Abstract
Pension reform in Poland is an urgent priority. Benefit expenditures have placed considerable pressure on public finances, hampering stabilization efforts. Demographic developments will exacerbate this pressure although probably less than some estimates have suggested. The main long-term argument for reform is that the current system induces severe micro-economic distortions. To analyze possible reforms, this paper employs a simulation model of a utility-maximizing household facing the detailed rules of the current Polish pension system. The reforms considered are designed to reduce the dead-weight losses associated with current arrangements and to alleviate the inequality engendered by the present system. [JEL H55, J26, J65]
The dire condition of the Polish pension system creates an urgent need for reform. Since the inception of the transition to a market economy, a precipitous rise in the number of benefit recipients has led to rapid growth in expenditures both for pensions (old-age and disability) and unemployment benefits (see Table 1). From 1988 to 1993, the budgetary cost of pensions, expressed as a percentage of GDP, has doubled. Over the same period, unemployment reached almost three million, resulting in a budgetary cost of about 2 percent of GDP. Since Polish local government bears a substantial share of the cost of assistance to those who have fallen out of the safety net (for example, the long-term unemployed), this is only part of the total cost.
Social Transfers
(As percentage of GDP)
Starting in 1992, pensions became taxable in a tax-neutral manner for the recipients (see below).
Social Transfers
(As percentage of GDP)
| 1988 | 1989 | 1990 | 1991 | 1992 | 1993 | |
|---|---|---|---|---|---|---|
| Pensionsa | 6.94 | 7.96 | 8.10 | 12.35 | 14.69 | 14.72 |
| Net pensions | 6.94 | 7.96 | 8.10 | 12.35 | 12.61 | 12.42 |
| Unemployment benefits | — | 0.06 | 0.58 | 1.65 | 2.12 | 2.00 |
Starting in 1992, pensions became taxable in a tax-neutral manner for the recipients (see below).
Social Transfers
(As percentage of GDP)
| 1988 | 1989 | 1990 | 1991 | 1992 | 1993 | |
|---|---|---|---|---|---|---|
| Pensionsa | 6.94 | 7.96 | 8.10 | 12.35 | 14.69 | 14.72 |
| Net pensions | 6.94 | 7.96 | 8.10 | 12.35 | 12.61 | 12.42 |
| Unemployment benefits | — | 0.06 | 0.58 | 1.65 | 2.12 | 2.00 |
Starting in 1992, pensions became taxable in a tax-neutral manner for the recipients (see below).
Demographic developments in coming decades, in particular the aging of the Polish population, are likely to exacerbate the situation. However, as we argue in Section II, compared with other countries, changes in the age profile of the population in Poland will be relatively small. 1 Further-more, current high levels of unemployment are likely to decline, which will help to alleviate the budgetary burden of total benefit payments.
This does not reduce the urgent need for pension reform, however. In addition to the important objective of reducing the budget deficit, there are two strong arguments for changes in current arrangements. These are, first, that the current pension rules induce serious microeconomic distortions through the large fluctuations in effective tax rates generated over the life cycle, and second, that present arrangements are quite inequitable, resulting in 15 to 30 percent of pensioners being below the poverty line while others enjoy relatively high living standards.
Before discussing specific measures that might be taken, three general points should be made. First, the new benefit system should be sustainable, in that benefits and contributions should be set at a level compatible with a long-run, intertemporally balanced budget. To limit deadweight welfare losses, this balance should be achieved without excessive fluctuations in tax or benefits rates.2 An intertemporally balanced budget may, however, affect transfers between generations, as the country’s demography and productivity change over time. As we shall show below, Poland is currently facing substantial shocks both to the age structure of the population and to the supply side of the economy; hence, some intergenerational transfers may be justified.
Second, as a practical matter, it is important to realize how different elements of benefit expenditures become closely interlinked in a transition economy. A large part of the increase in pension benefit expenditures is due to a surge in early retirements. While some of this may be caused by workers’ anticipation of a tightening of retirement rules and reduction in benefit levels, a good deal must also be the consequence of actual or feared unemployment. Unemployment benefits have been significantly pared back in the last few years, and their duration has been reduced. The chance to switch to state benefits, such as retirement or disability pensions, for which there is no fixed term and whose levels are relatively generous has, therefore, become increasingly attractive. Reforms of the safety net should take account of potential spillovers between different layers of the system.
Third, benefit cutbacks should, as far as possible, be implemented while taking into account equity considerations to avoid creating poverty amongst the old or the unemployed. The average retired household in Poland enjoys a relatively high standard of living. Nevertheless, concerns remain that reform of the pension system will exacerbate the problem of a significant part of the pensioner population being below the poverty line. Three factors make old households in Poland vulnerable.
As in other former centrally planned economies (CPEs), the very generosity of the pension system had discouraged the accumulation of personal savings. Hence, pensioners are not now in a position to supplement government pensions with their own savings, as is common in other countries.
What personal savings the old had accumulated in the past were further eroded (if not indeed eliminated) by the hyperinflation of the late 1980s, which was accompanied by large negative real interest rates. In the absence of a stock market or significant owner-occupied housing, most savings in Poland were held in nominal assets such as bank accounts and hence were affected by the jump in prices.
Real wage declines and growing unemployment mean that extended families will not be in a position to assist aged relatives sufficiently to make up for the withdrawal of support by the state. If anything, transfers are occurring in the other direction, with older people supporting young unemployed ones (see Cox, Jimenez, and Okrasa (1993)).
To study the various possible policies, this paper employs two different tools of analysis. To look at aggregate effects, we simulate the impact of changes in demography, productivity, and labor market imbalances for the next thirty years. To examine more microeconomic effects, we use a simulation model of a dynamically optimizing household to examine the welfare implications of a number of possible reforms of the current Polish pension and unemployment benefit systems. Comparable models have been developed by Seidman (1983, 1986), Auerbach and Kotlikoff (1984, 1985) and Craig and Batina (1991). The model employed here is a reasonably realistic description of households facing either the current Polish retirement and unemployment benefit systems or various alternative systems that have been proposed.
Factors explicitly modeled include consumption-leisure choices; endogenous retirement decisions; liquidity constraints; unemployment; a lump-sum pension element; a labor-income-based pension element; income testing of pension benefits; the reduction of pension entitlement upon early retirement; and dependence of pension benefits on the number of contributing years.
To analyze benefit reform in Poland systematically, we begin, in Section I, by giving a concise description of current regulations. In Section II, we measure the implications of demographic changes on Polish pensions. Section III sets out the simulation model of a single household that will subsequently be used to assess different possible rule changes. Section IV discusses a number of possible reforms in the system, focusing on their welfare implications and the effects they might have on the government budget, labor supply, and savings. Section V concludes the paper by summarizing recommendations based on the simulations reported before.
I. The Benefit System in Poland
This section describes the organization and the interactions between the various layers of the benefit system in Poland. It draws heavily upon descriptions found in Barr (1992), Diamond (1993), and Preker (1994).
Pension System Provisions
We start with a short description of the basic provisions of the Polish pension benefit system.3 In theory, three conditions must be met before a worker is eligible to receive a retirement pension. First, one must have contributed for 25 (20)4 qualifying years. Reduced benefits are available under certain circumstances to those who have worked only 20 (15) qualifying years. Second, recipients must cease full-time work. This latter constraint generally does not bind, however, since it is possible to work almost a full working week while still drawing a pension. Third, recipients must be older than 65 (60).
In fact, the rules of the system allow for numerous exceptions, reflecting past concessions to particular interest groups. As a result, for large sections of the work force the age requirement does not apply. For example, groups deemed to be in hazardous or otherwise demanding occupations may retire at 60 (55) (e.g., miners, firemen, and, more surprisingly, teachers, journalists, academics, customs officers, and artists). Some professions, including teachers and miners, have the right to retire if they have accumulated enough qualifying years (30 for teachers, 25 for underground miners) whatever their age. The exceptions to the age requirement apply to such a substantial fraction of the population that the average age for new retirees in 1990 was 58 for men and 57 for women and may have dropped below 55 in 1993.5 Preker (1994) also points out that some occupational groups receive favorable treatment in their contributions. For example, farmers have lower contribution rates and the members of the military are exempt from contributing.
