Colloquium 1995 Panel II : Living Standards and Social Welfare

Living Standards and Social Welfare Michael Ellman Depending on which international institution's statistics you choose to chart the economic and social evolution in Eastern Europe and the former Soviet Union, the same country can be presented as 'progressing nicely toward full recovery' or 'in serious decline'. The truth, believes Professor Michael Ellman, lies somewhere in between. While most of the international statistics available are produced by professionals using sound methods, they often do not take into account the unreliability and fast-changing nature of the markets they are measuring. Two examples of this are the tendency of producers to conceal part or all of their output to escape taxation and the fact that - in Russia, for example - the networks which provide trade and economic statistics are not in place and do not cover all regions and sectors of activity. Michael Ellman is Professor at the Department of Microeconomics, University of Amsterdam.

Introduction It is well known that after 1989 there was a sharp fall in living standards and social welfare throughout the region, with a big difference between countries. Data on this were presented at this Colloquium two years ago by Michel Gaspard. (1) The purpose of this paper is to update his contribution and also discuss certain social indicators which he did not analyse. Data Problems The data available to analyse living standards and social welfare are not very reliable. There are three reasons for this; structural change, change in statistical methods, and the survival of old problems. Statistical time series are most useful when the structure of whatever it is that is being measured remains unchanged. The reason for this is that one can be reasonably sure that a measured change is a real change and not merely a statistical artefact resulting from a change in the structure of what is being measured. In Central and Eastern Europe, however, a number of important structural changes have taken place which makes the meaning of many statistics problematic. Particularly important structural changes are the ending of shortages and queues, the growth of the private sector and the emergence of a multiplicity of actors in foreign trade. The transition from widespread shortages and queues to the general availability of goods for those with money, means that the conventional measures of "real income" (i.e. money incomes deflated by a price index) do not properly capture the relative incomes of the population before and after this very important systemic change. For example, the IMF has more than once published figures showing a growth in Soviet real incomes (or "statistical real incomes") in 1987-91, despite the fact that welfare obviously declined in this period. The confusion arises because inadequate account is taken of the worsening shortages in the USSR in 1987-91. Another fundamental structural change has been the growth of the private sector. This has probably led to an understatement of the level of output in recent years, and possibly also to its rate of growth. The reason is that a substantial part of the private sector is eager to conceal part or all of its output/income in order to escape taxes. Furthermore, the end of the state monopoly of foreign trade has also had consequences for foreign trade statistics. Whereas it was fairly easy to collect accurate foreign trade statistics when foreign trade was carried out by a small number of state organisations, it is much more difficult when it is carried out by a large number of actors. This is one of the reasons for the notorious inaccuracy of Russian foreign trade statistics.

The end of communism has been accompanied by important changes in statistical methods. One such change has been the transition from the MPS (material product system) method for the compilation of the national income accounts to the SNA (system of national accounts) method. This change, which began many years ago in Central Europe and which has not yet been completed in parts of the CIS, is a complex matter which raises many difficulties and automatically makes national income statistics from before and after the change non-comparable (unless they are specially adjusted). Another statistical change is the transition from complete enumeration, which characterised the communist statistical system, to sampling. The latter is in general more sensible and produces reliable results, provided that the sample is representative. Naturally, national statistical offices require time to learn how to use the new methods. During this learning period they may fall victim to various childhood diseases. Some old statistical problems have survived the transformation. For example, the Soviet family budget survey, although based on a large sample, was notoriously unrepresentative and unreliable. The same is true of its continuation, the Russian family budget survey. Similarly, the cause of death data in many countries, derived from death certificates, is of distinctly low quality. For example, an examination by the German Federal Statistical Office of 2,500 randomly selected GDR death certificates found inconsistencies between the examiner's report and the recorded cause of death in nearly 40 percent of cases. In 15 percent of them major errors were evident. (2) It is likely that such errors continue to be widespread in Central and Eastern Europe. Because of these statistical problems, it is difficult to get a clear and unambiguous picture of what has really happened in the last few years. This is one of the factors explaining why different people and organisations have quite different images of what has happened to living standards and social welfare during the transition. For example, IMF publications show a decline in Russian real wages between 1987 and 1994, but the decline is not very sharp and much less than the fall in statistically measured output. (3) Similarly, in a recent draft paper about Russia two IMF economists state that "Taking into account estimates for street and informal trade, retail sales declined only marginally in 1991 and 1992, and have been rising anew since 1993." (4) On the whole it is fair to say that the IMF picture is of a decline in living standards in some countries in some periods, but basically nothing dramatic. On the other hand, according to UNICEF, (5) "The mortality and health crisis burdening most Eastern European countries since 1989 is without precedent in the European peacetime history of this century. It signals a societal crisis of unexpected proportions, unknown implications and uncertain solutions." My view lies between these two positions. I do not share the self-satisfied view of the IMF. It seems to me clear that at any rate in some countries there have been serious declines in social welfare which are usually ignored in IMF analyses. (This is understandable in a financial organisation which is primarily concerned with monetary variables.) On the other hand, although I think that UNICEF has played a very positive role in collecting data on social welfare in Central and Eastern Europe and publicising them, I do not agree with some of the colourful adjectives UNICEF has used to attract publicity for the problems. For example, the idea that the recent increase in the crude death rate in Russia is "apocalyptic" (6) is completely absurd in the light of the demographic history of Russia/USSR in the twentieth century. (7)

