Journal Issue
Share
Article

Reforming Tax Expenditures in Italy: What, Why, and How?1

Author(s):
Justin Tyson
Published Date:
January 2014
Share
  • ShareShare
Show Summary Details

I. What Are Tax Expenditures?

Tax expenditures are government revenues foregone as a result of differential, or preferential, treatment of specific sectors, activities, regions, or agents. They can take many forms, including allowances (deductions from the base), exemptions (exclusions from the base), rate relief (lower rates), credits (reductions in liability) and tax deferrals (postponing payments). International comparisons are complicated by different methodologies and assessments as to what constitute a tax expenditure, but the practice is pervasive, and tax expenditures in Italy are clearly elevated.

Figure 1.Tax Expenditures in Selected Advanced Economies

(Percent of GDP)

Sources: OECD (2010); United States, National Commission on Fiscal Responsibility and Reform (2010).

Note: Higher values may arise from more comprehensve reporting (recent estimates for Italy are higher due to change in benchmark).

II. Why Should Tax Expenditures Be Reformed?

Tax expenditures can have major consequences for the fairness, complexity, efficiency, and effectiveness of not only the tax system itself but, since they often serve purposes that might be (or are also) pursued through public spending, of the wider fiscal system.

  • Tax expenditures can compromise fairness. Tax expenditures can be a poor way of pursuing equity objectives: in a progressive tax system, for instance, any policy that reduces taxable income will benefit most those in the highest marginal tax bracket and convey no benefit to those out of the tax system, a potential reason for using tax credits (or spending measures) instead. The tax expenditures associated with the reduced VAT rates in Italy, for instance, in themselves increase progressivity—but much of the benefit will go to the better off, so that the same equity objectives could likely be pursued at less cost through social spending.

  • Tax expenditures can be inefficient and poorly targeted. Tax expenditures can create unintended or unwelcome distortions: the current deduction of mortgage interest, for instance, may have been appropriate when imputed income from owner-occupation was effectively taxable, but now may encourage leveraged housing finance.

  • Tax expenditures are vulnerable to lobbying. Special interest groups may find it easier to argue for tax breaks than for explicit spending support. Tax expenditures often bypass the scrutiny accorded to spending in the regular budget and may not require annual renewal in the budgetary process—this lack of transparency may explain some of the appeal they hold for policy makers.

III. Are All Tax Expenditures Bad?

Not all tax expenditures are necessarily bad. For example, tax expenditures could benefit from administrative economies of scale—as they usually deliver their “rewards” through a reduction in taxes that would have been paid anyway, spending ministries do not need to allocate resources to administering substitute programs in either cash or kind.

Not all tax expenditures are unjustifiable tax loopholes (Buckley, 2011). In fact, tax expenditures can also be a way of achieving fair and efficient taxation: for example, the largest single tax expenditure in Italy is a tax credit that serves the same purpose as basic tax-free thresholds serve elsewhere. The table below provides a comparison of tax expenditures and direct spending, summarizing the main costs and benefits.

Table 1.Comparison of Tax Expenditures and Direct Spending
Tax ExpendituresDirect Spending
Accessibility for beneficiariesSimple, because of their automatic nature.More complex, requiring selection.
Administrative and compliance costsHigh, if exemptions are properly monitored.Medium, due to necessity of a selection and allocation system.
Possible abusesEvasion, avoidance, and rent seeking.Arbitrariness, inefficiency, and capture of the allocating body.
FlexibilityWork with permanent laws, thereby generating stability but also inertia.Work with budgets, evaluations, and regular reallocations. 1/
Transparency and accountabilityTheir automatic nature does not contemplate control mechanisms or accountability.Must be approved by the legislature, as with all government expenditure.
Expenditure controlExpenditure determined ex-post; uncertain and unlimited, which can cause fiscal imbalances.Programmed and controlled spending, limited by budget law.
EquityOnly potential taxpayers benefit, and those with the highest income often benefit the most.Discretion can provide more equitable access, enhancing targeting of beneficiaries.
Source: Villela, Lemgruber, and Jorratt (2010) in April 2011 Fiscal Monitor.

This has not been the case in Italy, where the budget contains many legal rigidities.

Source: Villela, Lemgruber, and Jorratt (2010) in April 2011 Fiscal Monitor.

