The Baltic countries have made considerable effort during the course of the transition to ensure the long-term sustainability of their pension systems and the adequacy of retirement income. As described above, these efforts initially involved steps to shore up the pay-as-you-go system inherited from the Soviet period, including by reversing the early expansion of this system, increasing retirement ages to more sustainable levels, and linking more closely lifetime contributions and retirement benefits. While these steps met with some success in improving the financial health of the Baltic pension funds, gains were partially undone by subsequent ad hoc benefit increases. Further, the adverse demographic trends facing the Baltic countries led them to consider more fundamental pension reforms, in particular, the establishment of a three-pillar pension scheme, including a mandatory, fully funded, defined contribution second pillar.
While a move toward a fully funded pension system can potentially make an important contribution to the objectives of pension reform, such a change is neither necessary nor sufficient to meet these goals. As noted in the paper, the existing PAYG pension system can, at least in theory, be made sustainable by an appropriate adjustment of payroll tax rates and expected lifetime pension benefits, although the average replacement rate implied by such changes may well be fairly low, reflecting the expected demographic developments. The introduction of a fully funded pillar can help the Baltics address this demographic challenge only to the extent that this reform allows an increase in their long-term sustainable growth rates, either through their impact on savings and capital accumulation or by enhancing labour market efficiency. Whether this can be achieved will largely depend, in turn, on detailed decisions regarding the implementation of the second pillar, most of which remain to be made. Moreover, some benefits being pursued by the introduction of a fully funded scheme could also be pursued through other mechanisms; for example, a move from direct to indirect taxation could serve to increase private savings.
As the primary benefit of the introduction of a fully funded element to the pension system is the possibility of increased savings, the transition costs of this reform should be financed, to a substantial degree, by fiscal adjustment. The paper argues that the alternative of financing reform by increasing public sector debt is unlikely to generate additional savings for the economy and, as such, would not contribute to the objectives of pension reform. However, the prefunding implied by fiscal adjustment—either in the form of higher taxation or cuts in government spending—is likely to be a difficult political step, as it can be seen as imposing an inequitable burden on the current generation of workers. This difficulty has undoubtedly influenced the decisions of both Latvia and Estonia to begin with a rather small second pillar, limiting both the potential costs and benefits of the reform.
The paper has a number of other implications for the optimal design and implementation of a fully funded pension pillar in the Baltics, including that:
-
Given the need to generate adequate returns on pension fund investments and encourage risk diversification, any fully funded pension should be allowed to invest abroad. Given that the Baltics are small open economies with access to international capital markets, attempting to stimulate domestic capital markets through tight limits on investment abroad is unnecessary and could well be counterproductive.
-
Any guarantees—implicit or explicit—on the returns of individual pension funds or accounts should be strongly resisted as they would introduce potentially severe problems of moral hazard and raise administrative costs. Regulation of pension funds or accounts should focus primarily on ensuring transparency in operations and full disclosure.
-
Indexation should allow retirees to partially share in the expected sizable productivity gains in the Baltics over the medium term, allowing both real increases in retirement income and a decline in the first pillar deficit. Both Latvia’s gradual move from CP1 indexation to wage indexation, as well as Estonia’s proposal to index to a weighted average of the wage fund and CPI, would appear to meet these objectives.
Finally, the success of the three-pi liar scheme will also depend on the ability of the three countries to continue to strengthen the long-term finances of the pay-as-you-go pillar. Key steps in this regard will include increasing the retirement age over time and resisting future pressures for unsustainable increases in benefits.