Chapter

III. Identifying Fiscal Consequences

Author(s):
Sheetal Chand, and Albert Jaeger
Published Date:
December 1996
Share
  • ShareShare
Show Summary Details

This section outlines the analytical and empirical approach adopted for assessing the long-run fiscal impact of a public pension system.15 The section also presents projections of the fiscal consequences of preserving current public pension arrangements.

The Approach

As virtually all of the major industrial countries operate some version of a public defined-benefit pension system (mostly on a PAYG basis), either in isolation or in combination with other pension arrangements, an approach to assessing the fiscal implications of such systems requires a modeling of the basic features of the public pension scheme for each country and then undertaking simulations.

In a public pension, defined-benefit system, pensions to the retired elderly are almost wholly financed from the contributions paid by the working population, usually expressed as a proportion of total wages, as well as from budgetary transfers. By definition, in a full-fledged PAYG system, the two determinants of the equilibrium average contribution rate needed to provide for pension payments are (1) the support ratio, defined as the number of contributing workers per pensioner, and (2) the average replacement rate, defined as the proportion of the average wage that is replaced by the average pension.16 Obviously, the fewer workers there are relative to the number of retired persons, the higher is the average contribution rate needed to cover the overall cost of pension outlays. Similarly, the more generous are pension benefits, as measured by a higher replacement rate, the higher contribution rates will have to be.

Demographic factors and employment conditions, together with the prescribed retirement ages, determine the overall size of the support ratio. To illustrate the effect of demographic factors, consider first a population with an unchanging age structure. For this benchmark case, the flow of payments from the working population would, at an appropriate predefined contribution rate, be just sufficient to cover payments. Continuing with the illustration, assuming that three fifths of the population fall in the working-age group, one fifth are children, and the remaining one fifth are retirees, the stable support ratio would be three workers to one retiree. If the replacement ratio is set at one half of the average wage, a contribution rate of one sixth of the wage— the equilibrium contribution rate—would suffice to cover benefit payments.17 Assuming that the share of labor amounted to two thirds of GDP, the flow of contributions as a percentage of GDP would then amount to just over 11 percent, which is the amount needed to cover the consumption needs of the pensioners at the assumed replacement level.

Next suppose an aging population, with the share of the elderly increasing. This will decrease the support ratio, and, if the replacement rate is kept unchanged, necessitate an increase in the contribution rate in order to finance pension outlays. However, the approach adopted in this paper also allows for changes in the replacement rate as a consequence of aging itself, adjustments in the benefit or contribution formulas, including the age of retirement, and movements in average earnings. With regard to aging, the methodology developed is fairly comprehensive, in that it attempts to take a fuller account of demographic developments, tracing effects through different age cohorts of the earnings history and other characteristics that may be unique to a particular age group. The average replacement rate over the projection period will thus vary, reflecting the changing income characteristics of the cohorts.

A full general-equilibrium approach is, however, not adopted. In particular, the projection model does not endogenize the labor force’s reaction, or the savings response of the private sector, to changes in public pension arrangements. Such responses could be significant; for example, an increase in contribution rates could be sufficiently distortive so as to lower the supply of labor, thereby increasing the system dependency ratio; in a similar vein, increased pension expenditure could crowd out private capital accumulation and make it more difficult for the economy to support a larger aged population.

A further important assumption that may affect the results of the study is the choice of the projection horizon. The present study uses 2050 as the cutoff year for the projections, roughly comparable to projection horizons used in other studies.18 The selected projection horizon reflects a somewhat arbitrary compromise between the objective of capturing the full fiscal ramifications of the transition to a new “population steady state” and the increasing margin of uncertainty that attaches to projections that reach farther and farther into the future. The most likely effect of extending the projection horizon beyond 2050 would be to enlarge the estimated net pension liabilities, as all countries except the United Kingdom are projected to have primary pension fund deficits in 2050.

To make the analysis more tractable, certain plausible assumptions are imposed regarding the rate of growth of the labor force (and the implied participation and employment rates) taking into account preannounced increases in retirement ages, the rate of capital accumulation and the implied rate of private saving, and the rate of total factor productivity growth, and hence the underlying nature and rates of technological progress. This approach has the advantage of not relegating the determination of key macroeconomic variables, whose interaction with pension arrangements is, to say the least, controversial, to a “general equilibrium black box.” Moreover, the assumptions made can be easily replaced by others, which can then be used to assess the sensitivity of the results, as is done subsequently.

