Journal Issue

From Crisis to Convergence: Charting a Course for Portugal

Dmitry Gershenson, Albert Jaeger, and Subir Lall
Published Date:
March 2016
  • ShareShare
Show Summary Details

4. Dealing with Public Debt Overhang: Fiscal Adjustment and Challenges Ahead1

Portugal has made significant progress in its fiscal consolidation, overcoming structural and legal challenges, in the context of a sharp fall in output. Still, the debt-to-GDP ratio remains high (130.2 percent of GDP at the end of 2014). Over the medium term, the authorities face the challenge of balancing the need for further fiscal adjustment to reduce debt vulnerability, with the potentially negative implications for growth. This chapter takes stock of fiscal consolidation from 2011 to 2014, and proposes measures to increase the efficiency of spending and make space for targeted reforms aimed at promoting faster growth.

Buildup of Imbalances Prior to the Crisis

An expansionary fiscal policy and the materialization of large contingent liabilities have resulted in a rapid run-up in Portugal’s public debt since 2007. Gross public debt nearly doubled from 2006 to 2014, when it reached 130 percent of gross domestic product (GDP).2 About two-thirds of this increase has been due to fiscal deficits—the result of loose fiscal policies since euro adoption in 1999, together with adverse movements in interest rates. The global financial crisis aggravated the fiscal position, as stimulus policies led the deficit to reach 11.2 percent of GDP in 2010 in the context of low growth. The contraction in output experienced during 2011–13 also contributed to the increase in the ratio of public debt to GDP during this period.

A significant expansion of social spending over the past decade was at the root of the fiscal deterioration. The fiscal space created by the decrease in sovereign yields as Portugal moved into the euro area was more than offset by permanent spending increases; public expenditure increased by more than 9 percentage points of GDP between 2000 and 2010. In particular, spending on social benefits rose sharply during this period (+5.7 percentage points of GDP).

The deterioration of the fiscal position was accompanied by aggressive off-budget spending, leading to a buildup of substantial contingent liabilities. Portugal adopted one of the largest public-private partnership (PPP) programs in the world in the decade prior to the crisis, with cumulative investment of 12 percent of GDP in 1990–2011 (IMF 2014a). Similarly, the state-owned enterprise (SOE) sector expanded considerably, often to circumvent stricter policies applied to the general government entities in order to minimize the short-term impact on the deficit and debt indicators. In the aftermath of the global financial crisis, the financial imbalances that accumulated in SOEs led to the reclassification of a number of these entities to within the general government perimeter, adding over 10 percent of GDP to the stock of public debt.

Program Intervention and Implementation to Date

The strategy under the IMF-supported Extended Arrangement envisaged a sharp fiscal consolidation, with a front-loading of measures in order to boost credibility and restore market confidence. The fiscal path aimed to stabilize public debt through a sharp improvement in the primary balance and to achieve a fiscal deficit of 3 percent of GDP by 2013 (in line with the Stability and Growth Pact [SGP] objective), compared with 9.1 percent of GDP in 2010 (according to European System of Account [ESA95]).

An ambitious program of structural fiscal reforms was also adopted to support the consolidation efforts. Reforms aimed to streamline the functioning of the public sector in order to reduce fiscal risks. Priority was given to (1) better monitoring and reducing arrears, and strengthening commitment controls to prevent expenditure overruns; (2) designing subnational government financial arrangements; (3) enhancing the management and reporting of fiscal risks arising from PPPs and SOEs; and (4) reinforcing tax compliance and modernizing tax and customs administration.

Portugal has made significant progress in its fiscal consolidation efforts over the period of the program. An overall primary structural adjustment of 8.8 percent of GDP has been implemented since 2010, resulting in a shift to a sizable structural primary surplus since 2012. However, the overall adjustment still fell short of the original program objectives.3

In addition, the composition of adjustment deviated significantly from the original program design. The program request in 2011 focused primarily on expenditure consolidation, given the unsustainable level of spending; thus, three-fourths of the fiscal consolidation planned for 2011–14 was expected to come through an across-the-board spending adjustment. However, efforts to reduce the public wage bill and rein in pension expenditures were hindered by recurrent adverse rulings by the Constitutional Court (CC). As a result, the authorities relied more than initially planned on revenue-based measures to achieve program targets, with nearly half of the structural primary adjustment achieved from 2011–14 coming from higher revenue, mostly resulting from major changes in value-added tax (VAT) and personal income tax (PIT).

The bulk of the spending consolidation that was achieved came from investment spending (gross fixed capital formation), as current spending proved difficult to contain (Table 4.1). General government gross fixed capital formation was reduced by about 3.3 percent of GDP under the adjustment program (2010–14), while current spending continued to increase (by 1.1 percent of GDP). Non-interest current spending declined by 0.9 percent of GDP from 2010 to 2014 thanks to reductions in compensation of employees (−1.9 percent of GDP). However, these savings were partially offset by further increases in social benefits, which rose by 1.1 percent of GDP.

Table 4.1.General Government Spending by Economic and Functional Classifications(On the ESA 2010 basis)



points of GDP)
(Percent of GDP)
Economic classification
Total expenditure45.350.251.850.048.549.951.7−0.1
Current spending41.345.444.645.645.346.745.71.1
Compensation of employees13.114.013.712.811.712.511.8−1.9
Goods and services5.−0.1
Interest payments3.
Current transfers2.
Social benefits16.718.818.618.919.620.419.71.1
Capital spending4.−1.3
Gross fixed capital formation3.−3.3



points of GDP)
Functional classification
Total expenditure45.350.251.850.048.550.1−1.7
General public services6.
Public order and safety1.
Economic affairs4.−1.6
Environment protection0.−0.2
Housing and community amenities0.
Recreation, culture, and religion1.−0.1
Social protection15.016.917.117.718.218.91.8
Sources: Eurostat (ESA 2010); and IMF staff calculations.
Sources: Eurostat (ESA 2010); and IMF staff calculations.

