The meeting drew on a number of papers which were presented by specialists from the Social Security Department at ILO headquarters and the ILO-CEET. Several core issues of the present reform of pension systems were discussed, including: social protection and pension systems in Central and Eastern Europe; guiding principles for social security; the current debate on social security reform; design of defined-benefit and defined-contribution plans; governance; coverage under social security pension schemes; the quantitative aspects of pension reform; and planning for the social security reform process.

The participants had a unique opportunity to access a wide range of information, not only on technical and administrative matters but also on normative issues which ultimately shape the development of social policy. Particularly appreciated were the papers presented by the ILO and the discussions with participants from other countries. Countries of the region have little direct experience of major reforms to social security systems, and must therefore draw on the experience of other countries and on the latest theoretical considerations in planning both the content and process of reform. That in itself is a difficult process, and is made more so by the turbulent changes occurring in the structure of the economy, in political procedures, in the bureaucracy, and in society as a whole.

The following conclusions can be drawn from the meeting:

• Improvements in governance are the first priority, but the process may be a fairly lengthy one. No specifics were reached at the meeting: this may reflect some uncertainty about what is being done and how to go about it. It may also reflect a reluctance to talk about issues affecting personalities and politics.
• An increased retirement age is also a high priority. This seems to be generally recognised, but difficult politically and socially to implement, and so few countries have gone very far. There are clearly difficulties in selling the idea of later retirement when life expectancy is low and has been declining.
• Implicitly, there may be a need to reduce (nominal) replacement rates and/or to bring them into line with actual replacement rates, which are lower. Again this is difficult politically, and may involve tightening eligibility requirements for disability pensions. An important question is whether this reform can be implemented when the design of the system is being changed.
• The position of retirees at the bottom end of the scale requires protection and reinforcement. The ILO benchmark for lower-end pensions is 40% of the earnings of an average (industrial) worker. It is unresolved whether this issue should be tackled separately (through means-tested social assistance to the elderly) or as part of the general pension scheme.
• The actual design of the general scheme is a second-order question. Some of the mixed systems and new designs being considered appear able to combine some of the best features of traditional schemes (pay-as-you-go or mandatory retirement saving), while avoiding some of their worst aspects. But they may be relatively complex to operate. Good governance is a prerequisite whatever design is chosen. The shift to a different type of scheme may permit other reforms Ë in retirement age, benefit rates, and/or minimum guarantees Ë to be incorporated at the time of the change with fewer political difficulties. The problems of managing and organising the change (especially in relation to cash flow and budget deficits) require careful attention. There was also a feeling that the choice between a pay-as-you go scheme and a mandatory retirement savings scheme is a false dichotomy: the real issues lie in the choice of an optimal mix.

KH,MR