Even where regulations exist, they are often easy to circumvent due to poor management. Diamond (1993) comments on the reluctance of the Pension Administration to adopt a modern computerized system. The absence of such a system makes it much harder to enforce rules on labor supply and earnings, especially given the lack of cooperation with the tax administration. It also allows pension officers to interpret and apply the existing ones in an idiosyncratic and sometimes lax fashion.
Once eligible for a pension, a worker’s benefit level depends on earnings and the number of qualifying years he has accumulated. Under the Social Insurance Act of October 1991, the pension entitlement equals 24 percent of the national average wage plus, for each qualifying year during which contributions were made, 1.3 percent of the worker’s pension base.6 The pension base is the average of labor earnings in the best 3 of the last 12 years of work. The authorities intend to increase the averaging period each year by one year until the pension base for a new retiree equals the average of the best 10 years of the last 20.7,8 Workers are given partial credit for periods during which no contributions were made (for example, periods spent in higher education, military service, or child care), receiving 0.7 percent of their pension base for each noncontributory year.
Under the July 1991 income tax law,9 pension benefits are fully taxable. There is a minimum pension equal to 35 percent of the average wage, and a maximum implied by the fact that the pension base applicable for benefit calculations has an upper limit of 2.5 times the average wage. Pensions are subject to a simple earnings test in that those earning more than 120 percent of the average wage receive no benefits, while pensioners earning 60–120 percent lose the flat-rate part of the pension, that is, 24 percent of the average wage.
Changing the rate at which pensions were adjusted for inflation had been the Polish Government’s preferred means of coping with the budgetary cost of pensions in recent years. In the 1980s, rapid inflation seriously eroded pensioners’ purchasing power. In 1988 and 1989, benefit levels were adjusted on an ad hoc basis to alleviate this problem. In 1990, pension payments were indexed by average wages, at a time when real wages were declining. Overall, since 1988, the purchasing power of pensions has decreased slightly but the decline has been less than the fall in real wages (see Table 2). The resulting income and substitution effects have probably affected household retirement decisions in opposite ways. We discuss below whether their net impact can explain the surge in early retirements or if the threat of unemployment is a more likely explanation.
The system has also created blatant inequalities among pensioners, depending upon the timing of retirement. Initially, after the collapse of communism, entitlements were based on the nominal wage received in the last year before retirement. Benefits received by those who had retired before and after the hyperinflation therefore varied greatly. The October 1991 law remedied the situation by implementing the benefit formula according to which both the pension base and the flat-rate components of the pension are indexed by average wages. However, this formula only applies for new pensioners. A recent World Bank study (Preker (1994)) reported that nearly 20 percent of the retired live in households with incomes below the minimum for new retirement pensions. Since granting an overall increase in pension benefits is not an option, one may consider a budget-neutral reform that would reduce the dispersion of benefits. Some rough calculations suggest that a floor equal to 75 percent of the average pension could be financed by imposing a ceiling set at 150 percent of the average pension. Introducing such a floor would undoubtedly reduce poverty among pensioners.
Pensions: Number of Recipients and Purchasing Power
Pensions: Number of Recipients and Purchasing Power
| 1989 | 1990 | 1991 | 1992 | 1993 | ||
|---|---|---|---|---|---|---|
| Number of pensioners | 6,827 | 7,101 | 7,444 | 8,495 | 8,730 | |
| Retired (thousands) | 2,264 | 2,353 | 2,775 | 2,982 | — | |
| Invalids (thousands) | 2,152 | 2,187 | 2,318 | 2,435 | — | |
| Farmers (thousands) | 1,356 | 1,506 | 1,790 | 1,990 | — | |
| Real average wage index | 111 | 81 | 81 | 78 | 78 | |
| (1988= 100) | ||||||
| Retirement pensions | ||||||
| % of real wage | 57.0 | 65.0 | 76.2 | 72.5 | 71.4 | |
| Real index (1988 = 100) | 110 | 91 | 107 | 99 | 97 | |
| Disability pensions | ||||||
| % of real wage | 44.5 | 51.1 | 56.1 | 54.1 | 52.8 | |
| Real index (1988 = 100) | 107 | 89 | 98 | 92 | 88 | |
| Farmers’ pensions | ||||||
| % of real wage | 35.2 | 46.4 | 45.3 | 39.8 | 40.0 | |
| Real index (1988 = 100) | 107 | 103 | 100 | 86 | 85 | |
Pensions: Number of Recipients and Purchasing Power
| 1989 | 1990 | 1991 | 1992 | 1993 | ||
|---|---|---|---|---|---|---|
| Number of pensioners | 6,827 | 7,101 | 7,444 | 8,495 | 8,730 | |
| Retired (thousands) | 2,264 | 2,353 | 2,775 | 2,982 | — | |
| Invalids (thousands) | 2,152 | 2,187 | 2,318 | 2,435 | — | |
| Farmers (thousands) | 1,356 | 1,506 | 1,790 | 1,990 | — | |
| Real average wage index | 111 | 81 | 81 | 78 | 78 | |
| (1988= 100) | ||||||
| Retirement pensions | ||||||
| % of real wage | 57.0 | 65.0 | 76.2 | 72.5 | 71.4 | |
| Real index (1988 = 100) | 110 | 91 | 107 | 99 | 97 | |
| Disability pensions | ||||||
| % of real wage | 44.5 | 51.1 | 56.1 | 54.1 | 52.8 | |
| Real index (1988 = 100) | 107 | 89 | 98 | 92 | 88 | |
| Farmers’ pensions | ||||||
| % of real wage | 35.2 | 46.4 | 45.3 | 39.8 | 40.0 | |
| Real index (1988 = 100) | 107 | 103 | 100 | 86 | 85 | |
Disability Pensions
Disability pensions in Poland can be viewed largely as a substitute for unemployment benefits. In CPEs, workers deemed to be unemployable were often declared disabled, since unemployment was illegal. Eligibility required little more than a doctor’s certificate. The same lax definition has remained in use since the transition, which explains the high proportion of disability pensioners in Poland: 35 percent of the total in 1992, with an average age below 46. In 1988, according to United Nations (1993), 30 percent of the 55–59 age group was officially disabled in Poland, compared with 16.1 percent in neighboring Germany.
Three levels of disability are recognized. Group 1 consists of severely disabled persons, unable to look after themselves; Group 2 is made up of those unable to work; and Group 3 consists of individuals with long-term medical problems who can only work part-time. As with old-age pensions, disability pension recipients are not permitted to work full-time, but working for a substantial fraction of the full working week is allowed. On average, in 1993, disability pensioners received benefits equal to 74 percent of the old-age pension, which compares favorably to unemployment benefits.
The Unemployment Benefit System
Unemployment has increased rapidly since Poland’s reform process began, rising from negligible levels at the start of 1990 to 2.9 million or 16.1 percent of the work force by March 1994.10 The latest information suggests that this increase has leveled off but a serious long-term unemployment problem is now developing. The average duration of unemployment has risen steadily since the start of the transition. New entrants to the labor market have suffered particularly, leading to substantial growth in youth unemployment.
Under the Polish unemployment benefit system, instituted in December 1989, unemployed workers received 70 percent of their previous wage for 3 months, followed by 60 percent for the subsequent 6 months, and 40 percent thereafter.11 The benefit entitlement was in nominal terms and hence decreased more rapidly than the above percentages suggest when inflation was high. The minimum and maximum benefit levels were the minimum and average wages. Since these were indexed, the minimum benefit has acted as a partial safety net in the event of large price movements.