Fiscal Balance

(Percent of GDP)

Sources: INE; and IMF staff estimates.

Overall Fiscal Balance

(Percent of GDP)

Sources: INE; and IMF staff estimates.

From a functional perspective, the reduction in spending was concentrated in the economic and security sectors (Table 4.1). Between 2010 and 2013, savings in the security sector and economic sector (driven by a sharp decline in transport spending) exceeded total expenditure reduction. Cuts in health (mainly on medical products and outpatient services) and education spending (particularly on primary and secondary education) were fully offset by higher social protection spending, driven by public pension outlays. General public services outlays, meanwhile, significantly increased on the back of higher debt service payments.

In 2000–10, real primary spending growth outpaced real output growth for all levels of government, particularly in the social security sector (Table 4.2), highlighting the role of weak expenditure control in the buildup of fiscal imbalances prior to the crisis. By comparison, real growth in primary spending in euro area countries was more in line with real output growth over the same period.

Table 4.2.Primary Spending in Portugal and the Euro Area by Level of Government1(Percent of GDP, unless otherwise indicated)
Nominal primary spending (Percent of GDP)2000–2010


General Government
EU 1843.644.845.347.744.14.1−3.6
Central Government
EU 1818.519.820.222.719.64.2−3.1
Local Government
EU 1815.515.716.016.515.00.9−1.4
Social Security Funds
EU 1821.621.621.121.921.30.2−0.6
Real primary spending growth (Percent)Average over


General Government
Central Government
Local Government
EU 182.30.8−−0.6
Social Security Funds
Real GDP growth (Percent)Average over 2001–20102010–2014 (Change)
EU 182.1−
Sources: Eurostat; and IMF staff estimates.

Spending by level of government includes transfers across levels of government, which are netted in general government data.

Italy was chosen as comparator country because its pre-2010 spending growth and the current debt level are similar to Portugal’s.

Sources: Eurostat; and IMF staff estimates.

Spending by level of government includes transfers across levels of government, which are netted in general government data.

Italy was chosen as comparator country because its pre-2010 spending growth and the current debt level are similar to Portugal’s.

Creating Fiscal Space for Growth-Enhancing Measures General Government Expenditure Targets

Empirical evidence suggests that spending rules are associated with expenditure containment and higher primary balances, and have a higher likelihood of compliance than other fiscal rules (IMF 2014b). Such rules would help mitigate spending pressures and ensure that future adjustment focuses on spending reform rather than further revenue measures, given Portugal’s already-high tax burden. Government revenue as a share of GDP in Portugal has risen at about three times the pace of the euro area since 2006 (in Spain, Portugal’s largest export market, it has actually fallen).1 In addition, excluding interest payments from the expenditure rule would neutralize the impact of shifts in financial market conditions on interest costs and help to ensure that savings from favorable current market conditions (see Box 4.1) are applied to debt reduction.

General Government Revenue

(Percent of GDP)

Sources: Eurostat; and IMF staff calculations.

Since 2012, Portugal’s medium-term budget framework (MTBF) has set expenditure ceilings for 10 high-level programs for the budget year plus one, and a binding overall spending limit for budget years plus two and three. Portugal has also integrated the “expenditure benchmark” of the SGP in its national legal framework, which requires real growth in primary expenditure to remain below the medium-term growth rate of potential GDP (for countries that have not yet reached their Medium-Term Objectives2). However, this has yet to be implemented.3

Real primary expenditure targets should preferably be set to achieve the annual structural adjustment called for under the SGP. This would provide a tool for implementing an expenditure-based structural adjustment (IMF 2015).4 Estimates show that keeping real primary expenditures unchanged in real terms from 2016 to 2018 would be consistent with the adjustment envisaged in the authorities’ April 2015 Stability Program, which targets a reduction in primary spending of 3.3 percent of GDP from 2015 to 2019.5

Institutional changes would be required to ensure that a spending rule could be operational and sufficiently binding. While the current medium-term fiscal strategy sets a nominal target for general government expenditure, it does not provide a breakdown by levels of government. Expenditure ceilings currently set in the MTBF only cover the central government, and exclude expenditure financed by own-source revenues (IMF 2014b). In order to ensure their effectiveness, real primary expenditure targets should cover all general government expenditure, and be sufficiently binding to anchor fiscal policy at all levels of government. The new Budget Framework Law 151/2015 requires that from 2019 the MTBF also covers social security (excluding unemployment spending) and expenditures financed with own-source revenues. This would require setting indicative aggregate spending targets for local governments6 and social security funds either in the medium-term fiscal strategy or a specific fiscal law, with the recently created intergovernmental coordination council being responsible for monitoring local government outturns. Mechanisms to monitor and adjust spending should be designed, such as incentives for local governments (linking the level of central government transfers to achieving the expenditure target),7 or alert mechanisms for health spending, with the obligation to adopt in-year corrective actions in case of spending slippages. An extension of the MTBF would also be needed to better capture central government spending—in particular, spending financed by its own revenue sources.

Specific policy measures to contain spending should also be identified to enforce the expenditure targets, with a focus on public sector wages and pensions.8 Public sector wages and pensions account for nearly 25 percent of GDP and more than half of non-interest government spending. Wage bill expenditure has been significantly reduced in Portugal under the program. However, the sustainability of these savings is uncertain, as many of the public sector wage cuts introduced under the program are set to be reversed over the medium term, in line with the CC rulings (see below). Public spending on pensions has continued to increase, although at a slower pace since 2010, and remains significantly above the euro area average.

