The paper examines macroeconomic and structural factors potentially explaining the country's underperformance. A comparison between the reform and baseline policy scenarios underscores maintaining a strong fiscal position, early reductions in primary expenditures, and reducing fiscal vulnerability. Assigning the financing and management of the health care system to the regions may increase the efficiency and the productivity of the health care system. The information on Italy's economy and legal as well as regulatory environment is available on the worldwide web, and the paper lists the related sites.

Abstract

The paper examines macroeconomic and structural factors potentially explaining the country's underperformance. A comparison between the reform and baseline policy scenarios underscores maintaining a strong fiscal position, early reductions in primary expenditures, and reducing fiscal vulnerability. Assigning the financing and management of the health care system to the regions may increase the efficiency and the productivity of the health care system. The information on Italy's economy and legal as well as regulatory environment is available on the worldwide web, and the paper lists the related sites.

II. Italy’s Fiscal Strategy in a Medium-Term Framework33

64. Over the course of the 1990s, Italy has undergone a remarkable fiscal adjustment, which enabled it to become one of the founding members of European Economic and Monetary Union (EMU). The deficit of the consolidated general government, which in 1990 stood at 11 percent of GDP (equal to its average over 1981-90), had been brought below 3 percent of GDP by 1997, the year when compliance with the Maastricht limits was formally assessed. Over the following two years the deficit declined further, to less than 2 percent of GDP in 1999. The fiscal consolidation has also allowed Italy to arrest the explosive upward trend in its debt-to-GDP ratio, which peaked in 1994 at 124 percent, and has since been reduced by 9 percentage points.

65. This remarkable fiscal adjustment has set the stage for shifting the focus away from immediate considerations and toward a comprehensive medium-term fiscal strategy. What is the appropriate fiscal position over the medium term? How quickly should the debt-to-GDP ratio be lowered further? Should the size of government—as measured by the ratio of public revenues and expenditures to GDP—be reduced? What would be the implications for the rate of growth of economic activity?

66. In view of Italy’s still high stock of public debt, and of the potentially sizable budgetary impact of the looming demographic shock, the appropriate medium-term fiscal strategy should be designed with an eye to longer-term developments. This chapter tries to highlight some key tradeoffs and constraints that emerge in designing a fiscal strategy for Italy over the medium and longer term.

A. The Budgetary Costs of Aging

67. Italy enters the new decade with two constraints that are, in important respects, more severe than in most industrial countries. First, the debt stock is still above 110 percent of GDP, leaving the budget exposed over the medium term to the risk of increases in interest rates—the improvement in the fiscal balance of the last few years owes much to a substantial decline in interest payments. Second, Italy faces one of the most adverse aging trends among OECD economies—the old-age dependency ratio is projected to increase from less than 30 percent in 2000 to 67 percent by 205034 (the highest in the OECD)—with sizable budgetary implications. This section reviews the fiscal pressures that are likely to arise from the projected aging of the Italian population.

68. The budgetary impact of population aging is likely to be most pronounced on social spending. At present, Italy’s overall level of social expenditures is not grossly out of line with that prevailing in other industrial countries, but its composition is heavily biased toward pension benefits. Overall social expenditures as a ratio to GDP are about 2 percentage points higher than the OECD average (although slightly lower than the European Union average). Within social expenditures, however, health expenditures are among the lowest in OECD economies, while pension expenditures, on the other hand, are the highest.35 An important reason behind the high level of pension outlays in Italy is the significant role played by benefits which are not strictly old-age-related: early retirement schemes and disability pensions, for example, have often been used as social buffers.

Pensions

69. The most direct impact of population aging will of course be felt on the evolution of pension benefits. Italy’s pension system, which at the beginning of the 1990s was one of the most generous among OECD countries, underwent three major rounds of reforms over the last decade.36 These reforms have gone a long way toward limiting the projected future increase in the ratio of pension expenditures to GDP: the authorities estimate that, in the absence of such reforms, they would have reached a staggering 23 percent of GDP by 2035, when the budgetary impact of the aging crisis will reach its peak—that is, nearly 9 percentage points of GDP above their current level. These reforms notwithstanding, pension expenditures are bound to increase over the next few decades starting from an already very high level.

70. Estimates of the likely increase in pension outlays vary significantly. Projections developed by the Italian Treasury point to an increase in the pension expenditure to GDP ratio by 1½ percent of GDP over the next thirty years. The OECD (2000) regards the Treasury’s forecast as perhaps optimistic, as the underlying demographic projections elaborated by ISTAT seem likely to overestimate fertility rates and underestimate longevity. The Fund staff, in a background study to last year’s consultation (SM/99/115, 5/20/99), estimated that pension expenditures as a ratio to GDP could increase by some 4 percentage points between 2000 and 2035.

