UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Problems of labour markets: East European economies in transition Ivanov, Sergei Leonid 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-0361.pdf [ 10.87MB ]
Metadata
JSON: 831-1.0087006.json
JSON-LD: 831-1.0087006-ld.json
RDF/XML (Pretty): 831-1.0087006-rdf.xml
RDF/JSON: 831-1.0087006-rdf.json
Turtle: 831-1.0087006-turtle.txt
N-Triples: 831-1.0087006-rdf-ntriples.txt
Original Record: 831-1.0087006-source.json
Full Text
831-1.0087006-fulltext.txt
Citation
831-1.0087006.ris

Full Text

«PROBLEMS OF LABOUR MARKETS: EAST EUROPEAN ECONOMIES IN TRANSITION» by SERGEI LEONID  IVANOV  M.A.,Moscow State University, 1989  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Geography)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA  June 1995  Sergei L. Ivanov, 1995  In  presenting  this  thesis in  partial  fulfilment of  the  requirements  for  an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for department  or  by  his  or  scholarly purposes may be granted by the head of her  representatives.  It  is  understood  that  copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of  Geography  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  22 August 1995  11 ABSTRACT  The main objective of the present study is to examine the role of labour markets in East European economic transition and the current labour market situation in different countries with a particular focus on the links between labour markets and the task of macroeconomic restructuring. The principal questions posed are, briefly: what is the role of labour markets in helping to ensure success for economic transformation?, what are the current problems threatening the success of the reforms?, and, finally, what are the goals and mechanisms of labour market transition?" The study is confined to the macro level and does not focus on the level of the region and/or enterprise. In dealing with the problems of goals and mechanisms, the study is policy-oriented. Methodologically the study sees the transitional development of labour markets in East European countries as a triply-influenced process. First, labour markets have inherited a number of features from pre-transitional economic and political systems. Second, broad economic policy pursued by the present governments undoubtedly shapes both current labour market issues and the scope of active employment measures to be implemented. Finally, changes within the labour force itself induced by the systemic changes are important. The methodological approach of analyzing East European labour markets development as a triply-influenced process is closely linked to the main research questions above. The link is provided by the two crucial issues which are examined throughout the thesis: the degree of labour market restructuring to be expected, and the origins and scale of East European unemployment. Part I of the thesis deals with introductory issues and provides some theoretical and empirical insights into economic transition in Eastern Europe. Part II analyzes the theory and practice of labour markets under central planning. Part III is devoted to structural problems of East European labour market transitions and offers some new methodological approaches and working hypotheses. Part IV offers an overview of labour issues in growth and development thought. Finally, Part V deals with transitional issues of East European labour markets.  Ill  TABLE OF CONTENTS Abstract  ii  Table of Contents  iii  List of Tables  v  List of Figures  ix  Acknowledgement  x  Foreword  xi  PART I. TRANSITION IN EASTERN EUROPE: AN INTRODUCTION AND OVERVIEW 1 Chapter 1.  Terminology, Research Purpose, Research Setting, Data and Methodology 2  Chapter 2.  Labour Market Issues in East European Transition: Theoretical Considerations and a Review of Literature  14  §1.2.1.Contrasting Macroeconomic Frameworks §1.2.2.General Macro Perspectives on Labour Market Policy  16 23  Chapter 3.  Historical Perspective on Transition in Eastern Europe: General Economic Outlook 28  PART H. THEORY AND PRACTICE OF LABOUR MARKETS UNDER CENTRAL PLANNING 47 Chapter 1.  The Heritage of East European Labour Markets  §2.1.1.Patterns and Interim Conclusions PART m . STRUCTURAL PROBLEMS OF EAST EUROPEAN LABOUR MARKET TRANSITIONS Chapter 1.  Occupational Structure of Employment: a Preliminary Analysis  §3.1.1.Scope of the Approach  47 88 93 94 95  IV  §3.1.2.Occupational Structure of Employment: an Overview and Descriptive Analysis Chapter 2.  Occupational Structure of Employment and Level of Economic Development in Eastern Europe  §3.2.1.Scope, Limitations and Main Assets of the Approach §3.2.2.Occupational Structure of Employment and Level of Economic Development: General Tradeoff §3.2.3.Maturity of Occupational Structure Chapter 3.  Concluding Remarks  99  110 112 119 125 144  PART IV. LABOUR ISSUES IN GROWTH AND DEVELOPM E N T THOUGHT: AN OVERVIEW  154  PART V. EAST EUROPEAN ECONOMIC TRANSITION AND LABOUR MARKETS  179  Chapter 1.  Main Characteristics of the Performance of Labour Markets under Transition 179  §5.1.1.Recent Labour Market Developments in Eastern Europe: an Introduction 181 §5.1.2.Transitional Macroeconomics and East European Labour Markets: General Implications 188 §5.1.3.Empirical Illustration of Regional Dimensions 205 §5.1.4.Unemployment  Compensation under Transition  216  Chapter 2.  Summary of Labour Market Policies in Eastern Europe 226  Chapter 3.  Conclusions  237  PART VI. RETROSPECTIVE VIEW AND AFTERWORD  242  ENDNOTES  250 258  BIBLIOGRAPHY  V  LIST OF TABLES PART I:  Table 1. Rates of Pre-Transitional Economic Growth in East European Countries, average GNP per period, %, 1950-1988.  29  Table 2. Rates of Growth of GDP in Selected Countries in 1981-1992,%.  34  Table 3. Short-Term Forecast of Real GDP in Eastern Europe and Former USSR in 1993-1996 (annual percentage change unless stated otherwise).  44  Table 4. Forecast of GNP Growth Rates in 1993-2000,%.  46  PART H :  Table 5. Labour Force Participation Rates in Estonia, Latvia and Lithuania in 1989, as a % of total population. Table 6. Employment in Industry by Occupation, %, various countries, various years. Table 7. Level of Education of the Labour Force in Different Countries, as a % of the total labour force, various years.  48 54 56  Table 8. Enrolment Ratios at Secondary and Higher Levels of Education, as a % of the population of the corresponding age groups, various countries and years.  57  Table 9. Type of Education: Secondary Level Enrolments, %, various countries, various years.  58  Table 10. Type of Education: Higher Level Graduates by Field of Study, %, various countries, various years.  59  Table 11. Inter-Sectoral Wage Differentials: Average Monthly Wages in Selected Sectors as a % of wages in manufacturing in the USSR, 1970-1989.  63  Table 12. Differentials of Average Monthly Wages by Branch and Worker Category in Hungary, 1981-1989, industry indicator - 100%.  64  vi Table 13. Per Capita Earnings and Social Consumption Fund Benefits in the USSR in 1960-1989(1960- 100%). Table 14. Labour Turnover by Sector in the USSR in 1965-1989,% of average employment. Table 15. Labour Mobility and Turnover Rates as a % of Total Employment, various countries, various years. Table 16. Vacancies, Job Seekers and Unemployment in Hungary in 1984-1989,thous. people. Table 17. Job Leavers by Reasons in Hungarian State and Cooperative Sectors, %, 1976and 1985. Table 18. Job Losses by Sector in Several Eastern European Countries as a % changes between 1989 and 1991. PART  III:  Table 19. Sectoral employment in Selected Countries, %, various years. Table 20. Sectoral Employment Structures for Average Countries, % Table 21. Sectoral Employment Structures for Selected Groups of Countries, %, various years. Table 22. Sectoral Employment in Selected Countries and Selected Groups of Countries, %, various years. Table 23. Sectoral Employment in Selected Countries and Selected Groups of Countries, %, various years. Table 24. Sectoral Employment in Selected Countries and Selected Groups of Countries, %, 1925. Table 25. Sectoral Employment Structures for Selected Countries, %, 1925. Table 26. Per Capita GNP in Some European Countries, USD, 1950.  Vll  Table 27. Alternative Estimates of Per Capita GNP in East European Countries, USD, various years.  108  Table 28. Integral Indicator of the Employment Structure (IIES) for the Hypothetical West European Country (HWEC), USA and East European Countries, 1960-1987.  132  Table 28 (continuation). Integral Indicator of the Employment Structure (IIES) for the Hypothetical West European Country (HWEC), USA and East European Countries, 1960-1987.  133  PART V.  Table 29. Some Macroeconomic Indicators for Eastern Europe, 1992.  179  Table 30. Employment in East European Countries by Form of Ownership, 1990-1992.  183  Table 31. Employment and Unemployment Changes in Eastern Europe, as a % of initial labour force, 1989-1992.  184  Table 32. The Matching Function in Eastern Europe, selected countries, various years.  193  Table 33. Estimates of Employment Responsiveness to Changes in Output in 1980-1985and 1986-1991,various countries.  195  Table 34. Officially Registered Vacancies and Unemployment, 1992.  206  Table 35. The Turnover of the Unemployment Pool in East European Countries, 1992, % of the population of origin.  207  Table 36. Characteristics of Long-Term Unemployment in Eastern Europe, 1992.  208  Table 37. Distribution of Inflows into Unemployment in Eastern Europe by Reasons of Separation, 1991 and 1992, selected countries  210  Table 38. Regional Differences in Unemployment and Vacancies in Bulgaria and Czechoslovakia in 1992.  211  Table 39. Regional Unemployment Differentials and Regional  vm Mismatch.  212  Table 40. Proportions of the Registered Unemployed Stock in Receipt of Unemployment Insurance and Unemployment Assistance Benefits, %, 1988.  220  Table 41. Unemployment Benefits in Some East European Countries, 1992 .  222  Table 42. Summary of Active Employment Measures in Eastern Europe.  228  Table 43. Vacancies and Unemployment by Skill Levels in Hungary, %, 1992.  229  Table 44. Structure of Unemployment in East European Countries by Gender and Age in 1992, %.  233  Table 45. Structure of Unemployment in East European Countries by Level of Education in 1992, %  233  Table 46. Duration of Unemployment Benefits in Bulgaria  235  ix LIST OF FIGURES  Figures 1-3. Sectoral Dynamics of Employment Macrostructure in West European Countries: 1970,1980,1987, %  122  Figures 4-6. Sectoral Dynamics of Employment Macrostructure for Hypothetical West European Country: 1970,1980,1987, %  124  Figures 7-8. Interrelationship of Sectors in Occupational Structure of Employment (Intersectoral Flows of Labour Force) in Normal (Figure 7) and Modified Systems of Coordinates, 1960-1987  130  Figure 9. Elasticity of Integral Indicator of Employment Structure (IIES) on GNP Per Head for Hypothetical West European Country, USSR and East European Countries, 1960-1987 136 Figures 10-12. Real and Hypothetical Employment Structures for the USSR and East European Countries, 1960-1987  137  Figure 13. Logic of Adjustment to Equilibrium Employment under High Unemployment, Low Inflation and Fixed Nominal Exchange Rate  198  Figure 14. Logic of Adjustment to Equilibrium Employment under High Unemployment, Positive Inflation Differential and Low Competitiveness  199  X  ACKNOWLEDGEMENT  Any study is a result of participation of many people. While the arguments are advanced by the researcher, who is alone responsible for their content, many others enable the completion of both project and thesis. In particular, some advance funds, others in the field provide advice and criticism, and still others that most precious resource for any writer, which is encouragement and sympathetic ear. This study was supported by the UBC Department of Geography both intellectually and financially. Financial, spiritual and inward participation of Mr. Arnold Landau and Mrs. Shirley Kliman from Incline Village, NV deserves special emphasis. Not only their initial financial contribution to my study, but more their love, sympathy and concern for my overall success in North America have been and still are their most valuable donation. This study was partly supported by the Research Support Scheme of the Central European University (Prague, Czech Republic), Research Grant No: 748\91. This is not customary to specifically acknowledge the contribution of one's academic supervisor. However since the writer is fortunate to be the first student from the former Soviet Union admitted into postgraduate study at the UBC Department of Geography, special recognition of Dr. Robert North's contribution seems appropriate. Although the topic of the thesis was not within the area of his direct scholarly expertise, the extraordinary vision and scientific background of my supervisor have made his criticism merely invaluable for the content of this study. In addition, Dr. Robert North had to cope with another very cumbersome problem, which is the author's imperfect English. Dr. Robert North's patience, tact and accuracy have made this study readable to native speakers. The research also draws on the expertise of practioners in the field. Dr. Igor G. Ushkalov from the Institute for International Economic and Political Studies of the Russian Academy of Sciences (Moscow, Russia) provided valuable assistance and advice on aspects of the field research of this study. The author has learned much working with him in 1989-91. Mr. Vsevolod Yu. Bulantsev from the Central Bank of Russian Federation (Moscow), an old friend of mine, was of great help during the work on Part III of the thesis. His kind cooperation has helped to formulate some basic concepts of the study. Colleagues and friends listen, ask questions and empathize. Drs. David Edgington, Walter Hardwick, Daniel Hiebert, and Trevor Barnes were particularly generous in sharing their time and providing fresh perspectives. Mrs. Margaret North helped me over bumps on the road. In alphabetical and no other order, Lyle Courtney, Rhys Evans, Rodrigo Hernandez, Alexander Kliachin, and Jean-Francois Proulx provided helpful perspectives from their fields. The final acknowledgement is to Irina Frolina, my wife and