Inequality Accurate data on inequality in Central and Eastern Europe at the present time are difficult to gather for three reasons. First, for measuring the relationship between high and low incomes it is naturally particularly important to have accurate data on the highest and lowest incomes. Actually, however, it is usually much more difficult to collect these than to collect accurate data on the middle of the income distribution. The reason for this is that at the top of the income distribution people may be particularly anxious to keep their income from the knowledge of the tax authorities, and at the bottom of the income distribution it may be difficult to get reliable data because the people are not officially registered, move house frequently, are only partially literate, are in trouble with the authorities, etc. Hence data about the relationship between the highest and lowest incomes are particularly unreliable. Secondly, the transition from a society in which a large part of inequality was formed by non-cash privileges (e.g. access to high quality cheap housing or restricted access shops) to one in which inequality is largely a monetary phenomenon, may confuse interpretation of the data (this is an example of structural change). Thirdly, the existence of substantial price and income differences within a country may generate a spuriously high level of "inequality". This is particularly the case in Russia, where prices (and incomes) vary sharply across that huge country. According to the World Bank researcher Milanovic, "On average, income inequality has increased by around 5 to 6 Gini points, from about 24 in 1987 to about 30 in 1993, with the greatest increases in the Baltics, followed by Russia, Bulgaria, the Czech Republic and Poland. (The Gini coefficient measures the inequality of income distribution, the maximum being 100 points. Absolute equality measures 0.) This five-to-six point increase is about half of what occurred in the United Kingdom during Margaret Thatcher's ten-year rule. In Eastern Europe and the FSU, however, the change occurred in four to five years, so the intensity of change was about the same. A Gini coefficient of about 30 is still not high, relative to many middle income countries, but exceeds that in many OECD countries, among them the Scandinavian countries, Belgium, Holland, and Germany." (8) In addition to the increase in inequality within countries, it is likely that there has also been an increase in inequality between countries. Whether or not this "Thatcherisation" of Central and Eastern Europe was desirable, is naturally a matter on which opinions differ.

Employment and Unemployment The sharp decline in employment in Central and Eastern Europe is a phenomenon to which the ECE in particular has drawn attention. (9) In 1990-92 employment fell by 29 percent in Bulgaria, 18 percent in Slovenia, 13 percent in Poland and 9 percent in the Czech Republic. Partly this led to a growth in measured unemployment, which by the end of 1993 had reached 16 percent in Bulgaria and Poland and 14 percent in Slovakia. (By the end of 1994, measured unemployment ranged from about 3.5 percent in the Czech Republic to 19 percent in the former Yugoslav Republic of Macedonia.) Partly it led to a decline in the participation rate. The decline in employment in industry and agriculture was necessary because of low levels of labour productivity in those sectors under the old regime and the sharp falls in production which have taken place. Further declines in employment and increases in unemployment can be expected in some countries. Unemployment has become a serious problem for particular groups of the population (e.g. school leavers, gypsies, the elderly), for particular occupations (e.g. missile engineers and coal miners), and for particular regions (e.g. in Russia for towns in the far north and former military-industrial towns, and in many rural regions throughout the area). The decline in employment and growth of unemployment can be evaluated in various ways. The emergence of unemployment as an ever present reality throughout the region may well have a positive effect on work effort and motivation. Many people consider that the employment of women under the old regime was excessive and the withdrawal of some of them from the labour force is a positive development. On the other hand, large scale unemployment among young men will obviously stimulate crime and threaten social order. (In 1991-93 the Russian homicide rate doubled, and in 1993 was about three times that of the USA and 15 times that of Italy.) Similarly, the expulsion of elder people from the labour force has often been very disagreeable for them, harming both their incomes and their self-esteem. Furthermore, the decline in demand for certain skills has naturally hit the people concerned very badly. A positive aspect of the decline in employment has been the recent rapid growth of industrial labour productivity in a number of countries. For example, in Poland in 1994 labour productivity in manufacturing rose by 12.5 percent (this followed increases also in 1992 and 1993). Similarly, labour productivity in manufacturing in Hungary rose by 7.9 percent in 1994 (after rising by 12.6 percent in 1992 and 16 percent in 1993). In the Czech Republic it rose by 5.7 percent in 1994 (after declines in 1990-93). These sharp increases resulted from a resumption of economic growth combined with restructuring. (10) They are positive developments for the international competitiveness of these countries.