71. Long-term projections of pension expenditures are unavoidably subject to a high degree of uncertainty, and they are particularly sensitive to assumptions on demographic variables (fertility and mortality), on labor market variables (unemployment, migration, and labor force participation rates), and on other key variables such as labor productivity rates. The staff’s simulations were carried out making use of the projection model developed by the Italian Treasury, and were based on the central ISTAT projections for demographic and labor market variables. In the baseline scenario, the staff assumed that labor productivity would grow at 1.2 percent per year—the same rate employed, for example, by the U.S. administration to study the implications of aging for the U. S. social security system. This rate was suggested by a catch-up model of labor productivity growth, indicating a tendency for labor productivity rates in Italy and other industrial countries to converge to the U. S. rates as the gap in labor productivity levels is eliminated. The staff’s study also considered two alternative scenarios: a “high productivity growth” scenario, with labor productivity assumed to grow at 1.7 percent per year, and a “low productivity growth” scenario, with labor productivity assumed to grow at 0.7 percent per year. The OECD’s Ad Hoc Experts Group on Public Finance Implications of Aging has recently decided to assume in its baseline scenario a labor productivity growth rate of 1¾ percent per year for all EU countries. In the staff’s simulations of the high productivity growth scenario this would still result in an increase in pension expenditures by 3 percent of GDP between 2000 and 2035.

Health care

72. The second important source of aging-related budgetary pressures is medical care for the elderly, which has so far played only a secondary role in Italy’s social safety net. This is in part a reflection of the heavy bias of Italian social spending toward pension benefits—as mentioned above, health care expenditures in Italy are among the lowest for OECD economies, whereas pension expenditures are among the highest. It also reflects the Italian social context, where care for the elderly has traditionally been provided mainly within the family.

73. The social context, however, is likely to undergo important changes in the future. In particular, the combination of lower fertility rates—which lead to a decline in the average family size—and higher rates of labor force participation by women will make it increasingly difficult for the family to continue as the chief provider of care for the elderly. Pressures will therefore inevitably arise for the public sector to play an increasing role in the supply of care services for the elderly. At the same time, the demand for such services is likely to accelerate—even in the absence of significant changes to the social context per capita medical costs would increase with population aging, as the elderly are more likely to need recourse to pharmaceuticals or hospital care. Moreover, the proportion of the very elderly is projected to increase substantially—between 2000 and 2050 the proportion of people aged 80 years and over in the total population is expected to rise from 3.7 percent to 11.7 percent, bringing their share in the 65 and over age group from 23 to 37 percent.37

74. The budgetary implications may be substantial, and the exact impact on public expenditure will of course depend on the policy decision with respect to service levels. A study published by the Italian Treasury considered two different scenarios.38 In the first scenario, standardized per capita consumption of health care services is assumed to increase at the same rate as labor productivity.39 Under the baseline demographic projections (ISTAT) population aging would then cause an increase in health care expenditures by 1.7 percent of GDP between 1995 and 2045. In the second scenario, standardized per capita consumption of health care services is assumed to increase at the same rate as per capita income. As in a context of aging population per capita income would grow at a slower pace than labor productivity, the increase in health care expenditure would be limited to 1 percent of GDP. Simulations developed by the OECD—based on different demographic assumptions than those of the Italian authorities—indicate that health care expenditures may increase by 1½- 2½ percent of GDP between 2000 and 2030.

75. The burden of the increased cost of health care services for the elderly could be limited by shifting a part to the elderly themselves through higher co-payments. The OECD estimates that the retirement age population in Italy has incomes in the range of 90-100 percent of average incomes, and that many households are likely to have built up assets prior to retirement. This picture may however change significantly as a higher proportion of the retired population will be accounted for by those who retire under the new, less generous pension regime. Moreover, special care should be taken to safeguard the needs of the retirement-age households with incomes close to the poverty line.

76. Overall, therefore, increased demand for health care services stemming from population aging seems bound to impose significant additional pressure on the budget over the coming decades, compounding the pressure created by rising pension expenditures.

B. Economic Growth and the Size of Government

77. The relative ease or difficulty with which the additional budgetary pressures generated by population aging can be accommodated will depend to an important extent on the rate of economic growth. Population aging itself will of course tend to depress the rate of real GDP growth through a decline in the overall level of employment.40 This direct effect could be partially offset by a combination of (i) a decline in the unemployment rate; (ii) an increase in the overall participation rate for women; (iii) an increase in the participation rate for both men and women in the older age brackets, through an increase in the average effective retirement age; and (iv) an increase in immigration. For these developments to materialize, however, further significant structural reforms would have to be implemented, in particular in the labor market.