long-time confidant. To one and all, including those whom I may have innocently forgotten to acknowledge, thank you.  xi FOREWORD  Firs:  In the old days, forty, fifty years ago... they used to make jam out of the cherries...They knew the recipe in those days. Ranevskaya: And what's happened to this recipe now? Firs: They've forgotten it. No one remembers it. A. Chekhov. Cherry Orchard, 1904  Nobody ever knew the recipe. Who could have imagined some five seven years ago that Mr. Gorbachev's inauguration would finally result in the collapse of the Soviet Union, the "black empire of evil" which always seemed unshakeable; "velvet revolution" in and the breakdown of Czechoslovakia, breath-taking horrors of Civil war in the former Yugoslavia, Azerbaijan and the Caucasus; the inglorious end of Nicolae Ceaucescu in Romania; and, finally, to the fatal mistakes made by Mr. Gorbachev himself...Time goes by, but there is still no recipe. About two years ago Russian President Mr. Yeltsin took to the airwaves to announce that he was dismissing the Supreme Soviet and calling early elections for a new parliament, actions that unleashed open rebellion and ended with the shelling of the White House two weeks later... This was the presidential recipe of the moment. Unfortunately there can be no general recipe for changing the situation to the better, improving the lives of the people, avoiding impoverishment of the nations, socially protecting everybody who deserves it, and keeping tanks off the streets. Nobody knows it. Here one of the most important postulates of Marxian dialectics works as well as ever: "Practice  Xll  is the only criterion for the truth"... This thesis, certainly, does not pretend to be a recipe. This is as much as I can do to help my country and my people at the moment. This thesis is offered to all honest people in Eastern Europe and Russia, who are trying to find their way in a frequently changing environment. May their lives improve and hardships disappear; may every next year bring a slightly better time. This thesis is dedicated to my beloved parents, who, among other people, are trying to adjust to new and unfamiliar conditions of rapidly changing values. It is to those people who continue to try, all the time in the face of great difficulty, to lead lives of humanity and to nurture values enhancing peace and understanding that this project is addressed.  1  PART I. TRANSITION IN EASTERN EUROPE: AN INTRODUCTION AND OVERVIEW. Chapter 1 of Part I presents the general background of Eastern European economic transition, states the purpose of the research, defines the relevant terminology,  and introduces  some general theoretical  and empirical  considerations essential for the analysis of East European labour markets undertaken in Parts II, III and V. Chapter 1 also sets forth the general methodological approaches used in this thesis. Chapter 2 reviews relevant literature and deals with the general theoretical background of labour market issues in East European transition. Chapter 3 studies the development of the general economic situation during the years under transition and addresses some crucial problems, such as the reasons for collapse in output, the main directions of government macroeconomic policy, rising unemployment and others.  2  CHAPTER 1. TERMINOLOGY, RESEARCH PURPOSE, RESEARCH SETTING, DATA AND METHODOLOGY. The usage of the term "transition" in the literature is far from uniform. Is transition one stage in a process, or the whole process leading to the triumph of the market economy over central planning? Is it a reflection of the desire of a large proportion of the population of the countries involved to bring their standard of living into line with the high levels of consumption  in the  developed countries, and used by them to signify the process they hope will bring this about? Does it express a new political and economic culture in which moral considerations of various types have come to the fore? The trouble arises because many authors use this term to mean what they want it to mean. For several years "transition" has been a locution embracing an impressive set of euphemisms that include "restructuring", "decentralization", "deregulation", "modernization", "intensification" and the like, each of which has a somewhat different meaning. The most common understanding of  "transition" is "a decisive repudiation of the command-  administrative system of management, and a progression to economic methods based on market mechanisms" (Standing, 1991). According to a variety of sources, the economic system that should emerge from this transition should combine: "high efficiency of production", with social goals underlying its development; "increased wage differentials with a system of social guarantees"; "the release of manpower from production in keeping with the goal of effective employment"; "revival and broad development of cooperative principles, with  3  the modernization of the public sector"; and "asocial market, strengthening its influence on production" (Standing, 1991; Boeri & Keese, 1992; Bova, 1991; Schopflin, 1992, and others). Needless to say, such a statement of goals is too broad and fuzzy. At an abstract level, who could be against structural adjustment toward a high efficiency of production? The question therefore still stands: "Transition from a centrally planned economy to what? And how?" In other words, what are the goals and mechanisms of transition? "Market economy" sounds like a good goal, but the existence of many kinds of market economies means that this generalization is not very useful with respect to definitive terminology. Many different models exist in Western Europe, for example, as far as the degree of state intervention is concerned. Obviously, there is no simple a-priori solution to this question, and it is impossible to determine which model should be chosen and followed from the very beginning. Most new governments in the post-communist countries tend to have a general liberal, laissez-faire attitude - maybe as a reaction to what was abandoned. But at the same time some of them have chosen the "big bang" approach to economic restructuring with a corresponding set of plans, decrees and other measures. One of the complications is that all post-communist countries are overwhelmed with West European, North American and Far Eastern advice from various specialists, who, unfortunately, often lack a full idea of the range and diversity of the specific practical problems these countries are  4  facing and how their economies would react to the proposed measures. The implementation of this advice however is often related to the questions of either Western financial aid or other political and/or economic matters essential for the course and direction of transition. Scholars and policy-makers from Eastern Europe are either not competent enough or not sufficiently aware of possible "domestic" solutions alternative to those offered by Western experts. There is diversity in Western and Eastern European understanding of "transition", sometimes very significant, especially as far as short- and medium-term goals of transition are concerned. In sum, it is unclear how the "transition" should be interpreted from a scholarly point of view. Therefore, in order to avoid needless terminological debate, the  meaning of  "transition" has been narrowed to economic  transformation only. In other words, the terms "transition", "economic transition",  "transformation",  "economic  transformation",  "transitioning  economies" and others of similar character are used as synonyms in the present study and are applied to the processes of macroeconomic change in the region taking place as a result of systemic state-led changes since 1989. The research setting is Eastern Europe, a collection of several associated countries which form an area of historical and economic unity. The notation "Eastern Europe" used in this study stands for the former socialist countries of Central and Eastern Europe including Russia and the Baltic Republics if not otherwise stated or implied. Since the scope of the present study is restricted to  5  economic transition only, and even to one narrow aspect of economic management, the terms "Eastern Europe", "East European countries", "postcommunist countries", "post-communist  Europe", and others of similar  character are used interchangeably without implying any substantial difference among them. Whatever the problems with defining transition in general, there is no doubt that attempts should be made to gain insight into one aspect of transition, namely labour market developments and the difficult choices which will have to be made in this field. This is vitally necessary because the rise of tension caused by labour market problems, specifically the growing level of unemployment, rapidly expanding large-scale poverty, as well as social and personal insecurity, may lead to social unrest. The main objective of the present study is to examine the role of labour markets in East European economic transition and the current labour market situation in different countries with a particular focus on the links between labour markets and the task of macroeconomic restructuring. The principal questions posed are, briefly: What is the role of labour markets in helping to ensure success for economic transformations? Secondly: What are the current problems threatening that success? And thirdly: What are the goals and mechanisms of labour market transition? It follows from the nature of the objectives that the study is confined to the macro level. It does not address the level of the region and/or enterprise.  In dealing with the problems of goals and mechanisms, the study is intended to incorporate policy-related comments. Every theoretical or empirical conclusion is intended to contribute to the solving of practical problems of the reforms, especially those related to the realm of "employment". Effective policies to prevent the persistence of high levels of unemployment in Eastern Europe, as well as the achievement of other policy goals, related to the labour market, must be based on a thorough assessment of past and present labour market conditions. This raises the problem of data availability and reliability. A common denominator of any Western study, devoted to problems of labour markets anywhere in the world, is a combination of and tradeoff between, on the one hand, household or enterprise surveys and, on the other hand, administrative data sources and data of national statistical agencies or their subsidiaries and analogies. One of the problems with East European statistics currently available is that each of the countries has just recently started changing bases for centralized data gathering in order to put it in conformity with internationally accepted standards. Every country of the region therefore has a far from appropriate balance between the two statistical sources. Balance, however, is urgently needed for developing a system for monitoring market conditions. Until recently, published labour force statistics relied almost entirely on information collected from the state enterprises by means of rather complex, compulsory statistical forms. These censuses were carried out on a yearly basis,  7 but enterprises in some sectors - mainly in the material production sphere were compelled to report on a monthly or quarterly basis depending on the sector, the number of employees as well as wages and earnings. Households were the statistical unit for the gathering of data on, say, living standards, time allocation and so on. Household surveys thus did not usually contain questions on labour market status. Even if such information was provided by the households, it was not utilized to calculate official labour market statistical indicators. One of the crucial features of changes taking place in the system of data gathering is that enterprises are no longer exclusively used for data collection. Without going deeper into the reasons for this change, we shall note that it has three main implications for our study in terms of the research data used. First, one can notice a discrepancy between the types of data used in different parts of the thesis. For example, while Part II, devoted to labour markets under communism, is based mostly on centralized labour force statistics, Part IV, dealing with the performance of labour markets under transition, utilizes household-generated data more widely. Even though these household survey data quite often result from an "urgency approach", made up of ad/resolutions (such as small sample surveys) and focusing on ad hoc indicators, the nature and scale of current labour market problems in the region foreclose the option of waiting until the reforms of national statistical systems are over. We do use such data in the study but are careful to point out their possible bias.  8 The second data-related implication is that of international comparability within the region, and with the West. It is obvious that different approaches to data collection  in different countries  not only preclude full-scale  comparisons, but raise crucial questions on statistical definitions in the context of transition as well. Of course, the majority of terms used in the study are defined. In all other cases, usage fully corresponds to that accepted in international practice. The same relates to statistical data used in the study. When international comparisons are conducted, we apply and describe methodology  which is used to adjust the data in order to make them  comparable. When statistical data are cited without any reference to differences from international practice, a lack of difference is implied. Third, the issues related to the improvement of data gathering methods and procedures seem to be only part of the general problem of developing labour-market monitoring mechanisms adequate for market conditions in Eastern Europe. In our opinion important issues arise with respect to the choice of indicators that could consistently summarize evolving trends in the transition phase. The indicators currently used in the West to analyze the degree of efficiency of labour markets are not necessarily the best for assessing employment and unemployment trends in the East. Rather, the characteristics of labour markets in transition countries suggest that some indicators are more valuable than others. This is the main reason why the present study offers a set of new, original estimates and indicators along with the traditional ones.  9 As a general principle in Eastern Europe there seems to be a strong case for developing mainly quantity-based as opposed to price-based indicators, even if the former require large statistical inputs. Sometimes a combination of the two brings interesting results (refer to Part III for details). Indicators largely based on price or wage statistics may not be particularly valuable, although perhaps less demanding in terms of the statistical information required for their computation, given the severe distortions characterizing prices and wages in East European countries. Much depends on the aims of indicators and what they are supposed to measure. When attempting to establish a sound methodology, research priorities should first be pin-pointed, in our case priorities regarding economic and social processes shaping  the  labour  market.  