Table 5 lists the main macroeconomic projections. Among the features of note in the table are: (1) for most countries, employment growth is projected to be negative as a consequence of aging, and even for the United States and Canada, employment growth, which is boosted by immigration, is projected to be low; (2) these employment projections constrain real GDP growth to be low, despite assuming a relatively high rate of multi-factor productivity growth and some capital deepening;19 and (3) the real interest rate is projected to be substantially higher than the projected real GDP growth rate, indicating that the projected long-run growth path is dynamically efficient,20 but also that fiscal accounts will have to exhibit sizable primary surpluses for public debt ratios to be sustainable.

Table 5.Projections of Averages of Macroeconomic Variables, 1995–2050(In percent)
Employment GrowthReal GDP GrowthReal Interest RateInflation Rate
Major industrial countries–0.11.43.53.0
United States0.31.73.53.0
Japan–0.61.13.53.0
Germany–0.81.13.53.0
France–0.21.33.53.0
Italy–0.60.83.53.0
United Kingdom–0.11.43.53.0
Canada0.11.63.53.0
Sweden1.33.53.0
Source: IMF staff estimates.
Source: IMF staff estimates.

The implications of a PAYG scheme can be assessed by first examining how required contribution rates will need to evolve over the projected period to provide for benefits. This is shown by movements of the “equilibrium rate” in Table 6, or the proportion of the average wage that has to be contributed to just cover pension benefits as determined by replacement rates and support ratios. The table shows substantial increases in the equilibrium contribution rate for all countries except the United Kingdom and Sweden. For the latter countries, the declining equilibrium contribution rates reflect the assumed indexation of all flat-rate pension benefits (for new and pre-existing pensioners) to CPI inflation and a less pronounced deterioration in the support ratios.21 While Canada also indexes flat-rate benefits to CPI inflation, the decline in the support ratio more than offsets the projected fall in the replacement rate, which causes the equilibrium contribution rate to continue rising.

Table 6.Projections of Pension Replacement Rates, Support Ratios, and Contribution Rates
19952000201020302050
Major industrial countries
Replacement rate37.537.135.435.835.1
Support ratio3.23.23.02.01.8
Equilibrium rate13.714.113.821.522.6
Projected rate13.413.413.413.413.4
United States
Replacement rate38.537.735.136.836.6
Support ratio4.24.34.12.52.3
Equilibrium rate9.18.88.615.015.9
Projected rate9.79.79.79.79.7
Japan
Replacement rate19.619.819.219.519.3
Support ratio2.62.32.11.81.5
Equilibrium rate7.78.79.310.812.7
Projected rate5.65.65.65.65.6
Germany
Replacement rate52.051.049.048.848.7
Support ratio2.32.12.01.21.2
Equilibrium rate22.625.024.741.141.6
Projected rate22.822.822.922.922.9
France
Replacement rate60.159.459.559.859.5
Support ratio2.52.62.41.61.4
Equilibrium rate24.323.224.437.741.2
Projected rate23.423.423.423.423.4
Italy
Replacement rate53.955.855.653.750.8
Support ratio1.31.21.40.90.7
Equilibrium rate42.645.540.461.968.2
Projected rate42.642.642.642.642.6
United Kingdom
Replacement rate17.517.416.814.410.6
Support ratio2.72.72.52.12.1
Equilibrium rate6.46.46.86.95.0
Projected rate6.26.26.26.26.2
Canada
Replacement rate29.228.425.622.620.1
Support ratio3.63.52.91.71.6
Equilibrium rate8.18.28.913.712.9
Projected rate5.75.75.75.96.0
Sweden
Replacement rate39.038.334.329.023.8
Support ratio2.62.72.51.81.9
Equilibrium rate14.814.114.015.912.8
Projected rate12.312.312.312.312.3
Source: IMF staff estimates.Notes: The replacement rate is defined as the average pension benefit as a percent of average gross wage. The support ratio is defined as the ratio of contributors to beneficiaries. The equilibrium rate is the contribution rate including net budget transfers (as a percent of wage bill) that maintains year-by-year financial balance of the pension system. The projected rate is the projected contribution rate including net budgetary transfers (as a percent of wage bill).
Source: IMF staff estimates.Notes: The replacement rate is defined as the average pension benefit as a percent of average gross wage. The support ratio is defined as the ratio of contributors to beneficiaries. The equilibrium rate is the contribution rate including net budget transfers (as a percent of wage bill) that maintains year-by-year financial balance of the pension system. The projected rate is the projected contribution rate including net budgetary transfers (as a percent of wage bill).