78. The productive potential of the economy may also benefit from reducing the size of government, and in particular from lowering the fiscal burden. The relationship between taxes and economic growth has been the subject of intensive investigation. Economic theory points to several ways in which higher taxes can depress output growth.41 Higher taxes can reduce the supply of production factors, discouraging investment or labor participation; they can lower the factors’ marginal productivity, by distorting the intersectoral allocation of investment and employment; and they can limit productivity growth by discouraging investment in research and development or the development of high-tech industries.

79. The potential impact on productivity growth also points to the possibility that tax policy may have a long-run impact on economic growth rates. Such a possibility does not arise in the context of the traditional neoclassical growth framework a la Solow. In the Solow growth model, the steady state rate of per capita real GDP growth is uniquely determined by the exogenous rate of productivity growth; the distortionary impact of taxes therefore has a permanent impact on the level of output, but only a temporary impact on the growth rate of real GDP. In endogenous growth models, instead, the long run rate of economic growth is determined by economic decisions such as the accumulation of human capital or investment in research and development, and can therefore be affected by tax policy. Empirical studies based on endogenous growth theory provide some evidence that tax policy may have a permanent effect on the rate of economic growth, although the estimates of the impact vary significantly.42 In the case of Italy, the weak growth performance of the past decade seems to be due in no small part to the marked increase in fiscal pressure over the period (see Chapter I).

80. The impact of tax policy on economic growth will crucially depend not just on the level of the tax burden itself, but also on its composition, which is key in determining the distortionary impact of the tax system. Income taxes, for example, create a stronger disincentive to save (and hence to invest) than consumption taxes.43 Daveri and Tabellini (1997) highlight the potential growth-reducing impact of labor taxes in the presence of a highly unionized labor market, and provide empirical evidence of a high negative correlation between tax rates on labor income and employment and growth in European countries.

81. It can of course be misleading to consider tax policy in isolation, as the impact of a given change in taxes on economic growth will depend on what changes are simultaneously affecting public expenditure and the overall public sector balance. As in the case of taxes, the composition of public expenditures plays a key role, as some categories of public spending are likely to have a positive impact on growth—for example, public spending on education, and investment in communication and infrastructure.44

82. Several empirical studies have found a negative correlation between the size of government as measured by the ratio of public spending to GDP or by the aggregate tax burden—above a certain threshold level—and economic growth. The evidence is, however, not unequivocal. Easterly and Rebelo (1993), Barro and Lee (1994), Barro and Sala-i-Martin (1995), and Sala-i-Martin (1996) have found a negative correlation between growth and such variables as government consumption, government investment, government deficits, overall government expenditures, and taxes. Barro and Sala-i-Martin (1995), however, also found a significant positive impact on growth of public spending on education.

83. Estimates of the precise impact of changes in the size of government on the rate of growth vary. Engen and Skinner (1992) found that an increase in the tax revenue-to-GDP ratio by 2.5 percentage points would lower long-term output growth rates by 0.2 percentage points. In their cross-country studies Easterly and Rebelo (1993) estimate a coefficient of 0.24 for overall government expenditure, whereas both Barro and Lee (1994) and Sala-i-Martin (1996) estimate a coefficient of 0.1 for government consumption. In the case of France, Habermaier and Lenseigne (1998) estimated that a 1 percentage point increase in the revenue ratio would reduce the employment rate by 0.3-0.5 percent. In their analysis-of the macroeconomic impact of rising age-related entitlement spending, they assumed a coefficient of ¼ on the employment rate, which in turn implied that a 1 percentage point increase in the revenue ratio would lower the level of real GDP by about 0.2 percent.

84. Some findings in the empirical literature suggest, therefore, that the impact of a reduction in the revenue ratio on real GDP growth may range between 0.1 percent and 0.5 percent. In the case of Italy, for the purposes of the simulations discussed below in Section C, this paper will adopt the relatively conservative assumption that a decrease in the revenue ratio by 1 percentage point leads to an increase in the real GDP level by 0.2 percentage points, phased over a five-year period. As mentioned above, some empirical evidence supports the hypothesis of a permanent impact of tax reform on the rate of growth of output—not just the output level—indicating that the growth benefit of reductions in fiscal pressure could be stronger than assumed in the simulations carried out below.

C. Two Illustrative Fiscal Scenarios

85. This section develops two scenarios that illustrate alternative fiscal strategies over the medium and longer term. Both scenarios span the period through 2045. The first scenario, which will be called a “baseline” scenario, is characterized by a moderate reduction in the revenue-to-GDP ratio in the coming years combined with cuts in primary expenditures, which allow for a gradual reduction in the debt-to-GDP ratio. The second scenario, which will be called a “reform” scenario, is characterized by early ambitious reform measures on the expenditure side, which allow for a faster reduction in the debt stock combined with a more decisive reduction in the revenue ratio; the more decisive expenditure and tax cuts also are assumed to pay off in the form of a higher (transitional) rate of growth of real GDP than under the baseline scenario.