As Chapter  methodologically this study sees the transitional  2 makes clear,  development of  labour  markets as a triply-influenced process. First, labour markets have inherited a number of features from the pre-transitional economic and political systems. Although those features were slightly different in form in every East European country, they derived from the same causative mechanism, and it is therefore possible to generalize about them (Part II). Second, broad economic policy pursued by the present governments undoubtedly shapes both current labour market issues and the scope of active employment measures implemented. Chapter 3 of Part I and Part V amplify this statement.  10  Finally, the processes within the labour force itself induced by the systemic changes are important. Although the thesis pays less attention to this component,  Chapter 1 of Part V furnishes some useful insights into the  development of labour market organizations. It should be stressed again, however, that numerous social problems related to labour market issues (like, for example, those of marginal groups, which can be of considerable size and policy importance: of discouraged unemployed workers, involuntary part-timers, and people on training and other re-qualification programs) are too complex to consider within the scope of the present study. The focus therefore is on economic aspects of the three basic concepts underlying any description of the labour market, i.e. employment, unemployment, and economic inactivity. We  leave aside other important relevant topics, including earnings, social policy indicators, industrial relations and training. Some of them are touched upon only incidentally in relation to the three basic concepts mentioned. It seems worthwhile to make a few brief terminological points with respect to these three basic concepts. The term "employment" used in this study comprises all persons above a certain age, who, during a specified brief period, were either in paid employment or self-employed. It also includes people not working during the reference period but having a formal attachment to their job, such as those on educational leave, the sick, and part-time or occasional workers having worked a minimum time. It includes the police and armed forces, but does not incorporate  unpaid family workers, people on  11 maternity or parental leave and employed students. The author realizes that the definition of employment used in the paper is not exactly that accepted in the West. Although this may be called inaccurate, it has been done deliberately in order to facilitate international comparisons between Western and Eastern countries. A constraint of such an approach is that the indicator it creates measures the number of jobs rather than the number of employed workers (though both are sometimes provided), so people holding multiple jobs are often included more than once. Nevertheless, this indicator and this concept of employment seem appropriate in light of the research tasks posed. In addition, double-counting seems unlikely to be a problem in Eastern Europe at present. Although multiple job-holding is becoming very common in the region, people rarely work for two or more employers legally. In an underdeveloped legal system with a virtual absence of state control and punishment incentives, the amount and variety of black-market activities are increasing enormously. Numbers of black-market workers are extremely difficult to estimate because many people already working legally take an "under-the-table" job. According to some estimates, the share of these "multiple job holders" may approach 7080% of all multiple job holders in Eastern Europe (Nesporova, 1993). Actual labour supply is identified with the "economically active population". This comprises all persons of either sex who furnish the supply of labour for the production of economic goods and services as defined by the UN system of national accounts and balances, during a specified period. The  12  economically active population includes the self-employed, people with employment status even if temporarily out of work (e.g. women on maternity leave, soldiers in obligatory military service, etc.), members of producers' cooperatives, unpaid family workers, and the unemployed. The notation of economically active and inactive population is understood approximately the same all over the world which makes the figures easily comparable internationally. There are some minor differences which stem from varying definitions of economic activity depending on legal status. The remaining population is classified as "economically inactive". This leaves some working population not included in the category of economically active, such as occasional workers, working pensioners or students, black-market workers, the unemployed seeking their first job, workers under or over a specified age, etc. "Unemployment" is understood as a fraction of the economically active population, who are neither self-employed nor in paid jobs. "Unemployment" comprises only those persons who are registered as unemployed with National Employment Agencies or their analogues. In this study we do not make any distinctions between voluntary and involuntary unemployment. The methodology used in analyzing the development of East European labour markets as a triply-influenced process is closely linked to the objectives stated. The link is provided by the two crucial issues which run throughout the thesis and furnish a bond among all its parts. We would call these issues  13 "leitmotif problems". The first is the degree of labour market restructuring to be expected. The second is the origins and scale of East European unemployment. Research tools used in the study include both descriptive and formal analyses, utilizing statistical methods, economic simulations, and some mathematical modelling.  14 CHAPTER 2. LABOUR M A R K E T ISSUES IN EASTERN EUROPEAN TRANSITION: THEORETICAL CONSIDERATIONS AND A REVIEW OF LITERATURE.  A common approach to transition has been prevalent in the majority of Eastern European countries. Its main ideas are that resource allocation and economic  outcomes  should be left to market clearing forces, that  macroeconomic policy should be geared primarily to monetary stability, and that government should concentrate on the creation and preservation of a legal framework in which "business" can be done. In relation to labour markets some authors call this a "supply-side" approach (Boeri & Sziraczki, 1993). The main idea is that by removing certain types of protective government regulation the employers' costs of change will be reduced, labour mobility between sectors, regions, enterprises and jobs will be accelerated, and relative prices will be adjusted accordingly. Economic transition  in that view represents an  enthronement of the market, or "market populism" (Standing, 1991). The main aim of the present chapter is to identify and contrast the labour market policies and labour developments in Eastern Europe that could be expected to correspond to the supply-side approach and those that one might expect from a possible alternative perspective. The specific results of the first years of economic transition outlined in the next Chapter serve as a practical benchmark for this contrast. Since the interpretation of the above questions is built partly on past work by others, the present chapter is also intended to serve as a literature review.  15 In other words, a basic task of the chapter is to conceptualize labour markets in transitioning Eastern Europe. Transformation in the region has seen the juxtaposition of two perspectives on labour market policy. The first is defined by macroeconomic policy which is pursued by the government. The second, pursued mostly by public organizations, emerging unions, and small parliamentary oppositions narrows down to issues of distribution, appropriate institutions and the labour process itself. The intention is to identify competing hypotheses and clarify the underlying assumptions and values of the alternative views. What is more, theoretical considerations submitted below not only touch upon specific East European issues, but also link the discussion to the broader perspective of world experience of labour market development and government employment policies. Before turning to theoretical considerations of labour market policies, it seems useful to sketch macroeconomic policy frameworks that might be seen as alternative routes to economic transition, in the sense of enabling the labour market structures to respond to external pressures.  16 §1.2.1. Contrasting Macroeconomic Frameworks.  The present section attempts to identify the principal strategies for macroeconomic restructuring both in general and in Eastern Europe, and the logic underlying them, so as to determine whether certain types of labour and employment policies could be anticipated in different macroeconomic policy contexts. The question is whether those contexts have implications for the success or failure of the labour market policies adopted. One of the main macroeconomic frameworks is Social Keynesianism. Its adherents say that macroeconomic policy (both monetary and fiscal) should be aimed at securing full employment, while microeconomic policy is expected to regulate inflation. The fiscal constituent of macroeconomic policy is usually viewed as  a "counter-cyclical stabilization device", involving tax cuts in  recessionary phases and tax increases in booms (Standing & Tokman, 1991). As some sources note, for many years supporters of Social Keynesianism saw little need for active (i.e. counter-cyclical) labour market policy (Ibid.). It was therefore presumed that with macroeconomic "fine-tuning" full employment could be ensured so to say automatically, as industrial enterprises would be sufficiently motivated  to train workers and to pay for the required labour  mobility. Practically it was assumed that industrial growth would be "structurally stable", i.e. the industrial structure would change only gradually and not require massive labour shifts (Dornbush & Fisher, 1991). In fact, Social Keynesianism seems to be more appropriate to an economy based on giant  17 industrial enterprises in a situation of industrial stability and a labour market in which the great majority of workers are in regular, full-time employment. It is  understandable then why the research and writings of Social Keynesianists were received with more enthusiasm in the countries of the former Eastern Block than the works of followers of any other economic theory. Social Keynesian views of public expenditure as a factor facilitating long-term growth and industrial development made them seem almost in sympathy with central planning principles. The major difficulty for Social Keynesianism has been its apparent tendency  to generate accelerating inflation  (the Phillips-curve inverse  relationship, Dornbush & Fisher, 1991).The result is that through the theories of "rational expectations" and "hysteresis" Keynesians have given an increasingly prominent role to statutory incomes policy as the means of overriding the inflation-unemployment inverse tradeoff and enabling governments to maintain full employment (Lindbeck, 1992).This is sometimes described as "Fordism" as understood by the French Regulationist School.  1  Since this framework is  somewhat conservative from the present point of view and not suitable for periods of industrial dislocation and structural change, it is quite difficult to trace the use of its principles during the transition process in contemporary Eastern Europe. Another macroeconomic framework - Market Keynesianism - has been developed in the USA in the post-war period. It focuses on the institutional  18 part of development as opposed to the welfare-state priority which was typical in post-war Western Europe. Of course "institutions" and "welfarestate" are not naturally contrasting terms. What is meant here by "opposing" them is that the institutional framework has given precedence to "pronounced divisions of social functions" (Ibid.), with management preserving the right to manage and with labour only allowed to bargain about the effects of management decisions. So, it seems that the cornerstone of labour market policies in such a context  is a law - the prime example being the US National Labour Relations Act (the Wagner Act) - that not only limits government intervention in private enterprises, but also puts strong curbs on the bargaining rights of trade unions. Such a policy could be expected to result in greater fluctuations in economic activity, meaning more bankruptcies and unemployment in recessionary periods and more dynamic spurts in times of economic growth (Standing, 1991). (More bankruptcies will occur if the state does not actively support the private sector). Relatively inefficient or unprofitable sectors would not be preserved, or at least not to the same extent as under Social Keynesianism. This framework does not seem viable in relation to the contemporary  Eastern Europe with its  contradictory legislation, collapse of executive power and high degree of income redistribution. One version of Keynesianism that has been taken up in some East European countries (the former Czechoslovakia and, partly, Poland) is the socalled "Scandinavian Model" of macroeconomic development. While it gives a  19 pivotal role to counter-cyclical fiscal policy, macroeconomic policy sets an employment level below full employment (Flanagan, 1987). What is meant by "below" is that magnitude which is precisely necessary to prevent the emergence or intensification of inflationary pressures. Even so, the overriding socioeconomic objective is full employment and a progressive redistribution of income and welfare (Ibid.). To ensure full employment, a crucial role is given to active employment measures (AEMs), where the "active" refers to their counter-cyclical character. The number of workers covered by such measures as training, retraining, and subsidized employment schemes is expected to rise in recessions and periods of structural change and to fall in periods of growth (Burda, 1988, 1993).  