Poverty There are a variety of estimates about the proportion of the population in the region in poverty and the changes in this proportion during the transformation. Naturally, the proportion of a population living in "poverty" always depends on where the "poverty" line is drawn. By drawing it lower one always reduces the proportion, by drawing it higher one increases it. Since the cut-off point is always essentially arbitrary, it is not surprising that there are a variety of different estimates in circulation. One estimate is that of UNICEF, which calculates that in 1993 in Bulgaria 26 percent of the population lived in extreme poverty, and that in 1992 the corresponding figures were 1 percent in the Czech Republic, 15 percent in Poland, 19 percent in Romania and 23 percent in Russia. Milanovic has compared the proportion of the population of the region living in poverty before the transformation began and after it. He used a common poverty line ($120/month/capita in 1990 international prices). He came to the conclusion that whereas in 1987-88 only eight million people (about 3 percent of the population) lived in poverty, by 1992-93 (or 1994) this figure had risen to 58 million or 18 percent of the population. This result, of 50 million new poor arising from the transformation, is a very striking, and very sad, result. Of these 50 million new poor, more than 40 million live in the three Slavic states of the FSU. In the Czech Republic, Slovakia, Slovenia and Hungary, poverty defined in this way scarcely exists, since the cut-off point chosen is low compared to incomes in these countries. The only positive finding by Milanovic is that, unlike in Latin America, the poor do not - yet - represent a distinct underclass. They are currently reasonably integrated into society. It will require the maintenance of the present situation for some time to generate an underclass that is only weakly integrated into the wider society. Mortality In Russia and Ukraine there has been a sharp increase in mortality. A small increase in mortality also took place in Bulgaria (in 1989-93). The crude death rate in Russia has risen from 11.4 (per thousand inhabitants) in 1991 to 12.2 in 1992, 14.5 in 1993 and 15.6 in 1994. This was an increase of 37 percent in the three years 1992-94, a remarkable demographic phenomenon. Although it was partially a result of an ageing of the population, it was mainly a result of increases in the age-specific death rates. This increase in mortality and its ultimate causes, have already given rise to extensive discussion. (11) Two ways of presenting the significance of these mortality data in easy to comprehend forms are to estimate excess mortality and life expectancy. It seems likely that in 1990-94 excess mortality in the whole of the former USSR arising from the increase in age-specific mortality rates was about 1.5 million. (This is about one and a half times excess mortality in the 1947 Soviet famine.) In 1987-94 Russian male life expectancy at birth fell by about eight years, from 65 to 57. The latter is a very poor figure in international perspective. Not only is it well below the advanced countries, but it is also below the middle income countries such as Turkey or Brazil, at the level of Indonesia in the late 1980s, and only two years above the Indian level of the early 1980s. The only positive aspect of the Russian mortality data is that it seems to have stopped increasing. In the first quarter of 1995 the crude death rate was 15.7, which was 3 percent below the level of the first quarter of 1994. This gives rise to the hope that Russian mortality peaked in 1994 and will fall this year. (It should however be pointed out that these date are not comprehensive since they exclude Chechnya, where mortality has not declined in 1995.) These figures are not just a curiosum for historical demographers. It is entirely possible that in the Duma elections in December 1995 and the Presidential elections of summer 1996 the nationalists and communists will make an issue of the "genocide against the Russian people" of which the West is allegedly guilty.

Morbidity "The past four years [i.e. 1990-93] have witnessed a persistence or resurgence of several 'poverty diseases', including infectious, parasitic and sexually-transmitted diseases. Children are the most likely victims of these ailments. The situation regarding the incidence of diphtheria, tuberculosis and hepatitis started to deteriorate in 1990 and 1991, worsened in 1992, and has now reached epidemic proportions in some areas. There are also signs that these diseases are slowly spreading to previously unaffected areas." (12) Russia is still experiencing diphtheria and syphilis epidemics and a steady growth of tuberculosis. In the first four months of 1995, in Russia the incidence of diphtheria rose by 75 percent over one year earlier, of syphilis by 146 percent and of tuberculosis by 9 percent. (13) As in most welfare indicators, there is a sharp difference both in levels and in trends, between countries of the region. For example, UNICEF has drawn attention to the decline in the incidence of tuberculosis after 1992 in the Czech Republic. (14) Improvements The sharp deterioration in welfare in recent years has been one of the results of the collapse of Communism and of the depression. Now that the depression has been replaced in many countries by economic growth, one would expect to see an improvement. Even when the economy grows, however, it is not always the case that consumption also grows. It could be that the additional resources are devoted to improving the external balance (which is necessary in Hungary) or to investment or to government expenditures. Furthermore, even if average consumption grows, it may be that the consumption of those at the bottom of the income distribution may not do so. Nevertheless, if real GDP in a country with a more or less stable population grows by 5 percent or more, then average consumption is likely to grow and also the consumption of most of the population is likely to grow. According to the EBRD Transition report update five countries of the region are likely to have a real GDP growth of 5 percent or more in 1995. They are, Albania, the Czech Republic, Estonia, Poland and Slovenia. In all these countries, it seems likely that average consumption will rise in 1995. For a large part of the population in a number of countries, the transformation crisis is receding into the past.