Determinants of Portuguese Interest Rates

Portuguese sovereign bond yields spiked in response to global financial turbulence in 2008–09, receding with market conditions but rising again as the European debt crisis took hold. Several factors have driven the Portuguese 10-year yield down since its peak in early 2012, including a package of economic reforms, macroeconomic recovery, elevated investor risk sentiment, and the establishment of a more robust European backstop. While public debt levels remain high, it is important to understand the relative contribution of each of these factors in order to assess the future path of interest rates and the fiscal burden of interest payments.

Ten-Year Sovereign Spreads over Bunds

(Basis points)

Sources: Bloomberg; and IMF staff calculations.

Measures of Investor Risk Aversion


Sources: DataStream; and IMF staff calculations.

1Volatility index for the DAX (German Stock Index).

2Chicago Board Options Exchange Volatility Index.

High euro area sovereign yields began declining in tandem in the summer of 2012 as details about the European Central Bank’s (ECB) bond-buying programs were clarified. Over the same period several key indicators of investor risk sentiment also declined to near historical lows including the Chicago Board Options Exchange Volatility Index, the volatility index for the DAX (German Stock Index), and the index of High-Yield European Bonds. Portuguese economic fundamentals also improved over this period. The Portuguese government undertook a number of structural reforms, in particular to resume economic growth, strengthen private sector balance sheets, and improve debt management. The fiscal deficit narrowed and the current account deficit moved into surplus, debt levels stabilized, and unemployment declined.

Credit Ratings and Spread Compression

(All countries spreads to Bunds, basis points)

Sources: Bloomberg; and IMF Staff Estimates.

Percent of Total Spread Decline on Portuguese Event Dates

Source: IMF staff calculations.

A few stylized facts suggest that global factors explain most of the decline in spread volatility. Credit ratings across European sovereigns, a rough proxy for creditworthiness, have not improved significantly during the crisis, but their relationship with spreads has been significantly compressed. Similarly, outlier trading days—when the Portuguese spread rose or fell by more than two standard deviations—saw other European sovereign spreads moving in the same direction. If news about local factors were driving big spread movements, we would expect these outlier trading days to be uncorrelated across countries.

A statistical analysis of the evolution of 10-year sovereign bond spreads over this period also suggests that global and euro area-wide factors played a lead role in the decline in spreads. Panel regressions including a range of domestic, pan-European, and global variables indicate that Portuguese sovereign yields could rise significantly if investor risk sentiment and conditions in Europe return to their average values over the past 15 years. The estimates suggest that improvements in Portuguese economic flow variables have been largely offset by persistent stock imbalances, which continue to weigh on investors’ assessments. In addition, quantile regressions indicate that Portuguese spreads are particularly sensitive to stock imbalances, especially the public debt-to-GDP ratio, when spreads are high, while sensitivity to measures of risk aversion holds at all spread levels.

Contributions to Spread Growth

(Log of spread in basis points)

Sources: Bloomberg; DataStream; Haver; and IMF staff estimates.

Several local factors, which were not included in the analysis due to data limitations, may be responsible for some of this decline, biasing our results toward global factors. These include perceptions of Portuguese program ownership, the establishment of a sizable cash buffer, the lengthening of average debt maturity, and the diversification of the investor base. These factors certainly played some role but are difficult to quantify in a consistent way in a panel study. Other methodologies, for example a risk-on/risk-off regime-switching model, might suggest that current spreads can be maintained for as long as current conditions persist.

To the extent that these results reflect greater investor confidence in the European backstop, the observed gains may be treated as largely permanent. Normalization of investor risk sentiment, however, is an ongoing source of concern. Monetary policy liftoff in the United States and other global sources of volatility could cause a shift in credit risk pricing. A decisive and durable reduction in stock imbalances would help to limit Portuguese fiscal exposure to these conditions and secure recent gains from other reforms.

Public Sector Employment and Compensation

Portugal has achieved a comparatively large reduction in public employment since 2011 (Table 4.4). General government employment declined by about 10 percent over this period, reflecting a reduction at all levels of government. Attrition was the main driver of employment rationalization, as several measures intended to help encourage mobility outside the public sector (including the mobility pool and termination scheme) proved less successful than expected.

Public Wages and Pension Expenditure

Table 4.3.General Government Employment (ESA 2010)(Number of employees)
2011201220132014Percent change

General government727,294699,480674,338655,820−9.8
Central government551,373529,844509,520497,020−9.9
Local governments163,178157,324153,072148,078−9.3
Social Security Funds12,74312,31211,74610,722−15.9
Percent change

Public Entities out of general government172,85557,98657,512−21.1
Central government56,13541,60841,348−26.3
Local governments16,72016,37816,164−3.3
Total Public sector772,335732,324713,332−7.6
Source: Directorate-General for Administration and Public Employment.

Some SOEs were reclassified in the general government in line with ESA 2010, and are not excluded consistently over the 2011–September 2014 period.

Source: Directorate-General for Administration and Public Employment.

Some SOEs were reclassified in the general government in line with ESA 2010, and are not excluded consistently over the 2011–September 2014 period.

However, efforts to contain the public wage bill by curtailing compensation spending proved less successful, largely due to successive adverse CC rulings. Although a first round of wage cuts was permitted by the court, they are expected to be fully reversed by end-2016, at a total cost of 0.3 percent of GDP. The suspension of both the holiday and the Christmas allowances for civil servants after 2012 (and the second wave of wage cuts adopted in the 2014 Budget Law) were also successively canceled (with no retroactive effect).9 As a result, average monthly public earnings now exceeds its end-2011 level, partly due to the payment of the Christmas bonus in 12 installments.