Initially, unemployment benefit eligibility did not depend upon past labor market participation and a large proportion of applicants in the first year of the scheme were not unemployed in an economic sense. In July 1990, eligibility was tightened by requiring that the unemployed person had worked for 180 days in the past year. The minimum benefit was also slightly reduced and some special extra payments to unemployed university graduates were reduced.
The October 1991 employment law introduced an indexed, flat-rate unemployment benefit of 36 percent of the average wage in the previous quarter (with a supplement of 35 percent of the average wage for those in training programs). An important feature of the new benefit was that it was only payable for 12 months, except to workers approaching retirement. Claimants ceased to receive benefits if they turned down two job offers and tighter rules were adopted to discourage working in the private sector (particularly in agriculture) while claiming benefits.
II. Reforming Pensions: Main Macroeconomic Issues and Basic Calculations
Demographic pressures are often cited as an argument for reform in the Polish system. The dependency ratio, defined as the working-age population divided by the number of pensioners, is expected to decline sharply between now and 2030. It might appear, therefore, that the tax burden faced by future generations will rise to intolerable levels. In fact, however, we suggest in this section that, if one adjusts for the effects of unemployment, adverse demographic trends are less of a problem than one might think. More important are structural problems with the Polish benefit system. In particular, the fact that social security contributions are levied on an extremely narrow base makes them highly distortionary. The contribution base is small partly because wages in Poland represent only a small fraction of GDP and partly because of the current cyclically high level of unemployment, which reduces the number of active workers available to finance benefit expenditure. One might even argue that the evolution of unemployment in the next two decades will largely determine whether the current system is viable.
Demographic Pressures
As in other European countries, the Polish population is aging. Until recently, this process was mainly the result of rising life expectancy. Between 1950 and the early 1990s, life expectancy at birth for men and women has risen from 64.8 and 70.5 years to 67.15 and 75.7 respectively. Since the start of the transition, however, the instantaneous fertility rate has dramatically collapsed (see Table 3). Though it is certainly too early to guess the long-term implications of this development, on current trends, the Polish population will decline rapidly, reversing the upward trend observed since World War II.
Births in Poland
(Per thousand women)
Births in Poland
(Per thousand women)
| Age | 15–19 | 20–24 | 25–29 | 30–34 | 35–39 | 40–44 | 45–49 | 15–49 |
|---|---|---|---|---|---|---|---|---|
| 1980 | 33 | 180 | 136 | 69 | 29 | 8 | 1 | 76 |
| 1992 | 29 | 150 | 118 | 58 | 24 | 6 | 0 | 53 |
Births in Poland
(Per thousand women)
| Age | 15–19 | 20–24 | 25–29 | 30–34 | 35–39 | 40–44 | 45–49 | 15–49 |
|---|---|---|---|---|---|---|---|---|
| 1980 | 33 | 180 | 136 | 69 | 29 | 8 | 1 | 76 |
| 1992 | 29 | 150 | 118 | 58 | 24 | 6 | 0 | 53 |
However, it is perhaps more likely that the sharp decline in fertility, like that of the late sixties, will prove temporary. It may well be that the recent social and political turmoil in Poland has led households to postpone childbearing without affecting their total, lifetime fertility rates. Adopting this supposition, we shall employ pretransition fertility rates in our simulations.
Dependency ratios also depend crucially on activity rates. As in other former CPEs, female labor force participation in Poland is higher than in most Western countries. In the simulations reported below, we assume that this will remain so in the future.
The activity rate of old workers (55 and older, see Table 4) is harder to predict, however. Polish workers over 65 seem, at first sight, to have activity rates in line with those observed in other middle-income countries. But in fact, according to the 1988 census, 90 percent of old workers are farmers. Although they are entitled to retirement benefits, they usually remain active, given that their benefits are much lower than those for occupational workers.12 Excluding farmers, the activity rate for old workers is actually fairly low in Poland.
A controversial question is whether a significant proportion of pensioners is employed in the black economy, taking advantage of the lax regulations on labor supply after retirement. The only available evidence consists of data from surveys of household consumption expenditure. Pensioners’ expenditures represent 60 percent of workers’ expenditure in 1992, against 51 percent in 1990.13 This increase is, broadly speaking, in line with the increase in the ratio of pension benefits to wages, suggesting that pensioners are not substantially supplementing their incomes by staying in the labor force.
Activity Rates by Age Group
Activity Rates by Age Group
| Males | Females | |||||
|---|---|---|---|---|---|---|
| Age | 1970 | 1980 | 1990 | 1970 | 1980 | 1990 |
| 50–54 | 94.05 | 86.36 | 81.70 | 75.40 | 71.54 | 71.07 |
| 55–59 | 90.95 | 79.71 | 70.88 | 67.85 | 56.54 | 49.63 |
| 60–64 | 83.05 | 60.31 | 52.83 | 50.90 | 36.65 | 33.99 |
| 65–69 | … | 42.57 | 40.00 | … | 27.62 | 26.65 |
| 70+ | … | 28.93 | 26.76 | … | 14.75 | 15.08 |
Activity Rates by Age Group
| Males | Females | |||||
|---|---|---|---|---|---|---|
| Age | 1970 | 1980 | 1990 | 1970 | 1980 | 1990 |
| 50–54 | 94.05 | 86.36 | 81.70 | 75.40 | 71.54 | 71.07 |
| 55–59 | 90.95 | 79.71 | 70.88 | 67.85 | 56.54 | 49.63 |
| 60–64 | 83.05 | 60.31 | 52.83 | 50.90 | 36.65 | 33.99 |
| 65–69 | … | 42.57 | 40.00 | … | 27.62 | 26.65 |
| 70+ | … | 28.93 | 26.76 | … | 14.75 | 15.08 |
It seems plausible, therefore, that, barring new distortions created by the benefit system, old workers’ activity rates will increase when the labor market tightens. This would undoubtedly improve the finances of the pension system.
Unemployment and Dependency Ratios
The recent rises in unemployment have affected dependency ratios through three different channels. First, as we have already mentioned, unemployment has generated a surge in early retirements. Second, it has reduced the tax base on which contributions are levied. Third, instead of contributing to the system, the unemployed constitute an additional burden.
Dependency ratios are usually defined as
We suggest instead the use of a modified dependency ratio defined as
where k is the ratio of the cost to the system of an unemployed person relative to that of a pensioner.14
We would argue that MDR is a better indicator of fiscal pressure than DR. To see why, let tR denote the replacement rate (the ratio of the average pension to the average wage, W), and let P and UB denote the total costs of pension and unemployment benefits respectively, while s is the share of wages in GDP. Then, the cost of benefits equals
which depends in a simple way on MDR.
Figure 1 illustrates the divergence between DR and MDR. In line with the calculations of Hambor (1992), the standard dependency ratio falls sharply from 2.9 in 1995 to 2.1 in 2020, with the largest decline occurring around 2010. In contrast, our calculation indicates that the dependency ratio, as measured by MDR, is likely to improve in the next fifteen years, although starting from a much lower base.
In other words, the main argument for pension reform is the need to limit the current cost to the budget rather than concerns about a future decline in dependency ratios. The very low level of our MDR measure also, in a sense, explains the difficulty the Polish authorities are currently experiencing in financing the benefit system. The present average retirement age of about 55 leads to a dependency ratio as measured by MDR of just 1.45, roughly half of the figure quoted by Hambor (1992). The low ratio shows why Poland is experiencing difficulties in paying for pensions even though poverty among the old is widespread.
Financing Benefits in the Long Term
To measure the impact on future generations, we simulate the cost of pensions and unemployment benefits under the following assumptions. Suppose that productivity and real wages increase by 4 percent a year between 1994 and 1999 and by 2 percent thereafter; the unemployment rate declines by 0.6 percent each year until it reaches 7.0 percent; the share of wages in GDP and replacement rates remain constant; and demographic variables change in line with past trends.
In Table 5, the “benchmark” row describes our forecast. It shows how the fall in unemployment translates into lower financing costs until around 2010–15, when the aging of the population has a serious impact. By 2040, the total cost of benefits settles down to 1½ percentage points higher than current levels.