Baseline scenario

86. It should be noted at the outset that the baseline scenario is not a “current policies” scenario, as it incorporates the assumption of a further moderate intensification of reform effort. It outlines a strategy that entails a moderate reduction in the revenue ratio and in the debt-to-GDP ratio over the medium-term, broadly in line with Italy’s Stability Program through 2003. On the expenditure side, some expenditure-reducing measures are assumed to be adopted in the education sector, to capitalize on the decline in the number of students, brought about by population aging.

Macroeconomic assumption

87. This scenario assumes that, after the current cyclical upturn has run its course, labor productivity remains at 1.2 percent per year—as assumed in the baseline pension projections scenario developed by the staff in last year’s background studies.45 This is projected to result in average real GDP growth of 0.5 percent per year over 2006-45, as population aging causes a decline in employment. Inflation, currently projected somewhat above 2 percent in 2000, is assumed to decline to 1.5 percent by 2002 and remain constant thereafter. The real interest rate, which is projected to increase to about 4½ percent by 2005, is assumed to gradually decline thereafter and stabilize at 3½ percent.46

88. As a result of population aging, pension and health expenditures are projected to increase markedly. Concerning pension outlays, the baseline scenario embodies the central projections in last year’s background study, which entail an increase in pension outlays by some 4 percent of GDP between 2000 and 2035. As for health expenditures, they are assumed to increase by 2½ percent of GDP over the same period (along the lines discussed above).

The fiscal strategy

89. Table 1 summarizes the fiscal strategy under the baseline scenario. The primary surplus is maintained constant at its current level of about 5 percent of GDP until overall balance is reached in 2006. This path, which is broadly consistent with the authorities’ medium-term strategy,47 corresponds to a gradual decline in the structural primary surplus from 6½ percent in 1999 to 5 percent in 2006.48 Once overall balance is reached, the budget is kept in balance. On the revenue side, the baseline scenario assumes that the revenue-to-GDP ratio is allowed to decline gradually by 2 percentage points over a four-year period, from 46.9 percent in 1999 to 44.9 percent in 2003, as envisaged under the authorities’ medium-term strategy.

Table 1.

Italy: Fiscal Developments Under the Baseline Scenario

article image
Source: Fund Staff calculations.

90. On the expenditure side, the decline in interest payments, which has facilitated the fiscal consolidation of recent years, continues, but at a much more moderate pace, as much of the benefits of EMU accession in terms of reduction of the risk premium have already been reaped. The path for the overall balance outlined above allows interest payments to come down gradually from 6½ percent of GDP in 2000 to 4½ percent of GDP by 201.0, and to just over 2 percent of GDP by 2045. Capital expenditures would decline somewhat, from 4 percent of GDP in 2000 to 3½ percent of GDP by 2003, as envisioned in the authorities’ program.

91. By around 2010, the budgetary impact of population aging begins to be felt more decisively. Pension and health expenditures, which by 2010 are already higher by over 1 percent of GDP than in 2000, are projected to increase by a further 2½ percent of GDP by 2020, and by almost an additional 3 percent of GDP between 2020 and 2035. This sharp increase in age-related entitlements, which quickly outpaces the yield of the endogenous decline in interest payments, poses a serious policy dilemma. Under this scenario, the stock of public debt is still some 80 percent of GDP in 2010, thereby limiting the possibility of accommodating the increase in health and pension outlays through a relaxation in the fiscal position (see below).49 In this scenario, it is assumed that only a small fraction of this additional budgetary pressure is absorbed within the expenditure side itself, more precisely through a moderate reform of the education sector. The reduction in the number of students, which also ensues from population aging, should allow for a reduction in education expenditures (estimated to be about 1 percent of GDP between 2010 and 2035).

92. In order to maintain budget balance, therefore, revenues will need to increase in line with the portion of the rise in health and pension outlays that cannot be compensated by savings on interest and education expenditures, The required increase in the fiscal pressure would be sizable: by 2035, the revenue-to-GDP ratio would be some 3½ percentage points above its 2005 level. Figure 1 shows the evolution of general government’s expenditure, revenue, and overall balance over the period, and clearly highlights the sharp increase in the revenue-to-GDP ratio between 2005 and 2035. It is important to note that this increase in the tax burden would take place as GDP growth slows down due to population aging—given the underlying decline in employment, the rise in the per capita tax burden on employed persons would be proportionately higher still than the increase in the revenue ratio.

Figure 1.
Figure 1.