This has not been the case in Italy, where the budget contains many legal rigidities.

While generalizations are difficult, the literature points to some guidelines for incorporating tax expenditures within an optimal tax framework. For income taxes, broadly defined bases allow for lower marginal rates, while generously defined exemptions can help with equity goals and also reduce compliance and administrative costs. Standard allowances can reduce equity but are simpler to enforce (see Alm, 1996, for a fuller discussion). Consumption taxes should be largely proportional—any divergences should focus on higher rates for goods and services that are unresponsive to price changes or that generate negative spillovers (Alm, 1996). Welfare objectives are likely best pursued through well targeted subsidies (Johansson et al., 2008).

IV. How Can Tax Expenditures Be Identified and Quantified?

Identifying and quantifying tax expenditures are critical for an understanding, and informed public discussion, of the overall impact of the tax system. The first step is the definition of a benchmark tax system. It is this norm that allows identification of provisions in the tax system that are part of this tax norm and those that are not. However, there is no consensus on how to define a tax norm.

The main distinction is between approaches that use a norm based on theoretical concepts of income, consumption, or value-added taxes and those that use a country’s own tax laws as a basis to define the benchmark, isolating differential or preferential treatment judged as tax expenditures (e.g., targeted provisions to address specific policy objectives). The former will classify as tax expenditures elements which might otherwise be considered part of tax design.2

Once tax expenditures have been identified, the second step is to quantify the budgetary cost of the individual tax expenditure “policies.” Most countries estimate the loss in government revenue due to the tax expenditure, based on the actual uptake of the differential treatment. Substantial work has already been undertaken on the identification and quantification of tax expenditures in Italy. A report, commissioned by the Ministry of Economy and Finance (MEF), identifies and costs 720 tax expenditures amounting to about 16 percent of GDP.3 The MEF, by choosing the conceptual approach to identifying the benchmark, provides very extensive coverage by international standards; however, it would be neither feasible nor desirable to eliminate all the tax expenditures identified.4

V. Once Identified, How Should Tax Expenditures Be Evaluated?

Clear and frequent evaluation of the identified tax expenditure is necessary for scaling back those not generating benefits commensurate to their cost. This is not always straightforward, in that the impact of tax expenditures on behavior is often hard to identify; but even in such cases, some sense can be given of whether it is plausible that the response is large enough to warrant the costs incurred. As a start, tax expenditures should be evaluated within the framework used to analyze the tax system as a whole:

  • Does the tax expenditure promote or hinder economic efficiency? To what extent does it influence taxpayer behavior in desirable ways or distort behavior in undesirable ways?

  • Is the tax expenditure fair? Are similar individuals treated similarly? Does the tax expenditure account for individuals’ different capacities to bear the burden of taxation?

  • Is the tax expenditure simple and easily administered? This includes calculating tax liability, filing taxes, government administration, enforcement, and ease of evasion.

  • Is the tax expenditure vulnerable to lobbying? Does the tax expenditure benefit very specific interests groups, and how frequently are costs and benefits scrutinized?

It is also important to compare tax expenditures with alternative spending measures and, more generally, to assess them in the light of what can be achieved on the spending side. For example, some of the tax expenditures with social objectives, such as the dependent relative tax credit, potentially overlap with the objectives of social assistance programs run through the social security system—the relative costs and merits of both delivery options need to be considered.

VI. What Types of Tax Expenditures Are There in Italy?

A. Personal Income Tax (PIT)

Establishing the relevant benchmark for the personal income tax is not easy, as it is a prime tax instrument for redistribution. While the literature offers some guidance on what is optimal (i.e., a natural benchmark), it requires a subjective assessment of the equity-efficiency trade-off. Some targeted allowances can be desirable, but in practice many are often not effective. In Italy, tax expenditures were measured both against a measure of comprehensive economic income (consumption plus change in net worth) and also against a dual income tax (DIT) system, which taxes labor income at progressive rates but capital income at a low single rate.5

Many of the largest identified PIT tax expenditures are arguably desirable parts of a normal tax system. Table 2 summarizes the largest PIT tax expenditures, collectively worth over 5 percent of GDP.