To assume that equilibrating adjustments will always take the form of further increases in contribution rates is problematic, as considerable resistance is likely to be encountered from contributing generations to further hikes in contribution rates, especially when these are already high. Keeping contribution rates and associated budgetary support constant at current levels could, however, lead to highly adverse future fiscal circumstances. To bring this aspect out fully in a manner that also sheds light on the possible need for pension reform, the approach pursued here is to keep contribution rates and any net budget transfers at their present levels—the “projected rate” in Table 6—and to then examine the potential fiscal consequences. This is undertaken by estimating the implied accumulation of unfunded liabilities.

The approach proceeds by first constructing a baseline using the “projected-contribution rate,” and then asks what the fiscal implications for the public pension fund are. Depending on the time horizon of the analysis and the demographics, the result could be a persistent deficit that results in a growing net liability position. Through simulations, the analytical framework can then be used to answer questions involving parametric reforms, such as by how much the contribution rate would have to rise to eliminate the increased net liability position, or what the potential impact would be of alternative policy choices, such as adjusting benefits and extending the retirement ages.

Evaluating the fiscal consequences of a change to a fully funded system raises a number of issues. To determine the fiscal costs, the study seeks to identify initially the size of the unfunded liabilities associated with existing and prospective pensioners in the current systems. The sizes involved will of course depend on the arrangements that are made as to how the transition from a defined-benefit, PAYG system takes place. The two polar alternative scenarios considered here are (1) the “sudden” transition, where the switch to a fully funded, defined-contribution system is made at one point in time and applies to all current pensioners, who are “cashed out,” and future beneficiaries; and (2) the “gradual” transition, where only new entrants to the labor force after the selected date of conversion become members of a fully funded, defined-contribution scheme.

In the latter case, the transition may take many years depending on the survival rates for beneficiaries still subject to the defined-benefit, PAYG system. Regardless of which approach is adopted, and the variants in between, the effect of taking the decision to move to a fully funded, defined-contribution system is to make explicit the unfunded liabilities of the PAYG system. Recognizing the full extent of these liabilities as a stock at one point in time raises more immediately the issue of how this debt is to be dealt with.

Projected Fiscal Positions Under Present Arrangements

The methodology described above is employed to assess how aging affects public pension expenditures under present arrangements that include known future changes in the structure of benefits. Table 7 and Chart 2 show projected levels of pension expenditures as a percentage of GDP. The table and Chart 3 present the balance in each of these years between the flows of expenditures and revenues (including interest on net assets) from contributions, including any budgetary support provided at the levels established at the start of the projection period as a fixed percentage of GDP. As shown in Table 7 and Chart 4, these annual flows will have implications for the net asset position of the (implicit) reserve funds associated with the pension arrangements. For the United States, Japan, and Sweden, these are explicit reserve funds; for the other countries, these reserves are implicitly attributed by the assumed methodology.

Table 7.Baseline Projections of Pension Expenditure, Balances, and Net Asset Positions of Public Pension Funds
Country19952000201020302050
Major industrial countries
Pension expenditure6.76.97.010.711.4
Balance0.50.2–0.3–6.6–15.5
Net assets8.35.6–1.1–61.6–209.7
United States
Pension expenditure4.44.34.27.47.7
Balance0.81.11.7–2.2–7.2
Net assets7.09.517.23.0–66.7
Japan
Pension expenditure5.76.57.58.910.7
Balance1.1–0.4–4.1–10.9–23.4
Net assets26.513.9–17.1–144–399.2
Germany
Pension expenditure10.011.111.018.418.7
Balance0.2–0.9–1.3–14.9–34.7
Net assets1.1–0.1–8.8–115.6–431.3
France
Pension expenditure12.512.012.619.421.3
Balance–0.5–0.4–13.2–31.5
Net assets–0.5–1.20.6–100.5–369.6
Italy
Pension expenditure16.017.115.223.325.7
Balance–1.1–1.1–8.8–18.4
Net assets–16.9–29.9–186.8–338.2
United Kingdom
Pension expenditure4.44.34.64.73.4
Balance–0.2–0.2–0.7–1.1–0.2
Net assets–0.2–4.3–130.5–14.5
Canada
Pension expenditure4.44.54.97.57.1
Balance–0.2–0.5–1.4–7.6–14.7
Net assets7.04.0–5.1–67.3–188.8
Sweden
Pension expenditure8.58.28.19.27.4
Balance1.30.40.2–3.0–3.8
Net assets25.821.918.3–16.3–56.7
Source: IMF staff estimates.
Source: IMF staff estimates.