AEMs are "also expected to be an instrument in  controlling cost-push inflation, primarily by facilitating various forms of labour mobility and reducing their cost" (Standing & Tokman, 1991). In the Scandinavian Model the above mentioned statutory incomes policy is ruled out as a macroeconomic means of controlling inflation, on the grounds that it introduces rigidities into the labour market. Instead, reliance is placed on mechanisms of centralized bargaining and consensus. Accordingly, the success of macroeconomic policy depends on the institutionalized  and socially  responsible participation of representative organizations of employers and workers (Lindbeck, 1992). This model therefore cannot work without social consensus, a fairly organized solidarity of representative organizations of employers and workers and the existence of a strong trade union movement  20 representing the majority of  workers. It would be a dubious model for  countries without a "government of social trust" and in a state of social and economic flux. The Scandinavian model also differs from conventional  Social  Keynesianism in its regulation policy. Under the latter the emphasis of regulation is on the preservation of macroeconomic stability. The emphasis in the former is on facilitating economic change without deserting the basic Fordist principle underlying all versions of Keynesianism - an institutional preference for wage-oriented, or negotiated, rather than market-oriented methods of regulation (Ruccio, 1989). In the model the labour market organizations are obliged to negotiate structural change, whether in the field of technology, organizational structure, industrial relocation, job structures or employment changes (Lindbeck, 1992). Another aspect of the macromodel is its dependence on an undervalued exchange rate as a means of boosting exports. Private ownership plays a crucial role with light taxation on capital and profits and hard taxation on consumption. Finally we come to "Market Populism", or the Supply-Side Perspective (SSP) which has been followed by the majority of Eastern European countries as a core framework of macroeconomic policy. As is well-known, in the 1970s 2  Keynesian methods, assumptions and values were attacked sharply (Friedman, 1977). Out of monetarism developed the SSP that crystallized in the orthodox  21 strategy for stabilization through structural adjustment that has been promoted intensely in both developing and industrialized countries in the 1980s. This orthodoxy is sometimes called "supply-sideism" because of the overwhelming concern for adjustments of the supply-side of capital and labour markets to what is supposed to be a stable macroeconomic policy environment (Standing & Tokman, 1991). According to this view, the Keynesian pursuit of full employment and rapid growth by macroeconomic policies induces not only accelerating inflation, but rigidities that arrest structural and labour market adjustments, undermining competitiveness and intensifying price distortions. According to the supply-side view, fiscal and monetary policy should not be counter-cyclical, but rather be geared to the reduction of government deficits, with the objective of securing a balanced budget (Friedman, 1968). Indeed, a premise of the SSP is that macroeconomic policy cannot determine the level of employment and unemployment, or the level of economic activity, and that there is a "natural" rate of unemployment set by the institutional and regulatory structure of the labour market and the behavioral adjustments that occur in it (Dornbush & Fisher, 1991). The SSP accepts the existence of wide income differentials to maximize incentives to factor mobility, and for similar reasons advocates low direct tax rates. It advocates privatization of economic activities and a dismantling of the institutional-redistributive welfare state, turning it initially into a residual welfare state geared to selective protection for the most disadvantaged. Given the balanced-budget objective (the main aim of so-called  22 "shock therapy"), only selective employment measures are feasible, either as counter-cyclical or as longer-term development measures. It is thought that mechanisms such as minimum wage rates are "rigidities" that introduce price distortions and raise the natural rate of unemployment. According to the SSP, unions interfere with market forces and should be curbed as monopolistic entities (Standing & Tokman, 1991). Instead SSP advocates that in the interest of flexibility, hiring and firing decisions should be the prerogative of management, with minimum restrictions, and that union interference in those matters should be prevented (Lane, 1987). Each macrostructural developmental framework entails  a particular  view of labour market policy, in terms of both the desired objectives and components. We could say therefore, that the transitional development of labour markets in Eastern Europe depends on three main factors: a) preceding development strategy; b) labour market components of the macropolicy (stabilization - adjustment package); and c) cultural, behavioral, historical and the like factors, determining the directions and dynamics of the "internal" development of the East European labour force. As has been mentioned, the present thesis mainly addresses the first two points.  23 §1.2.2. GENERAL POLICY.  M A C R O PERSPECTIVES ON LABOUR M A R K E T  Accompanying loss of faith in Keynesianism and the enthroning of market mechanisms, there has emerged a corresponding orthodoxy on appropriate development strategy. As happened earlier with some other countries, the typical stabilization and "structural adjustment" package has been superimposed on Eastern European countries. According to some sources, there are two complementary orthodoxies in place: an outward-oriented development policy and a supply-side stabilization. The latter essentially means a tight monetary policy to combat inflation and hold down domestic demand, so limiting imports (Blanchard, 1991). That means lowering many people's living standards and employment in the short term. Coupled with that is an outwardlooking development policy, which means trade liberalization, by nominal and real exchange rate devaluation (Ibid.). In recent years the pressure for adjustment to external economic shocks and the belief in outward-oriented development have led to a sustained critique of many labour market policies long perceived as desirable attributes to development. As a consequence of the supply-side critique, we have witnessed the juxtaposition of two perspectives on labour market policy:  "social  adjustment" and "supply-side" models (Standing & Tokman, 1991). These are not just theoretical models, since each has been applied in practice. For a long period the dominant role in economic thinking could be summarized as the social adjustment perspective (Ibid.). The underlying  24 assumptions were Keynesian, backed by a social democratic ethos and a belief that markets could and should be circumvented or moderated by institutional and other regulatory devices in the interest of both equity and long-term economic  growth. For well-known reasons this perspective was thrown away in the 1980s, the result being the triumph of the SSP, based on an overwhelming faith in markets untrammelled by protective regulations, collective organizations or other institutional interventions (Ruccio, 1989). The pursuit of outward-oriented development became a global panacea. The SSP has been very influential over the past decade, but one can anticipate a continuing debate between its adherents and those of a reemerging social adjustment model more concerned with distributional issues, institutional structures and the labour process itself (Standing, 1991). It is interesting to look at the contrast between the views of the two perspectives on labour market policies with relation to Eastern Europe. Has the development of labour markets in the region so far corresponded to either of the competing hypotheses? The essence of social adjustment strategy is "growth with social protection" (Lindbeck, 1992). Among its elements are the following labourrelated aspects. First, to alleviate poverty, there are usually food subsidies and price support systems to encourage the production of domestically-consumed food. Supply-side critics say that these are market distortions. Second, there is usually some minimum-wage protection machinery. Critics say that this is a  25 market distortion, which raises wages above the market clearing level, deters employment  by favouring capital-labour substitution  and increases the  inequalities between the formal and informal sectors of the economy (Howitt, 1985). Third, there are institutional forms of labour security protection -safetyand-health standards, employment security regulations, bans on moonlighting, etc. Critics say that all these represent non-wage labour costs and rigidities that impede labour mobility and thus efficient resource allocation (Boeri & Keese, 1992). Fourth, a social adjustment approach would foster freedom of association and legitimacy of collective organizations. Trade unions should be given an active role to restrict all kinds of discrimination. SSP critics see such organizations as' market distortions in that they limit the ability of firms to react to market forces, limit the realization of high profits boosting investment and limit "individual freedom" (Feiner & Roberts, 1990). Fifth, in the social adjustment model public expenditure is regarded as complementing private expenditure, as potentially productive and as a means of mobilising and retaining resources for national development. Critics, again, consider this to be a distortion, arguing that it results in "financial crowding out" of private investment and growth while being unproductive, unresponsive to market forces and sets an inappropriate standard for wages and conditions of employment (Flanagan, 1992). By contrast with the social adjustment model, the labour-related policies associated with the SSP focus on price mechanisms and overcoming what are  26 depicted as price distorting statutory interventions. Thus government should cut subsidies to basic consumption goods because they distort market signals. Critics from the social adjustment strategy may say that the removal of subsidies results in social distortion and a loss of welfare (Ibid.). One  of the most  important  differences  refers, of course, to  unemployment. The supply-side school expects a period of resource reallocation -through trade liberalization, market deregulation, etc. - to lead to a transitional period of high unemployment. Advocates and critics alike agree that such unemployment should be minimized. But SSP adherents would expect the unemployment  level to fall to its "natural" level as a result of market  mechanisms, explaining any prolonged high unemployment as due to "market failure" or "voluntary unemployment". Social adjustment proponents would call for adequate institutional mechanisms to secure labour reallocation, and thus would attribute high unemployment to inadequate labour market policies and inadequate institutional mechanisms for securing labour reallocation (Standing & Tokman, 1991). One of the most exciting prospects of the social adjustment model is the reorientation of active employment measures (AEMs). Whether or not one  accepts "post-Fordism" as a general trend, the growth of labour market flexibility implies abandonment of the old social Keynesianism version of full employment. But if we can foresee a need for worksharing, more labour mobility and a trend towards lifetime flexibility, we will have to focus on  27 measures to make labour market services more flexible. Although the debate around flexibility of East European labour pertains rather to the realm of fiction than everyday reality, one part of it deserves special attention, namely training universally favoured by both supply-siders and social adjustment adherents. But there is a difference in emphasis. The SSP view is that schooling and training should be geared to the needs of industries with externally tradeable product and that policy-makers should take a fairly straightforward economic rate of return approach to educational investment. The social adjustment perspective would see education much more as a social end in itself, as well as a means of promoting development (Boeri & Keese, 1992). So, as we have already mentioned, labour markets are influenced not only by the direct labour market policies of a government, but also by the broader economic policy it pursues. This puts the questions of labour markets (and in particular - unemployment) on a large and very complex scale and shows that those issues are in fact interconnected with the whole system of government attitudes, value systems and actions. Therefore there are questions about the initial results of economic transition in the region. They are incorporated in the next chapter which should be examined before moving on to the narrower and more specific issues of labour markets.  28 CHAPTER 3. HISTORICAL PERSPECTIVE ON TRANSITION TN EASTERN EUROPE: GENERAL ECONOMIC O U T L O O K .  