Table 2.Summary of Largest PIT Tax Expenditures
DescriptionValue (billions)Percent GDPComment
Tax credit for wage income from employment, pensions, self-employment and similar income.37.732.41This regime is a substititue for the fact that Italy has no lower personal income tax threshold.
Various financial subsitute taxes (lower rates) on interests, dividends, capital gains and other forms of return.13.170.84Substitute tax regime by-passes the requirement to tax capital income according to the progressive PIT schedule by mimicking a Dual Income tax regime.
Tax credit for dependent relatives.10.500.67This benefit is sometimes considered a measure of ability to pay and part of the benchmark system rather than a tax expenditture.
Tax exemption (excluded from base) for contributions to welfare and pension schemes for employees.10.100.64Pension contributions are excluded and pension income is taxed. Avoids double taxation.
Lower PIT rates for payment of separation allowances and “golden handshakes”5.100.33Unwinds tax progressivity that would come from what is effectively receipt of multi-year income in one period.
Tax exemption (excluded from base) for compulsory contributions to welfare and pension schemes for self-employed.4.310.28Pension contributions are excluded and pension income is taxed. Avoids double taxation.
Tax credit for medical expenses and health assistance services.2.360.15Considered to have welfare objective.
Sum of largest PIT tax expenditures.83.265.32
Sources: MEF, 2011; and IMF staff estimates.
Sources: MEF, 2011; and IMF staff estimates.

The largest single item is a universal tax credit that compensates for the lack of an income tax threshold, which Italy does not have but many other countries do. Consequently, it can be seen as a key component of a progressive income tax system. Similarly, the 2nd largest tax expenditure in this category allows capital income to be taxed at a low and (almost) uniform rate. As such, it is an intrinsic part of the DIT system. The third largest item is a tax credit for dependent relatives, which in many countries is considered a measure of ability to pay and hence contributes to the fairness of the tax system. Arguably, of the largest PIT tax expenditures, the tax credit for medical expenses and health services is one that most likely could be replaced by a better targeted expenditure program.

B. Corporate Income Tax (CIT)

To measure CIT tax expenditures, a concept of net business profit was applied. The largest CIT tax expenditure is a measure to avoid double taxation of foreign sourced dividends. Other measures, although less a feature of a normal CIT system, could have their merits in the current environment, such as the deductibility of social security costs from the regional business value added tax (IRAP). This measure was introduced to reduce the labor tax wedge and incentivize hiring by firms. Other CIT tax expenditures, such as those to incentivize restructuring and mergers, should be weighed against other policy options for achieving this objective, such as a streamlining of court processes.

An important element of corporate/business taxation in Italy is the Allowance for Corporate Equity (ACE). It is an important example of a provision that looks like a tax expenditure but can also be considered a key element of tax design. By providing a tax deduction for a notional return on additional equity injected into companies, this system reduces the cost of such finance and eases the tax incentive to use debt rather than equity finance. It constitutes an important step toward greater neutrality for business investment and financing decisions, taking Italy closer to a form of “dual income tax”.

Table 3.Summary of Largest CIT Tax Expenditures
DescriptionValue (billions)Percent GDPComment
Foreign-source dividends received by a resident corporate taxpayer are 95% exempt from CIT.8.380.54Measure to avoid double taxation.
Subsitute tax (16 percent) for capital gains from goodwill, trademarks and other intangible assets resulting from extraordinary operations, such as restructuring and mergers.7.430.47Tax provision to promote firm dynamism and incentivize new activity.
Full deduction from IRAP tax base of SSC costs related to permanent workers; full deduction from PIT and CIT tax base of IRAP on labour costs (plus partial deductability of interests costs). 1/6.690.43Provision mainly to reduce labor tax wedge.
Substitute tax for capital gains arising from “extraordinary” operations, suchs as mergers, divisions, and transfers of companies.6.400.41Measure to favor restructuring, which brings higher depreciation charges (lower taxes) from corporates in the future.
Subsitute tax on capital gains (lower rate) from revaluation of assets held on the balance sheet at historical cost.4.180.27Generates current revenue for the authorities in exchange for higher depreciation charges (lower taxes) from corporates in the future.
Sum of largest CIT tax expenditures.33.082.11

Full deduction of IRAP labor costs from PIT and CIT was introduced in Law 214/2011 and is not explicitly costed in the table, which instead includes estimates based on earlier lump-sum deductions per employee.