Chart 2.Projected Pension Expenditure of Public Pension Funds

(In percent of GDP)

Source: IMF staff estimates.

Note: Projected public pension expenditure under present pension arrangements (see also Table 7) taking account of legislated future increases in statutory retirement ages and based on projections of macroeconomic variables reported in Table 5.

Chart 3.Projected Balances of Public Pension Funds

(In percent of GDP)

Source: IMF staff estimates.

Note: Balance of public pension fund Is defined as difference between projected pension expenditure (Chart 2) and projected revenues from contributions, budgetary transfers, which are kept constant as a percent of GDP at 1995 level, and interest on net pension fund assets.

Chart 4.Projected Net Asset Positions of Public Pension Funds

(In percent of GDP)

Source: IMF staff estimates.

Note: These figures represent projected net assets of implicit pension reserve funds (see also Table 7).

The qualitative thrust of the results confirms the findings of previous studies such as Leibfritz and others (1995). All the listed countries, with the exception of the United Kingdom and Sweden, exhibit rising shares of public pension expenditures in GDP. Maintaining contribution rates at the projected levels results, again with the exception of these two countries, in a considerable worsening of public pension fund balances, with the deficits becoming larger as the projection period is lengthened. The problem is especially pronounced for Germany, Japan, France, and Italy, for whom, in addition to the adverse effects of aging, the progressive deterioration in the net asset positions underlying the public pension schemes is a major contributory factor.

Over the projection period, the deteriorating balances overturn a positive net asset position in 1995 for several countries, resulting in a growing stock of accumulated net liabilities. The need to meet the debt-service charges on the accumulated liabilities further aggravates the pension fund balances. While noteworthy for most countries in the sample, it is especially marked for Japan, Germany, France, and Italy, for whom the negative net asset position reaches a multiple of three to four times GDP in 2050. The United Kingdom and Sweden show the least deterioration because of relatively moderate decreases of support ratios and—as already indicated—flat-rate pension benefits of new and preexisting pensioners are indexed only to CPI inflation, while contributions are indexed to the rate of nominal wage growth. This latter approach represents a potent combination for preserving the financial integrity of a pension scheme, although at the expense of a major deterioration in the relative income position of pensioners.22

Under the assumed conditions, with the exception of the United States, France, and Sweden, all the listed countries would have a negative net asset position by 2010. Subsequently, the United States, France, and Sweden also begin to show negative net asset positions. In the year 2050, even though there is some reversal in elderly dependency ratios, the earlier buildup in the negative net asset positions becomes a dominant factor that, as noted earlier, contributes to a worsening of pension fund balances and thus a further deterioration in the net asset positions.

Sustainable Fiscal Stance and Projected Liabilities

The projected deterioration in pension fund balances under present arrangements and the associated increase in net liabilities aggravates the fiscal position of most major industrial countries. Table 8 shows the net public debt position of the major industrial countries and Sweden at the end of 1994, which ranges from about 33 percent of GDP for Japan to almost 113 percent of GDP for Italy. If account is taken of unfunded pension liabilities, the implied public debt position deteriorates for most of the countries considered. This is evident from column 2 of Table 8, which converts the net asset positions of the previous table to 1995 present values using the interest rates shown in Table 5. The estimated net pension liabilities range from a low of nearly 5 percent of GDP for the United Kingdom to about 110 percent of GDP for Japan, Germany, and France. When added to the end-1994 net public debt positions, the resulting combined net debt position is much worse, exceeding 100 percent of GDP for all countries except the United Kingdom and Sweden.