The present chapter elaborates on some results of economic transition in Eastern Europe, shows their relevance to the analysis of labour markets, and fits them into the theoretical framework outlined in the previous chapter. The conclusions offered here are relevant, in principle, to the whole group of East European countries, although in many cases empirical data, sufficiently reliable to draw general conclusions, are not yet available. What is more, the process in question is still in statu nascendi and therefore any too far-reaching generalizations would be fraught with risk. In addition, and this is quite obvious, the experience of transformation in the post-Soviet Asian Republics is quite different from that in Poland and Hungary. In the latter case, the marketoriented transformation process is to a large extent already qualitatively advanced, and the starting-point of that process was considerably different because of the previous significant progress in market-oriented reforms of the centrally planned economy in those countries. Irrespective of how these reforms are assessed in detail there is no doubt that they have facilitated the process of transition to a market-based economy in general (Nuti, 1992; Kolodko, 1993). Keeping in mind these and other differences, it is nevertheless worthwhile attempting to answer questions about general conditions and prospects for stabilization as well as about causes of severe output collapse in the post-communist economies. The necessity of such an overview is in part  0J3 1988  1986-  29  ^ m  ON  +  :  w  :  m'  •  ^  si.  m  :  w  1985  s  1980  1976-  1981-  >  in cn"  ^ si.  T t  vd  ^  ^  +  CN  w  oo  ^  *  si.  i  ^  '•  cn cn  i  ^  ON  w  d  + s_.  ^  si.  5i VO  vo  3  o  ^  T t  >i.  r-'  ^  o  :  u.  CD EX  cn cn  ^ si.  Pi  "  ^ -a  1975  1972-  i  ^  00  + w  r-  ^  -*'  si.  si.  o in'  ^  ON  ^  si.  -*'  si-  >n ^ -H"  +  t—H s^.  o in  ^ si.  <tf 5 c  1971  1968-  cn si.  in'  ^ si.  ^ + in w  CN  ^  VO s i .  s +o  t-^  ^ si.  in  ^  VO s i .  1967 1963  1960-  1964-  O N <—-  5i o  ^  OO  ^  oo  +  w  ^  °. + in w  ^  in  ^  si.  -*'  si.  T *  rt'  Pi  O r-'  ^ si.  Pi  rt  6 .S i  r-'  c  'Si  vo % S •=! "o <*-  m cn  ^ ^  +  +  OO s ^  vo  r-'  ^ si.  q vd  I  GO  3  O  ^ si.  1959  1957-  X ^> ^ *—1  +  s^  o  Si Pi  d  ^ +  m  ^  VO s i .  T t  ^  CN +•  sD  ^  d i  a^  .2 3  18*  1956  1953-  cfl  GO  m  ^  in  ^  VO s i -  vd  si.  ON  in'  ^  VO  ^  ^  si.  rs'  CN  si.  ON'  si.  ON  cn'  ^  CN  ^  si.  OO  w  CO  u ... g  1952  1950-  CN 3 S in d  cn  cn  in  ON  ON  • ^  VO •>*  q r-*  .a -a W  frt  VO  u •!->  a  5  3  rN  m  , ^ op ccj  T3  5 : &  30 to make the point that many contemporary sources tend to ignore the fact that during most of the period of so-called "real socialism" there was a rather high rate of economic growth (Kolodko, 1986; Wellisz, 1991; Micklewright, 1991). Of course, one must be aware of the errors and methodological unreliability of the official statistical data reflecting the economic growth in 1950-1989,but on the other hand the resulting revisions are not qualitatively significant (Kolodko, 1993, see Table 1). This tendency to neglect the scale of pre-transitional economic growth in Eastern Europe is in part a result of the low rate of economic growth during the decline phase of "real socialism" and in part is deliberately formulated out of political and ideological motives. It is worth stressing once more that the downgrading and/or negation of past economic growth can put the contemporary processes in a different context.  3  One of the most important macro-characteristics of economic growth in the centrally planned economy was its cyclical character, understood as "endogenous oscillations of the growth rate around a long-term ascending trend" (Kolodko, 1986; Table 1). A distinct deceleration in economic growth in the late 1980s was an undisputed fact. What is more, it was accompanied by increasing destabilization expressed in, among other things, price inflation, growing shortages, and deepening external deficit. (For a detailed summary of these macroeconomic indicators see for example Bergand and Sachs, 1992 and Table 29-31). It is  31 understandable that a negative feedback among those processes took place and its results are still felt today. The decreasing economic activity strengthened the destabilization trends, and the growing destabilization, in turn, bolstered the trend towards a slowdown of growth and, finally, contributed to stagnation (Kolodko, 1989). However, a decline in output in those countries did not take place until 1989. (One version of statistics is shown in Table 2). Except for the Romanian and to a lesser extent the Hungarian and Polish cases, a fall in GNP appeared only in 1990. A clear collapse was visible in that and the two following years, parallel 4  with a simultaneous decline in growth dynamics in the developed market economies. A very important fact is that the recession appeared only with entry into the phase of systemic transformation. This could lead to the conclusion that the decline in economic activity was directly connected with the systemic transformation. Therefore, the question arises as to whether it was really so and to what extent the fall in output in the post-communist economy was a function of the transformation and its three components, i.e. macroeconomic stabilization, institutional  changes, and microeconomic restructuring of  production capacities. To what degree can that decline be explained by other factors?  5  In the classical business cycle theory, recession follows boom and precedes the next depression. From such a perspective the emphasis is put on  32 a certain regularity of the process. However, more important than the relative regularity is the mechanism of shifting from one phase to another. In the business cycle this mechanism has an unambiguously endogenous character.  6  Already in the recovery phase, the prerequisites for future economic overheating and a relative or absolute decline in output level are inherent (Dornbush and Fisher, 1991). Conversely, in the recession phase, the prerequisites of transition to a recovery phase are present which in the classical cycle was preceded by a depression phase. In the case of post-communist economies such a sequence does not occur, although a comparison with the classical cycle may to a certain extent arise. The most important statement to be made at this point is that in the postcommunist economies the business cycle is not the same as in a market economy. The post-communist economies - by their very nature - are neither market nor planned economies.  The recent output contraction in transitioning economies is characterized by features that make that phenomenon different from typical recessions, which have been well-described in the past. First, it follows a phase of low economic activity (with propensity to stagnation) characteristic of the cyclical character of growth in the socialist economy (see evidence below). Second, it manifests itself in, among other things, an absolute decline of output and investment level (though not necessarily in other macroeconomic aggregates, such as exports Kolodko, 1993). Third, the mechanism of transition to the post-recession  33  recovery does not work. Fourth, the sequence of phases that follow the recession during transition is not predetermined: it sometimes may be a sequence comparable to that of the classical business cycle, which means that immediately after the decline in output the economy enters the recovery stage. It also may happen - which is more probable and its symptoms are already visible in at least some post-communist countries - that recession will be followed by a depression that, after passage through recovery, will lead to boom. It is only then that growth will occur and its course will more or less follow the modern business cycle in capitalism. One may expect that the events will follow - and in principle already are following - the latter scenario, which is supported by a number of facts. In Bulgaria, the Czech Republic and Poland the very low production activity has already persisted long enough to permit talk about the appearance of the postrecessionary depression phase. The latter is characterized both by the lack of any clear trend toward a further output decline and the absence of signs of growth trends. In other words, there is stagnation on a very low level accompanied by further growth in unemployment. This has already become a specific feature of depression in post-communist economies, distinguishing this phenomenon  from its classical course. The origins and character of  unemployment in the region - the most important indicator of labour market performance - are different from those in the West, these differences being dominated by the specific features of recession.  7  34  CN ON ON  VO  O  m o oo in  CN CN  CN  CN  o CN  CN  ON  VO  O  CN  CN  CN  CN  CN  £  1  P-  £3 PH ON ON  in r-'  ON CN  ON  in  o o o CO 00  in  o d  in  CN  ON  CN  cn  in CN  r--  cn  CN  O  ^  >.  o ho  o  oo  in  ON ON  co co  o  vO |  m  in  oo CN  oo  ON  CN  CO  CN  O  CN  CN  8 D  8 .2 ON OO ON  CN ON  i  ON  oo  CN  d  ON  d  CN  o  o  o  in  in  CN  in  ON  d  ON  oo  ON  CO  CO  oo co  CN  co CO  o CN  oo  d  in co  in  co  vo o  CO  in  CN  CN  CN  co  ° cS  a  a  1  oX oo oo  D CN  I  VO  oo ON  co  co  oo  CN  00 CN  CN  d  vo  CN  co  VO CN  O  CN  VO  CN  CN  in co  co  EG  w  S  3  t-H  o  PH  piO PH  in oo oo  '3 CN  d  ON  CN  co  CO  CO  CN  in  Q. CO  oo co  ON  I  CN  CO  ON  V5  > o  o O  i—I PH  PQ  N U  O  CO  < CO  o P< p<  PH  3  CO  O PH  u ww o w  p  Q U  , N U  Q  cT c ^ E ON O  ON  o M  -o o  ai o  <u -t—»  o  PH  CQ  o co  35 However, the sequence and intensity of particular phases will be ultimately decided by the economic policy pursued in individual countries, especially by the systemic transformation policy in all its three components. This is illustrated by, among other things, the experience of the 1990-1992 period. The decline in output was so drastic that even in its optimistic forecast of the future the World Bank assumed that the output level prior to that collapse would not be regained earlier than in 1996 for the so-called northern part of the region (Czech and Slovak republics, Hungary and Poland) and in 2000 for its so-called southern part (The Transformation, 1991). As is apparent from many relevant sources, there was a tendency to downplay the scale of decline in production, a consequence either of excess optimism or of the embellishment of reality for political reasons. For example, Berg and Sachs (1992) suggest a much smaller fall in GNP in Poland than in fact occurred (3.8% instead of 7-8%).In addition, these authors explain the fall mainly as a result of the decline in output previously exported to the USSR. But in this case how can one explain the output decline in Albania which was for almost 30 years outside COMECON? The above remarks touch upon a key issue of the present chapter, namely the causes of recession and of the depression accompanying the marketoriented transformation of the post-communist economy. A great variety of assessments and explanations, frequently contradictory to each other, arose in  36  the 1990-1992period, which makes it rather cumbersome to offer the detailed analysis of literature which is expected in such a thesis. What follows below is a generalization of the main ideas and the author's interpretation of the events in terms of the diversity of existing explanations. The partial recognition of the considerable output decline reflected in official statistics was often accompanied by a suggestion that this had no negative significance (Van Brabant, 1990). Here, a particularly unconventional explanation is that it took place because of the existence of over-industrialization at the beginning of post-communist era (Berg and Sachs, 1992). The process of structural adjustment has therefore to take place not so much through a more rapid development of the service sector as through a drastic decline in industrial output and shift of resources to the service sector. In the light of such an approach, the recession is not a negative phenomenon but, above all, an "economy-clearing" process, improving the overall economic structure of the country in question. According to another interpretation, the basis from which the decline in output fall was calculated had been permanently exaggerated in the past, hence the real decline in output was allegedly less than that shown by statistical data. Similarly, where the output had fallen, this was allegedly caused by the elimination of production that should not have been undertaken in the first place because of its unprofitability (Winiecki, 1991). There are individuals and institutions who are interested in exaggerating  37 the favourable results of the transformation and downplaying its costs. In extreme cases they talk about the so-called "perception error", implying that the situation is better, but that the people in the countries involved in the transition process are incapable of perceiving it.  7  Sometimes the assessments are distorted the other way around. Accordingly, because the whole process of systemic transformation is entangled in a huge redistribution both of resources (wealth) and flows (incomes), the whole hitherto existing network of political and economic interests has been affected by it. These interests, it is asserted, exert influence on the various assessments , which means that such assessments are often of a political rather than a scholarly character (Kolodko, 1993). Thus far four separate sources of recession in the post-communist economies have been identified in the literature. First is the systemic and structural legacy of the socialist economy that was gradually losing its momentum until the appearance of stagflation trends in its decline phase. It may be assumed, although it can hardly be proved, that if the countries of Eastern Europe had not entered the systemic transformation phase, they would have ended up in a recession anyway, although its nature and depth and duration would have been considerably milder. Second, there were the external shocks, some of which were of a specific character in that they were not exogenous in the full sense of the word, but at least partly, self-imposed by political decisions.  