This review provides a detailed analysis of the different components of the Czech pension system. It assesses the system according to OECD best practices and guidelines, and draws on international experiences to make recommendations for improvement.

While various changes have taken place over the last twenty-five years, the architecture and principles in the design of the Czech pension system have been mainly unaltered since the 1995 Pension Insurance Act. The system works relatively well in ensuring net replacement rates close to the OECD average for average-wage workers. Moreover, there is a high level of redistribution within the pension system, at least within generations, leading to a very compressed benefit structure. Replacement rates are high for low earners while high earners receive very low internal returns on their contributions. For the latter, the boundary between contributions and taxes is blurred.

Overall, the main weaknesses of the system are the following. First, the average income of older people relative to that of the total population is relatively low despite high contribution rates. Second, the build-up of pension entitlements is very complex, making it difficult for contributors to understand their accrued rights (and anticipate their future pension level) and for the administration to manage the system. Third, ageing pressures will lead to financial imbalances driven by higher pension spending. Fourth, funded private pensions play a limited role, which prevents reaping the benefits of a diversified structure.

The agenda for a broad and substantial pension reform should thus include several components. The way to calculate pensions should be simplified so as to identify entitlements when they accrue, enable workers to better anticipate their retirement income and make the management of the system easier and more precise. In addition, eligibility to an old-age pension should be possible for people with shorter contribution periods. Pension financial prospects should also be improved. Moreover, the current financing structure exclusively relies on contributions while some pension components might be better financed by taxes. Finally, voluntary private pensions should be better designed to boost their capacity to complement public pensions.

The review offers various policy options to improve the Czech pension system. The key recommendations are:

• Simplification. Calculate earnings-related entitlements using a constant effective accrual rate across earnings levels (up to a ceiling). Adjust then the basic pension benefit to achieve redistributive objectives.

• Eligibility. Drastically reduce the minimum number of years required to be eligible to both the basic pension and the earnings-related component at the statutory retirement age, and make the basic pension benefit proportional to the validated contribution period.

• Financial sustainability. Implement the legislated increases in the retirement ages and their convergence between men and women, and link the unified retirement age to gains in life expectancy. Raise the minimum early retirement age and adjust early retirement ages to life expectancy as well.

• Financial structure. Consider financing some redistributive components of public pension expenditures by taxes to improve pension finances, boost pensions for people earning more than the average wage, lower mandatory contributions or help finance funded pensions.

• Private pensions. Improve the performance of pension funds by encouraging or nudging participants to switch to funds without capital guarantee and by promoting the access to an appropriate default investment strategy, as a pre-requisite to make them contribute more to a well-functioning voluntary system.

These are now presented in greater detail.

• Calculate earnings-related entitlements using a constant effective accrual rate across earnings levels (up to a ceiling). Adjust then the basic pension benefit to achieve redistributive objectives.

This is the main feature in the recommended simplification of the benefit calculation. It will allow to clearly identify accrued entitlements earned for a given contribution period rather than waiting for the end of the career to know what was earned in that period. This will also pave the way to creating individual accounts where pension entitlements steadily build up, thus enabling people to better anticipate their future pension level. At the same time, the management of the system will be facilitated as entitlements when accruing do not depend any more on the future part of the career.

The current core formula, although much more complicated in its details, is very close to the sum of a benefit equal to three times the current basic pension level and accruals based on a constant rate of 0.39% across all earnings levels up to the contribution ceiling. These parameter values would preserve the strong progressivity of the current system. The redistribution level is a political choice, however. For a given level of spending, a less (more) progressive system is obtained from a lower (higher) basic pension level combined with a higher (lower) accrual rate.

• Drastically reduce the minimum number of years required to be eligible to both the basic pension and the earnings-related component at the statutory retirement age, and make the basic pension benefit proportional to the validated contribution period.

The Czech Republic is an outlier in requiring that either 30 years are contributed or 35 years are validated to be eligible to an old-age pension at the statutory retirement age. This applies to both the basic and the earnings-related components. Currently, those who have contributed or validated fewer years have to wait until five years after the retirement age to claim an old-age pension, provided in addition that they have contributed at least 15 years or validated 20 years. This raises the risk of having to retire very late to access a small pension or to have contributed for nothing.

So far, such a long required period has not raised much concern thanks to high recorded employment in former Czechoslovakia and generous validation of non-contributed periods. However, the transformation of the economy from the 1990s, the welcome tightening of some instruments to validate pension entitlements as well as labour market trends are likely to raise the share of older people not covered by old-age pensions or with short contribution periods. Such a long required period might thus increase the risk of old-age people facing low income in the future. Ideally, eligibility for the basic pension should be possible with only one year of contribution, with the benefit level being pro-rated according to the length of the contributed period.