General Government Employees: Average Monthly Earnings


Source: Directorate-General for Administration and Public Employment (DGAEP).

Several other structural reforms have moved slowly and will generate much more limited savings than envisaged. The Single Wage Scale was initially expected to generate significant savings (of about 0.4 percent of GDP) by streamlining career paths and allowing a permanent reduction of the pay level. Savings (of about 0.1 percent of GDP) were also expected from the adoption of the Single Supplements Scale through a reduction in the number of supplements. However, design issues as well as extensive phasing in (these reforms will be applied only to new employees and those with mobility between careers) will limit savings in the short and medium term.

The CC decisions have also created significant legal obstacles to reining in the public sector wage bill going forward. In particular, the CC requires (1) the burden of fiscal consolidation to be shared between civil servants and the rest of the population, and (2) wage bill consolidation to take place through structural reforms of public employment, rather than nominal cuts in wages. These legal restrictions necessitate a more strategic approach to reining in the wage bill over the medium term. The main challenge for Portugal will be to increase capacity in public administration, while reducing both public employment and the wage premium relative to the private sector.

There is considerable scope for rationalization of the public sector to reduce duplication and streamline public services. The public sector in Portugal is highly fragmented, with 6,095 separate institutional units (IMF 2014a/b). The central government, in particular, comprises 331 units, including 298 autonomous funds and services that carry out policy and operational functions. In the absence of strong control to ensure compliance with budgetary and financial regulations, this fragmentation was one of the root causes of fiscal slippages prior to the financial crisis. There has been some progress on consolidating financial reporting of these various entities under the program, including reclassification of 141 SOEs within the general government (IMF 2014a/b), privatization or dissolution of about half of local SOEs, and a reduction of nearly one-third in the number of parishes responsible for administration at the municipal level. However, there is still a very large number of SOEs outside the general government, and a reassessment is needed on the role and functions of the various levels of government in Portugal going forward (Cangiano 2013). In this regard, the authorities should prepare a comprehensive strategy to modernize the public sector, building on an in-depth assessment of the appropriate level of public employment needed to deliver the desired level of services.

A further reduction in public employment, particularly in overstaffed sectors, is needed over the medium term. Increasing the rate of natural attrition would provide a gradual approach to rationalizing public employment, and would generate savings in both the short and medium term. While across-the-board attrition would be easier to implement, it may have an adverse impact on service provision in certain sectors. As a result, the authorities should target overstaffed sectors, including the education sector, where staffing needs should fall in line with the declining school-age population (see below). This should be complemented by enhancing the requalification pool and scaling up the scheme for termination by mutual agreement, which has resulted in fewer departures than expected thus far. In addition, the increase in required working hours (to 40 per week) should be enforced across the public sector, in particular in local governments.

A key objective of these reforms should be to reduce the disparity between public and private sector wages (Manuel Campos and Centeno 2012). At present, less qualified civil servants receive relatively high pay compared with peers in the private sector and with more qualified civil servants. In addition, the wage scale is relatively flat and depends mostly on years of experience, rather than performance. This makes it difficult to attract highly qualified staff, as private sector opportunities (with lower entry salaries but steeper increases for performance) are considerably more attractive for those people (IMF 2013). In order to become more attractive for high-skilled workers and benefit from ongoing improvements in the quality of tertiary education,10 the civil service should identify specific skills that are needed in the public sector, and revise the relatively flat wage structure that proves costly and impairs talent acquisition (IMF 2013).

Finally, further measures are needed to contain the wage drift embedded in the current system.11 When progression is automatic, average wage levels rise as the public sector workforce becomes more experienced, even in the absence of wage and employment increases (IMF 2010). In the past, Portugal has been characterized by a powerful automatic progression system, which translated into early attainment of higher wage levels in most occupational careers, particularly for high-skilled workers (Centeno and Coutinho Pereira 2005). The authorities have already replaced the automatic progression mechanism with a system in which promotions and salary increases are now linked to the results of the staff appraisals, and subject to budget constraints. To prevent any relaxation of these changes, the authorities could further slow automatic progression by lengthening the maximum duration for a civil servant in each scale level. This could be anchored on the recent extensions of the retirement age (see below), and would make career progression more gradual, to avoid an early attainment of high wages.

Pension System

Portugal has implemented a number of reforms to the pension system in recent years. These measures include (1) the introduction of a “sustainability factor” in 2007 linking initial benefits with improvements in life expectancy at retirement; (2) the increase in the legal retirement age to 66 years, with further automatic increases linked to the evolution of life expectancy; (3) the suspension of pension indexation (excluding the minimum pensions); and (4) the creation of a solidarity surcharge levied on higher pensions. For the public sector, a “convergence law” entered into force in March 2014, aligning the rules for the public sector pension scheme (CGA) with the changes for new entrants to the general social security system.

However, the impact of these changes on pension expenditure has been limited by legal roadblocks and an excessive backloading of savings.12 The reforms introduced to date will generate savings only over the long term due to extensive grandfathering rules that protect current retirees. Therefore, public pension spending is expected to increase to 15.0 percent of GDP in 2030, before gradually declining to its 2013 level (13.8 percent of GDP) only by 2055 (EC 2015). On average in the EU, gross pension spending will spike at 11.7 percent of GDP in 2035, and then decline to 11.3 percent in 2055. In addition, the CC rulings have invalidated a number of changes to the pension system implemented under the program, imposing a requirement that reforms must account for equity and intergenerational solidarity issues, and not undermine the acquired rights of people who are already pensioners.