38 Third, there were the effects of macroeconomic stabilization and of getting rid of the economy of shortages. Every stabilization effort tends to be accompanied by a decline in economic activity because of the suppression of aggregate demand (Bruno, DiTella and others, 1989). This decline was unavoidable under a stabilization policy oriented not only towards reduction of the inflation rate but also toward the elimination of shortages and the introduction of a market-clearing price system. Fourth, there was the elimination of the negative value-added output through trade liberalization and opening-up of the economy to external competition. There is also a fifth case - in our opinion the most important one which in some countries has proved to be the most momentous factor contributing to recession and depression. This cause is the economic policy pursued and, more precisely, the errors in that policy, consisting first of all in the false sequencing of measures and overshooting of the macroeconomic stabilization as well as mismanagement of the state sector at the early stage of transition (Kolodko, 1992).  8  As mentioned above, there is a strikingly large gap between the expectations of various processes occurring during systemic transformation and the reality. According to the most frequent expectations, a minor and shortlasting recession was to be followed by a quick entry into a dynamic growth phase. Such an error was committed by some very important institutions - such  39 as the World Bank and the IMF - as well as by a number of individual scholars and experts. In all such cases one can talk about the errors of excessive optimism. Its first and obvious cause pertains to the excessive official optimism on the part of new elites and their governmental bureaucracies. Inadequate professionalism, and a lack of imagination and responsibility, together contributed to a formulation of unattainable goals such as the assumption of a mere 3.1%decline in national product and a 5% reduction in industrial output in Poland for 1990, or the expectation of an entry into the recovery phase as early as the end of 1992 in Russia (Kolodko, 1992). The record shows that politicians in those countries, often knowingly, proclaimed quantitative goals the attainment of which was at least doubtful. These claims were made for tactical reasons and for political strategy, and they were to a certain extent understandable, but nonetheless hardly justifiable. It is understandable that the government of a given country, even if it could have foreseen it, would not be willing or able to declare that its policy would within two years lead to a GNP decline of 20%, with a 40% reduction in industrial output and the emergence of a high rate of open unemployment. The announcement of such economic programs would result in the overthrow of the governments well before they were able to put their program to the test (Kolodko, 1992). The second group of causes stems from an incorrect identification and diagnosis of the realities of the post-communist economy. Today the situation is incomparably better explored and identified than two years ago. Here it is  40 worth mentioning such simplifications as comparing the state of affairs in postcommunist Eastern Europe with the situation in post-war Western Europe, with its expectations of rapid growth based on an apparent analogy with the recovery from the devastation caused by the World War II (The Transition..., 1991; Wolf 1992). Echoes of such an approach, though becoming weaker and weaker, still persist, especially in some Western circles. However, greater significance is to be attributed to the approach that treats the post-communist economies in transition as having a system and structure typical of the developing countries (Kolodko, 1992; Dornbush, 1991). To a large extent such an approach influenced the attitude towards the transformation processes adopted by the World Bank, the IMF and the US government, as well as by the experts sent by them to Eastern Europe, the majority of whom were entering this part of the world for the first time in their lives. While it might be understandable that in official documents of international financial organizations this group of countries has passed from the dwindling category of "planned economies" into the category of "European developing countries", objections should be raised to the treating of those countries as not differing much in their socio-economic characteristics from the so-called less developed countries (LDCs, see Chapter 4 below, Kolodko, 1992). In accordance with this approach - and it is to that approach that ultimately the whole transformation policy has been subordinated - the structural and institutional problems to be solved by post-communist Europe resulted only  41 from quantitative differences with respect to price distortions as well as from underdevelopment of market institutions, in particular the financial ones (Eichengreen and Uzan, 1992). The third cause of false expectations stems from their methodological and substantive weakness. Many forecasts, especially the econometric ones, were based on dubious and often hardly justifiable assumptions. For example, Borensztein and Montiel (1991),using regression analysis, claimed that as much as 75% of investment in the planned economy had been ineffective. The fourth cause of excessive optimism has been the naive expectation of large amounts of Western economic aid and its salutary, stimulating impact on production. This belief consisted not so much of groundless expectations of the absolute size of the aid, but of the effectiveness of that aid. It is worth 9  mentioning that by 1992 the scale of Western aid to European transitioning economies  (excluding the former USSR) even in real terms had already  exceeded the aid granted by the US to Western Europe under the Marshall Plan (Eichengreen and Uzan, 1992). However, the impact is much less than had been expected, which can not be explained without reference to the mechanics of distribution of that aid.  10  The fifth group of causes consists of political and ideological considerations influencing actual political choices and the manipulation of public opinion which tended to be uninformed and gullible. It is otherwise difficult to explain the promise of a one-digit inflation rate one month after the  42 introduction of the stabilization package or the notorious 500-days Plan in Russia. The above discussion has attempted to characterize the causes of excessive optimism concerning growth under the transition. For the sake of the analysis in the following chapters, it must be noted that some expectations were justified. In particular, the positive effects of certain institutional changes such as those relating to deregulation and competition-promoting policy, as well as to the general liberalization and financial reforms in the majority of countries in question and to the efficiency-improving effects of macroeconomic stabilization, could and should have been expected (Sachs, 1991;Portes, 1992; Church, 1992; Kolodko, 1993). The inaccurate forecasting was a result of an improper sequencing of institutional changes and of a distortion of the stabilization policy in some countries of the region (Nuti, op.cit.). It is with this in mind that one should view the forecasts of the IMF formulated in agreement with the respective governments  in the spring of 1991, regardless of the fact  that they too were to a certain extent fraught with some of the above-discussed distortions (Table 3). Subsequently, the IMF has formulated more sceptical estimates, above all with respect to the economic growth rate. The reports of the World Bank, OECD, and EEC have similarly toned down their forecasts, drawing similar conclusions from gradually accumulated experience (Kolodko, 1993). Earlier, some very optimistic forecasts were offered. At the time when  43 the IMF assumed for 1991 a GNP decline in six countries of 1.5% only and then - for 1992 - a growth of 2.8%,some sources (Borensztein and Montiel, op.cit.)  predicted for Poland and Hungary a growth of 6-7% and for  Czechoslovakia - 3.25%. Almost at the same time, the World Bank, using different forecasting techniques, also assumed a prompt entry into the phase of growth and its acceleration in the latter half of the decade (Summers, 1992). According to those forecasts presented in Table 4, the GNP was expected to grow in Poland for example by 6% in 1993 after a 2% growth in 1992. In reality, it fell by almost 8% in 1992 and rose by approximately 1 % in 1993. The magnitude of error was considerable, disqualifying both the assumptions adopted and the forecasting techniques. How otherwise can one treat a forecast of GNP growth in Poland by nearly 14% in 1991 and 1992 in a situation where this indicator has actually fallen by almost 10% over the same period? (Gomulka, 1990). Practical experience, as illustrated above, leads to a conclusion that the transition from stabilization to growth is not automatic. It requires a skilful economic policy to be implemented by the state and efforts to be made on the part of the society as well as outside support. Putting aside the greater than originally expected difficulties of macroeconomic stabilization, the creation of a growth mechanism requires a greater effort in the area of structural adjustments than in the case of underdeveloped market economies. Critics of the traditional (SSP)approach to structural adjustments believe  w ^  CM  Easitern Europe  tern Europe ormer USSR  1992  CN i  1991  -<* i  1990  -3.8  -8.6  1989  °) -0.9  CO CN  1988  1993-96  44  CN  OO  CN  m i  45 that its advocates accept as given the fact that the market by itself will secure the availability of the organizational, managerial and technological skills necessary for the effective functioning of the private sector. As experience shows, this is not usually the case, and the arbitrary assumption of the existence of the basic factor of sustainable growth - human capital with the desired skills - cannot be accepted, implying that it is the state which has to help develop this factor. It seems that in this regard the post-communist economies are in some cases in a better position and in some other cases in a worse position in comparison with the underdeveloped market economies (Sziraczki, 1990; Nee, 1989). Very important is the fact that educational level of the post-communist societies is relatively high, but often ill-suited to the needs of the market economy (Kolodko, 1993).  46  o o o  fN  CM  cn d  CN  fN  cn  m  fN  fN  CN*  CN  NO  d  ...  S3  O 1  c 3 S c ON ON ON  ON CN  in  q  CN  in  in  in  O O O H  TT'  TI-'  I  cn  cM  o 3 & '3 CM 3 '5  oo ON ON  Tt  q  ON  CN  CN  in  q "*'  in  oo oo oo cn CN  * &  ON ON  ON CN  oo i q  in  o  Tt  T*  Tt  oo oo oo cn ^t- C N  If  .  OH •J  CM  o CM 0NO ON ON  Tt  cn  CN  Tt'  in  q in"  q cn  CN Tj-"  in cn cn rr C N  |.<3  Tt  0) C3  E «  m ON ON  cn  CN  q in  cn  oo cn  o S <U N  C  w  cn cn CN cn -<* C N  £  'I"  1  i I cM  /II  o o c  is"  l-H '3 (D 03 O X ! +-» O <D N 1/5  ON ON  Tt CN  in  Tt  cn  in  cn  cn  cn  O  —I  -vt  CN  U  •a  t-c  cM  <u  I CM  3  cn ON ON  cn  CN  CN  Id  o  in  NO  CN  VO  cn  fN  cn  oo  ,o  O =3  y-4  CM  TT  CM  £  CM  g>  ttf)  N U  o  OH  z  Pi  o  PQ o i c PQ i—I  w  CM  ca O  PQ S  |i Z  i  o o o  o  l-l C/5  47  PART II. THEORY AND PRACTICE OF LABOUR MARKETS UNDER CENTRAL PLANNING. CHAPTER 1. T H E HERITAGE MARKETS.  O F EAST EUROPEAN  LABOUR  The present chapter attempts an overview of the main features of East European labour markets prior to the current wave of reforms and tries to single out particular areas which are affecting the nature and outcome of transition. As was mentioned elsewhere in this thesis, economic growth in CPEs typically came from more extensive use of inputs, including labour, and to a lesser extent from their more intensive use (that is, from increases in productivity). The mobilization of women in the work force was a key factor behind high participation rates of the population of all ages when compared with OECD countries. Male participation rates were more in line with OECD rates and exhibited a similar decline over time (Boeri and Keese, 1992). Table 5 illustrates labour force participation rates in the Baltic States in 1989. Average figures are typical for all East European countries. In theory, employment  was guaranteed for all; hence, official  unemployment was either negligible (in the case of Hungary, see below) or nonexistent. The compulsory socialist paradigm in labour economics had its roots like any other branch of economic study - in the complete theoretical identification of state with nationwide ownership. This identification was used for further  48 Table 5. Labour Force Participation Rates in Estonia, Latvia and Lithuania in 1989, as a % of total population. Age groups  T  Estonia M F  T  Latvia M F  Lithuania T M F  16-19  36 39 32  37  39 35  29 33 25  20-24  79  80 77  81  82 81  76  79 73  25-29  93  96 90  94  97 91  92  97 88  30-49  96 97 95  96  97 94  95  97 93  50-54  90  92 88  90  93 88  88  92 84  55-59  67  83 54  66  86 51  62  86 43  60-64  43  52 38  41  52 34  33  44 25  65-69  29  36 25  26  35 22  19 27 14  70+  9  16-54/59 16-64  6  5  87  88 85  84  86 81  83  85 77  78  83 72  8  8  86  87 85  80  85 77  13  13  11  4  Source: Oxenstierna, 1991. direct substantiation of the "absence of hiring and exploitation" in CPE , of the 11  growing "immediacy" of the social nature of labour , of the claimed allocation 12  of consumption goods "in accordance with the quantity and quality of work",  49 of full employment as an "incontestable historical achievement of socialism" and so on (Ivanov and Ushkalov, 1991a). The question of unemployment under socialism was replaced by the full employment doctrine. Searching for general methodological support in Marx's Capitalist Law of Population, the "Socialist Law of Population" was understood as "full employment of the whole fraction of the population able to work" (Paunov, 1993). Full employment was treated as a situation in which the labour of every citizen able to work was utilized. Questions about the rational (natural) rate 13  of unemployment were either completely ignored or treated as inseparable from the full employment assumption (i.e.socialist macroeconomic equilibrium may be reached either only with (or at least in spite of) full employment - see for example Logvinov, 1972; Penkin, 1990) . This approach, usually bound to 14  postulates about the obligatory character of labour , ultimately led to a very 15  high labour participation rate by every social and demographic fraction of the population able to work, mainly by simply increasing the number of jobs available and the incentives for labour supply (such as numerous non-pecuniary benefits for working people, for example, free day care and primary education as well as Summer camps and other forms of care for children, etc). In a situation characterized by a stable or declining number of people able to work (partly due to the full employment, partly to general demographic trends) labour shortages and an excess of jobs became apparent. A specific, resourceconstrained labour market emerged.  