Table 8.Net Pension Liabilities and Sustainability of Fiscal Stance
Sustainable Primary Balance Required to
Net Public Debt at End, 19941Net Pension Liability, 1995-20502Combined Net Debt LiabilityPrimary Balance 19953Stabilize net public debt in 19954Stabilize net public debt and prevent buildup of pension debt5Adjustment Needed in Primary Balance for Fiscal Sustainability6
(1)(2)(3)(4)(5)(6)(7)
Major industrial countries57.260.0117.20.71.02.92.2
United States63.325.789.00.41.11.91.5
Japan33.2106.8140.0–0.20.33.63.8
Germany52.5110.7163.22.41.14.52.1
France42.4113.6156.0–0.30.74.04.3
Italy112.975.5188.43.32.14.61.3
United Kingdom37.74.642.30.40.70.80.4
Canada71.667.8139.40.22.74.74.5
Sweden54.520.474.9–5.10.11.06.1
Source: Davis (1993), Tables 3.1 and 3.4.

Adjusted for net assets of public pension fund at the end of 1994. Estimate of net public debt for Germany includes unification debt as of the end of 1994.

Net present value of difference between projected primary expenditure and revenue of public pension fund during 1995-2050, adjusted for net asset position of public pension systems at the end of 1994.

May 1995 WEO projections of structural primary balance of general government.

Primary balance required to stabilize net public debt in 1995.

Sustainable primary balance in column (5) plus contribution gap from column (4) of Table 9.

Difference between columns (6) and (4).

Source: Davis (1993), Tables 3.1 and 3.4.

Adjusted for net assets of public pension fund at the end of 1994. Estimate of net public debt for Germany includes unification debt as of the end of 1994.

Net present value of difference between projected primary expenditure and revenue of public pension fund during 1995-2050, adjusted for net asset position of public pension systems at the end of 1994.

May 1995 WEO projections of structural primary balance of general government.

Primary balance required to stabilize net public debt in 1995.

Sustainable primary balance in column (5) plus contribution gap from column (4) of Table 9.

Difference between columns (6) and (4).

If one judges the sustainability of a nation’s fiscal stance in terms of the conventional criterion of stabilizing net public debt at its current levels, it would appear that the present fiscal stances of Italy and Germany are sustainable: their projected primary balances in 1995 exceed the amounts needed to stabilize the ratio of net public debt (excluding pension liabilities) to GDP. Because assumed real interest rates are higher than projected real growth rates, primary fiscal balances would have to be positive to offset the growth in debt ratios induced by the interest rate. Those other countries for which the primary balances are negative will therefore have to engage in substantial fiscal consolidation just to stabilize their public debt ratios. This is seen by comparing the estimated sustainable primary balances shown in column 5 of Table 8 with their current levels. Sweden, for example, would have to improve its primary balance by an estimated 5.2 percent of GDP, while Canada would need to improve it by 2.5 percent of GDP.

However, if account is taken of the projected buildup in net pension debt, and the sustainability criterion is modified to include as well the prevention of any buildup of pension debts, primary balances would have to be even more positive, as is shown by the estimates in column 6. The full amount of the fiscal consolidation needed to both stabilize current net public debt positions and prevent any buildup of pension debts is indicated in the last column of the table. This shows that fiscal consolidation needs become much more stringent for most of the countries. For Germany and Italy, indications under the conventional criterion that their primary balances are adequate are reversed: improvements in primary balances of 2.1 percent of GDP for Germany and 1.3 percent of GDP for Italy are required. For the major industrial countries as a group, an improvement of 2.2 percent of GDP is needed to ensure fiscal sustainability. As a percent of GDP, the greatest need for fiscal consolidation is estimated to arise for Japan, France, and Canada at about 4 percent to 5 percent of GDP and Sweden at 6 percent of GDP. The United States and the United Kingdom manifest smaller needs, both because their current primary balance is close to the levels required to stabilize existing public debt ratios and because the projected buildup of pension debt under current pension arrangements is more modest.

The sought-after improvements in primary balances can be obtained in several ways through adjustments in different revenue and expenditure components. A convenient way of expressing the adjustments needed to prevent any buildup in pension-related debt is to express them as increases in contribution rates at the outset of the projection period—so-called “sustainable contribution rates.”23 Column 1 of Table 9 presents overall public pension contribution rates (defined as a percent of GDP), including net budget transfers, that are expected to prevail in 1995 for the major industrial countries and Sweden. The second column presents the average projected contribution rates ascribable to the entire projection period 1995-2050. By assumption, the average projected contribution rates do not differ from those shown in the first column, ranging from a low of 3.8 percent of GDP for Canada to a high of 16 percent of GDP for Italy.24 It is noteworthy that these contribution rates are clearly insufficient for the equilibration of the public pension schemes as is indicated by the sustainable contribution rates shown in column 3.