Italy: General Government Revenues, Expenditures, and Balance under the Baseline Scenario, 2000-2045

(In percent of GDP)

Citation: IMF Staff Country Reports 2000, 082; 10.5089/9781451819793.002.A002

Source: Fund staff calculations.

Higher deficit scenarios

93. To what extent could the increase in revenues be limited by allowing the overall balance to deteriorate? The room for maneuver is circumscribed by the need—under the Stability and Growth Pact—to prevent the overall deficit from rising above 3 percent of GDP (including in case of a cyclical downturn) and to continue reducing the debt-to-GDP ratio at a satisfactory pace (until this ratio has fallen well below 60 percent). Figure 2 shows the time path of the debt-to-GDP ratio and the revenue ratio under different “constant deficit rules”: Starting from 2006, the overall balance is maintained at a constant level as in the baseline, but this constant level is alternatively a deficit of 1, 1.5, or 2 percent of GDP. The top panel of Figure 2 clearly shows that a deficit of 1.5 percent of GDP would already be deemed “excessive,” as it would imply that the debt-to-GDP ratio would start rising again by around 2025, when the debt stock is still above 80 percent of GDP. The bottom panel of Figure 2, moreover, highlights an additional drawback of these alternatives: under any of these deficit rules, the eventual increase in the revenue ratio would be even higher than in the baseline. The reason is that maintaining a higher deficit over this long horizon would slow down debt reduction, and backfire in the form of higher interest payments, eventually requiring a rise in the revenue burden.

Figure 2.
Figure 2.

Italy: Debt and Revenue Dynamics under Different Deficit Rules in the Baseline Scenario, 2000-2045

(In percent of GDP)

Citation: IMF Staff Country Reports 2000, 082; 10.5089/9781451819793.002.A002

Source: Fund staff calculations.

94. Moving to a higher deficit relatively early—by 2006 in the scenarios in Figure 2—is clearly self-defeating, as it would lead to an even slower decline in the debt ratio accompanied by an even higher increase in fiscal pressure. An alternative strategy would keep the budget in balance for a more extended period before the overall balance is allowed to deteriorate in response to the increase in expenditures. Given the constraints under the Stability and Growth Pact, however, a strategy of this kind would not be very effective in limiting the increase in the revenue ratio.50

Reform scenario

95. A second scenario outlines a significantly more ambitious reform strategy, entailing a faster reduction in both the stock of public debt and the fiscal pressure. This more ambitious strategy would yield more pronounced savings on interest payments; and, helped by a lower tax burden, a higher rate of real GDP growth. It requires, however, important expenditure measures—over and above those assumed in the baseline scenario—some of which would necessarily involve tackling key expenditure areas such as pensions, health, and public sector employment.

Macroeconomic assumptions

96. Real GDP growth is assumed to be the same as under the baseline scenario over the WEO projection horizon to 2003. Thereafter, economic growth is assumed to respond positively to the lower level of fiscal pressure compared to the baseline scenario. More precisely, it is assumed that a reduction in the revenue ratio by 1 percentage point compared to the baseline scenario would temporarily boost real GDP by 0.2 percentagé points, with the effect phased gradually over a five-year period. This would result in an average annual real rate of growth of GDP that is 0.03 percentage points higher over 2006-2045 than under the baseline scenario. The assumptions for the real interest rate and the inflation rate are the same as in the baseline scenario. The underlying increase in pension and health outlays brought about by population aging is assumed to be the same as under the baseline scenario, but it is assumed to be partially offset by reform measures in both sectors (see below).

The fiscal strategy

97. The fiscal strategy underlying the reform scenario is summarized in Table 2, and the path for general government revenue, expenditure, and overall balance is shown in Figure 3. This scenario assumes that the structural primary surplus is maintained at the level projected for 2000—that is, 6¼ percent of GDP—throughout the present decade, leading to an overall surplus of 2½ percent of GDP by 2010. The stronger fiscal position over the medium term allows for a faster decline in the stock of public debt, which would drop below 60 percent of GDP after 2010.

Table 2.

Italy: Fiscal Developments Under the Reform Scenario

article image
Source: Fund Staff calculations.
Figure 3.
Figure 3.

Italy: General Government Revenues, Expenditures, and Balance under the Reform Scenario, 2000-2045

(In percent of GDP)

Citation: IMF Staff Country Reports 2000, 082; 10.5089/9781451819793.002.A002

Source: Fund staff calculations.

98. The faster decline in the debt ratio is mirrored in interest payments, which fall from 6½ percent of GDP in 2000 to 3½ percent of GDP in 2010 and to less than 2 percent of GDP in 2045. As in the baseline scenario, capital expenditures would decline somewhat, from 4 percent of GDP in 2000 to 3½ percent of GDP by 2003, as envisioned in the authorities’ program.