Moreover, Portugal’s public pension system remains inequitable, and will face major adverse demographic pressures in the long run. In addition, demographic trends are unfavorable, with the dependency ratio in Portugal expected to more than double by 2050 to become the highest in the EU.

Further pension reforms should aim to limit indexation and shorten the transition period to the new pension system, as gross spending on public pensions is expected to increase in the medium term (from 13.8 to 15.0 percent of GDP between 2013 and 2035, according to EC 2015). First, indexation mechanisms (based on GDP growth and CPI) that were suspended during the adjustment program should be revisited (except for minimum pensions). While low growth and inflation are expected to contain pension dynamics in the near term, a more sustainable formula is needed to prevent a procyclical increase in pension spending going forward. Second, the authorities should further reduce grandfathering for those who are not yet retired, and tighten some of the pension eligibility rules, particularly for those who would receive pensions from the CGA (the pension scheme for civil servants hired before December 2005). Finally, employees’ contribution to CGA could increase over time to limit subsidies received from the central government; for example, an increase by 1 percentage point of employees’ CGA contribution would provide about 0.1 percent of GDP of additional revenues annually.

Projected Old Dependency Ratios


Sources: Eurostat; and IMF staff estimates.

In addition to the adjustment of benefits to account for improvements in life expectancy introduced under the program, an economic adjustment factor could be applied to pension bonuses that were reinstated by the CC, conditioning the payment of these bonuses on achievement of a certain level of GDP growth, as in Hungary.

Finally, pension reforms could contribute to higher labor force participation. Portugal has already introduced bonuses in order to postpone retirement. The following reforms can raise labor participation further, particularly for low-skilled workers: (1) increase the reward for additional years of contributions for low-income workers, and (2) the minimum pension could be set to increase in strict proportion to the number of years of contributions, hence eliminating the current step increases (at 20 and 30 years of contributions), which create incentives for informality.13

Supporting Growth through Targeted Fiscal Measures Growth-Friendly Tax Policy

Growth-friendly tax policy recommendations typically consist of corporate tax reform and increased incentives for labor force participation. Corporate income taxes are generally considered to be the most harmful type of tax for economic growth, followed by personal income taxes, then consumption and property taxes (OECD 2010). Corporate taxes are deemed to discourage capital accumulation and productivity improvement, and often introduce a bias toward the use of debt financing (IMF 2014a/b). Therefore, a shift from direct income tax to indirect consumption and property tax is usually considered to be supportive of growth, as consumption and property taxes create less of a distortion of saving and investment decisions. Given that the VAT is generally regressive in advanced economies, tax reforms should be carefully designed to balance distributional and efficiency objectives. For example, minimizing the use of exemptions or reduced VAT rates and using the proceeds to increase social benefits is found to significantly reduce inequality while boosting tax revenues (IMF 2014a/b). In addition, increasing tax incentives for private research and development spending is often recommended as a way to support innovation and boost productivity growth.

In 2012, labor taxation in Portugal was significantly lower than the euro area average and lower than in other southern European countries, while the implicit taxation of capital was significantly higher. Recent developments have been positive, moreover, with a decline in the implicit taxation on capital and an increase in the share of indirect taxation, in particular VAT.14 Indeed, incidence analysis shows that the incidence of corporate taxes tend to fall on labor to the extent that labor is relatively less mobile than capital.15 Reducing taxation on capital could therefore be more progressive than usually considered, and could contribute to lower taxes on labor and capital, hence improving competiveness and employment. However, the implicit taxation of labor increased further during the program period through higher PIT.

In 2013, Portugal adopted a comprehensive reform package to address deficiencies in the corporate income tax (CIT) regime. The comparatively high CIT rate was reduced from 25 percent in 2013 to 21 percent in 2015, while the reform of tax provisions applicable to holding corporations should help to attract and retain large multinational companies in Portugal. In addition, several previously ineffective investment tax incentives were revised in 2014 to offer higher CIT credits, particularly for investment in poor regions, information and communication technologies, and small and medium enterprises (SMEs). These reforms should help create a business environment that is conducive to higher investment.

Implicit Tax Rates

Tax incentives for research and development in Portugal compare favorably with peer countries, with little need to expand the current framework. The Portuguese tax system appears supportive in this regard, both for large firms and for SMEs, as well as for profit-making and loss-making companies. The 2013 CIT reform also introduced intellectual property incentives for income derived from patents, and extended the carry-forward period for research and development (R&D) spending.

Tax Subsidy for R&D Expenditures, 2013

(Rates, 1-B-index)1

Source: OECD Science, Technology and Industry Outlook 2014.

1The OECD defines the tax subsidy rate as a measure of the before-tax income needed by a firm to break even on US$1 of R&D outlays.

Going forward, reforms should focus on eliminating capital tax-induced distortions to growth, and reducing informality and tax avoidance. The authorities have committed to further reductions in the standard CIT rate. However, priority should be given to elimination of the state CIT surcharge, as its progressive rates create distortions in investment decisions. To partially finance these tax cuts, the authorities could remove tax incentives for investment in the medium term, as removing the preferential tax treatment created by the surcharge should help to improve the quality of investment (OECD 2010). A reduction in VAT exemptions and in the use of reduced rates could also partially offset the CIT reform fiscal costs, and would further shift the tax burden from income to consumption. In terms of equity, VAT exemptions and reduced rates are blunt redistributive instruments, because the rich generally spend more in absolute terms on exempted goods and thus enjoy significant benefits. Targeted transfers and progressive PIT transfers are usually more effective tools to help the poor and the most vulnerable (IMF 2014a/b). Revenue administration reforms could also support growth by further lowering compliance costs, and reducing tax avoidance.

