Pension reforms in the large accession countries

Pension reforms in the large accession countries

In the Central and Eastern European candidate countries for EU accession, the reform of pension systems remains one of the fundamental tasks of the transformation process. During the era of command economies, the social security systems were particularly susceptible to government wilfulness. The top priority has therefore been to construct a new legislative foundation for the governmentprovided social security systems that is consistent with the principles of liberty, the rule of law and the free-market economy. This included, for example, the introduction of objective criteria for pension assessment and transparent methods of financing pensions. It was also important to open up opportunities for private retirement saving. Especially private pension provision on a funded basis can contribute substantially to the process of catching up with western economies: It helps reduce individuals’ dependence on the state, provides enhanced incentives for endeavour and performance in the official economy and for saving, and mitigates an overtaxing of the government-provided social security systems. The private pension tier can also boost the development of efficient capital markets. To be sure, there is a mutual dependence: An upturn in private funded retirement provision requires capital markets that are able to absorb the funds to be invested.

After some initial hesitation, the Central and Eastern European (CEE) countries have confronted these economic- and social-policy tasks with growing enthusiasm. Since the middle of the past decade, advised by experts especially from the World Bank and the OECD, a number of these countries have reformed their government pension systems from the ground up and supplemented them with privately financed pillars. The goal is to ensure pension security by means of a balanced threetiered system of public, occupational and personal pensions. The reforms undertaken in Poland and Latvia in particular have been widely studied as models for such an approach.

Bringing pension systems into line with the new economic order is only one motive for the pension reforms in CEE. Another purpose is one that is shared with the EU-countries: to make pension systems storm-proof for the future. Although populations in Eastern Europe are still younger on average than in the EU, they are ageing ever more rapidly because of their in some cases extremely low birth rates. While today in CEE there are barely 30 pensioners for every 100 persons of working age, by 2050 the number could rise to almost 80 pensioners. The trend would be further aggravated if, as some fear, Eastern Europeans were to migrate west by the millions, although we do not expect this to happen.

Even today the sustainability of pensions for the long term is thus already a central issue for the countries aspiring to EU accession. They too must ensure an adequate standard of living for a growing number of older people without overburdening the gradually shrinking working age groups.

All this gives us occasion to examine the pension reforms in the accession countries more closely in this article. We begin by reviewing the major external factors affecting the reform policies – the past process of transformation and the future demographic trend – and outlining the reform alternatives available against this backdrop. In the subsequent, main section we discuss the pension systems in the three largest accession countries: Poland, Hungary and the Czech Republic. Our focus here is on private retirement saving through pension funds. This reflects the special importance of pension funds for the longerterm stability of retirement income provision and for the financial markets.

Transformation has triggered reforms

With respect to old-age security, as in other policy areas, the CEE countries started with a daunting legacy when the process of transformation began in 1989/90. While the socialist regimes had held sway, pension systems had been entirely under state control and were integrated into government budgets. The public pension schemes were structured along socialist principles and subservient to national interests; free-market incentives and tenets such as the equivalence of eventual pension benefits to pension contributions were not features of these systems. To finance pensions, contributions were collected from employers (often at flat rates based on the wage bill). The size of individual pensions was unrelated to these contributions; in many cases it was determined by individuals’ pay in the final year or years before retirement. Those in occupations promoted by the government or in sensitive professions enjoyed special privileges. All the same, under socialist rule most pensioners belonged to the financially disadvantaged strata of society. In many cases, current pensions were inadequately indexed to general wage and price increases and thus lost significantly in real value over time. In order to supplement their meagre benefits, many pensioners continued to work.

During the macroeconomic-restructuring crisis at the beginning of transformation, the monopolistic state pension systems suffered particular strain. In this phase of the decline and decimation of socialist companies, production and employment slumped dramatically everywhere while the development of new, free-market structures through privatisation, new business start-ups and foreign direct investment made only halting progress. The contribution base of the state-run social security systems contracted with corresponding abruptness. Meanwhile, however, their expenses rose because pension benefits had to be increased to keep pace – at least partially – with the strong initial surge in the cost of living. As well, the widespread practice of sending the unemployed into early retirement – designed to curb rampant unemployment – led to substantial additional pension costs.

As a result, the transformational crisis made itself felt in a strong increase in the ratio of numbers of pensioners to employed contributors (and taxpayers). From 1989 to 1995 this so-called “system dependency ratio” rose from less than 40% to more than 60% in Poland and from 51% to fully 82% in Hungary. Meanwhile the “old-age dependency ratio” – the ratio of the number of people aged 60 years or more to the working-age population (aged 15-59) – remained largely unchanged. Expenditure for public pensions was accordingly steep. In the mid- 1990s in Poland, where unemployment was especially high but pensions had been raised sharply at the beginning of the decade from their earlier low levels, spending for state pensions consumed 14 ½% of GDP; in Hungary the figure was roughly 10 ½% of GDP. To place this in perspective, the OECD average in 1995 was around 10%, ranging from 5% in Canada to about 15% in Italy.