Effects It seems likely that the situation described above has had and will have four effects: It is a tragedy for many of the people directly affected and their relatives and friends. Those who are part of the additional deaths or additional cases of various unpleasant diseases are of course victims of disagreeable social and epidemiological processes. This situation is partly the responsibility of the West. Having devoted enormous resources to struggling against Communism, and having encouraged the people of Central and Eastern Europe to destroy Communism, the West declined to launch a "new Marshall Plan" (although it did provide substantial gross resources for the countries of Central and Eastern Europe). With the exception of the former GDR, the people of Central and Eastern Europe did not receive the massive net transfers from the West than many of them had previously hoped for and expected. The social costs of the transformation were one of the reasons for the electoral victories of the post-communists in a number of countries (Lithuania, Bulgaria, Poland and Hungary). It seems likely that in the forthcoming Russian Duma and Presidential elections these social costs will also be a factor influencing the results. Russia, however, does not have a post-communist party, but a genuine old-fashioned Communist party. This is the strongest political party in the country and is likely to fare relatively well. So may other nationalist and anti-Western groups as the (surviving) victims of the policies of the past few years express their discontent. It will strengthen the self-confidence of the local liberals (15) and of the international financial institutions. The five countries listed above as likely to fare well in 1995, are all countries that have pursued fairly orthodox economic policies, usually in collaboration with the international financial institutions. Those countries in the region which have done something very different have not fared better, usually just the reverse. Officials from the countries which are the principal shareholders in the international financial institutions can be satisfied that by and large those institutions have fulfilled the task assigned to them.

Footnotes M. Gaspard, Incomes and living standards in Central and Eastern Europe and the former Soviet republics: recent developments, current situation and outlook, Economic developments in cooperation partner countries from a sectoral perspective (NATO, Brussels, 1993). N. Eberstadt, Demographic shocks in Eastern Germany, Europe-Asia Studies 1994, vol 46 no. 3 p. 529. Russian Federation, IMF Economic Reviews 16, 1994 (IMF, Washington DC, 1995) table 17 p.78. (This table does show a very sharp fall in real pensions.) V. Koen & M. Marrese, 'Stabilization and structural change in Russia, 1992-94' mimeo March 1995 pp 10-11. Public policy and social conditions, Regional Monitoring Report no. 1 (UNICEF, Florence, 1993) & Crisis in mortality, health and nutrition, Regional Monitoring Report no. 2 (UNICEF, Florence, 1994). Public policy and social conditions, Regional Monitoring Report no. 1 (UNICEF, Florence, 1993) p. 20. It is characteristic of UNICEF's knowledge of Russian/Soviet affairs, that when comparing current mortality increases with World War II, it uses for Soviet excess mortality in World War II not the currently available best estimate (26-27 million) but the Stalinist falsification (7 million). See Crisis in mortality, health and nutrition, Regional Monitoring Report no. 2 (UNICEF, Florence, 1994) p.46. Transition (World Bank) vol. 5 no. 8 October 1994. Economic Survey of Europe in 1993-1994 (ECE, UN, New York & Geneva 1994) pp 84-85. For these figures see Transition report update (EBRD, London, 1995) p. 9. Crisis in mortality, health and nutrition, Regional Monitoring Report no. 2 (UNICEF, Florence, 1994); M. Ellman, The increase in death and disease under 'katastroika', Cambridge Journal of Economics vol/ 18 no. 4 August 1994; J. Nell & K. Stewart, Death in transition: The rise in the death rate in Russia since 1992 (Innocenti Occasional Papers EPS 45, UNICEF, Florence, 1994); J. Shapiro, The Russian mortality crisis and its causes, A. Aslund (ed) Russian economic reform at risk (Pinter, London & New York, 1995). Crisis in mortality, health and nutrition, Regional Monitoring Report no. 2 (UNICEF, Florence 1994) pp. 54-55. Zdorov'e naseleniya i sreda obitaniya 1995 no. 5 p. 19. Crisis in mortality, health and nutrition, Regional Monitoring Report no. 2 (UNICEF, Florence, 1994) p. 55. The word "liberal" is used here in the European sense.