Alternative simulations in the next two rows give some idea of how the budgetary cost could be limited. In the first simulation, we assume that participation rates for the 50–65 age group increase by 20 percent. In the long run, this implies a reduction in costs of about 1 percent of GDP. In the second simulation, we suppose that, starting in 2010, pensions are imperfectly indexed on productivity, pensioners receiving only 80 percent of productivity gains. These simulations suggest that limiting the budgetary cost of benefits should not prove too serious a problem.
However, a quite different question is whether the microeconomic distortions associated with the pension system can be held in check. The main problem is that the tax base for social security contributions in Poland is currently too narrow. Indeed, wages in Poland represent only a fraction of GDP: 47 percent for gross wages in 1992, and 33.8 percent net of social security contributions, according to Glowny Urzad Statystyczny (1994). The tax base is narrowed further by the various exemptions mentioned above. In the next section, we discuss the distortions implied by the pension system in more detail, calculating the impact on the welfare of a representative household.
The Cost of Pensions and Unemployment Benefits
(Percentage of GDP)
The Cost of Pensions and Unemployment Benefits
(Percentage of GDP)
| Horizon | 0 | 5 | 10 | 15 | 20 | 25 | 30 | 35 | 40 |
| (in years) | |||||||||
| Benchmark | 16.7 | 15.7 | 15.1 | 15.7 | 17.3 | 18 | 18 | 18 | 18.3 |
| Change in | -.7 | -.7 | -.9 | -1 | -1.2 | -1.1 | -1 | -1 | -1.1 |
| participation | |||||||||
| Indexation | — | — | — | -0.35 | -0.7 | -0.9 | -1.2 | -1.6 | -1.7 |
The Cost of Pensions and Unemployment Benefits
(Percentage of GDP)
| Horizon | 0 | 5 | 10 | 15 | 20 | 25 | 30 | 35 | 40 |
| (in years) | |||||||||
| Benchmark | 16.7 | 15.7 | 15.1 | 15.7 | 17.3 | 18 | 18 | 18 | 18.3 |
| Change in | -.7 | -.7 | -.9 | -1 | -1.2 | -1.1 | -1 | -1 | -1.1 |
| participation | |||||||||
| Indexation | — | — | — | -0.35 | -0.7 | -0.9 | -1.2 | -1.6 | -1.7 |
III. A Simulation Model of Polish Pensions
This section describes the simulation model that will be used to analyze various policy options. It also discusses how the design of pension schemes affect the household budget constraint.
Description of the Model
The model represents an intertemporally optimizing household, choosing leisure, consumption, and retirement date subject to various constraints. The constraints include a lifetime wealth constraint, liquidity constraints that prevent borrowing in anticipation of future income, and unemployment constraints that limit total household labor supply. The basic maximizing problem for the household is
where Ct and Lt are respectively consumption and leisure in period t and T is the maximum possible life span. The parameters δ, α, ρ, and αo are fixed; δ, α, and p, represent, respectively, the rate of time preference, the coefficient of intertemporal substitution, and the leisure-consumption substitution elasticity. The maximization is carried out subject to the following constraints.
Lifetime Wealth Constraint>
where Pt is the price of the consumption good, wt is the wage rate, Zt is lump sum transfers, rt is the gross interest rate, and τwt, τst, and τc are tax rates on labor and savings income and consumption goods respectively, all at time t. Here, total labor endowment in each period is normalized to unity. We discuss below how pension benefits affect that budget constraint.
Liquidity Constraints
where St is liquid private sector savings at period t and equals the partial sum of the budget constraint.
Employment Constraints
where L*t = l,2,…, T are fixed constants that represent the minimum leisure that the household can consume in given periods.
The solution of this program is considerably complicated by the presence of all the various constraints. Indeed, wages and interest rates are no longer the shadow prices used by the household to optimize its utility.
The Microeconomic Impact of Pensions
Like any pay-as-you-go (PAYG) system, the current Polish pension scheme may be regarded, to a first approximation, as a system of forced savings. The size of the distortions created depends on the return to this forced savings; whether or not households are liquidity constrained; and the design of the pension system itself.
Consider, first, the return on pensions. Wage taxes paid by individuals during their working life can be thought of as savings. One may then compare the return on these savings (pension benefits) to the return on other savings (net interest rate). If they are equivalent, a PAYG pension system is said to be “actuarially neutral” or “fair.” Distortions generated by an actuarially neutral PAYG system are minimal since individuals can accumulate negative savings, thereby offsetting the impact of pensions. As detailed in Appendix II, a PAYG pension system is actuarially neutral if the interest rate equals the sum of the growth rates in real wages and in the labor force, that is, if the neutrality condition r = ν + n, holds. If the interest rate exceeds ν + n, the pension system is equivalent to a tax levied on wages, with an income effect because of the implicit transfer to current pensioners. Formally, if τ is the pension contribution rate, the marginal tax rate affecting household labor supply decisions is τ[1 - g(r - ν - n)], where g(0) = 1. Table 6 provides estimates of g as a function of the interest rate, assuming a working life of 45 years and a 15-year retirement period.
A Measure of the Distortionary Impact of Pensions
A Measure of the Distortionary Impact of Pensions
| r - ν - n | -2% | -l% | 0% | 1% | 2% |
| g | 1.67 | 1.30 | 1.00 | 0.76 | 0.43 |
A Measure of the Distortionary Impact of Pensions
| r - ν - n | -2% | -l% | 0% | 1% | 2% |
| g | 1.67 | 1.30 | 1.00 | 0.76 | 0.43 |
In what follows, we make the admittedly strong assumption that the neutrality condition holds. In Section II, we argued that the Polish labor force is unlikely to change dramatically. Hence, whether or not the neutrality condition holds depends on the relative magnitude of real interest rates and wage growth. In transition economies, real interest rates are typically very high but this partly reflects large, inflation-related risk premiums, and wage and productivity growth are also likely to be rapid as the process of restructuring progresses. Hence, assuming that r = μ. is not too unreasonable.
Even when the neutrality condition holds, a PAYG system may have significant effects if there are binding liquidity constraints. In our simulations, liquidity constraints generally bind at the start of the household’s life cycle. This implies that, for young households, the shadow interest rate exceeds the net interest rate. In these circumstances, even actuarially fair pensions can generate significant welfare losses.
Finally, consider the design of the pension system. Important questions are whether the system is actuarially fair and how the system influences labor supply and retirement decisions. As explained above, Polish pension benefits for person i at time t essentially depend on Wi, the average wage in Poland; the number of working years, Ti; and the ratio, Mi, of wages earned to average wages in the economy over the averaging period. Following the analysis of Diamond (1993), we model pension benefits at period t as
The first term on the right-hand side corresponds to a lump-sum component of pensions, while the second one is a labor-income based component. For Poland, the legal values for parameters a and b are 0.23 and 1.3 percent. The value of Mi then depends mainly on the seniority rate, that is, the premium paid by firms for experience (see Appendix II). The limited evidence available indicates that experience was less rewarded in former CPEs than in Western economies. Flanagan (1994) finds a value of about 3 percent in the Czech Republic and we employ this figure in our baseline.
The first question we wish to examine is how the Polish benefit levels (given by equation (9)) compare to what pensioners would receive in a fully PAYG system in which the neutrality condition holds (see Appendix II). This comparison will tell us whether the current system is actuarially neutral for new entrants. We report the results, for different retirement ages and seniority rates, in Table 7. With a 2 percent seniority rate, an individual retiring after 45 years of work in the current Polish system receives 0.73 times the average wage. In an actuarially fair, fully PAYG system, assuming that social security contributions equal 30 percent of the wage and the dependency ratio is 3, such an individual would receive 0.90 times the average wage. Thus, for this worker, the current Polish system is less than actuarially fair. It only becomes neutral if the averaging period is shorter or the pensioner works for fewer years.