The outsized public social security systems thus became a serious roadblock on the path to creating market economies. This state of affairs is not changed by the fact that, early in the transformation process in some countries, these institutions were reorganised into social insurance systems and wholly or partially removed from the national budgets. Funding them required growing contributions from employees and employers, with negative consequences for labour costs, motivation to work and investment in new jobs. Thus, in Hungary in 1996, social security levies amounted to more than 60% of gross remuneration, with half of these contributions flowing into the national pension scheme. In Poland during this period, pension contributions alone constituted 30% of total employment costs. Yet, social security funds ran mounting deficits that had to be covered out of the general budget. These two countries’ already high national debt grew further as a result. Pension finance was not as huge a problem in the Czech Republic. Here too there was a trend towar ds early retirement and growing numbers of pensions, but employment did not nosedive as badly.

Much-needed measures to counter the ballooning of expenditures – such as the abolition of special privileges enjoyed in socialist times and the introduction of higher retirement ages and tougher early-retirement rules – at first frequently met with strong political opposition and were thus only implemented with some delay. In this early stage, pension policy thus usually restricted itself to unavoidable ad-hoc interventions, like the adjustment of pension benefits to the amount of funds available. One of the first countries to reform its state pension scheme systematically was the Czech Republic. A new pension insurance law passed in 1995 raised the retirement age, expanded the contribution periods on which pensions are assessed, and modernised the benefit legislation (for example by introducing widow’s pensions). Admittedly, experts were aware from the outset that such parametric reforms (i.e. non-structural adjustments intended only to improve the current financial position of public-tier pension systems) could not possibly be enough in countries striving for a rigorous change-over to free-market economies. Rather, all along the ultimate task in pension reform, analogous to that in other parts of the economy, was to pry open state monopolies and establish markets for private provision of retirement income.

Great demographic challenges

The CEE countries are not being spared the demographic shift towards ageing, shrinking populations. In fact, in this region the decline in birth rates, as the major cause of the rising average age, has been particularly dramatic since the onset of transformation at the end of the 1980s. Estonia, Lithuania and the Czech Republic currently have some of the lowest birth rates in the world, at less than 1.2 births per woman. In Hungary and Poland as well, birth rates stand well below the replacement rate of 2.1, the rate required to hold population size steady. With a sustained trend towards fewer births and increasing life expectancy, these countries will inevitably experience a marked longterm shift in their populations’ age structure. This process is accentuated as the baby-boomers born in the 1950s and 1960s grow older. While the working-age population is thus contracting, the number of pensioners is growing apace. The old-age dependency ratio, and thus ultimately the system dependency ratio, will grow all the more swiftly in the countries involved. The expected demographic changes will exceed even the scope of the distortions witnessed during the past decade in the course of economic restructuring. Most importantly, these changes will be lasting ones.

The process in question will largely set in at the end of this decade. Until then CEE, notably Poland, still benefits from the higher birth rates of past decades. The old-age dependency ratios, which at present are mostly still significantly lower than in the EU-15, will as yet change little between now and 2010.

Though the point of departure is relatively favourable, however, the process of population ageing, once under way, becomes very rapid. In part this is the result of the very low birth rates of the recent past, which even in future are likely to rise only slightly. It also results from the retirement of the baby-boomers, beginning in about 2015. According to UN projections, the old-age dependency ratio in the Czech Republic (the accession candidate with one of the most-rapidly ageing populations) will more than double by 2030, thus reaching the average value for today’s EU countries. In Hungary, by contrast, the increase in the ratio of seniors to people of working age will initially be less drastic (not least because life expectancy is still low for the time being). The starting level in Hungary, however, is higher. Poland too is expected to age somewhat less quickly and, like Hungary, could retain a younger population (as measured by th e old-age dependency ratio) than western EU countries. until as late as 2050. Still, at that time there would be more than 70 older people for every 100 workers – almost three times as many seniors as today. A very high rate of labour emigration would aggravate the situation further. For example, if in the years after joining the EU, Poland were to see a westward exodus of 2 million young people of working age, the old-age dependency ratio in 2030 would increase from 46% to 51%. However, we consider such a scenario to be highly unlikely.

The pension systems in CEE – not unlike those of the EU 15 – thus face enormous challenges. This is especially true for the pay-as-you-go systems, since these have no financial reserves for times of intense demographic strain. The last decade has shown that, when the system dependency ratio increases, this directly affects the fiscal position. This problem is the more serious the more generous the promised public pensions and the more dominant the role of state-provided pensions in the country’s pension system. For this reason, demography is one of the powerful arguments for permanently restricting the role of payas- you-go state-administered pension insurance in favour of increased private provision on a funded basis.

