Stephen Browne, UN Resident representative in Ukraine at the World Bank Consultative Group meeting in Paris, March 1995

The debate is not new but it seems to have generated almost emotional heat: social spending in post-communist Central and Eastern Europe is said to be too high and thus-so the horror stories go-cripples desperately needed economic growth, debilitating economic restructuring and recovery. So far the argumentation has been economic, but recently it has acquired a political twist. Too many people in Central and Eastern Europe-so the latest version of the story goes-are dependent on social protection benefits and see the successors of the former communist parties as guarantors of their benefits and so voted them back to power in several countries. 1

This is seen in some quarters as further threat to the political and economic transition. There is little hard evidence to support either of these arguments.2

Let us first look at the facts: is social spending too high in Central and Eastern Europe?

The level of social spending or-better-redistribution is an indicator of social values and choices (i.e. what standard of living a country wants to afford to its less fortunate, sick, disabled, unemployed, young and old citizens): no more, no less. There is simply no economic theory that can stipulate the 'right' level of social redistribution. One can, however, calculate what minimum level of social expenditure is needed in a given society to keep people from poverty-if one can agree on a minimum humane standard of living.

First, (measured) GDP has dropped throughout the region since 1989 achieving compound losses at the end of 1994, in many countries of between 30 and 40%. Some-but certainly not all-of this loss is said to have been compensated by increased activities in the informal economy, though due to its very nature the subject in general, and pre- and post-transition comparisons in particular largely elude statistical analysis. What is often overlooked in this context is that increased informal sector activity is automatically accompanied by a loss in social protection, which is in turn a source of uncertainty and so detrimental to quality of life. What we can observe is that measured poverty skyrocketed as a consequence of the contraction of overall (known) economic activity. However, even here there is inevitably some debate on what absolute poverty line is 'adequate' for a given country. But if one uses consistent poverty definitions (i.e. a poverty line) to compare the development of poverty within individual countries over time, even sceptical observers such as the World Bank are forced to conclude that poverty has multiplied since 1989. It is believed to have more or less doubled in Poland, multiplied in Russia (see table), increased by a factor of 7 in Latvia (the number of those living on incomes below the poverty line has risen from 5 to 35%), and at least quadrupled in Ukraine, to encompass about two-thirds of the population after the hyperinflationary year of 1993 (50 to 100% per month). Estimates for other countries are also in or approaching the 40 to 50% range.3

These figures, combined with rapidly emerging open and hidden unemployment, demonstrate the enormous challenge for the various national social protection systems. Although social protection systems are supposed to protect people in times of economic crisis, as a consequence of which social spending should be expected to be anti-cyclical-that is, spending should increase in times of economic distress-what has happened in Central and Eastern Europe is different

While social expenditure ratios (SERs) at GDP were allowed to rise during the economic downturn, real expenditure levels declined compared to 1989 due to the inflationary impact on wage and benefit levels. Table 2 demonstrates that SERs would have increased much faster than they did, had total social expenditure maintained its 1989 real levels. Hence governments have employed-in part substantial-measures to contain the recession-triggered increase of social expenditure. Even if the crude estimates provided in table 2 must be interpreted with care due to notorious data uncertainties, the combined total reduction of explicit social transfers and consumer subsidies (which must be regarded as implicit social protection expenditure) during the last few years may have reached levels of between 8 and 10% of GDP (which is the equivalent of a typical national pension system) in some countries (for example, Slovakia and Bulgaria). In addition, hidden expenditure on social employment has also been reduced substantially. In total, the social expenditure reduction forced upon governments by economic depression has reached levels which would be totally inconceivable in any OECD country. The overall reduction of real spending, combined with increasing numbers of beneficiaries has pushed down per capita benefits dramatically.

Recent ILO model calculations showed that a country with a 'typical' Eastern European demographic and economic structure and a poverty level of 30% (which must be considered a conservative assumption) would require a social expenditure ratio of at least 25% of GDP to provide social protection benefits above the poverty level.4 The above present national expenditure level thus indicates that room for manoeuvre with respect to cutting social expenditure has probably already been exhausted. Any further benefit level cuts might trigger a serious political reaction.

The public is already dissatisfied with the performance and benefit levels of individual social protection schemes. The most important classical benefits, such as pensions and short-term benefits, usually reach their target population due to the diligence of benefit administrators who keep the systems alive even under adverse social and political conditions, but high and generally increasing dependency rates are compensated by low and decreasing pension levels (achieved by less than full inflation indexing) which in many cases approach subsistence minima (with the possible exception of Poland). In Ukraine, even general statutory maximum pensions are now below the subsistence minimum. The performance of the newly introduced or reactivated benefits is generally less than satisfactory-unemployment and social assistance benefits are only reaching a relatively small proportion of the target population. Beneficiary rates of below 60% in the unemployment schemes and below 50% in the social assistance schemes appear usual. Poverty in families affected by hidden unemployment is emerging as a major social problem in countries with low registered unemployment as the 'traditional' anti-poverty mechanisms offer no replacement for incomes lost due to 'voluntary' unpaid leave. The quality of care provided by underfunded national health care systems is deteriorating and important health status indicators are pointing downward for the first time in decades.