99. On the revenue side, over 2000-2003 the same strategy as under the baseline reform scenario is followed, with a reduction in the revenue ratio by 2 percent of GDP over the four- year period. The reduction in the fiscal burden is, however, continued beyond 2003: through 2010, with priority being given to maintaining a sizable primary structural surplus, the reduction in the revenue ratio is limited to a further ¾ of 1 percent of GDP; after 2010, once the debt-to-GDP ratio has been brought below 60 percent, the reduction in the fiscal burden is accelerated. By 2020, the revenue ratio is 5 percent of GDP lower than in 1999. This reduction in the revenue ratio should be brought about in a way that secures a structural improvement in the tax system, reducing its distortionary impact—in particular, given Italy’s starting position, the focus should be on reducing the high tax burden on labor (see SM/88/00, 5/15/00).

100. As mentioned above, meeting the dual objective of a stronger fiscal position and a more decisive reduction in the fiscal burden requires up-front action on the expenditure side. It is therefore assumed that measures to contain expenditures are implemented at an early stage in key sectors, including pension and health care.51 Thanks to these reform measures, the increase in pension and health expenditures between 2000 and 2010 is limited to ¾ of 1 percent of GDP. Important cost-cutting measures are also assumed to be carried out in other sectors, with primary expenditures other than health and pensions declining by nearly 4 percent of GDP between 2000 and 2010, compared to 1¼ percent of GDP under the baseline scenario.

101. As under the baseline scenario, it is by 2010 that population aging begins to impose significant pressure on the budget, with pension and health outlays projected to increase by nearly 2 percent of GDP between 2010 and 2020, and an additional 1¾ percent of GDP between 2020 and 2035. The cumulative increase is some 2½ percent of GDP less than under the baseline scenario, as up-front reform measures reduce the long-run costs of age-related entitlements. Under the reform scenario, moreover, the economy is now characterized by a stronger initial fiscal position (an overall surplus of 2½ percent of GDP compared to budget balance under the baseline scenario), a lower stock of public debt (60 percent of GDP compared to 80 percent of GDP), and a lower level of public expenditures.

102. The more ambitious strategy would, therefore, have created sufficient room to accommodate the remaining budgetary pressure from age-related entitlements through a gradual weakening of the fiscal position. From a surplus of 2½ percent of GDP in 2010, it is allowed to swing to a deficit of about 1 percent of GDP by 2035, when the aging crisis reaches its peak.52 The stock of public debt would continue to decline through 2030, when it reaches some 30 percent of GDP. Subsequently, it would rise by some 6 percent of GDP as the transitory aging shock is absorbed, before stabilizing at the end of the projection period.

The impact of higher productivity growth

103. Fiscal projections over such a long time frame tend to be especially sensitive to changes in the underlying assumptions. In particular, it is important to evaluate the sensitivity of the two scenarios to different real GDP growth assumptions. Figure 4 summarizes the behavior of the key fiscal variables under the assumption that labor productivity grows at an annual rate of 1¾ percent, rather than 1.2 percent, resulting in average real GDP growth of 1 percent per year over 2006-45 (see Section II.A above for further details on the productivity assumption). The higher real GDP growth rate by itself eases the fiscal burden of population aging, in particular by limiting the increase in pension expenditures as a percent of GDP. Moreover, for a given real rate of interest, a higher real GDP growth rate allows for a faster decline in the debt-to-GDP ratio and, correspondingly, in interest payments.

Figure 4.
Figure 4.

Italy: Revenues and Expenditures Under Different Scenarios and Productivity Growth Assumptions, 2000-2045

Citation: IMF Staff Country Reports 2000, 082; 10.5089/9781451819793.002.A002

Source: Fund staff calculations.

104. Under the baseline scenario, assuming that the same fiscal strategy outlined above is maintained, the debt-to-GDP ratio would still be only marginally below 80 percent by 2010. The increase in age-related entitlements, however, would be contained to 4½ percent between 2000 and 2035. Together with the somewhat faster decline in interest payments, this would limit the rise in the revenue ratio necessary to maintain budget balance to 1½ percent of GDP between 2005 and 2035.

105. Under the reform scenario, the dividend from higher economic growth could be allocated to a more sizable reduction in fiscal pressure. Assuming, for example, that the path for the overall fiscal balance were the same as under the lower productivity growth assumption, the revenue-to GDP ratio could be lowered by as much as 6½ percentage points compared to the 1999 level.