Two models for privatisation of pension schemes

There is no single accepted way of accomplishing the restructuring required by the overly large public pension systems in CEE and in many EU countries. In essence, there are two basic competing models:

1. Sweeping privatisation of pension schemes.Under this model, similar to the Chilean reform of 1981, state pensions are reduced to a subsistence minimum and paid for with tax revenue. Instead of the government, private pension funds become the main providers of pensions. These institutions finance the benefits from capital reserves that are accumulated in personal retirement accounts. Retirement saving is mandatory. However, savers are free to choose from an array of eligible funds and can also switch between funds. Existing entitlements under the former state-run scheme – which is eventually closed down – are converted to interest-bearing securities and transferred to the personal savings accounts.

    

2. Partial privatisation.In this solution, a pay-as-you-go public tier is retained in a modified form. However, the public system is downsized, and complemented by the creation of occupational pension funds, contributions to which may or may not be compulsory. This makes it possible to construct a balanced, threepillar system of pension provision, with public pension insurance as the first tier, occupational pensions as the second pillar, and personal retirement savings as the third tier. To date the only European countries with such a balanced mix are the Netherlands and Switzerland. Important differences to the other approach are the larger role played by the reformed public pension pillar and the slower development of the second pillar. However, the biggest difference is that, because pay-as-you-go financing continues, the contributors who finance the scheme gain new benefit entitlements. The existing government-administered pillar is thus not eliminated. In consequence, the demographic burden with which this public system is saddled (i.e. unfunded liabilities) does not become evident. Furthermore, new and existing entitlements can simply be added together.

Neither approach is intrinsically better than the other. Rather, each has different advantages and drawbacks. The first method (full privatisation) offers the opportunity of swift and consistent restructuring of retirement provision. This can generate a strong impetus for savings and the development of financial markets. However, in order for full privatisation to work, the pre-conditions for quick expansion of these markets must exist or be created. What is more, a fast change-over to a fully private system places a greater onus on the current working popu lation. Workers must finance today’s pensions with their taxes, but this does not earn them pension credits, apart from the entitlement to the low minimum pension. The tax burden associated with a sweeping privatisation is thus less marked in countries with a relatively young population and correspondingly low legacy liabilities in the public pension system than in jurisdictions with a higher share of pensioners and older workers.

Granted, existing pension obligations can also be met by government borrowing, as was effectively the case in Chile. But this is possible only if it does not require an excessive expansion in national debt (as a percentage of GDP) and if enough latitude exists for such borrowing. (Here the accession countries could face potential conflicts with the Maastricht convergence criteria for joining EMU. One of the conditions of membership is that public-sector debt not exceed 60% of GDP.) In a gradual privatisation, these problems are less serious. Moreover, the aim of risk diversification makes a balanced three-pillar architecture desirable, as its constituent systems involve distinctive kinds of risk. Thus, while the public systems are subject primarily to the threat of political intervention, funded systems entail the risk of unsatisfactory investment returns.

Will to reform is slowly growing

Plans for root-and-branch reform of public pension systems and the creation of private funded schemes were discussed in some CEE countries early on – for instance, as early as the beginning of the 1990s in Poland. Naturally enough, the issue of whether and to what extent to privatise pension insurance was at first very contentious – and in some countries it still is. The notion of privatisation was championed especially by the respective CEE ministries of finance and by economic experts, while social politicians and representatives of the state pension schemes tended to reject this course of action. It therefore took several years for legislative bodies to set such reforms in motion.

The trailblazers were Estonia, Latvia, Hungary and Poland. Reform packages involving partial privatisation were passed in Hungary in 1997 and in Poland in 1997 and 1998, then enacted in the years that followed. By contrast, in the Czech Republic no agreement has been reached to this day on taking similar steps to overhaul the pension insurance system, despite a relatively early start to the debate. Thus far a second, occupational pillar in the Czech Republic exists only in the form of a (still contested) draft law.

In contrast to the creation of a second pillar, the establishment of a third tier – that of voluntary personal retirement saving – and construction of the necessary legal framework for supervision and taxation of the pension systems was generally less controversial. Thus, as early as 1993 and 1994 Hungary and the Czech Republic created the legal basis for the activity of investment funds, mutual funds and private pension insurance. In both countries, personal provision for retirement has since then also been promoted by government incentives in the form of direct subsidies and tax privileges. In the following section we take a closer look at the status of, and outlook for, retirement income provision in the three large CEE countries. Given that reforms have been pursued with differing intensity, these countries present a varied picture.

Modern three-pillar system in Poland

Of the large CEE countries, Poland has to date achieved the most extensive renewal of its pension system. The reform laws passed in 1997 and 1998 by the Sejm in two packages, both enacted in 1999, spell out the establishment of a three-pillar system. Both its redesigned first pillar and the newly created second tier feature a remarkable architecture. The guiding ideas behind their structure are to achieve high acceptance for the pension reform and to ensure that the system can be financed without overburde ning the contributors and taxpayers.