Total Return on Pensions
Total Return on Pensions
| Averaging period | 3 years | 10 years | 20 years | 30 years |
|---|---|---|---|---|
| T = 45, R = 15, PAYG = 0.90 | ||||
| s = 2% | 0.73 | 0.64 | 0.65 | 0.60 |
| s = 4% | 0.93 | 0.84 | 0.73 | 0.64 |
| s = 6% | 1.17 | 0.99 | 0.81 | 0.67 |
| T = 40, R = 20, PAYG = 0.60 | ||||
| s = 2% | 0.64 | 0.61 | 0.57 | 0.53 |
| s = 4% | 0.80 | 0.72 | 0.63 | 0.55 |
| s = 6% | 0.97 | 0.83 | 0.68 | 0.57 |
| T = 35, R = 25, PAYG =0.42 | ||||
| s = 2% | 0.56 | 0.53 | 0.50 | 0.47 |
| s = 4% | 0.67 | 0.61 | 0.54 | 0.48 |
| s = 6% | 0.81 | 0.70 | 0.57 | 0.49 |
Total Return on Pensions
| Averaging period | 3 years | 10 years | 20 years | 30 years |
|---|---|---|---|---|
| T = 45, R = 15, PAYG = 0.90 | ||||
| s = 2% | 0.73 | 0.64 | 0.65 | 0.60 |
| s = 4% | 0.93 | 0.84 | 0.73 | 0.64 |
| s = 6% | 1.17 | 0.99 | 0.81 | 0.67 |
| T = 40, R = 20, PAYG = 0.60 | ||||
| s = 2% | 0.64 | 0.61 | 0.57 | 0.53 |
| s = 4% | 0.80 | 0.72 | 0.63 | 0.55 |
| s = 6% | 0.97 | 0.83 | 0.68 | 0.57 |
| T = 35, R = 25, PAYG =0.42 | ||||
| s = 2% | 0.56 | 0.53 | 0.50 | 0.47 |
| s = 4% | 0.67 | 0.61 | 0.54 | 0.48 |
| s = 6% | 0.81 | 0.70 | 0.57 | 0.49 |
Another crucial point is that, outside the averaging period, pension contributions are purely distortionary since they do not imply any additional right to an offsetting future benefit.15 As emphasized by Diamond (1993), this is likely to reduce labor supply outside the averaging period and increase it during that period. Table 8 illustrates the magnitude of the phenomenon when the pension tax, τ, equals 30 percent.16 It also shows that the shorter the averaging period, the larger the distortions created on wages. The large changes in tax rates over the life cycle implied by a short averaging period are likely to disrupt household labor supply considerably.
Marginal Return on Pensions during the Averaging Period
Marginal Return on Pensions during the Averaging Period
| Averaging period | 3 years | 10 years | 20 years | 30 years | T years |
|---|---|---|---|---|---|
| T = 45, R = 15 | 262.5% | 57.8% | 13.9% | -0.8% | -10.5% |
| T” = 40, R = 20 | 316.7% | 74.0% | 22.0% | 4.7% | -4.0% |
| T = 35, R = 15 | 349.2% | 83.8% | 26.9% | 7.9% | 2.5% |
Marginal Return on Pensions during the Averaging Period
| Averaging period | 3 years | 10 years | 20 years | 30 years | T years |
|---|---|---|---|---|---|
| T = 45, R = 15 | 262.5% | 57.8% | 13.9% | -0.8% | -10.5% |
| T” = 40, R = 20 | 316.7% | 74.0% | 22.0% | 4.7% | -4.0% |
| T = 35, R = 15 | 349.2% | 83.8% | 26.9% | 7.9% | 2.5% |
The design of the pension system also affects retirement decisions. Polish workers face a wide range of minimum retirement ages, depending, for example, on the industry in which they work and on their health status. When choosing a retirement age, workers strike a balance between additional wages and increased benefits if they postpone their retirement; a shorter period of entitlement; and the disutility of work. For the purpose of the model, we shall assume that workers can retire when their age is in the range of 50 to 68.
The choice of retirement date is further complicated by the fact that the rules on retirement do not oblige pensioners to cease supplying labor after retirement, although full-time working is not possible. The current widespread unemployment, as we have argued, certainly prevents Polish households from taking advantage of that possibility on a large scale. However, the problem will become more acute as unemployment recedes. There is some empirical evidence that pensioners are paid less than other workers performing the same tasks (see Sziráczki and Windell (1992) for Hungary and Bulgaria). We therefore suppose that there is a limited loss of earnings upon retirement in that the wage profile is assumed to shift down after retirement by a factor of one half compared with what it would be if the household had not retired. We review below some measures, most notably income tests on total retirement income, that induce later retirement, thereby influencing total lifetime tax payments net of benefits.
The above framework can be compared with other models that have been developed for the study of savings behavior and pensions and unemployment benefit systems. The main strength of the model described above is the inclusion of a range of constraints upon household choices, over and above the budget constraint equating expenditures to total lifetime wealth—most notably, the limits on borrowing and labor supply. Given the deteriorating labor markets and primitive state of consumer lending in Eastern Europe, these features are clearly important. Hubbard and Judd (1986) formulate a relatively simple model of a liquidity-constrained household in order to study the impact of such constraints on savings and consumption decisions. The present model, with numerous complications such as endogenous retirement decisions and bequests, concentrates more on public finance aspects.
Probably the greatest weakness of the model employed in this paper is the limited role played by uncertainty. Time of death is stochastic in this model but no other sources of randomness are introduced. The study by Rust (1989), and the implementations of similar ideas by Lazear and Moore (1988) and Stock and Wise (1990), show just how stringent are the simplifying assumptions required if one is to incorporate stochastic elements. The largely deterministic nature of the model is particularly a problem in the treatment of unemployment, which is treated here simply as a constraint on labor supply. Fears of future unemployment among households still fully employed have probably contributed greatly to the surge in early retirements mentioned in the introduction to this paper. Here these can only be examined through fully anticipated employment constraints.
IV. Simulations and Policy Analysis
This section describes the simulation results. After discussing the methodology of our simulations, we first examine the impact of liquidity constraints and then discuss two possible sets of policy reforms. These are, first, changes in the current method used in calculating the wage base by averaging over wage income received earlier in life, and second, cuts in pension benefits for early retirement.
Baseline Simulation and Liquidity Constraints
This section describes the impact on households of a PAYG system. We emphasize, in particular, the impact of intertemporal transfers by the government. Four basic simulations are performed: first, as a baseline, a simulation without pensions; second, a pension system with lump-sum benefits and lump-sum financing; third, a pension system with lump-sum benefits and wage-tax financing; and fourth, the same as the third but with liquidity constraints.
The simulation results are presented in two forms. First, Table 9 summarizes total lifetime quantities such as utility and consumption of goods and leisure. Changes in utility are measured as a percentage of the wage income discounted over the life cycle. Second, Figures 2, 3, and 4 show the profiles for these variables over the adult life of the household. Pensions are calibrated in such a way that they represent 65 percent of average wages in the economy.
Simulations for a Notional PAYG System
Simulations for a Notional PAYG System
| Life utility |
Total savings |
Labor supply |
Consumption | |
|---|---|---|---|---|
| Baseline | reference | 257.3 | 41.8 | 22.3 |
| Flat pension and lump-sum financing | -0.00% | 92.0 | 36.3 | 22.1 |
| Flat pension and wage tax financing | -8.40% | 8.6 | 33.8 | 19.5 |
| Liquidity constraints | -7.80% | 40.8 | 33.7 | 19.8 |
Simulations for a Notional PAYG System
| Life utility |
Total savings |
Labor supply |
Consumption | |
|---|---|---|---|---|
| Baseline | reference | 257.3 | 41.8 | 22.3 |
| Flat pension and lump-sum financing | -0.00% | 92.0 | 36.3 | 22.1 |
| Flat pension and wage tax financing | -8.40% | 8.6 | 33.8 | 19.5 |
| Liquidity constraints | -7.80% | 40.8 | 33.7 | 19.8 |
In the second simulation, slight welfare losses result only from the timing of transfers by the government. This is an example of the changes in steady-state utility that are always available to governments in overlapping generations models when the interest rate exceeds the growth rate of the economy. To see this, suppose the economy has zero population growth with a stationary age structure and that r > 0. Then the government can obtain the same steady-state net tax revenue while switching the tax burden from young to old, whereas steady-state household utility will increase with such a switch since discounted lifetime income will rise. Again, such steady-state welfare gains are somewhat illusory since they ignore the welfare losses of the transitional generation that is old when the new policy is enacted.