Table 9.Sustainable Contribution Rates and Contribution Gaps(In percent of GDP)
Projected Contribution Rate in 19951Projected Average Contribution Rate 1995-2050Sustainable Contribution Rate 1995-20502Contribution Gap3
(1)(2)(3)(4)
Major industrial countries6.56.58.31.8
United States4.74.75.50.8
Japan3.93.97.23.3
Germany10.310.313.73.4
France12.112.115.43.3
Italy16.016.018.52.5
United Kingdom4.24.24.30.1
Canada3.83.85.82.0
Sweden7.17.18.00.9
Source: IMF staff estimates.

Including net budget transfers.

The sustainable contribution rate is defined as the constant contribution rate over 1995-2050 that equalizes the net asset position in 2050 with the initial net asset position in 1995.

Defined as the difference between sustainable contribution rate and projected contribution rate in 1995.

Source: IMF staff estimates.

Including net budget transfers.

The sustainable contribution rate is defined as the constant contribution rate over 1995-2050 that equalizes the net asset position in 2050 with the initial net asset position in 1995.

Defined as the difference between sustainable contribution rate and projected contribution rate in 1995.

The sustainable contribution rate is for the most part substantially higher than the projected average rate, especially for Italy, where it is estimated to be 18.5 percent of GDP, Germany (13.7 percent of GDP), Japan (7.2 percent of GDP), and France (15.4 percent of GDP).25 Subtracting the projected contribution rate for 1995 from the sustainable rate yields estimates of the “contribution gap” that are shown in the fourth column of the table. There is virtually no gap for the United Kingdom and relatively small gaps for Sweden and the United States, as compared to the other countries listed in the table.

In terms of maintaining fiscal viability, alternative options are a once-and-for-all move to the sustainable contribution rate and year-by-year adjustments of contribution rates to equilibrate the pension fund. The alternative profiles of contribution rates (again as a percent of GDP) over the projection horizon in the selected industrial countries are brought out in Chart 5.26 The chart shows that year-by-year adjustments of contribution rates, while seemingly more attractive in that initial-year rates would be less than their sustainable levels and would aggravate the burden of high contribution rates in the outer years raising issues of efficiency and intertemporal equity. If increases in contribution rates are considered inevitable at some point, the efficiency gains from tax smoothing could, in principle, provide a powerful economic rationale for moving to sustainable contribution rates early on. However, from other quantitative studies, the efficiency gains of tax smoothing associated with earlier moves to sustainable contribution rates are likely to be small.27 By contrast, intertemporal equity considerations could provide a forceful argument for an early implementation of sustainable contribution rates as it would equally distribute the burden of paying pensions across generations.

Chart 5.Projected Public Pension Fund Contribution Rates

(In percent of GDP)
(In percent of GDP)

Source: IMF staff estimates.

Note: For the definitions of equilibrium and projected public pension fund contribution rates, see notes in Table 6. The sustainable contribution rate is defined as the constant contribution rate that equalizes the net asset position in 2050 with initial net assets in 1995.

Sensitivity Analysis

Using the earlier results as a reference point, it is useful to examine the sensitivity of net pension liabilities and contribution gaps to the real GDP growth rate and the real interest rate. Table 10 presents the results of assuming that the baseline rate of real GDP growth falls by 1.5 percentage points as a result of an equivalent decline in the rate of technical progress; the effect of a reduction in the real rate of interest to 2 percent (from 3.5 percent in the baseline) is also shown.