The new first pillar is based on the so-called ” notional defined contribution” (NDC) model, a pay-as-you-go system. NDC is a modern concept that also underlies the systems adopted in Croatia, Latvia and (in the EU) Sweden. Its basic principle is to make the pay-as-you-go approach work more like private funded systems. Thus, it is characterised by the strict equivalence of contributions and benefits. Above the minimum pension that is designed only to stave off abject poverty, the amount of an individual’s NDC pension essentially depends only on the level of contributions made per year to personal retirement accounts (which are ” appreciated” by a method identical for all contributors) and on the insured’s age at retirement. In Poland the rate at which the contributions (and the value of accrued rights from the old system) are increased currently amounts to 75% of the growth in average pay. This reflects the goal of a long-term reduction in the present level of public pension. The current benefit level of 60% of final salary is regarded as unsustainable.1 In order to cushion the foreseeable demographically driven rise in benefit expenditure, Poland has added the so-called ” Demographic Reserve Fund” to the new system. This fund is to be financed from possible surplus contributions and, in the years from 2002 to 2008, with 1%-point of gross pay.

The clear, uniform rules, the direct personal credit for contributions and the associated transparency lead to important advantages. Notably, contributors have no incentive for early retirement. Instead, individuals can easily foresee the negative consequences of a shorter working life on the level of their pension (the statutory retirement age in Poland is 65 for men and 60 for women, but in practice the average age at retirement is still 59 and 55 years, respectively). Reinforced property rights encourage acceptance of the system – which can be an important factor especially during a scheme’s introduction – and discourage political meddling. The NDC model also permits the efficient countering of financial bottlenecks that result from demographic change, for instance an increase in life expectancy. This can be accomplished by raising the minimum retirement age and/or by adjusting the level of pension benefits to the greater longevity.

Participation in this new pension insurance scheme is mandatory for all employees born after 1969 (including young newcomers to the uniformed civil service). Those born between 1949 and 1969 had the choice of joining or remaining in their existing scheme. Employees who were over 50 in 1999 and those already in the civil service when the new system was launched in early 1999 did not have a choice, but stayed in the existing system. Individuals covered by the reformed first pillar contribute 12.2% of their gross pay (up to the pensionable maximum of 250% of the average working income).

In the reformed structure, the new second pillar forms the core of the bulwark against future demographic strain. Here private savings are accumulated in personal accounts kept at private pension funds. Membership in this system is compulsory for all employees born after 1969. The working population born between 1949 and 1969 was given the opportunity of making voluntarily contributions. Contributions presently amount to 7.3% of gross pay. Eventually this should enlarge the second pillar to the approximate size of the first.

Savers can choose between currently 17 different licensed pension fund providers. However, the financial transactions are routed through the reformed Social Insurance Authority (ZUS). Employers must remit the entire mandatory contribution of 19.52% (half of which is paid by employers and half by employees) to the ZUS. The ZUS diverts the savings portion (7.3%-points, or three-eighths of the total contribution) to the chosen pension fund. Where employees have not named a fund, the supervisory authority randomly selects o ne for them.

Polish pension funds – a success story

Pension funds are administered by fund companies that have the legal structure of public limited companies. The fund companies are overseen by a public agency, the Pension Funds Supervisory Office (UNFE), and are subject to minimum capital requirements. They invest the pension contributions in the capital market, subject to quantitative investment limits. A maximum of 40% of their accumulated assets under management may be held in shares of listed companies. The upper limit for investment abroad is 5%, although this is to be raised to 30% in the foreseeable future in connection with accession to the EU.

The funded pension plans are run on the defined-contribution principle, i.e. the amount of benefits paid depends solely on the amount of contributions and the realised return on investment. However, performance standards apply. For every quarter in which a fund misses the average two-year return of all funds by more than 50% or by more than 4%-points, the fund company must pay the difference. The attractiveness of the compulsory pension scheme is boosted by the incentive of deferred taxation (i.e. EET principle): The contributions come from pre-tax (i.e. tax exempt) income. The returns earned by the pension funds through investment of their assets are also exempt. Only at the disbursement stage are the pensions taxed.

Probably as a result of the attractive terms, the compulsory pension fund system appears to be very popular with Poles. The interest in participation was high from the start and considerably surpassed expectations. As long ago as the end of 1999 the ZUS registered about 10.5 million members, compared to an expected 7 to 8 million. Coverage thus reached slightly more than 90% of the country’s eligible population of 11.5 million. Of the 7.7 million employed in the middle age groups who had the right to join voluntarily, more than 85% opted for this new multi-pillar pension system. At the end of last year the total number of insured stood at just over 10.6 million individuals. At that point the pension funds had PLN 19.4 billion (EUR 5.54 billion) in assets under management, equivalent to almost 3% of GDP. By the end of 2002 this figure is projected to climb to about PLN 35 billion (EUR 8.75 billion). In spite of the vigorous growth in accumulated contributions, the number of certified pension funds has fallen from 21 at the end of 1999 to 17.

Compared to the largely attractive terms offered by the second pillar, the rules for the third tier (voluntary personal retirement savings) are relatively restrictive. It encompasses personal-saving products such as annuity insurance and investment saving accounts as well as voluntary occupational pensions. However, the voluntary occupational schemes in particular, which among other vehicles can involve direct insurance (group insurance policies) and occupational pension funds, is less attractive because of its restrictive rules. Thus, companies can offer their employees only uniform pension plans. Worse, there is little tax incentive to participate. Although employer contributions to the voluntary tier are to a limited extent considered as tax-deductible business expenses and are not subject to other social security contributions, the entire contributions are subject to the employees’ income tax.

