goretti, C. and M.C.Mercuri (2004), “Vincoli Finanziari, Decisioni di Bilancio, e Azione Amministrativa”Senate and ISAE, Rome,

    Grilli, V. (2004) “Audizione: Indagine Conoscitiva sugli Effetti e le Tecniche di Controllo dei Flussi di Finanza Pubblica in Ordine all’Andamento del Debito, con Particolare Riferimento alla Componente Non Statale” Senate, Economic Programming and Budget Commission, Rome, March11, 2004.

    • Search Google Scholar
    • Export Citation

    ISAE (2004), “Rapporto annuale sull’attuazione del federalismo,”February 2004, available at http://www.isae.it/secondo_rapp_annuale_federalismo_febbraio_2004.pdf.

    • Search Google Scholar
    • Export Citation

    ISAE (2003), “Rapporto annuale sull’attuazione del federalismo,”February2003, available at http://www.isae.it/primo_rapp_annuale_federalismo_febbraio_2003.pdf

    • Search Google Scholar
    • Export Citation

    JoumardI. and P.M.Kongsrud (2003) “Fiscal Relations Across Government Levels” Economic Department Working Paper No. 375, December2003, OECD Paris.

    • Search Google Scholar
    • Export Citation

    MessinaG. (2001), “Decentramento fiscale e perequazione regionale. Efficienza e redistribuzione nel nuovo sistema di finanziamento delle regioni a statuto ordinario”, Banca d’Italia, Temi di discussione, No. 416, August.

    • Search Google Scholar
    • Export Citation

    MoroF. (2002), “Prime valutazioni degli effetti finanziari di alcune interpretazioni dell’art. 119 della Costituzione”, in SVIMEZ, Sul finanziamento delle funzioni pubbliche di regioni ed enti locali secondo le nuove norme costituzionali, Quaderno n. 15 di “Informazioni SVIMEZ”, Rome, May.

    • Search Google Scholar
    • Export Citation

    OECD and World Bank (2002) Survey on Budget Practices, available at http://ocde.dyndns.org/

    Reforme (2004), “Federalismo 2004: Aspetti Quantitativi e Confronto con le Esperienze Europee”, available athttp://www.ref-online.it/altre_pubblicazioni.asp#8at

    • Search Google Scholar
    • Export Citation

    RoddenJ.A., EskelandG.S., LitvackJ., (2002), Fiscal Decentralisation and the Challenge of Hard Budget Constraints, MIT Press, Cambridge, Massachusetts.

    • Search Google Scholar
    • Export Citation

    SVIMEZ (2001), “Riforme federaliste e politiche per lo sviluppo delle aree economicamente svantaggiate”, Quaderno n. 19 di Informazioni SVIMEZ, Rome.

    • Search Google Scholar
    • Export Citation

    Ter-MinassianT., ed. (1997) Fiscal Federalism in Theory and Practice, Washington, International Monetary Fund.

    VitalettiG. (2004) “Il Federalismo Fiscale, ovvero: Dalle Imposte-Manna alle Imposte-Talenti?” presented at the conference “Dal Federalismo Legislativo al Federalismo Fiscale”, Turin, July2004, forthcoming in“Verso un nuovo federalismo fiscale.”

    • Search Google Scholar
    • Export Citation
130

Prepared by Annalisa Fedelino (FAD)

131

Local governments are defined to include regions, provinces, municipalities and metropolitan cities. In this paper, we focus on regional governments for sake of simplicity.

132

See, among others, Bordignon, Manasse and Tabellini (1996), Balassone, Franco and Zotteri (2002), ISAE (2003 and 2004), Bordignon and Cerniglia (2004), Giarda (2004), Reforme (2004), Franco, Messina and Zotteri (2004), and Vitaletti (2004).

133

Italy comprises 15 ordinary statute regions and five special statute regions; the latter were activated in 1948 and enjoy special autonomy and powers. In this paper, “regions” refers to ordinary statute regions only.

134

Bordignon and Cerniglia (2004) present a detailed comparison between the old and reformed text of the Constitution and a commentary. See also the link under “federalismo fiscale” in www.lavoce.info, which contains a number of articles in Italian on these issues.

135

This also applies to municipalities, provinces, and metropolitan cities.

136

The reform of the Constitution was confirmed by a referendum where only about 33 percent of the eligible voters participated, and 60 percent of them voted in favor (thus making it approximately a thin 20 percent of the electorate). Since mid-2001, the central government majority has changed (from center-left to center-right); but about 75 percent of local administrations are currently run by the center-left.

137

Constitutional amendments require a rather complex and lengthy legislative procedure: each chamber of parliament (Chamber of Deputies and Senate) needs to read and approve twice the proposed amendments without modification. A national vote (“referendum”) may also be needed

138

Law 281of 1970 limits local governments’ borrowing to 25 percent of own revenue.

139

Beginning in 1995, municipalities, provinces, and health units were allowed to adopt accrual accounting practices similar to those followed by the private sector. Health units have done so, but practices across municipalities and provinces continue to differ.

140

According to information provided by the Budget Commission, Chamber of Deputies.

141

Countries with strong federalist systems, such as Australia, Canada, New Zealand, and the United States, mostly rely on market discipline to foster compliance from lower levels of government.

142

As reflected in mimeo material provided by Budget Commission, Chamber of Deputies.

Other Resources Citing This Publication