Table 10.Sensitivity of Estimated Net Pension Liabilities and Contribution Gaps to Real GDP Growth and Interest Rate Assumptions(In percent of GDP)
Effect on Baseline Projection of
Baseline ProjectionLower real GDP growth1Lower real interest rate2
Major industrial countries
Net pension liability65.1–6.939.9
Contribution gap1.80.20.5
United States
Net pension liability25.7–5.824.3
Contribution gap0.8–0.10.4
Japan
Net pension liability106.8–38.543.4
Contribution gap3.3–0.80.6
Germany
Net pension liability110.7–40.078.4
Contribution gap3.4–0.70.8
France
Net pension liability113.6–22.985.6
Contribution gap3.30.20.8
Italy
Net pension liability75.578.659.4
Contribution gap2.54.10.8
United Kingdom
Net pension liability4.637.20.5
Contribution gap0.11.6
Canada
Net pension liability67.89.345.2
Contribution gap2.01.00.4
Sweden
Net pension liability20.420.720.8
Contribution gap0.90.90.4
Source: IMF staff estimates.

Rate of technical progress set at zero instead of 1.5 percent (baseline assumption), resulting in a reduction of real GDP growth by the same amount.

Real rate of interest set at 2 percent instead of 3.5 percent (baseline assumption).

Source: IMF staff estimates.

Rate of technical progress set at zero instead of 1.5 percent (baseline assumption), resulting in a reduction of real GDP growth by the same amount.

Real rate of interest set at 2 percent instead of 3.5 percent (baseline assumption).

Interestingly, lower GDP growth has a beneficial effect on net pension liabilities and contribution gaps for those countries for whom pension arrangements involve full or partial indexation of benefits to nominal wages, such as Japan, Germany, and France (the former two to net wages, the latter, for earnings-related pensions, to gross pensions). For example, in the case of Germany, the estimated net pension liability declines by 40 percentage points. On the other hand, for the pension arrangements of Italy, the United Kingdom, Canada, and Sweden, for whom indexation of benefits is to the CPl, the effects of lower GDP growth are adverse. For the United States, the reduction in GDP growth has only a small (negative) effect on the net pension liability.

Where nominal gross or net wage indexation prevails, lower GDP growth decreases the projected nominal amounts of primary deficits as contributions and expenditure decrease by roughly equiproportionate amounts.28 As the projected smaller primary deficits are discounted using the same interest rate as in the baseline and are then expressed as a percent of the unchanged level of 1995 GDP, the net pension liability declines in response to lower GDP growth. In contrast, under CPI indexation, pension expenditures may decline significantly less than contributions, resulting in increased projected primary deficits and thus increased net pension liabilities. These sensitivity results indicate that depending on the indexation scheme, the level of estimated net pension liabilities (and contribution gaps) may vary inversely with the assumed GDP growth rate, suggesting that these estimates may have to be treated more cautiously as indicators of fiscal stress in public pension systems than is customary in the literature.29

The effect of lower real interest rates is found to be uniformly adverse for both net pension liabilities and contribution gaps. The adverse effects are especially severe for Japan, Germany, France, and Italy. The reason is that the projected initial buildup in net assets from imposing sustainable contribution rates, which would subsequently be drawn down to pay for the aging-induced excess of payments over contribution flows, benefits less from interest rate compounding when the rate is lower. While lower real interest rates are desirable for their favorable economic effects, they do not directly help resolve pension-related problems. In general, the simulations demonstrate that traditional public pension schemes that index contributions to wages and benefits to CPI inflation are more viable when real GDP growth rates and real interest rates are higher.

Delaying Reforms

In light of the political difficulties associated with pension reform, it is useful to examine the added costs of postponement. Table 11 indicates the increment to the sustainable contribution gaps associated with delay. If reform is postponed by five years, the sustainable contribution gaps would rise but by a relatively modest amount; the largest increment would be for Italy, equaling almost 1 additional percent of GDP. Obviously, the longer the delay, the greater the costs in terms of the amount by which sustainable contribution rates would have to be raised. If, for example, reforms are delayed for 15 years, the contribution gap would in most cases increase by more than two thirds. These results suggest a limited window of opportunity for public pension reforms.

Table 11.Effect of Postponement of Pension Reform(In percent of GDP)
Increase in Contribution Gap If Reforms Are Postponed by
Baseline5 years10 years15 years30 years
Major industrial countries1.80.30.71.44.5
United States0.80.10.20.51.5
Japan3.30.61.42.78.9
Germany3.40.61.32.79.5
France3.30.61.32.47.9
Italy2.50.51.11.96.5
United Kingdom0.10.10.10.3
Canada2.00.10.50.93.4
Sweden0.90.10.20.41.5
Source: IMF staff estimates.
Source: IMF staff estimates.

    Other Resources Citing This Publication