Need for further reform in Poland

While the mandatory pension system has on balance been a success, the fund companies in particular point out a number of flaws in the system. This includes considerable payment arrears on the part of the government’s social security agency, an indication of problems – especially significant during the start-up stage – in the administration and distribution of contributions from the large member base. Thus, there is still a host of inactive accounts to which no contributions have ever been credited. Moreover, the performance standards are seen as req uiring reform, as the quarterly (rather than annual) measurement of performance increases costs and can cause funds to adopt too short an investment horizon.

In addition, the privatisation of state-owned companies and their flotation on the stock market is to continue to be fast-tracked in order to expand investment opportunities for the funds’ burgeoned assets.  

It would also be appropriate to improve the legal environment for the third pillar, especially the tax regulations. According to projections by the World Bank, the implicit significant downsizing of the first tier means that, in the long run, the total level of mandatory pensions will hardly exceed 62% of final income. Voluntary retirement saving thus assumes a key role in preserving a satisfactory standard of living for retirees. This makes it desirable to institute reforms such as the extension of deferred taxation to include private pensions.

Hungary: Status of the new system still uncertain

Hungary was the first CEE country to set up a three-tiered pension system. The legal foundations of this structure became effective in 1998, one year before the corresponding reform in Poland. However, the modernisation of Hungary’s old-age pension system is not progressing at a steady pace. Important parameters for the design of the public pillar and the new second pillar have been modified in the past several years, in some cases more than once. In particular, it is not yet clear whether membership will be made mandatory again for all new entrants into the workforce, as the new government seems inclined to do (membership had initially been compulsory but this approach was suspended).

For Hungary’s public pension system, the reform law of 1998 specifies a reduction in scope of the legacy pension scheme and regulations for the new first pillar, which likewise operates on a pay-as-you-go basis. Until 2013 the new first pillar is governed by provisions derived largely from those for its old counterpart. From 2013 the first pillar is then to be restructured and turned into more of a national defined contribution system. The restructuring and pruning of the public pension system is being achieved mainly by the following measures. First, as long ago as 1996 it was decided that the retirement age would be raised incrementally from 60 years for men and 55 for women to a uniform 62 years (by 2008). Second, the method for indexing current pensions was revised. Since 1999 they are no longer indexed to the trend in employment incomes, but according to a composite index that is based (since 2000) half on the increase in the net wage bill and half on retail prices. Third, from 2013 a new benefit formula will be used to calculate the pensions by which the relatively generous public pension benefits are to be pared down and the degree of equivalence strengthened.

Until 2013 the pensions in the new first pillar will amount to 75% of corresponding pensions from the ” old” pension system. The difference of 25%-points is to be largely made up through the second-pillar pension (also see below). From 2013 the pensions from the first pillar will be computed by a modified method: They will then be determined only by average gross income over the beneficiary’s working life and the number of years of contribution. Pensions will increase by 1.22% of average pay for every year of contribution. The approach is to ensure that, after 40 contributory years, the first-pillar pension will amount to 74% of the old-style pension.

The intended relative weighting of retirement income from the new first and second pillar in the ratio of three to one is also reflected in the initial model for the distribution of mandatory contributions. The total contribution was set at 32% of gross employment pay (up to the contributory ceiling of 200% of average pay). In 1998, 26%-points flowed to the first pillar and 6%-points to the second. This ratio was to be adjusted to 24-to-8 by the year 2000 (the three-to-one ratio). Instead, however, the adjustment took a different course: The very high employer contributions of 24%, which are channelled exclusively to the public (i.e. first) pillar, were gradually reduced by a total of 6%-points. The mandatory contribution level at present is thus 26% (18%-points from the employer, 8%-points from the employee). For employees covered by the new system, 20% (the employer contribution plus 2%-points of the employee share) go to the first pillar and 6% to the second. The employee has the option of raising this contribution rate to 10%.

The complementary second pillar was originally intended to be a compulsory scheme for all new entrants into the working population. Those already in work when the reform began initially had the choice of joining voluntarily by the end of August 1999. The architects of the reform envisioned that this would involve primarily people less than 40 years of age. In fact, the law set a maximum age of 47 years for voluntary membership (which on average would have translated to a minimum contribution period of 15 years). However, Hungary’s highest court declared the age limit to be unconstitutional. As a result, many late-career individuals joined the new system whom its rules do not fit well. This was one reason why last year the then centre-right government made sweeping changes to the terms of membership.

Thus, at the end of 2001, membership in the system was made optional for the time being. As well, for a limited period of time members are allowed to opt out of the scheme: Until the end of 2002, voluntary and formerly-compulsory members may quit the two-tier system and return to the traditional pension scheme. By opening up this return route, the government of the day intended in part to stabilise the financing of the old public tier. To date, however, only relatively few individuals have used this opportunity (about 1½% of members as of the end of 2001). The socialist government that has been in office for several months seems to plan to make membership compulsory again.