Sources: MEF, 2011; and IMF staff estimates.

Full deduction of IRAP labor costs from PIT and CIT was introduced in Law 214/2011 and is not explicitly costed in the table, which instead includes estimates based on earlier lump-sum deductions per employee.

Sources: MEF, 2011; and IMF staff estimates.

C. Value Added Tax (VAT)

A uniform rate with a broad base is a good benchmark as the VAT is not a good instrument either to address externalities, or to deal with redistribution. The information requirements needed to operate a differential VAT regime that improves (rather than worsens) economic welfare is so onerous as to make it impractical. Moreover, the welfare objectives are most probably better achieved through special excise taxes or direct subsidies rather than by a multi-rate VAT system (Johansson et al., 2008). Consequently, the top-down policy gap outlined below is a useful concept, as it measures a deviation from a benchmark that is deemed optimal.

For the VAT, the shortfall of C-efficiency6 from 100 percent can in principle be decomposed into terms relating to both the compliance gap and a “policy gap,” reflecting the extent to which consumption is not actually taxed at the standard rate. For Italy, C-efficiency is estimated at around 41 percent in 2010—revenue was 41 percent of what it would have been had the then-standard rate been applied to actual taxable consumption. Combining this with a compliance gap of around 30 percent, as the studies of the Revenue Agency suggest, implies a policy gap also of around 41 percent. These calculations are illustrative, in that they derive from distinct data sources that are not fully comparable. Nonetheless, they give some sense of the relative potential of design and compliance improvements: halving the compliance gap, maintaining all tax rates unchanged, would thus raise about 1.3 percent of GDP; halving the policy gap, keeping the standard rate unchanged, would raise about 2.7 percent of GDP. (The policy gap can in turn be decomposed into elements reflecting rate differentiation and the operation of exemptions: Box 1 illustrates for Italy).7

Box 1.Decomposing the IVA Policy Gap

De Mooij and Keen (2012) show that the policy gap can be further decomposed down as:

(1-policy gap) = (1-exemptions)×(1-rate dispersion)

where the first term captures the impact of exemptions (sectors, activities), and the second measures the effect of non-standard VAT rates on collections (usually lower rates). The recent report of the MEF on tax expenditures in Italy implies rate dispersion at about 0.25 percent; combined with a policy gap of 41 percent, this implies an “exemptions” gap of 0.22 percent in 2010. In percent of GDP (2010), this means lost revenue of 2.5 due to exemptions (including those mandatory under EU rules, which of course Italy cannot unilaterally remove), 2.9 due to lower rates, and 2.6 due to non-compliance. This leads to actual revenue collection of just over 6 percent of GDP—as opposed to potential revenue, at an unchanged standard rate, of 15 percent of GDP.

Decomposing Potential VAT

(percent GDP)

The largest VAT tax expenditures in the MEF report are the reduced rates (2.5 percent of GDP), consistent with a top-down decomposition of the VAT policy gap.8 However, the estimated cost of exempt regimes from the decomposition of the policy gap (2.5 percent of GDP) is much larger than that found by the MEF (0.06 percent of GDP)—this is likely due to the unquantified miscellaneous exempt regimes, both national and EC-mandated. Both groups of exempt regimes should be scrutinized. The MEF report also quantifies the cost of VAT reduced rates and some of the exempt regimes; these are summarized in Table 4.

Table 4.Summary of Largest VAT Tax Expenditures
DescriptionValue (billions)Percent GDPComment
VAT reduced rate (10 percent).24.601.57Including food items, restaurants and hotels, pharmaceutical and medical products, energy products and recreational services.
VAT reduced rate (4 percent).14.600.93Including food items, books and newspapers, and school and company canteens.
VAT Exemptions
Italy Specific
Special regime/s for agricultural producers0.310.02
Special regime for publishing sector0.240.02
Special regime for tobacco and match producers
Supplies of goods to (solely) charitable, educational or research bodies0.170.01
Exemption for funeral services0.120.01
Miscellaneous other regimes
EC regulationsIncluding financial sector, travel agencies, used cars, secondhand goods, antiques, sales by auction etc.
Sum of largest VAT tax expenditures.40.042.56
Sources: MEF, 2011; and IMF staff estimates.
Sources: MEF, 2011; and IMF staff estimates.