• Implement the legislated increases in the retirement ages and their convergence between men and women, and link the unified retirement age to gains in life expectancy. Raise the minimum early retirement age and adjust early retirement ages to life expectancy as well.

Population ageing pressure will translate into increased pension spending from about 2030 and thus lead to a deterioration of pension finances. Social security contribution rates are relatively high and the income of people older than 65 relative to that of the total population is relatively low. Increasing effective retirement ages is the key instrument to deal with the financial pressure for several reasons, even if additional measures might be needed to close the financing gap. First, the male statutory retirement age will remain significantly below the OECD average. It should thus be raised to 65 years in 2030, as planned, and then linked to life expectancy as is the case in six OECD countries. One key question is the pace of adjustment in this link, which should be decided in conjunction with other instruments that contribute to ensuring financial sustainability, such as the accrual rates, the basic pension level, contribution rates and perhaps tax sources. The initially legislated pace of increase in the retirement age, which was decided in 2011 and cancelled in 2016, was too fast, implicitly transmitting about 2 years of increase in the retirement age for each year of improvements in life expectancy at age 65. For example, assuming retirement takes place at the statutory retirement age, if about two-thirds of changes in life expectancy were transmitted to increasing the retirement age, the balance between time spent working and in retirement would be more or less stabilised.

Second, the Czech Republic is among the few OECD countries that still have gender-specific retirement ages. The planned convergence of retirement ages between men and women by 2037 is welcome. Third, almost one-third of people retire before the statutory retirement age in the Czech Republic. With the planned increase in retirement ages, early retirement will still be possible from age 60. The risk is that too many people might retire very early, generating potentially low pensions. The minimum age of early retirement should thus at least be increased in line with the statutory age, reaching at least 62 in 2030, and then be linked to life expectancy.

• Consider financing some redistributive components of public pension expenditures by taxes to improve pension finances, boost pensions for people earning more than the average wage, lower mandatory contributions or help finance funded pensions.

In the Czech Republic, redistribution takes place exclusively within the pension system as all pension revenues come from contributions levied on wages. Social security contributions amounted to almost 15% of GDP in 2018, out of a total of general government tax revenues of 35% of GDP, which is closely associated with the high labour tax wedge. Many countries finance part of pension spending through taxes. This helps lowering the labour income tax wedge – the difference between wage cost and take-home pay – which tends to improve economic performance by raising employment and competitiveness. The high tax wedge might become more detrimental to the Czech Republic as wage levels are catching up towards EU averages. A discussion might thus be welcome about the opportunity to shift – on top of health insurance contributions which generate benefits that are not earnings-related - part of the financing of some redistributive instruments, such as credits for non-employment periods or basic pensions, to general taxes. Using additional tax revenues might allow to increase accrual rates, in particular for high earners who have low replacement rates, or to leave more space to develop funded pensions.

• Improve the performance of pension funds by encouraging or nudging participants to switch to funds without capital guarantee and by promoting the access to an appropriate default investment strategy, as a pre-requisite to make them contribute more to a well-functioning voluntary system.

There is a need to improve the net performance of pension funds before encouraging people to contribute more to the system in order to increase their future pension income prospects. Three-quarters of participants are currently in transformed funds that have to provide an annual non-negative return guarantee and therefore invest very conservatively, thereby achieving low performance. The improvement of the net performance could be achieved by encouraging more participants to switch to participating funds, given that these funds have more flexibility to pursue growth investment and have thus far achieved higher returns on average. In addition, the Czech authorities should promote the access to an appropriate default investment strategy as many people may be unwilling or unable to choose their own strategy. Life-cycle investment strategies can be well suited to protect people close to retirement from the impact of extreme negative shocks in financial markets, while allowing younger participants to invest more in risky assets that yield higher expected returns over the long term. This could be built through the combination of several participating funds, the weights of which would vary as people get closer to the retirement age.

The other main recommendations to improve the public pension system are the following.

• Remove the double penalty related to the impact of non-validated periods on pension benefits. Some non-contributed periods generate no entitlements and, in addition, lower both already accrued and future entitlements by lowering the reference wage. When a period is not validated it does not generate, by definition, any accrual. However, the Czech system is unusual in lowering, on top, both already accrued and future entitlements by counting wages as zeroes for these periods in the calculation of the reference wage. In the current system, non-validated periods should not be accounted for in the calculation of the reference wage. The reform proposed above which is based on a constant accrual rate for each period would solve this problem.