The introduction of a deduction for corporate equity would reinforce recent changes to reduce corporate debt bias and help bring down corporate indebtedness. Tax systems typically favor corporate debt over equity, as interest payments are deductible for CIT purposes while dividend payments to shareholders are not (IMF 2011). The large debt burden in Portugal continues to have a significant adverse impact on both corporate investment and bank balance sheets. Portugal has progressively limited the deductibility of interest for companies in order to reduce the debt bias and prevent debt shifting by the multinational companies, in line with common practice in the EU.16 This could be complemented by introducing an allowance for corporate equity (ACE), which provides a deduction for the normal return on equity, equivalent to the rate of return on government bonds (a proxy for risk-free rate of return on capital). Estimates for selected advanced countries suggest that an ACE would have a significant impact on both corporate deleveraging and output gains (De Mooij, Keen, and Orihara 2015). Tentative calculations indicate that an ACE for advanced countries would involve an average budgetary cost of 0.5 percent of GDP. Revenue cost can be significantly mitigated through adequate design, by applying the ACE only to new investment, as recently implemented in Italy.

Other tax policy reforms can complement labor market reform. Disincentives to labor participation in the current system are particularly high for low-income families that may lose social benefits if a second earner enters the labor force. Possible reforms include the introduction of an earned-income tax credit that would separate employers’ labor cost from workers’ take-home pay.

More Efficient Productive Public Expenditure to Support Growth

Public infrastructure investment is typically considered to have a positive impact on output growth, in the short term through demand effects and the crowding in of private investment, and in the long term by raising productive capacity of the economy.17 In addition, debt-financed projects could have large output effects without increasing the debt-to-GDP ratio, if clearly identified infrastructure needs are met through efficient investment. However, these arguments appear less applicable in the case of Portugal. First, the high level of public debt limits the scope to increase borrowing for further investment without a potentially large adverse impact on financing costs. Second, public investment in Portugal increased significantly in the two decades before the crisis, and was significantly higher than in the euro area until recently. This has brought about a large and high-quality public capital stock in Portugal. Moreover, the level of the public capital stock is likely underestimated due to the large role of public-private partnerships and SOEs, which are reported outside the general government in public investment.18 As a result, there is little need for a further scaling up of public investment.

Public Investment and Capital Stock

(Percent of GDP, 2005 PPP US$-adjusted)

Sources: Center for International Comparisons (2013); OECD; WEO; and IMF staff estimates.

1Data include PPPs managed by the central government, and not local governments.

Instead, priority should be given to maintaining the quality of capital stock. If Portugal and other euro area countries were to maintain their current level of public investment in the long term, estimates show that Portugal’s capital stock would converge to the euro area level by 2022 (even with the drop in public investment in Portugal in 2013–14). However, in the absence of higher spending on maintenance, Portugal’s public capital stock would significantly deplete afterward and return to its 1993 level by 2030. The near-term priority, therefore, should be to ensure that maintenance spending is sufficient to prevent rapid deterioration in the quality of infrastructure, particularly in the transport sector where private and public maintenance spending have been low in comparison to other advanced economies.

Maintenance Spending

(Percent of GDP)

Sources: OECD International Transport Forum; and IMF staff calculations.

In addition, there is considerable scope to improve both the efficiency of education outlays and education outcomes in Portugal. Education spending rose significantly before the adjustment program (+0.7 percent of GDP in total over 2000–10), particularly on secondary and tertiary education. Despite a subsequent decrease, it remains above the EU average in percent of GDP, and significantly higher than in other southern European countries with higher GDP per capita. Average class size in Portugal was well below the 2012 EU average in primary education, with significantly lower student-teacher ratios at the secondary level. However, this has not been accompanied by a commensurate improvement in outcomes. Despite higher average growth in attainment rates in Portugal than the EU average over 2000–12, education attainment remains lower, both for tertiary and non-tertiary educations. Portugal’s Programme for International Student Assessment (PISA) scores significantly improved between 2003 and 2012, but remain below the EU average in mathematics and science.

Education Indicators

Sources: Eurostat; UNESCO; OECD PISA; and IMF staff calculations.

1 P: primary; S: secondary; T: tertiary.

2 Relative to full-time, full-year adult workers with tertiary education.

The improvement in education outcomes came at a high fiscal cost. Teachers benefit from a significant wage premium compared with other workers with tertiary education, while the EU average shows a large compensation gap for teachers. In addition, staff costs as a share of current spending (92.1 percent, for public institutions only) were the highest in the EU in 2011 (the EU average was 77.2 percent) (OECD 2014).

Evolution of Student Population in Portugal, 1998–2012

(Millions of students by level of education)

Source: OECD PISA.

Going forward, there is a need to better align the level of staffing with the shrinking school-age population. The overall number of students in the education system in Portugal fell by 2.4 percent between 1998 and 2012, and is projected to decline further. The number of students at the primary level is expected to shrink by about 20 percent during 2013–30 and by a further 15 percent during 2030–60, with a commensurate decline in the number of secondary students as they move through the system (IMF 2013). Student projections for Portugal differ significantly from comparators and the EU average, which are expected to see a limited decline (Spain, EU average) or a small increase in their students’ population (Ireland, Italy). Further adjustment to the school network and to the number of teachers will be needed as a result, particularly in rural areas where the pace of population decline is much more accelerated.