Lump-Sum Pensions and Wage Tax Financing
Citation: IMF Staff Papers 1994, 004; 10.5089/9781451930887.024.A005
The third simulation describes the distortionary impact of wage taxes levied to finance pensions. As emphasized, for instance by Auerbach and Kotlikoff (1987), a key issue concerning the financing of pensions is the linkage between contributions and benefits. When the linkage is loose, these simulations17 show that distortions may well be substantial.
Lump-Sum Pensions and Wage Tax Financing with Liquidity Constraint
Citation: IMF Staff Papers 1994, 004; 10.5089/9781451930887.024.A005
Finally, the last simulation shows that liquidity constraints have a very substantial impact on savings and, counterintuitively, a positive impact on welfare. The relative generosity of the Polish retirement income system effectively removes the standard life-cycle motive for saving, namely, that of providing for retirement consumption. In the absence of liquidity constraints, high pension benefits together with upward-sloping lifetime wage profiles typically lead to consumption patterns that require heavy borrowing early in life. Thus, in our simulations without liquidity constraints, households only begin to have positive net financial wealth late in life, after the age of 40.18
Imposing constraints on borrowing restores the more conventional hump shape for life-cycle saving, with liquidity constraints ceasing to bind when households are in their early thirties. An important implication of this analysis is that the current Polish benefit system, with its relatively generous replacement ratios, would have an extremely deleterious impact on aggregate saving if consumer credit were freely available. Since financial liberalization, while not yet achieved, will presumably be implemented before long, one could expect aggregate savings to suffer greatly unless changes are made in the current system, such as some reduction in the generosity of retirement benefits relative to preretirement incomes.
As a last point, the presence of liquidity constraints radically alters the time profile of household labor supply. Liquidity constraints also have spillover effects in the labor market since households supply more labor in an attempt to alleviate the constraint on liquid resources. In that sense, liquidity constraints alleviate the negative impact of taxes on labor supply—hence, the higher utility.
Averaging Periods
To analyze the effect of different pension benefit formulas, we conduct eight simulations for a range of different averaging periods (see Table 10 and Figures 5–8). It is assumed in these calculations that the retirement age is constant and set at 65. We perform these simulations in such a way that the average replacement rate is constant. Changing the averaging period therefore only affects the marginal wage tax rate. As shown in Section III, the principal economic impact of calculating pension entitlements based on labor income earlier in life is to reduce the total effective tax rate on wages. Given the heavy burden of taxation on wage income in Poland, this would seem highly desirable.
Simulations with Different Averaging Periods
Simulations with Different Averaging Periods
| Life utility |
Total savings |
Labor supply |
Consumption | |
|---|---|---|---|---|
| Without liquidity constraint | ||||
| Whole life cycle | -4.3% | 30.8 | 35.9 | 21.0 |
| Best 20 years | -5.3% | -306.0 | 35.7 | 13.6 |
| Best 10 years | -5.8% | -303.9 | 35.5 | 13.4 |
| Best 3 years | -6.4% | 15.6 | 35.0 | 20.3 |
| With liquidity constraint | ||||
| Whole life cycle | -3.7% | 62.2 | 35.8 | 21.4 |
| Best 20 years | -4.1% | 57.3 | 35.6 | 21.3 |
| Best 10 years | -4.8% | 51.6 | 35.4 | 21.1 |
| Best 3 years | -5.7% | 53.5 | 35.0 | 20.7 |
Simulations with Different Averaging Periods
| Life utility |
Total savings |
Labor supply |
Consumption | |
|---|---|---|---|---|
| Without liquidity constraint | ||||
| Whole life cycle | -4.3% | 30.8 | 35.9 | 21.0 |
| Best 20 years | -5.3% | -306.0 | 35.7 | 13.6 |
| Best 10 years | -5.8% | -303.9 | 35.5 | 13.4 |
| Best 3 years | -6.4% | 15.6 | 35.0 | 20.3 |
| With liquidity constraint | ||||
| Whole life cycle | -3.7% | 62.2 | 35.8 | 21.4 |
| Best 20 years | -4.1% | 57.3 | 35.6 | 21.3 |
| Best 10 years | -4.8% | 51.6 | 35.4 | 21.1 |
| Best 3 years | -5.7% | 53.5 | 35.0 | 20.7 |
Pensions with 45-year Averaging Period
Citation: IMF Staff Papers 1994, 004; 10.5089/9781451930887.024.A005
The first simulation assumes an averaging period of the whole working life,19 while simulations 2, 3, and 4 assume 3-, 10-, and 20-year averaging periods respectively. Simulations 5–8 assume the same averaging periods but assume that liquidity constraints are binding. As noted in a previous section, the Polish authorities intend to increase the averaging period gradually from the best 3 of the last 12 years before retirement to the best 10 of the last 20.
Averaging over a short period provides a strong incentive for intertemporal substitution of labor supply. Tax payments are significantly boosted during the short averaging periods.20 The simulation result summaries in Table 10 show, however, that cutting the length of the averaging period significantly reduces the household’s total lifetime tax payments minus benefits, since they reduce their labor supply and consumption. In general, averaging over a longer period seems preferable in the current Polish situation.
Pensions with 20-year Averaging Period
Citation: IMF Staff Papers 1994, 004; 10.5089/9781451930887.024.A005
Retirement Decision
So far, we have assumed in our simulations that households retire at the age of 65 and that the pension system is actuarially fair when they retire at that age. An important issue is whether the current Polish pension system rules are effective in deterring early retirement.
Table 11 shows that the penalties imposed by the current system on early retirement are ineffective when the averaging period is short. The basic reason is that with a short averaging period, households can circumvent the penalties by working more during the averaging period. The effect, however, is to generate serious deficits in the pension scheme since the system is certainly not actuarially fair for those who retire early.
Unemployment
Finally, we can analyze in the same framework the influence of unemployment on household behavior. Unemployment has two major sequences. First, it reduces households’ income, thereby providing an incentive to postpone retirement because the consumption of leisure is lowered. Second, since the period of unemployment has a negative effect on future pensions, incentives to retire are increased.21
Pensions with 20-year Averaging Period
Citation: IMF Staff Papers 1994, 004; 10.5089/9781451930887.024.A005
Simulations with Endogenous Retirement Decisions
Simulations with Endogenous Retirement Decisions
| Optimal retirement age |
Budget deficit (as % of total wages) |
Total savings |
Labor supply |
Consumption | |
|---|---|---|---|---|---|
| Whole life cycle | 64 | -1.7% | 38 | 34.6 | 21.0 |
| Best 20 years | 60 | -10.7% | -2 | 31.4 | 20.9 |
| Best 10 years | 58 | -25.4% | -90 | 29.2 | 21.4 |
| Best 3 years | 56 | -45.1% | -179 | 26.8 | 21.8 |
Simulations with Endogenous Retirement Decisions
| Optimal retirement age |
Budget deficit (as % of total wages) |
Total savings |
Labor supply |
Consumption | |
|---|---|---|---|---|---|
| Whole life cycle | 64 | -1.7% | 38 | 34.6 | 21.0 |
| Best 20 years | 60 | -10.7% | -2 | 31.4 | 20.9 |
| Best 10 years | 58 | -25.4% | -90 | 29.2 | 21.4 |
| Best 3 years | 56 | -45.1% | -179 | 26.8 | 21.8 |
Pensions with 3-year Averaging Period
Citation: IMF Staff Papers 1994, 004; 10.5089/9781451930887.024.A005
Our simulations show that the second effect is likely to dominate for older workers (see Table 12). In these simulations, unemployment translates into a labor income reduction amounting to 20 percent. Workers react by working more at an earlier stage in the life cycle, accumulating greater savings, and retiring earlier. In this way, our model describes the increased attractiveness of pensions when unemployment rises. It lacks realism, however, in its implication that savings provide a partial buffer against labor income shocks. This is obviously only correct if such shocks are properly anticipated.