50 Under these conditions of a resource-constrained labour market and a surplus of vacancies, policies of managed and centrally controlled wages (see details below) and labour allocation (Part III) resulted in obvious overmanning. On the one hand, cheap labour supported the technological backwardness of industry and services since labour inputs were always preferred and consequently substituted for capital in the production function. On the other hand, the labour force was underutilized (Part III) and unproductive. In reality many of those in formal employment were idle. Estimates of over-employment or labour hoarding are extremely difficult to make, but one 16  of the approaches suitable for doing so will be suggested in Part III. Naturally, estimates of labour hoarding vary according to the concept used and the method of estimation. Along with the present thesis, various studies (Boeri and Keese, 1992; Karpisek, 1991; Nesporova, 1991) suggest that on average between 25% and 30% of total employment in Eastern Europe was effectively hoarded labour. In the situation in question, the behaviour of employees was motivated not by competitive relations but by the presence of guaranteed employment (the right to work transformed into the right to work at a given concrete workplace) which had its natural effects on the whole economic system. The effect of labour hoarding was actually twofold. The first effect is related to overexpenditure of the state budget, since almost 1/3 of those employed were being paid without participation in the growth of output. The issue of labour productivity is one of the most important here.  51 There were many attempts to study the determinants of productive (in)efficiency of labour in Eastern Europe. Although most authors (e.g. Prasnikar and others, 1992; Kemme, 1987; Tyson, 1980; Sacks, 1983; Vanek, 1970; Whitesell, 1985;Nishimizu and Page, 1982) agree that labour productivity was an Achilles' heel of the CPEs, empirical estimates and policy formulation were hampered by a lack of rigorous studies at the level of the firm. Available estimates have been based primarily on a few case studies. The abovementioned sources consider East European labour productivity in industry to be 30-50% of that in developed capitalist economies. These are quite rough approximations because the area of labour productivity is plagued with measurement problems, particularly in relation to the adjustment of the relative prices of inputs and outputs for quality differences for the purpose of comparisons. Our study of the tradeoffs between sectoral employment and GNP per capita conducted in Part III provides some useful insights into the problem of labour productivity: by 1987 productivity (in aggregated terms) in Eastern Europe ranged from just over 1/3 to 1/2 of that in a middle-income OECD country like Austria and was also lower than that in a low-income OECD country like Greece. Some studies (Ray, 1991) report estimates for industry suggesting that during the 1980s the level of labour productivity for Czechoslovakia and Hungary was 40% to 60% below the level in Austria. Furthermore, it appears that there has been a long-term decline in productivity relative to the West (Ibid.).  52 The second effect of labour hoarding was the overgrown party-state and managerial employment sectors. In such a labour market situation employment varied around the "friction level", i.e. no potential workforce was available and frictional vacancies and unemployment became an organizational problem, aggravated by the structural discrepancy between the supply of and demand for labour. Thus, labour demand was formed not on a real market basis, but according to the priorities set by central planning. The resource-constrained labour market was coupled with a lack of competitive relations on the demand side and low labour productivity. The monopolistic buyer of labour - the state had the full power of applying command ways and means in order to abolish the natural consequences of the limited labour supply. As the result of policies pursued in accordance with this general scheme, an ineffective ("immature") occupational structure of employment at the macro level emerged (see Part III). Therefore, the ease with which restructuring and reallocation of labour occur under the transition presumably depends on two main factors: occupational structure and qualifications of the labour force. There is some evidence that the relatively high concentration of employment in manufacturing and agriculture disguised a large number of workers in tertiary jobs (Boeri and Keese, 1992; Nesporova, 1992). Socialist enterprises, being large in size, carried out a number of services in place of separate service industries, relating to transport and distribution, repairs and  53  maintenance, and the provision of food and other services to their workers. Therefore it is possible that, as restructuring proceeds, not all of the potential job transfers will involve a change in occupation. This would imply a somewhat lower cost of adjustment in terms of unemployment duration and retraining than some analysts have projected (see Part III below). One should note that the available data on occupations may not be a very good proxy for labour force skills and qualifications. As an alternative, information on educational qualifications can be examined, although substantial differences in institutional features of educational systems make quantitative international comparisons in this realm hazardous. It has been suggested that the work force in Eastern Europe is well qualified (Oxenstierna, 1990; Vodopivec, 1990). At existing levels of earnings this could both favour the inflow of foreign capital and also imply good adaptability of labour  during economic  restructuring. Table 7 presents a comparative overview of the structure and level of education using a range of available indicators. The groupings in Table 7 on educational attainment are somewhat arbitrary, but it would seem fairly clear that the proportion of the labour force in Eastern Europe with more than just a basic level of education lies around the middle of the range of OECD countries represented in the table. In terms of higher education, workers in the East appear somewhat less qualified than those in many Western European countries (with the exception of Austria, Italy and Spain - Boeri and Keese,  1988 16.0  OH  1989 in  12.0  CZ 1990 21.0  1985 14.9  Os  Year  Professional & technical  Administrati ve & clerical  | '•  oo  CN CN  d  CO  d  oo  CN  oo  CN CN  oo  oo  o o  — I1  38.1  38.1  oo  38.3  34.0  oo  53.4  26.7  oo  49.6  27.5  oo  50.7  CO  o o 54.5  o" oo  d  Share of production workers in total employment  58.4  t>  55.7  d  72.1  co co"  75.3  59.0  23.7  18.7  60.3  oo  vd  71.3  ON vd  vd  in  Total  co"  (-»'  o  Other & not defined  co  d  73.3  12.2  1985  vd  75.6  19.4  18.3  18.9  OS Os  16.6  1989  sw  OS Os  79.3  10.6  in in"  13.8  US  co in  Production & transport, etc.  1989  1990 oo  1990  NL  AU  -*  1989  WG  CM  Sales  PL  o  BL  54  55  1992). The data on educational attainment (Tables 7-9) represent the training received by a different cohort of workers in the past. Enrolment rates (Table 8) provide an insight into the level of education of new or potential entrants to the labour force and are lower both for secondary and higher education than in Western Europe, even when comparisons are made with  lower-income  countries. This may indicate that Eastern Europe is lagging behind OECD countries in the provision of education . 17  The possibility of finding a job depends not only upon the level but also the type of education or training being received. As Tables 8 and 10 show, a much higher proportion of secondary students in Eastern Europe attend vocational schools than in the West, even when compared to countries which have traditionally been oriented to vocational training such as Germany and Sweden. The proportion of higher level graduates in the field of engineering also tends to be higher. The examples of the mentioned countries and Japan suggest that a strong emphasis on vocational training is not inimical to growth and may, in fact, favour better economic performance, which depends, of course, on the exact nature of the technological training. However, the rather narrow base in the training of workers in Eastern Europe may hinder the restructuring process. Many of their skills may become obsolete and substantial amounts of retraining may be required . 18  The pre-transitional government responses to the problems connected with the allocation of labour and the balance between demand and supply are  56 Table 7. Level of Education of the Labour Force in Different Countries, as a % of the total labour force, various years.  Year  BoL  V  S  H  O  AL  BL  1990  44.6  15.8  30.0  9.6  ...  100  CZ  1989  26.0  21.0  43.8  9.2  ...  100  HG  1990  38.4  23.1  26.9  11.6  ...  100  PL  1988  34.2  29.5  27.9  8.4  ...  100  RN  1990  35.8  31.4  24.0  8.8  ...  100  AU  1990  28.8  57.8  6.3  7.1  ...  100  FR  1989  35.3  ...  46.0  14.6  4.0  100  GR  1989  52.6  ...  35.3  11.4  0.7  100  IR  1989  26.8  ...  55.5  17.5  ...  100  IT  1990  26.6  ...  66.2  7.2  ...  100  NL  1989  12.6  ...  61.3  19.7  6.5  100  SP  1990  48.4  ...  46.1  5.5  ...  100  Note: BL-Bulgaria,CZ-Czechoslovakia, HG-Hungary, PL-Poland, RN-Romania, AU-Austria, FR-France, GR-Greece, IR-Ireland, IT-Italy, NL-Netherlands, SPSpain. BoL-basic or less, V-vocational, S-secondary, H-higher, O-other, AL-all levels. "Other" usually includes people currently studying. Source: Boeri and Keese, 1992.  57 Table 8. Enrolment Ratios at Secondary and Higher Levels of Education, as a % of the population of the corresponding age groups, various countries and years. Year  Age group Net Secondary  Age group Higher  BL  1989  14-17  59  75  20-24  26  CZ  1989  14-17  ...  87  20-24  18  HG  1989  14-17  73  76  20-24  15  PL  1989  15-18  76  81  20-24  20  RN  1989  14-17  ...  88  20-24  9  AU  1989  10-17  ...  82  20-24  30  FR  1988  11-17  83  97  20-24  37  GR  1989  12-17  IR  1988  12-16 79  97  20-24 26  IT  1988  11-18  ...  78  20-24  29  NL  1989  12-17  82 103  20-24  32  SP  1987  11-17  ... 105  20-24  32  85  97  20-24 28  Note: "Net" includes only students falling within the specified age group. BL-Bulgaria,CZ-Czechoslovakia, HG-Hungary, PL-Poland, RN-Romania, A U Austria, FR-France, GR-Greece, IR-Ireland, IT-Italy, NL-Netherlands, SP-Spain. Source: UNESCO Statistical Yearbook, 1991.  58 Table 9. Type of Education: Secondary Level Enrolments, %, various countries, various years.  Secondary Level Enrolment General Vocational  Country, Indicator  Year  Bulgaria  1989  60.3  39.7  Czechoslov.  1989  53.8  46.2  Hungary  1989  76.1  23.9  Poland  1989  77.4  22.6  Austria  1989  28.5  71.5  W.Germany  1988  36.3  63.7  Italy  1989  40.6  59.4  Netherlands  1988  44.3  55.7  Sweden  1988  35.6  64.4  U. Kingdom  1989  9.8  90.2  Source: UNESCO Statistical Yearbook, 1991 for all countries except Hungary. For Hungary the data were taken from Boeri and Keese, 1992.  59 Table 10. Type of Education: Higher Level Graduates by Field of Study, %, various countries, various years.  Country, Indicator  Year  Engin., arch.  Ind., trade  Bulgaria  1989  28.1  4.7  Czechoslovak  1988  38.9  2.6  Hungary  1989  21.7  Poland  1988  Austria  Nat.Sc, Medi-Other Math. cine  29.4  30.3  6.6  17.8  34.0  4.2  7.3  35.3  31.5  16.9  2.9  15.6  29.5  35.1  1988  12.5  8.5  14.3  14.4  50.3  W.Germany  1987  20.9  7.1  24.3  7.1  40.6  Italy  1988  11.5  9.6  25.8  2.8  50.3  Netherlands  1987  16.4  4.1  11.7  14.5  53.3  Sweden  1989  1.4  25.5  U. Kingdom  1987  29.5 14.3  5.5 13.9  7.5  23.1 16.8  8.3 46.8  Source: UNESCO Statistical Yearbook, 1991 for all countries except Hungary For Hungary the data were taken from Boeri and Keese, 1992.  of interest. The central authorities in all East European countries attempted to use wage regulation as a tool for allocating labour between firms and industries. The successive reforms of wage regulations, however, failed to work in this  60  respect (seebelow). This reduced the central management's ability to manipulate labour allocation according to its own priorities. The basic contradiction faced by the central power was as follows: whereas the prevailing system of resource allocation (other than labour) was increasingly under the direct control of the state, it could not spread the same regulations over the labour market, which remained relatively liberalized with significant labour mobility and the right of enterprises to determine their manning level as well as setting individual wages. Contrary to what might be thought a priori, there is some evidence to suggest that labour mobility (or "labour rotation", "labour turnover", understood as the proportion of employed changing their jobs every year) in Eastern Europe compared quite favourably with that in the West. However if in the West European countries labour turnover often has an inter-sectoral character, Eastern Europeans often changed jobs within the same sector of the economy. As shown below in Table 15, around 1/5 of the total work force in Czechoslovakia, Hungary and Poland used to change jobs every year. This is understandable given the "overhang" of vacancies in the countries. These data may even be biased downwards since in case of Poland they do not include parttimers. While the data in Table 15 are not strictly comparable across the countries,  given differences  in the concepts of hires and quits and  methodological differences in measurement, it does appear that labour turnover on the eve of East European transition was broadly similar to rates observed in some Western countries, particularly if comparisons are made by industry.  