Ambitious structural reforms have been launched to improve the quality of education services. Portugal has significantly stepped up its education reform agenda in recent years, including both the rationalization of the school network and a reduction in staffing. Future priorities include a new vocational training system to better match private sector needs (EC 2014), a program to reduce early dropout rates and increase the rate of education attainment, and a transfer of decision making to the local level (OECD 2015). Policy reforms could also include reducing grade repetition, better supporting disadvantaged students and schools, increasing family choices, strengthening teacher training and evaluation, and further decentralizing decision making (OECD 2014).

Number of Students

(Index: 2013=100)

Source: European Commission.

Improving Management and Implementation of Fiscal Reforms

The IMF-supported program provided institutional setup to discuss and monitor fiscal reforms, and offered leverage to ensure effective implementation. A reform unit was created within the prime minister’s office to centralize and manage policy discussions on structural reforms.1 Various ad hoc committees were also put in place to carry out public policy evaluations, such as the expenditure review in 2012, and more recently on pension reforms. Finally, the program provided the authorities with a framework for monitoring and assessing the implementation of structural reforms, including in the fiscal sector.

Institutionalizing the management of fiscal reforms will be essential in order to maintain this momentum over the medium term. This would include establishing mechanisms whereby changes are anticipated and options for reforms evaluated on a daily basis (Cangiano 2013), helping to ensure buy-in by civil servants, and reinforcing a reform culture within the public administration. Reliance on ad hoc and temporary commissions and working groups to prepare policy reforms should be reduced, in order to ensure continuity in the reform agenda and to promote horizontal coordination between services on a routine basis. A reform unit within the Ministry of Finance, under the authority of the minister, could coordinate the public sector reform agenda and provide regular reporting to the Prime Minister’s Office. This unit could usefully build on the current structure and staff of the Office of Planning, Strategy, Evaluation, and International Relations in the Ministry of Finance, and should rely heavily on the expertise and staff of core departments within the Ministry.

The Ministry of Finance should continue moving toward a more policy-oriented approach, particularly with regard to budgeting. In many advanced countries, recent reforms have promoted a broader strategic perspective, with a medium-term approach that focuses on improving the efficiency of service delivery, assessing the impact of policies, and managing and mitigating risks (Allen and others 2015). In Portugal, the budget function is currently geared toward detailed line-item budgeting and control, and lacks a more programmatic view of the budget and public sector reforms. The reorganization of the Budget Office at the Ministry of Finance should provide an institutional framework that is more conducive for analyzing drivers of public spending. The Ministry of Finance should aim to build on these changes by embedding its spending review into the annual budget process, and putting in place a performance-based budget framework to better evaluate the efficiency of spending. Recent changes in fiscal reporting, fiscal forecasting, and budgeting go in the right direction (IMF 2014a/b), and the adoption of a new Budget Framework Law provides a blueprint for public sector reform in the years to come.


Portugal has achieved a sizable fiscal consolidation since 2010, but relied more than initially planned on revenue measures and investment spending reductions, as it proved difficult to rein in current expenditure. Public debt has continued to rise, meanwhile, partly reflecting the materialization of large contingent liabilities. The level of public indebtedness exceeded 130 percent of GDP at the end of 2014 and remains a significant vulnerability, with only a gradual decline projected over the medium term under current fiscal policies.

Further fiscal adjustment is essential to accelerate the downward trajectory of public debt, and minimize scope for a significant worsening of debt dynamics, should downside risks materialize. While growth has resumed, the recovery is projected to be modest, creating a challenge for policymakers to balance the need for further fiscal adjustment with the potentially negative implications for growth.

The adoption of expenditure targets and institutional reforms would help underpin spending rationalization going forward. Identifying specific policy measures to rein in spending will be critical, with a focus on public sector wages and pensions. Targeted tax policy measures could also help to address bottlenecks to growth created by high corporate indebtedness and labor slack. While there is little fiscal space or need to further scale up public investment to support growth, education reform can have a positive impact on the skills composition of the labor force. Finally, in order to sustain reform momentum, the authorities need to institutionalize fiscal reform management and move toward a more policy-oriented approach. The implementation of the new Budget Framework Law offers a critical opportunity to move forward on this agenda over the medium term.


    AllenRichardYaseminHurcanPeterMurphyMaximilienQueyranne and SamiYlaoutinen. 2015. “The Evolving Functions and Organization of Finance Ministries.” IMF Working Paper 15/232International Monetary FundWashington.

    • Crossref
    • Search Google Scholar
    • Export Citation

    CangianoMarco. 2013. “Considerations on Public Sector Reforms in Portugal.” Bank of Portugal Seminar 2013 pages 150167. Lisbon.

    • Search Google Scholar
    • Export Citation

    CentenoMario and ManuelCoutinho Pereira. 2005. “Wage Determination in General Government in Portugal.” Bank of Portugal Economic Bulletin.Lisbon.

    • Search Google Scholar
    • Export Citation

    CentenoMario and ManuelCoutinho PereiraMichaelKeen and MasanoriOrihara. 2015. “Taxation, Bank Leverage, and Financial Crisis.” IMF Working Paper 13/48International Monetary FundWashington.

    • Search Google Scholar
    • Export Citation

    European Commission (EC). 2014. “The Economic Adjustment Program for Portugal 2011–2014.” Occasional Paper 202Brussels.

    European Commission (EC). 2015. “The 2015 Ageing Report. Economic and Budgetary Projections for the 28 EU Member States (2013–2060)”. Luxembourg.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF). 2010. “Evaluating Government Employment and Compensation.” Technical NoteWashington.

    International Monetary Fund (IMF). 2011. “Tax Biases to Debt Finance: Assessing the Problem, Finding Solutions.” Staff Discussion Note 11/11Washington.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF). 2013. “Rethinking the State—Selected Expenditure Reform Options.” Country Report 13/6Washington.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF). 2014c. “Fiscal Transparency Evaluation.” Country Report 14/306Washington.