D. Conclusions

106. Under the baseline assumption for productivity growth, a comparison between the reform and baseline policy scenarios underscores three key points. First, maintaining a stronger fiscal position over the medium term would later on allow to absorb the transitory shock of population aging through a temporary increase in the debt stock, rather than through a marked increase in fiscal pressure. Second, decisive and early reductions in primary expenditures, together with the lower interest payments stemming from a faster decline in the debt stock, would create the room for a steady decline in the fiscal burden, with a positive effect on the productive potential of the economy. Third, the reform strategy would reduce fiscal vulnerability more rapidly, notably as it relates to sustained increases in interest rates.

107. Figure 5 compares the paths of the debt-to-GDP ratio and the revenue ratio (as well as primary current expenditures as a percent of GDP) under the two scenarios. It should be noted that the effects of the required increase in the fiscal pressure under the baseline scenario would have to be borne by a shrinking working population—creating the risk that the disincentive effect of higher taxes would exacerbate the decline in the labor force brought about by population aging. A traditional tax smoothing argument would also lead to prefer the temporary increase in public debt under the reform scenario as a response to a temporary expenditure shock (see, e.g., Barro, 1979)—an option that is, under the baseline scenario, not viable within the constraints of the Stability and Growth Pact.

Figure 5.
Figure 5.

Italy: Public Debt, Revenues, and Primary Current Expenditures, 2000-2045

(In percent of GDP)

Citation: IMF Staff Country Reports 2000, 082; 10.5089/9781451819793.002.A002

Source: Fund staff calculations.

108. On the other hand, the pursuit of an ambitious reform strategy would not be costless and requires sizable reductions in primary expenditures. Some measures could be aimed at limiting the budgetary impact of population aging through reforms of the pension and health care system. In the case of pension reform, key measures that could be considered include shortening the transition to the new (Dini) regime, and increasing the effective retirement age.53 Issues pertaining to the health care system are discussed in Chapter III.

109. Table 3 provides a more detailed picture of the expenditure compression required under the two scenarios over the first ten years of the projection period. For each scenario, the table first separates out the two expenditure categories that are most directly affected by population aging, pension and health. The remainder of primary current spending is then broken down into four major categories: education, wages and salaries, goods and services, and other. Since the table combines a functional classification with an economic classification, care must be taken to avoid double-counting. Outlays on wages and salaries, and goods and services, therefore, are defined to exclude education and health care. For each one of these expenditure categories, and under each scenario, the table provides the value in percent of GDP and the real percentage changes.

Table 3.

Italy: Primary Current Expenditures Under the Two Scenarios

article image
Source: Fund Staff calculations.

Excluding health and education.

110. Although the reform scenario shows a slower increase in both pension and health care outlays, it would still require, relative to the baseline scenario, further compression in other expenditures. Under the reform scenario, primary current expenditures (other than on pensions and health care) would have to decline in real terms by ½ of 1 percent per annum on average over 2000-05, and remain broadly unchanged in real terms over the following five years (as compared to increases in real terms by 1¼ percent and 1½ percent on average, respectively, over the corresponding two five-year periods under the baseline scenario).

  • Alesina, A. and D. Rodrik, 1994, “Distributive Politics and Economic Growth,” Quarterly Journal of Economics, Vol. 109.

  • Barro, R., 1979, “On the Determination of the Public Debt,” Jornal of Political Economy, Vol. 87.

  • Barro, R., and J. Lee, 1994, “Sources of Economic Growth,” Carnegie-Rochester Conference Series on Public Policy.

  • Barro, R. and X. Sala-i-Martin, 1995, Economic Growth (New York: McGraw-Hill).

  • Daveri, F. and G. Tabellini, 1997, Unemployment, Growth and Taxation in Industrial Countries.

  • Easterly, W. and S. Rebelo, 1993, “Fiscal Policy and Economic Growth: An Empirical Investigation,” Journal of Monetary Economics: 32, pp. 417-458.

    • Search Google Scholar
    • Export Citation
  • Engen, E. and J. Skinner, 1996, “Taxation and Economic Growth” NBER Working Paper No. 5826 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Engen, E. and J. Skinner, 1992, “Fiscal Policy and Economic Growth,” NBER Working Paper No. 4223 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Ford, R. and D. Laxton, 1999, “World Public Debt and Real Interest Rates,” Oxford Review of Economic Policy: 15.

  • Habermeier, K., and F. Lenseigne, 1998, “French Public Finance: Modeling Long-Term Prospects and Reform Options,” IMF Working Paper 98/12 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Italian Treasury, 1997, “Sanità, scuola e pensioni. Le nuove previsioni basate sugli scenari demografici ISTAT,” Quaderno Monografico n. 13.

    • Search Google Scholar
    • Export Citation
  • Mendoza, E., G. M. Milesi-Ferretti, and P. Asea, 1997, “On the Ineffectiveness of Tax Policy in Altering Long-Run Growth: Harberger’s Superneutrality Conjecture,” Journal of Public Economics: 66 (October).