Endogenous Retirement Decisions with Unemployment
Endogenous Retirement Decisions with Unemployment
| Optimal retirement age |
Labor supply |
Consumption | |
|---|---|---|---|
| Without unemployment | 58 | 30.9 | 21.9 |
| With unemployment | 52 | 25.0 | 20.6 |
Endogenous Retirement Decisions with Unemployment
| Optimal retirement age |
Labor supply |
Consumption | |
|---|---|---|---|
| Without unemployment | 58 | 30.9 | 21.9 |
| With unemployment | 52 | 25.0 | 20.6 |
V. Conclusions
The title of this paper echoes that of Diamond’s (1977) seminal paper on the analysis of social security. That paper set out arguments to justify state intervention in the savings market through the provision of social security. Diamond’s arguments of wealth redistribution, market failure, inadequate voluntary saving, and administrative efficiency remain convincing. However, budgetary constraints are forcing numerous countries and particularly the reforming economies of Eastern Europe to reassess the liabilities of their welfare systems. If cuts are to be made, those cuts should be designed as carefully as possible, so as to limit the welfare impact and avoid aggravating poverty. This paper has aimed to develop a framework of analysis for pension and benefit reform based on a realistically complicated computer simulation model of a household facing alternative rules and regulations.
Our study has three main findings. First, despite adverse demographic trends, financing Polish pensions should prove sustainable when unemployment recedes.
Second, the current design of Polish pensions and the high marginal tax rates on wages that it generates imply substantial labor market distortions. A particularly serious problem is the narrow tax base upon which contributions are levied, since wages only represent a small fraction of GDP. From this standpoint, it may be advisable to consider a partial financing of pensions with other taxes. A consumption tax, for instance, would provide a larger tax base and could alleviate tax evasion.
Another key issue for Polish pensions is the definition of the averaging period used in the calculation of pension benefit entitlement. Increasing the averaging period spreads over a larger number of periods the reduction in effective marginal wage tax rates that such an averaging system generates. Currently, in any year outside the averaging period, the high social security tax rates constitute deadweight distortionary taxes. Since tax distortions generally rise with the square of the change in rate, spreading the reduction in effective taxes over a longer period should in itself increase efficiency.
While increasing the averaging period is not feasible for older workers since information on their incomes earlier in life is not available to the social security authorities, younger workers could be included immediately in a system with long-term averaging. Our simulations suggest that an averaging period of at least 20 years would generate significant benefits to the Polish economy.
Third, the current design of Polish pensions does not sufficiently penalize early retirements. In particular, if the averaging period is short, households are better off when working very hard for a small number of years and retiring early during the life cycle. Such behavior undoubtedly threatens the balance of the Polish pensions budget. It therefore appears advisable to incorporate more incentive for households to delay their retirement.
APPENDIX I
Parameterization
This appendix discusses the choice of parameters for the baseline simulations. In setting the tax parameters, we draw heavily on the detailed discussion of recent reforms in the Polish income tax system in Gorecki and Wisniewski (1992). At the start of 1992, the Polish authorities introduced a new personal income tax system according to which income is taxed in three bands, at the rates of 20, 30, and 40 percent.22 The bands are wide, representing almost three times the average wage, so it seems reasonable to take the 20-percent rate as the baseline case. The bands are imperfectly indexed, so the current flatness of the income tax structure is likely to change in the future. Hence, we adopt a baseline wage tax level of 25 percent and a baseline savings tax rate of 25 percent. Social security taxes are paid directly by firms through a 45-percent levy on the pretax enterprise payroll, which translates into a proportional tax rate of 33.7 percent.23 On consumption goods taxes, the Polish authorities replaced the system of turnover taxes with a Western-Europe an-style value-added tax in 1993. The tax has a reasonably wide coverage of different goods, but the effective rate for consumption goods as a whole is unlikely to exceed 8 percent.
Utility function parameters for Poland are a matter of educated guesswork. There are almost no significant empirical studies and what past studies do exist used data from before a very substantial regime change involving a relaxation in quantity constraints on consumption, savings, and labor supply behavior. The best approach is, therefore, to use reasonable values, as suggested in the literature on other economies. We set the elasticity of intertemporal substitution, p, at 0.8. There is some controversy over reasonable values for this parameter for the U.S. economy; a well-known study by Boskin (1978) found high substitution elasticity, while other work (for example, Carlino (1982)) suggests lower values. The figure here is fairly high but still lower than Boskn’s estimate. We chose a subjective discount rate, δ, of 3 percent. The important figure is in fact the difference between δ and the after-tax real interest rate. Since we set the pretax interest rate at 5 percent, choosing 3 percent for 5 implies a gradually increasing path for full consumption (given the fact that marginal utility is decreasing in consumption). We set the consumption-leisure elasticity of substitution also at 0.8. This figure implies a very small elasticity of labor supply, in line with estimates in other countries. The last utility parameter, the consumption-leisure parameter, αo, largely determines what the household’s expenditure shares devoted to leisure. It is fairly arbitrary since we normalize leisure endowment to unity and have little idea what proportion of leisure is actually consumed.
APPENDIX II
The Return on a Pay-As-You-Go Pension System
Let labor force growth, the real wage growth rate, and the net real interest rate respectively equal n, ν, and r. Assume also that wages grow with seniority at rate s for a given individual and that taxes are levied on these wages at the rate τ. Finally, in what follows, geom(ρ,q) represents the sum 1 + q + q2 + … + qρ-1.
If Lο individuals start working at time t - 0 for T periods and their initial wage equals Wο, then the discounted value of wages over their entire working life is
In this economy, at time t, one can easily show that the labor force and the number of pensioners respectively equal
At time t, the wage bill in the economy is equal to
and the average wt is defined by WBt - LFt,Wt. Since the amount of taxes available for financing pensions equals τ Wbt, the average pension can be calculated as
Therefore, the ratio of Bt to the average wage is simply the product of the tax rate and the dependency ratio.
Finally, one can compare the discounted value of pension benefits with the value of discounted wages and verify that the pension system is actuarially fair if r is equal to ν + n, that is, the net interest rate equals the sum of the growth rate in productivity and in labor force.
Turning now to the specifics of the Polish pension scheme, one can calculate the ratio, Ratt, of wages during the averaging period to average wages as a function of the length of the averaging period (Aν) and of the seniority rate. Table A1 performs this calculation with T = 45.
Ratio of Highest Wages to Average Wages
Ratio of Highest Wages to Average Wages
| Aν (years) | 3 | 10 | 20 | 30 | 45 |
|---|---|---|---|---|---|
| s = 2% | 1.46 | 1.37 | 1.25 | 1.14 | 1.0 |
| s = 4% | 2.00 | 1.76 | 1.48 | 1.25 | 1.0 |
| s = 6% | 2.59 | 2.14 | 1.67 | 1.33 | 1.0 |
Ratio of Highest Wages to Average Wages
| Aν (years) | 3 | 10 | 20 | 30 | 45 |
|---|---|---|---|---|---|
| s = 2% | 1.46 | 1.37 | 1.25 | 1.14 | 1.0 |
| s = 4% | 2.00 | 1.76 | 1.48 | 1.25 | 1.0 |
| s = 6% | 2.59 | 2.14 | 1.67 | 1.33 | 1.0 |
REFERENCES
Auerbach, Alan J., and Laurence J. Kotlikoff, “Social Security and the Economics of the Demographic Transition,” in Retirement and Economic Behavior, ed. by Henry A. Aaron and Gary Burtless (Washington: The Brookings Institution, 1984), pp. 255 –76.