• Eliminate age-specific credits for unemployment periods. Long periods are currently validated for employment breaks related to childcare. For unemployment, the period that is credited for pension entitlements depends on the length of the unemployment-benefit period plus some extra months for periods without unemployment benefits, both of which are more generous after age 55. Age-specific measures might contribute to stereotypes about working at older ages, which ultimately lower employment prospects at the end of the career and encourage early retirement.

• Raise the contribution base of the self-employed to better harmonise contributions and entitlements between employees and the self-employed with similar earnings. The self-employed accrue less pension entitlements as their contribution base is low. In line with the proposal from the Commission on Fair Pensions, the contribution base should be substantially increased from the current level of 50% to 75% of profits, ensuring a better harmonisation with the equivalent of gross wages. Such an increase would also reduce inequity between different forms of work and limit the misuse of self-employment to reduce labour cost, which might develop in the future as new forms of work gain in importance.

• Avoid encouraging self-employment through lower contributions, which generate lower pension entitlements. If there is a political objective to support self-employment, make any subsidy explicit by financing the contribution gaps compared with employees having similar income through general taxes. The increase in the contribution base may be perceived as generating a too high burden of contributions paid by the self-employed. However, lower pension contributions generating lower pension entitlements should not be used as an instrument to promote self-employment. The cost of any support to self-employment should be made transparent by subsidising part of these better harmonised contributions through general taxes.

• Index pensions in payment to price inflation. Pension replacement rates are high for low-income earners in the Czech Republic. Future pension spending could be reduced if needed by indexing pensions in payment to prices, which would still preserve the purchasing power of pensioners.

• Separate the account of the pension system and the government budget more clearly. There is no clear financial separation between the central government budget and the pension account. Net pension surpluses have been transferred to the government account without recording net accumulated revenues. Likewise, in the future, public transfers will have to cover any pension deficits, without properly accounting for the pension-related financial transfers across generations. The effective separation of the pension account from the central government budget would greatly improve transparency and strengthen the credibility of pension management, especially given financial strain from ageing trends. This would in particular make the transfers from the government budget to cover family- and unemployment-related pension components more transparent.

• Index the social assistance benefit level to nominal wage growth. Social assistance provides an income guarantee (living minimum), which is very low in comparison to other OECD countries and might not effectively prevent relative income poverty in old age. Its level is nominally fixed, and it substantially declined from more than 30% of the average wage in 1991 to 10% in 2019.

• Introduce a higher social assistance benefit level for people reaching the statutory retirement age. The social assistance scheme does not provide specific benefits to older people. Providing such benefits would open the possibility to provide a higher benefit level for those that have reached the statutory retirement age. While minimum income schemes might generate work disincentives, such concerns are of less importance after the statutory retirement age. Having no specific safety net for older people tends to limit the old-age benefit level as any increase applies to the whole eligible population and therefore generates a high cost.

Additional recommendations to strengthen the role of the funded pension system are the following.

• Better align fees charged to participants with the costs incurred by the pension management companies. There is no consensus as to whether fee caps are appropriately defined. The Czech authorities should analyse the cost of investing in different asset classes to form a view of whether the current caps are in line with the objective of having well-diversified portfolios to reach better risk-adjusted returns. Once this is done, a regressive scale for management fees could be adopted to pass on economies of scale to participants as assets under management grow.

• Encourage participants to contribute more. Contributions to supplementary pension schemes are generally too low to help individuals complement their state pension significantly. Many participants only contribute the minimum incentivised amount and less than one quarter receive employer contributions. Redesigning some elements of state support could improve incentives to raise voluntary contributions. Employers could also play a bigger role, by contributing more systematically on behalf of their employees and implementing an automatic escalation mechanism, whereby employee contributions would increase gradually in line with pay rises up to a pre-set maximum. Moreover, providing information about expected benefits from the entire pension system could help people identify how much they need to contribute to the funded scheme in order to achieve a given target retirement income.

• Lengthen the minimum saving period. The minimum saving period of five years to withdraw retirement benefits and keep the state financial incentives is too low. Countries with comparable schemes have a minimum saving period of at least 10 years.

• Consider introducing automatic enrolment. Participation is the lowest for people aged under 30. People tend to delay enrolment in funded pension arrangements because of procrastination and inertia. Automatic enrolment into an occupational pension plan or a participating fund would increase take-up among younger people. Appropriate default contribution rates and investment strategies should be defined.

• Extend the take-up of products providing lifelong retirement income. If the previous measures are implemented successfully and the level of assets accumulated at retirement becomes significant, more rules should be put in place to ensure that people actually use their retirement savings as a complementary source of income during retirement. This would imply discouraging the lump sum pay-out option and increasing the attractiveness of lifelong retirement income products.