Pension funds successful in Hungary as well

The second pillar operates through pension funds. Unlike Poland, though, these are run not by public limited companies, but by (pensionfund) mutual associations, or not-for-profit co-operatives, that do not need to carry share capital. Each member is a co-owner of the respective fund. Pension funds may be established by (sizeable) companies, trade unions, associations and third-pillar pension fund providers (the latter are typically banks and insurance companies). Banks and insurers dominate today’s market for pension funds. Pension funds may be of the closed type, i.e. serving a defined or limited membership (of at least 2000 individuals), or they may be open. The fund is in principle selected by the employer, who is also responsible for transferring the contributions to the fund. The employer must offer a fund, but the employees need not accept this offer. The fund’s administration and especially its asset management can be outsourced.

Investment is subject to quantitative rules. At least 10% of assets must be invested in vehicles of the lowest risk category; no more tha n 30% may be held in the riskiest asset class. A maximum of 50% may be invested in domestic equities. Foreign assets are capped at 30%, with at most 10%-points to be invested in non-OECD countries. Even tighter restrictions on investments abroad were lifted at the beginning of 2002. Compliance with these and other regulations is monitored by a public supervisory authority.

As in Poland, the Hungarian pension funds too basically offer pension plans only on a defined-contribution basis. Benefits are generally paid as annuities, with identical life tables required to be used for men and women. However, part of the accumulated capital may also be capitalised when the pension to be financed with it exceeds a specified threshold (set at 50% of the pension from the first pillar).

Neither the public system nor the two new private pillars are currently taxed. Contributions come out of pre-tax income; pension benefits, including all payments from the second pillar, are tax-free. This blissful state of affairs will come to an end in 2013, when a new system will be launched wherein benefits will be taxed on disbursement (EET principle).

In order to ensure the greatest possible security of the funded pensions, the supervisory authority monitors the funds’ performance. Its benchmark is the index for long-term government bonds. Funds whose investment return fails to reach 85% of this index for three years in succession can expect a government inquiry. The funds must also build a capital reserve to buffer swings in performance. Furthermore, the funds are under a very far-reaching performance obligation: The pensions which they disburse are to reach a minimum of 25% of a comparable public pension benefit on retirement at the statutory pension age. In particular, after 15 contributory years the pension paid by pension funds, when added to the first-pillar pension, is supposed to amount to at least 92% of a corresponding old-style pension. In contrast to Poland, the Hungarian pension funds or asset managers cannot be held liable for missed targets, as their corporate assets are not kept segregated from the members’ assets.

For this reason the second-pillar pension funds are required to reinsure themselves with a guarantee fund. This fund of last resort is to guarantee the minimum benefits. In other words, it is to top up pensions due from those pension funds which do not meet the performance standards. Most notably, it is to step in if a pension fund becomes insolvent. The contributions to the guarantee fund come directly out of the individuals’ contributions. In addition, the government originally guaranteed the minimum pension, but the Hungarian parliament rescinded this commitment in 2001. This calls the already-dubious guarantee structure into even greater doubt.

Although work is still in progress to give Hungary’s second pension pillar its final form, the system can be counted a success. At the end of 2000 the second-pillar funds already boasted some 2 million insured members, more than 90% of whom had joined voluntarily. One year later the member base of 2.25 million had grown to more than half of the country’s active workforce. The assets managed by the compulsory pension funds as of the end of 2001 amounted to HUF 291 billion (EUR 1.19 billion), or slightly more than 2% of GDP. The number of registered pension funds has declined from an initial 31 to 22.

After just four years, the fast-growing pension funds of the second pillar thus already surpassed the third-pillar pension funds both in terms of membership (1.15 million) and assets under management (HUF 283 billion, or EUR 1.15 billion). This is all the more remarkable as pension funds in Hungary have been licensed under the voluntary personal pension scheme since 1994. These pension funds operate in a legal and regulatory framework similar to that which applies to the secondpillar funds. However, the portfolio regulations are more liberal (for example, there is no m inimum level of low-risk investments, and the upper limit for the equity weighting is 60%). Another major difference is that the voluntary funds have never been required to issue performance guarantees and thus do not need to pay into the guarantee fund.

The number of funds in the third pillar decreased from 270 in 1994 to little more than 100 by the end of 2001. This reflects a prolonged market consolidation that has seen most of the initial array of very small and less efficient funds assimilated by larger entities in take-overs and mergers.

The third pillar of pension provision, which inter alia can employ insurance products, has benefited in past from very easy tax treatment. Not only were the personal pensions collected tax-free, but in some cases pension contributions were actually subsidised4. As this took a large toll on the budget, some early changes have already been made, with more to follow: From 2013 the pensions in the third pillar are also to be taxed.