D. Excise Tax

The benchmark used in the MEF study is the legal tax system, although in principle, as excise taxes are designed to address externalities, one could assess the extent to which taxes fall short of the Pigouvian rate. Most excise tax expenditures are designed to provide some degree of competitive advantage to sectors that could be better achieved through other reforms—for example, structural reforms to reduce the cost of energy (see Table 5).

Table 5.Summary of Largest Excise Tax Expenditures
DescriptionValue (billions)Percent GDPComment
Reduced excise tax for public transport companies.1.670.11Sectoral benefits.
Reduced excise tax for agricultural sector.0.960.06Sectoral benefits.
Tax exemption for houses with monthly energy consumption0.550.04
between 3 and 120 KWH.
Sum of largest Excise tax expenditures.3.170.20
Sources: MEF, 2011; and IMF staff estimates.
Sources: MEF, 2011; and IMF staff estimates.

E. Other taxes

Some very large tax expenditures have already been removed with recent reforms. For example, the proposed revaluation of property cadastral values will eliminate previously identified tax expenditures related to property and transfers taxes (Eyraud, 2013).

VII. Which Tax Expenditures Should Be Reformed and How?

Italy has a strong foundation to build on in terms of reforming tax expenditures. As mentioned, much work has already been undertaken on the identification and quantification of tax expenditures by the MEF. In addition, the draft tax reform law (Delega Fiscale) provides for the annual publication of a list of tax expenditures according to criteria and a methodology that will be supported by an external review body. This goes beyond the standards set out in the IMF Code of Good Practices on Fiscal Transparency and the OECD Best Practices for Budget Transparency.9

Several other steps should be considered.

  • Ideally, all tax expenditures should be regularly and systematically reviewed, just like regular government expenditures. Indeed, in Italy tax expenditures reviews should be combined with spending reviews to allow policy makers to analyze holistically government support to particular sectors, activities, regions, or agents. Otherwise, there is a risk of overlapping objectives and expenditures between different programs. The text table provides an illustrative estimate of agricultural sector support, but the transport sector and social welfare programs are other areas where potential duplication should be evaluated.

  • Tax expenditures often do not require annual scrutiny and renewal by Parliament. To incentivize proper consideration of costs and benefits, the authorities could consider legislating expiry clauses for all tax expenditures (especially those benefiting particular groups). In addition, any extension of existing tax expenditures, or granting of new ones, should only be possible as part of the annual budget process.

  • Particular caution is needed in aggregating tax expenditures, because the aggregate revenue gain from eliminating two tax advantages may differ from the sum of the gains from eliminating each in isolation.

It is hard to identify up front how much could be saved from a thorough review. However, using the study already done, policy makers should consider priority reviews for tax expenditures that are not fully quantified (e.g. VAT exempt regimes); are poorly targeted (e.g. VAT reduced rates); are better covered by expenditure programs (e.g. for medical services); benefit only certain groups that also receive spending support (e.g. reduced VAT and excise tax for public transport companies and agriculture); and are distortionary (e.g. preferential treatment for government bonds). Table 6 provides an illustrative guide for priority review (it should be noted that savings will not equal the total estimate of the tax expenditure cost if it is to be replaced by a better targeted spending program or if the reduction is partial).

Table 6.Illustrative Table of Tax Expenditures for Priority Review
DescriptionValue (billions)Percent GDPComment
VAT Reduced Rates39.202.50Not all of this can be saved, but some reduced rates could be eliminated and others harmonized. Expenditure programs may be needed to offset some redistributive impact.
VAT Exempt Regimes2.52Top-down estimate. Some are EC requirements.
Fuel excise tax exemptions and reduced rates3.170.20Accelerated structural reforms could provide alternative support to productivity.
Income tax credit for medical expenses and health assistance services.2.360.15Objectives potentially better achieved by spending progam.
Income tax credit for interest paid on mortgage for principal residence (or construction of principle residence).1.340.09A complete repeal could create budget difficulties for some households, so a review should proceed with caution focusing first on new mortgages.
Reduced income tax rates on the interest and bonuses from government securities and other forms of public debt e.g. postal bonds.1.380.09Supports government bond market over other forms of investment.
CIT: substitute taxes to encourage restructuring, mergers and transfers.13.830.88Objectives potentially better achieved through structural reforms. Estimates should be treated with caution as some may be one-off.
Total (illustrative)61.286.44
Sources: MEF, 2011; and IMF staff estimates.
Sources: MEF, 2011; and IMF staff estimates.
Selected Support for Agricultural Sector
DescriptionValuePercent
(billions)GDP
Tax Expenditures
Reduced excise tax for agricultural sector.0.960.06
Special VAT regime for agriculture.0.310.02
Reduced IRAP rate for agricultural sector.0.230.01
CIT exemption for certain agricultural cooperatives.0.170.01
Reduced transaction taxes for rural property.0.160.01
Budget Support
National funds for rural development1.50.10
EC Comon Agricultural Policy (mainly single payment)5.90.38