Auerbach, Alan J., and Laurence J. Kotlikoff, “Simulating Alternative Social Security Responses to the Demographic Transition,” National Tax Journal, Vol. 38 (June 1985), pp. 153 –68.
Auerbach, Alan J., and Laurence J. Kotlikoff, Dynamic Fiscal Policy (Cambridge: Cambridge University Press, 1987).
Barr, Nicholas, Poland: Income Support and the Social Safety Net: Policies for the Transition, World Bank Report No. 9661-POL (Washington: World Bank, June 1992).
Boskin, Michael J., “Taxation, Saving, and the Rate of Interest,” Journal of Political Economy, Vol. 86 (April 1978, Part 2), pp. S3 –S27.
Carlino, Gerald A., “Interest Rate Effects and Intertemporal Consumption,” Journal of Monetary Economics, Vol. 9 (March 1982), pp. 223 –34.
Cox, Donald, Emmanuel Jimenez, and Wlodek Okrasa, “Family Safety Nets during Economic Transition: A Case Study of Poland,” Paper Presented at a Conference on Privatization and Socioeconomic Policy in Krakow, Poland. (unpublished; October 1993).
Craig, Ben. and Raymond G. Batina, “The Effects of Social Security in a Life Cycle Family Labor Supply Simulation Model,” Journal of Public Economics, Vol. 46 (November 1991), pp. 199 –226.
Diamond, Peter, “A Framework for Social Security Analysis,” Journal of Public Economics, Vol. 8 (December 1977), pp. 275 –98.
Diamond, Peter, “Pension Reform in a Transition Economy: Notes on Poland and Chile,” in The Transition in Eastern Europe, ed. by Olivier Jean Blanchard, Kenneth A. Froot, and Jeffrey D. Sachs (Chicago: University of Chicago Press, 1993), Vol. 2, pp. 71 –103.
Flanagan, Robert J., “Were Communists Good Human Capitalists? The Case of the Czech Republic,” (unpublished; Stanford, California: Stanford University, 1994).
Glowny Urzad Statystyczny, “Rocznik Statystyczny,” (Warsaw: various issues).
Gorecki, Brunon, and Marian Wisniewski, “The Income Tax Reform in Poland” (unpublished; Warsaw: Warsaw University, 1992).
Hambor, John C, “Issues in Eastern European Social Security Reform,” U.S. Treasury Department Research Paper No. 9201 (Washington: U.S. Treasury Department, June 1992).
Hubbard, R. Glen, and Kenneth L. Judd, “Liquidity Constraints, Fiscal Policy and Consumption,” Brookings Papers on Economic Activity I (1986), The Brookings Institution, pp. 1 –59.
Lazear, Edward P., and Robert L. Moore, “Pensions and Turnover,” in Pensions in the U.S. Economy, ed. by Zvi Bodie, John B. Shoven, and David A. Wise (Chicago: University of Chicago Press, 1988), pp. 163 –88.
Popović, Bojan, “New 1980 and 1990 Estimates of Activity Rates by Age and Sectoral Distributions for Some Countries of Eastern and Southern Europe,” Bulletin of Labour Statistics, No 2, (Geneva: International Labour Office, International Labour Organisation, 1994).
Preker, Alexander S., “Republic of Poland: The Pension System, Old-Age, Survivors and Disability,” (unpublished; Washington: World Bank, 1994).
Rust, John, “A Dynamic Programming Model of Retirement Behavior,” in The Economics of Aging, ed. by David A. Wise (Chicago: University of Chicago Press, 1989), pp. 359 –98.
Seidman, Laurence S., “Social Security and Demographics in a Life Cycle Growth Modei,” National Tax Journal, Vol. 36 (June 1983), pp. 213 –24.
Seidman, Laurence S., “A Phase-Down of Social Security: The Transition in a Life Cycle Growth Model,” National Tax Journal, Vol. 39 (March 1986), pp. 97 –107.
Stock, James H., and David A. Wise, “Pensions, the Option Value of Work, and Retirement,” Econometrica, Vol. 58 (September 1990) pp. 1151 –80.
Sziráczki, Gyorgy, and James Windell, “Impact of Employment Restructuring on Disadvantaged Groups in Hungary and Bulgaria,” International Labour Review, Vol. 131, No 4–5 (1992), pp. 471 –96.
Tymowska, Katarzyna, and Marian Wisniewski, “Poland’s Social Security System,” (unpublished; Warsaw: Warsaw University, 1991).
United Nations, Statistical Office, Demographic Yearbook, (New York: U.N. Nationals Statistical Office, Department of Economic and Social Affairs, 1993).
William Perraudin is Woolwich Professor of Financial Economics at Birkbeck College, London and Research Associate of the Department of Applied Economics, Cambridge University. Thierry Pujol is an Economist in the European I Department. The paper was begun while William Perraudin was a visiting scholar in the European I Department. The authors thank Ehtisham Ahmad Michael Deppler, Liam Ebrill, Stanislas Gomulka, and Christian Mulder for helpful conversations. All errors are the authors’ responsibility.
The share of the elderly population was 14.6 percent in the last (1988) census, which compares favorably with the situation in Germany (20.7 percent in 1987), the United Kingdom (20.2 percent in 1981), Hungary (18.9 percent in 1990), etc. (see United Nations (1993)).
Grandfather clauses are particularly distortionary.
The main state pension scheme is administered by the Social Insurance Institution or ZUS. Two other state pension schemes exist for private-sector farmers (KRUS) and priests.
Figures for women are given in parentheses after the corresponding figures for men.
Parliament initially approved more generous percentages but backed away from them given the budgetary implications.
Prior to the October 1991 law, the final wage applicable to benefit calculations was simply a workers’ earnings over the last twelve months of his working life. The move toward a longer average implies a significant streamlining of the Pension Administration. Anecdotal evidence suggests, however, that this move is urgently required since firms and workers arbitrage the current system by drastically increasing wages when retirement approaches.
Note that, for a given worker, the averaging rules depend upon the year of retirement, not age. This complicates optimal retirement decisions somewhat.
The new law came into force on January 1,1992, replacing four payroll and income taxes with a single personal income tax. For a detailed description, see Gorecki and Wisniewski (1992).
The deterioration has exceeded that in other Central and Eastern European countries. Unemployment rates in the Czech Republic and Hungary, for example were 3.5 percent and 12.1 percent respectively at the beginning of 1994.
The December 1989 law allowed anyone out of work to claim the unemployment benefit. In July 1990, eligibility was tightened so that that benefit can only be obtained if no job or training program is available and if the claimant has worked for 180 days in the past year.
On average, farmers’ pensions represent approximately 66 percent of occupational workers’ pensions.
For Poland, “back-of-the-envelope” calculations suggest that k equals approximately one third.
Outside the averaging period, the marginal tax rate equals the pension tax rate, τ. Over the averaging period, if αν denotes the length of the averaging period, the marginal tax rate equals Tb R/aν - τ.
The Diamond (1993) example is admittedly more striking: out of exhaustion, an overworked worker, near retirement, was involved in a serious traffic accident.
Here, the contribution rate equals 30 percent.
Note that the simulations assume that households receive no bequests. Including bequests would probably not change matters since these would normally be received around the age of fifty.
This is the approach taken in the U.S. social security benefit system.
Although effective tax payments are, of course, cut since the household receives extra pension benefits in reward for extra supply of labor.
The same argument would apply for any job that does not increase future pensions during the averaging period. Therefore, an old worker who could retain his job only if he accepts a drastic wage cut has an incentive to retire instead.
The tax replaced the old payroll tax as well as the less important income equalization tax and various small taxes on individual crafts and small manufactures.
Of the 45 percent, 43 percent goes toward the social insurance expenditures of the ZUS, while 2 percent is used as a contribution toward the costs of unemployment benefits. This distinction has no impact on the economic effects of these taxes.