Realignment of Hungarian pension system not yet complete

While it is off to a good start, the new system of retirement income provision in Hungary will require substantial further effort to make it viable for the long term. To be sure, politicians feel that the 1998 reform has ushered in a successful overhaul of the pay-as-you-go public scheme. In fact, official forecasts predict surpluses from 2008 to 2050. Yet, these projections are based on the very high contribution rates that were set initially, but which are quite unlikely to be sustained for the long term. Moreover, the official figures are predicated on optimistic assumptions concerning the trend in employment which the OECD regards as unrealistic. The OECD’s own predictions paint a different picture: Even on the premise of a still very high total contribution rate of 29%, the pay-as-you-go system runs into persistent deficits that mount sharply from about 2030 and grow to roughly 1.5% of GDP per year by 2040. The public pension system in its planned proportions can thus hardly be financed on a stable basis for the long run. This urges further streamlining measures. For example, the OECD proposes another increase in the retirement age to 65 years.  

Also, the status of the new two-tiers system ought to be clarified as soon as possible. In its present state it is not tenable for long, as that would imply the lasting co-existence of two public pay-as-you-go tiers. This would be inefficient even with both systems administered by the same government agency. It would thus seem appropriate for Hungary to make membership compulsory again for young workers, although this is by no means a prerequisite for building a second pension pillar. These considerations aside, it will be important to further buttress the new second pillar. A swift reinstatement of the now-shelved increase in the contribution rate from 6% to 8% would go a long way towards achieving this. If the status quo were maintained, on the other hand, the pension funds should find it difficult to assume their significant assigned role of safeguarding slightly more than one-quarter of the retirement income of their members. This is especially true with respect to the many older members with a relatively short contribution period. Further development of the Hungarian capital market would also be helpful. Much like Poland, Hungary suffers from a dearth of sufficiently liquid shares of sizeable domestic firms in which pension funds could invest. This exacerbates the benchmark-induced trend towards a high weighting of domestic bonds in pension fund portfolios (2001: almost 87%). However, the supply of government and corporate bonds too can barely keep pace with the expanding demand.

Czech Republic: Three-pillar system thus far exists only on paper

At the moment the Czech Republic makes do with a two-pillar pension system. Alongside the public scheme, a voluntary personal pillar supported by government in centives figures prominently. Like its counterpart in Hungary, the Czech pension system remains in the midst of a stage of reform and tinkering.

Membership in the state pension scheme is compulsory for all employees, many agricultural workers and some self-employed. To finance the benefits (by the pay-as-you-go method), contributions are levied in relation to employment income. The contribution rates (as of 2001) are 6.5% of contributory income for employees (one-quarter of the total contribution per person), 19.5% for employers (the other threequarters) and 26% for the self-employed. The statutory retirement age is currently still low, but under the pension reform is being raised in increments from 1995 to 2007: from 60 to 62 years for men and to 57 to 59 years (depending on the number of children) for women. The insured become eligible for a ” full” pension (50% of average past pay) after 25 years of contribution, with additional working years translating into a higher pension. Contributions and benefits are tax-free.

The Czech pension insurance system is highly redistributive in nature6. At first it also provided considerable incentives for taking early retirement, and to a lesser degree it still does today. Due to resulting distortions coupled with a rise in unemployment, the financial position of the Czech system – which after the first reform measures had initially been comparatively good – deteriorated visibly in the second half of the 1990s. The government has begun to respond to the need for action and launched a number of measures to strengthen the link between levels of contributions and of benefits (e.g. reducing non-contributory periods; increasing the benefit penalties for early retirement based on actuarial rules). Further planned reforms are to create an occupational pension fund tier. In the event that these plans are implemented, one reform option being contemplated (for which the blueprint already exists) is to redesign the public pension pillar as a notional defined-contribution (NDC) system.

A second pillar of funded provision does not exist in the Czech Republic at present. However, in 2001 the government presented a draft bill that would introduce a system of voluntary occupational pension funds. Under the proposed law, the prospective occupational pension funds in the Czech Republic would satisfy most international standards, including those set out by the planned EU pension fund directive7. This entails, among other features, the use of individual retirement savings accounts in which to hold the pension funds’ assets, and strict separation of fund assets from the capital of the provider institutions, especially from that of the sponsor companies. Both employers and employees are to make contributions. The pension funds are to pay benefits on a defined-contribution basis, with employees able to choose among pension plans with different risk-return profiles. According to the proposed legislation, benefits would not be paid until retirement age. In order to protect beneficiaries’ interests, there are plans for strict supervision by the government. In a departure from international standards, however, the planned occupational pension funds would work in an insurance-like manner. Thus, they would have to offer a minimum rate of return on the accumulated contributions and as a rule would be permitted to disburse benefits only in the form of life annuities.

The new occupational funds would function alongside the already well established pension funds of the personal pillar. These third-pillar pension funds are operated by public limited companies, which must have a certain minimum share capital. However, this capital is not held separately from the pension fund assets. The personal pension funds often only defined-contribution plans, without having to provide guarantees. Benefits may be paid out as life annuities or as a complete lump-sum disbursement of the accumulated capital. Akin to the situation in Hungary, the personal pe nsion funds in the Czech Republic enjoy generous support through tax-free benefits and government subsidies for contributions (both employee and employer contributions receive a tax break).