    • Search Google Scholar
    • Export Citation
  • OECD, 2000, 1999-2000 Annual Review-Italy (Paris).

  • Perotti, R., 1991, “Income Distribution, Politics, and Growth,” American Economic Review, Papers and Proceedings, Vol. 82.

  • Persson, T. and G. Tabellini, 1991, “Is Inequality Harmful for Growth?” American Economic Review, Vol. 84.

  • Sala-i-Martin, X., 1996, “Transfers, Social Safety Nets, and Economic Growth,” IMF Working Paper 96/40 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Tanzi, V., and H. Zee, 1998, “Taxation and the Household Saving Rate: Evidence from OECD Countries,” IMF Working Paper 98/36 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Tanzi, V., and H. Zee, 1997, “Fiscal Policy and Long-Run Growth,” Staff Papers, International Monetary Fund, Vol. 44 (June).

  • World Bank, 1994, Averting the Old Age Crisis (Washington).

33

Prepared by Marco Annunziata.

34

Based on World Bank demographic projections. The old-age dependency ratio is defined as the ratio of the population aged 65 years and older to the population of working age (20-64).

35

In 1994, pension expenditures in Italy were 2 percent of GDP higher than in France and Sweden, and some 3 percent of GDP higher than in Germany. See World Bank (1994).

36

See Chapter I of SM/99/115, 5/20/99, and references therein.

39

Standardized per capita consumption is defined as the per capita consumption of health care services that would prevail keeping unchanged the gender and age structure of the population as well as the type of health services provided.

40

This would occur both because of the increase in the proportion of the population older than the statutory retirement age, and because of the lower participation rates prevailing in the older (but still below the retirement age) age brackets.

41

See, for example, Tanzi and Zee (1997), and Engen and Skinner (1996), Tanzi and Zee (1998) provide evidence of a strong negative impact of taxes (in particular income taxes) on household savings.

42

See Engen and Skinner (1996) and the references cited therein.

44

See Tanzi and Zee (1997) for a review of the arguments and of the literature. Several studies also focus on the impact of income distribution on growth, through the redistribution role of fiscal policy—see for example Alesina and Rodrik (1994), Persson and Tabellini (1994), and Perotti (1992).

45

See Chapter I of SM/99/115, 5/20/99, for further details.

46

This is based on the assumption that real interest rates in Italy over the long run will be tied to the world interest rate, which has been estimated at around 3.5 percent since the early 1980s. For a discussion of the determinants of the world real interest rate see Ford and Laxton (1999).

47

As outlined in Italy’s Stability Program for 2000-03, which however envisages that approximate overall balance would be achieved by 2003.

48

As the output gap is projected to narrow from 3.1 percent in 1999 to 0.7 percent in 2003 and to be eliminated in 2005.

49

Under the Maastricht Treaty, a country’s debt ratio should be “sufficiently diminishing and approaching the reference value at a satisfactory pace” as long as the ratio itself is higher than the reference value of 60 percent of GDP.

50

For example, maintaining budget balance through 2015 and then allowing the overall balance to deteriorate up a maximum deficit of 1 percent of GDP would halt the decline in the debt ratio while limiting the increase in the revenue ratio by only ½ percent of GDP.

51

Last year’s Selected Issues paper (SM/99/115, 5/20/99) outlines key reform measures that could be implemented in the pension system.

52

The deficit would therefore always remain smaller than the limit (1½ percent of GDP) calculated by the European Commission as the level that would allow Italy to accommodate historically observed cyclical fluctuations without breaching the 3 percent of GDP Maastricht limit for the general government deficit.

53

Increasing the effective retirement age could be achieved in different ways, for example, through an increase in the statutory retirement age, through an accelerated and across-the- board tightening of the requirements for seniority pensions, through a decline in the number of disability pensions, and through a strengthening of disincentives to early retirement—including modifying the “transformation coefficient” used to compute the initial pension benefit at retirement. See SM/99/115, 5/20/99 for more details.

Italy: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Italy: General Government Revenues, Expenditures, and Balance under the Baseline Scenario, 2000-2045

    (In percent of GDP)

  • View in gallery

    Italy: Debt and Revenue Dynamics under Different Deficit Rules in the Baseline Scenario, 2000-2045

    (In percent of GDP)

  • View in gallery

    Italy: General Government Revenues, Expenditures, and Balance under the Reform Scenario, 2000-2045

    (In percent of GDP)

  • View in gallery

    Italy: Revenues and Expenditures Under Different Scenarios and Productivity Growth Assumptions, 2000-2045

  • View in gallery

    Italy: Public Debt, Revenues, and Primary Current Expenditures, 2000-2045

    (In percent of GDP)