    International Monetary Fund (IMF). 2014b. “Fiscal Policy and Income Inequality.” Policy PaperWashington.

    International Monetary Fund (IMF). 2014a. “Expenditure Rules: Effective Tools for Sound Fiscal Policy.” Fiscal Monitor.WashingtonApril Chapter 2.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF). 2014e. World Economic Outlook.WashingtonOctober Chapter 3.

    International Monetary Fund (IMF). 2014d. “Growth-Friendly Fiscal Policy.” G20 Note. Washington.

    Manuel CamposMaria and MarioCenteno. 2012. “Public-Private Wage Gaps in the Period prior to the Adoption of the Euro: an Application Based on Longitudinal Data.” Bank of Portugal Working Paper. Lisbon.

    • Search Google Scholar
    • Export Citation

    Organisation for Economic Co-operation and Development (OECD). 2010. “Tax Policy Reform and Economic Growth.” Tax Policy Studies. Paris.

    • Search Google Scholar
    • Export Citation

    Organisation for Economic Co-operation and Development (OECD). 2013a. “PISA Results: Excellence through Equity: Giving Every Student the Chance to Succeed (Volume II).” Paris.

    • Search Google Scholar
    • Export Citation

    Organisation for Economic Co-operation and Development (OECD). 2015. “Education Policy Outlook 2015, Portugal Country Snapshot.”


Prepared by Maximilien Queyranne (FAD) and Matthew Gaertner (EUR). The authors would like to thank Marco Cangiano, David Coady, and Ruud de Mooij (all FAD) for helpful suggestions, and the authorities for useful comments.


The data in this chapter reflect the September 2015 excessive deficit procedure (EDP) notification.


The 2014 outturn has been revised to reclassify the loan provided to the Portuguese resolution fund for the recapitalization of Novo Banco as a fiscal outlay, increasing the deficit to 7.2 percent of GDP from 4.5 percent as initially reported.


The increase in general government (GG) revenue as a share of GDP reflects not only changes in tax policy and revenue administration performance (numerator), but also GDP changes (denominator).


The Medium-Term Objective (MTO) is set by the European Commission for each EU member state under the preventive arm of the Stability and Growth Pact. It consists of a country-specific medium-term budget target, defined in structural terms, intended to ensure progress toward compliance with EU fiscal rules. Portugal’s objective was set in 2013 at –0.5 percent of GDP, to be met by 2016. All countries must reach their objective or be on an appropriate adjustment path toward it, determined in reference to the size of the output gap.


As the expenditure benchmark does not apply to countries subject to the Excessive Deficit Procedure.


Evidence suggest that expenditure target defined in relationship with GDP are less binding, as GDP targets are often set too high for ensuring fiscal constraints. See IMF 2015.


The Stability Program targets a decrease in public debt to 107.6 percent of GDP by 2019. In addition to the larger reduction in primary spending, this also reflects a larger projected decline in interest spending and stronger medium growth than under staff’s baseline scenario. The Stability Program assumes real GDP growth of 2.4 percent in 2019, compared with staff’s projection of 1.2 percent.


Local governments have been developing MTBFs, but there is no consolidation for the local government sector as the whole.


The 2013 Local Finance Law created an early warning mechanism for local governments that break the debt ceiling rule. This mechanism could also be used to monitor expenditure growth.


This also aligns with the fiscal plans outlined in the Stability Program, which target expenditure savings of 2.5 percent of GDP from 2015 to 2019 from rationalization of the public sector wage bill and social security reform.


The holiday and the Christmas allowances for pensioners were also canceled, but they are not classified in the wage bill category.


In 2012, 28 percent of 25–34-year-olds had achieved tertiary education, compared with 11 percent of 55–64-year-olds.


There are usually two components in the wage drift: a positive one that increases wage spending (impact of discretionary promotion, automatic progression, and promotion related to civil servants passing competitive exams), and a negative one that reduces wage spending (savings due to lower level of compensation for new employees compared with higher level of compensation for employees retiring).


The following measures were canceled: (1) the suspension of the holiday and the Christmas allowances for public pensioners, (2) the new calculation formula for surviving dependents’ pensions, (3) the reduction in pension benefits granted within the pension scheme for the public sector, and (4) the progressive sustainability contribution on pensions. As a result of the cancellation of the progressive sustainability contribution, minor increases in employee’s social security contributions and in the standard VAT rate were also cancelled (European Commission 2014).


For example, with a minimum contribution history of 15 years, minimum pensions increase at 20 years of contributions, and then again at 30 years of contributions. Hence, low-income workers, who likely receive a minimum pension, do not gain from contributing in years 16–19 and 21–29.


In addition, a comprehensive reform of the property tax was adopted, which included a general update of the values of properties to increase revenues.


Recent empirical evidence on the long-run incidence of corporate taxes suggests that between 45 and 75 percent of the corporate tax burden falls on wages. See IMF 2014a/b.


The cap has been set to €1 million in net interest payments, and has been limited to a percentage of the EBITDA (from 70 percent in 2013 to 30 percent in 2017). In addition, the CIT rate cuts adopted in 2014 will contribute to limit the bias toward debt as less tax is saved at lower rates.


See IMF 2014a/b. Demand effects are usually expected to be stronger where accompanied by economic slack and monetary accommodation, as in Portugal.


Central government SOEs investment in fixed assets was about 0.4 percent in 2012 and 2013, net of government investment grants for private and public investment (0.3 percent of GDP).


Following the end of the Fund-supported program, this unit was dissolved, and its functions integrated in the Ministry of Finance.

Other Resources Citing This Publication