Now let us have a closer look at the 'economistic' arguments. Social security taxes and contribution rates in the region are high and so allegedly reduce the demand for labour and crowd out public funds for investment. But labour costs in the region are extremely low by international comparison, the wage share of GDP being in the order of 30 to 40% or even lower (considerably lower than most Western European countries), so they are not a threat to competitiveness. High social security contributions might indeed trigger increased 'grey market activities' on the margin of the labour market; this problem needs to be addressed. But cutting expenditure is only one solution: the income side of social protection must also be looked at. One reason-probably the most important one for high tax and contribution rates-is that contribution and tax collection remains low.

A substantial part of the financial problems of national social protection systems, as well as the fiscal problems of CEE countries are in fact problems of management and organisational governance. The non-collection of contributions or social protection levies are causing a loss of revenues of between 25% and 30% of currently collected contributions in several countries, for example in Bulgaria,5 (this is no less than the equivalent cost of a fully-fledged national health service). The reasons for the non-collection of social security dues are manifold, ranging from non-payment of wages by enterprises (due to a cash shortage or simple fraud) and the overextensive use of administrative leave by state enterprises to simple administrative and managerial inefficiencies in the social security or tax collecting agencies. Non-collection condemns tax and contribution rates to a vicious circle: the fewer the contributors, the higher the rates must be-the higher the rates are, the greater the incentive for non-compliance, and so on.

Finally, there is the semi-political argument that there are simply too many beneficiaries, notably early retirees who presently wield enormous power being sufficiently numerous to tip the political scales. Apart from the objectionable political connotation (should people receiving transfer incomes have the right to vote?), this is a chicken and egg problem. And again the egg is the economic downturn. Given present benefit levels not many people would volunteer for early retirement or opt for the disability outlet from the labour market. The real reason for the increase in the number of pensioners is neither demographic nor due to the generosity of entitlements, but simply the dramatic drop in the demand for labour and the de facto employment rate in most countries.

Unemployment benefit systems and social assistance schemes are still not functioning properly and so pension systems are being used as labour market correctives. This is far from ideal and is certainly inefficient; but where is the short-term alternative? Advocating non-solidarity based individualised savings systems like the Pinochet-model in Chile for Central and Eastern Europe is certainly no answer in the short run and almost certainly not in the long run either. Crowding out solidarity-based financing systems in times when societies are demanding huge pay-as-you-go transfers to people dependent on them and need all the solidarity they can muster, while inflation is running high, might have good economic reasons to back it up (like the forced creation of resources for high risk domestic investments), but cannot be considered politically or socially sound.

There is no doubt that the social protection systems in CEE countries are in dire need of reform. The ILO has stressed this point in all its technical advisory reports during the last five years-sometimes with annoying persistence as far as our client governments are concerned: pensionable ages must be increased, administrations must be overhauled, anti-poverty benefits must be made more efficient, and the health delivery and financing system requires a complete overhaul almost everywhere. All this might reallocate resources but is unlikely to decrease social spending below the present level-provided, of course, that we still want to finance a decent society in which everybody is guaranteed at least a survival level of income. What the countries and social protection systems of the region need most urgently is major investment in the overhaul of their social and fiscal governance systems, i.e. investment in all procedures that make the state function and social protection work, most prominently among them the revenue collection and enforcement systems.

One of the crucial roles of a national social protection system during transition is to rearguard political and economic restructuring by the financing of difficult and painful restructuring in the labour market and the economy at large. Cutting current expenditure levels at this point is a 'quick and dirty solution' sacrificing the social rearguard of reforms on the altar of rapid fiscal consolidation, and cutting corners at the expense of national consensus. It should not be overlooked that national consensus on economic and political reforms in some countries is already shaky, and consensus does not come in parts, it either carries economic reforms and political systems or it abandons them. Nobody knows the 'breaking point' at which consensual approval turns into collective rejection, but the writing is certainly on the wall. MC

Notes

1 This argument is spelled out at length in a recent article by Jeffrey Sachs-inter alia former economic adviser to the Russian Government-entitled 'Post-communist Parties and the Politics of Entitlement' in TRANSITION, a newsletter issued by the Transition Economics Division of the World Bank.

2 A response to the economic argument is spelled out in more detail in Cichon and Hagemejer: 'The Level, Structure, Trends and Limits of Social Protection Expenditure in Central and Eastern Europe, mimeo, ILO, 1994 (to be published in 1995).

3 Cf. inter alia UNICEF: Crisis in Mortality, Health and Nutrition, Regional Monitoring Report No. 2, 1994.

4 Cichon and Hagemejer, 1994, op.cit. pp. 18-20. The poverty line used here is equivalent to the one usually used by UNICEF i.e. 35% of the 1989 average wage.

5 ILO-CEET: The Bulgarian Challenge, (Budapest, 1994); and Cichon and Yakushev: 'Safeguarding Change: The Russian Social Protection System and Its New Role,' first draft, Budapest and Moscow, 1995.