The personal retirement savings pillar probably owes much of its success to the subsidisation by the state; 20% of the average contribution represents public money. About 2.5 million Czechs (out of a working population of 4.5 million) currently have a personal pension plan. The assets administered by the pension funds at the end of 2000 ran to more than CZK 40 billion (EUR 1.2 billion), or about 2% of GDP. In part the popularity of the personal pension funds is also explained by the fact that, thanks to their brief minimum contribution period of five years and the option of withdrawing the accumulated capital from age 60, many older Czechs used them as savings vehicles.

Speedy implementation of reform plans would be best

Notwithstanding the success of the personal pension funds, the restructuring of the Czech Republic’s pension system should thus not be put off. Especially in the Czech Republic, whose population will age more quickly and markedly than those in other accession countries, it is high time for reform. The lead time required to build sufficient capital for funded pensions should not be underestimated.

The establishment of a second, supplementary pillar is all the more crucial as the first pillar in its present structure is in danger of toppling. For the past several years, large amounts of funds have already had to be diverted from the national budget to finance public pensions. In absence of further reforms, these subsidies would have to mount sharply or the already high contribution rates would need to be raised strongly. The Czech ministry of finance predicts that by 2030 alone, contribution rates would have to climb from currently 26% to 44% if the public pension fund is to remain in equilibrium under present conditions.

Conclusion

The pension reforms in the large accession countries – like in many CEE jurisdictions – generally revolve around setting up a three-pillar pension system consisting of a lean public tier, a parallel pillar of high-performing (occupational) pension funds, and a third tier of personal retirement savings. If one looks beyond this broadly similar goal, however, the various countries have different ideas about the relative importance and architecture of the individual pillars. As well, reform is not progressing at the same speed throughout the region. Poland has gone furthest, establishing a pension system whose structure may serve as a model to guide reforms in other countries, including the present European Union. The Polish reform is characterised first of all by the consistent conversion of the first pillar into a lean defined-contribution system. This satisfies the demographic imperatives more effectively than the still relatively traditional public schemes in the Czech Republic and Hungary. In their present form these latter two systems will hardly be sustainable. The strongest sign of this is that the pension contributions in both countries still amount to a very high 26% (for employees with average income). Without the prospect of a substantial reduction in this burden, such contribution rates must sooner or later become a serious obstacle to economic growth. This threatens to lead to a vicious circle, as steady economic growth would in turn facilitate the provision of future retirement incomes.

The necessary further scaling-back of public pension systems should be accompanied by redoubled efforts to accumulate private capital for funded pensions. One argument for this approach is the need to top up declining levels of public pension benefits with the aid of private schemes. But more private retirement saving through pension funds is also beneficial from a macroeconomic standpoint. As major investors, pension funds can play a big role in impro ving the efficiency of financial markets, particularly in the accession countries with their still-junior equity and bond markets.

The new architecture of pension provision in Poland with its relatively strong second pillar can thus be considered comparatively resistant to demographic shock. Poland has opted for a compulsory second pillar with significant mandatory contributions (7.3%). In Hungary, by contrast, the final legal status of the new second-pillar pension funds is not yet decided, and the standard contribution rates are lower (6%). In the Czech Republic it seems that the notion of a compulsory private scheme in particular is meeting strong resistance. On the other hand, membership need not be compulsory if a high degree of coverage can be attained by other means, for example through tax incentives.

Relatively generous tax privileges have been instrumental in making the third pillar a popular success in the Czech Republic and Hungary. Such personal saving for retirement is also highly important and should not only be maintained, but if possible expanded further. This begs the question of how capable fiscal policy in these two countries is of promoting not only the third, but also the second pillar so as to maximise its use by the population.

Given its present condition, the Polish pension fund market in particular should harbour much further potential in the years ahead. This is indicated by the more than 10 million participants in the second pillar and the Polish economy’s continued favourable medium-term outlook for growth. Both factors point to sustained expansion in the inflow of contributions. Experts estimate that by the end of 2010, assets managed by pension funds in Poland could grow from currently over PLN 30 billion to as much as PLN 200 billion (25% of present GDP). In Hungary too, the pension fund industry can be expected to grow briskly if reforms are instituted without delay.

To be sure, dynamic growth in pension fund assets is likely to put the absorptive capacity of financial markets in the accession countries to a severe test. That and the value of diversification make it desirable for pension funds to be allowed to invest part of their assets abroad. Some accession countries have recently widened the legal scope for foreign investments or are about to do so. Most of all, however, accession to EMU (when it comes) will trigger a quantum leap. The acceding countries’ pension capital will then have access to the large euro area. Equity investors in particular will only then have at their fingertips a full universe of investment choices, with the corresponding opportunity to diversify.

For more DB Research analyses see the

Deutsche Bank Research website.

---