VIII. Concluding remarks

Tax expenditures are revenue foregone due to special tax treatment, such as exemptions and lower rates. Although international comparisons are complicated by different methodologies and assessments as to what constitute a tax expenditure, tax expenditures in Italy are clearly elevated. Although some forms of tax support may be justified, such as the universal income tax credit in Italy that substitutes a minimum threshold, tax expenditures are often a poor way of pursuing policy objectives, create distortions, and escape public scrutiny. Instead, government support given through tax expenditures, in particular the VAT reduced rates and exemptions, should be reviewed regularly in the budget process alongside normal expenditure. An added benefit would be a simpler tax system that reduces administration costs and strengthens compliance.

References
Annex I. Identifying and Quantifying ‘Tax Expenditures’

The first step is the definition of a benchmark tax system. It is this norm that allows identification of provisions in the tax system that are part of this tax norm and those that are not. However, there is no consensus on how to define a tax norm, meaning that certain tax provisions may be regarded as tax expenditures in some countries, but not others. The OECD outlines three broad approaches for defining a benchmark tax system (OECD, 2010b):

  • Conceptual approach, which uses an “optimal” tax system as the norm based on theoretical concepts of income, consumption, or value-added taxes. This norm may be modified to address data limitations and technical problems in applying a pure theoretical concept;

  • Legal approach, which largely uses a country’s own tax laws as a basis to define the benchmark, isolating differential or preferential treatment judged as tax expenditures (e.g., targeted provisions to address specific policy objectives); and

  • Expenditure subsidy approach, which seeks to cost only differential or preferential treatment that is clearly analogous to an expenditure subsidy.

Once tax expenditures have been identified, the second step is to quantify the budgetary cost of the individual tax expenditure “policies.” Again, there are several ways to estimate costs:

  • Foregone revenue method is an estimate of loss in government revenue due to the tax expenditure, based on the actual uptake of the differential or preferential treatment;

  • Earned revenue method is an estimate of additional revenue that would accrue from elimination of the tax expenditure, taking into account behavioral changes; and

  • Equivalent direct expenditure method estimates the cost of replacing the tax expenditure with a direct expenditure (subsidy) outside the tax system.

This paper, which develops the findings of IMF (2012), has benefitted from comments and inputs received from Ruud De Mooji, Luc Eyraud, Kenneth Kang, and Michael Keen. The author would also like to thank the staff and advisors of the Italian Ministry of Economy and Finance for helpful discussions. David Velazquez provided research support, and Barbra Licudine and Amara Myaing provided editorial help.

See Annex I for a more detailed description.

Since 2010, annual State Budget documents have included a list of all tax expenditures, but only at the level of central government and measured against current legislation—a narrower benchmark than used in the MEF report.

A DIT differs from a ‘comprehensive’ income tax in distinguishing between capital and labor income (rather than subjecting the sum of the two to a single progressive scale). To avoid arbitrage opportunities and facilitate implementation, the textbook prescription is to set the corporate income tax rate equal to the single rate on capital income. See IMF (2012) for a fuller discussion.

C-efficiency is the ratio of VAT revenues to the product of the standard VAT rate and consumption.

Box developed from IMF (2012).

The MEF estimate of the cost of reduced rates (2.5 percent of GDP) differs slightly from the estimate in Box 1 (2.9 percent of GDP) due to differences in the estimated base of taxable consumption.

IMF (2007) Article 3.1.3 and OECD (2002).

Other Resources Citing This Publication