61 Inappropriate wage differentials, in turn, impeded productivity growth. For decades East European countries pursued a low-wage policy, the rationale being that the "decommoditization" of labour required a gradual shift away from the cash-for-work nexus. The intention of successive generations of policymakers was that the so-called "social consumption funds" (financing non-wage benefits and allowances) would increasingly provide for workers' social and cultural needs, and would grow in size relative to wage funds. Low manufacturing wages were also intended to accelerate industrial accumulation. This led to a chronic undervaluation of labour by industrial enterprises, perpetuating labour-intensive production. There was little pressure to economize on labour. In recent years some changes in this realm occurred in almost all East European countries. Any intention to liberalize wages was hampered by the fear of wage-cost inflation, and some residuals of this fear can still be felt, for instance, in Poland and Russia. This is what led to the introduction in 1989 of a punitive tax on wage increases above 3 % in the USSR. Low wages affected productivity directly (influencing work motivation) and by raising labour turnover. As we noted, the turnover was too high in one respect and too low in others. For many years sectoral labour turnover was high thanks to the combination of low wages, full employment, and employment security, which allowed workers to job-hop in response to nonwage inducements  or dissatisfaction with their job. However, sectoral,  geographical and occupational-skill mobility (as opposed to high labour turnover  62 within a certain sector) have all been low, impairing the growth of labour productivity and thus the accumulation process. Geographical mobility has been restricted by the housing shortage, the need for residence permits and the structure of wages (Helgeson, 1986). Sectoral and occupational mobilities, as was already mentioned, have been limited by wage policy and by poorly developed mechanisms for labour market and enterprise-based training. Table 11 presents inter-sectoral wage differentials in the former USSR. The situation of large wage differentials favouring heavy industry, which persisted long after industrialization was under way, is typical for all former CPEs. By the 1970s this situation was hindering necessary shifts into the underdeveloped services and consumer goods industries, yet in the 1980s relative wages in services actually fell (the same table). Occupational wage differentials also fell during CPE times, according to official statistics. Some sources (e.g. Standing, 1991) reported even an inverse correlation between the years of schooling and wage rates: wages in some occupations requiring high education were lower than those for manual jobs (Table 12 illustrates this statement for Hungary). Although this situation presents a disincentive to the acquisition of education, there are some reasons for doubting the data (Ibid.). As will become apparent from what follows below, the main reason for doubting statistics on occupational differentials is that non-wage payments rose relative to wages, and higher-status employees usually received higher non-wage benefits (payments) than manual workers  63 Table 11. Inter-Sectoral Wage Differentials: Average Monthly Wages in Selected Sectors as a % of Wages in Manufacturing in the USSR, 1970-1989. 1970  1980  1985  1986  1987  1988  1989  A  75.8  80.5  86.5  89.0  89.4  87.5  88.6  C  71.3  74.5  70.8  70.9  70.2  68.6  71.0  H  70.9  71.8  69.7  69.2  69.6  69.8  68.5  P  81.1  73.3  71.2  72.2  74.6  71.2  66.6  Note: A-agriculture; C-commerce; H-housing & communal services; P-public education. Source: Narodnoye Khozyaistvo SSSR (Economy of the USSR), 1971-90. (Gordon, 1988). So total earnings may have grown despite any narrowing of wage differentials. From the very beginning of "socialism" in Eastern Europe, occupational wage differentials reflected an emphasis on job status, i.e. relative wages were shaped by an administrative perception of what was "socially useful", relying on distinctions between "productive" and "unproductive" labour. Status-based differentials tend to be rigid because there is no obvious way of altering the relative status of one job, let alone the whole structure. Determining new principles of wage differentiation in the state sector of the economy is an extremely difficult task, not relatively easy as some authors claim (Flanagan, 1992). It would mean a period of chaotic realignment of relative wages, which undermines principles of "socialist social solidarity" originated from overall "levelling".  64 Table 12. Differentials of Average Monthly Wages by Branch and Worker Category in Hungary, 1981-1989, industry indicator - 100%. Branch  Manual workers 1981 1984 1989  Non-manual workers 1981 1984 1989  Mining  153  147  137  148  139  130  Electricity production  105  108  110  104  106  103  Metallurgy  116  117  119  112  110  113  Machinery  96  97  96  95  96  94  Construction materials  97  96  95  98  96  95  Chemicals  106  110  117  107  112  117  Food processing  96  96  101  95  94  101  Light & other industry  86  85  82  93  Industry  100  100  100  100  100  100  Construction  103  102  103  106  105  107  92  88  87  100  98  96  109  105  97  98  93  84  Internal trade  81  80  78  90  90  90  Foreign trade  97  94  121  103  106  122  Water supply  103  104  104  91  107  102  Agriculture & forestry Transport & telecomm.  Source: Cukor and Kovari, 1991.  92  91  65 Table 13. Per Capita Earnings and Social Consumption Fund Benefits in the USSR in 1960-1989(1960- 100%).  Income source Year  1960  1970  1975  1980  1985  1989  Earnings  100  160  196  232  260  271  SCFB  100  207  279  . 347  417  510  Note: SCFB - Social consumption fund benefits - budget allocation, largely financed by central government and covering education, health, housing, pensions, benefits, subsidies, etc. Earnings - nominal money earnings net of taxes excluding social consumption fund benefits. Source: Standing, 1991. A very important tangle of contradictions, resulting also from the East European policy on incomes and contributing to low labour productivity, was the high and rising ratio of non-wage to wage forms of renumeration of labour. At the macro level this ratio rose continuously from 1950s in all countries in question. As Standing (1991) reports, in the USSR between 1960 and 1985 per capita earnings (net of taxes and excluding benefits and allowances) rose by 160%, while per capita benefits and allowances as measured by the social consumption funds rose by 317% (Table 13).  66 Shcherbakov (1991) puts it differently: whereas in 1965 in the USSR for every rouble of wages 46% were received from the social consumption funds, by 1984that proportion had risen to almost 70%.Indeed, the trend was for the consumption funds to grow at the expense of money wages and the income derived from private smallholdings. As a result, the proportion of incomes derived from wages was ever declining in all East European countries. Similarly, a system had arisen whereby substantial amounts of material goods were distributed at the workplace. A worker's position was thus determined not so much by the results of his/her work and the level of his/her pay as by his/her access to goods to buy with wages and to various services made available on favourable terms. Such privileges as the use of a works surgery or hospital, summer cottages, canteen, dress-making facilities, creche or holiday accommodation, which had a considerable effect on the quality of life, were by no means equally available to all workers. The allocation of social consumption funds was determined largely by the official unions. This sort of management role assumed by unions definitely lessened their ability to apply bargaining pressure on the enterprises, which, as shown by evidence elsewhere, generally encourages "dynamic efficiency" (Freeman and Medoff, 1984). One of the main desirable intentions of the high non-wage, low wage pattern of renumeration was an effective response to the artificial labour shortages by reducing labour turnover. This intended result is consistent with data showing a decline in labour turnover in the USSR in the years when the  67 non-wage to wage ratio was rising (Table 14). However, a cost of that policy is likely to have been a lower effort on the job and lower labour mobility of a beneficial kind. As many non-wage benefits were allocated according to status, the system was bound to have deterred inter-enterprise mobility by those who had secured a niche for themselves. The trend towards non-wage renumeration may have resulted from an overemphasis on the "reproductive function" of the payment system; low wages were long seen as essential for rapid accumulation, the social consumption funds being seen as the surest way of maintaining workers' capacity for labour. The direct subsidized or free provision of such expensive services like health, schooling and holiday facilities meant that wages could not rise much above a basic subsistence level. That such services were effectively part of the wage hampered the development of paid services and meant that education, health and other services remained tied to the minimal levels required for industrialization, rather than responding to the needs of wider social development. That in turn checked skill formation. To complete the circle, low wages - combined with the low standard of services on offer - restricted demand for paid services, so discouraging investment in those sectors (Standing, 1991).  As discussed above, the supply-demand contradiction (labour hoarding coincided with labour shortages) ensured the number of vacancies was very high and so a correspondingly high degree of "job-hopping" may not be surprising. Some authors (Boeri and Keese, 1992; Portes, 1992) mention that while rigidities in wage differentials between enterprises may not have provided any incentive for mobility, workers may have thought to improve their working conditions through frequent changes of workplace. Unfortunately it is impossible using the available data to disaggregate total separations into changes in occupation and changes of industry which would allow for a better understanding of the potential for labour to be rapidly redeployed in response to restructuring.  68 Table 14. Labour Turnover by Sector in the USSR in 1965-1989,% of average employment.  I  1965  1970  1975  1980  1985  1989  20.5  21.2  19.0  16.1  12.7  13.5  15.3  16.4  14.4  10.6  12.1  SF CN  34.2  33.0  28.1  22.6  18.1  16.5  R1T  10.4  12.1  11.4  10.9  10.1  10.4  RdT  ...  ...  23.0  19.4  15.5  15.0  CM  22.2  25.9  22.1  20.5  14.9  12.6  CR  ...  ...  ...  14.4  15.7  S  ...  19.2  17.6  19.3  23.1  Note: I-industry; SF-state farms; CN-construction; RIT-rail transport; RdT-road transport; CM-communications; CR-commerce; S-services."Commerce" includes public catering, material supply and procurement. Source: Standing, 1991. Although the idea of low wage dispersion among industries as a barrier to vertical labour mobility in Eastern Europe has become quite widespread in Western literature, we would like to offer a somewhat different point of view. In so doing, it is worth stressing again that in the 1970s the problems of labour allocation were exacerbated by the emerging general labour shortage. This shortage certainly was less acute in the USSR. The USSR, however, also faced it since extensive labour market reserves became exhausted and the labour force participation rate reached its potential maximum (Sziraczki, 1990). This led to fierce competition for labour among enterprises and the importing of unskilled  69 Table 15. Labour Mobility and Turnover Rates as a % of total employment, various countries, various years.  Year  H (I)  Q  cz  1989  19.1 (19.5)  19.6 (19.4)  HG  1989  20.2 (...)  22.6 (...)  PL  1989  16.2 (15.8)  19.8 (19.3)  RN  1989  C)  (10.5)  .32.2 (22.3)  31.8 (22.9)  (5.4)  (8.0)  18.9 (15.5)  17.7 (15.0)  (23.2)  (22.6)  FR  1990  IT  1987  JP  1988  UK  1988  Of total quits: EC RG D M TR OTH ...  9.1  ...  0.7  0.7  12.5  0.5  5.5  15.9  8.0 2.7  0.9  9.8  ... 0.3 4.6 4.5 1.7 3.5  11.7 0.7 3.4  1.0  Notes: CZ - Czechoslovakia; HG - Hungary; PL - Poland; RN - Romania; FR France; IT - Italy; JP - Japan; UK - United Kingdom. H(l) - hires (industry); Q(I) - quits (industry). EC - end of contract; RG - resignations; DM - dismissals; TR - transfers; OTH -other. Czechoslovakia - data do not take into account farming cooperatives. For industry - manual workers only. Poland - full-time workers in the state sector only. Romania - the categories "transfers" and "other" are included in "resignations". France - establishments with 50 or more employees. Italy - establishments with 500 or more employees. Source: Boeri and Keese, 1992.  70 labour from abroad (e.g. from Vietnam). Enterprises facing labour shortages reacted not so much by increasing wages, but by an increase in non-pecuniary benefits, by spending more on search activities, by lowering recruitment standards, and by relaxing work and disciplinary standards. In our opinion these developments in enterprises' hiring behaviour and management had an effect similar to a "hidden increase in wages".The local and central authorities were urged to intervene in the labour market. One of the most illustrative examples of the intervention was the import of labour from Asian socialism-oriented countries, particularly Vietnam. The direct import of labour from abroad, starting in the 1970s, has two interconnected implications for our analysis. First, external supply of labour meant that authorities failed to influence the labour market by "indirect means". Second, central control over the labour market was tightened by resorting to direct interventions, which meant the beginning of a qualitatively new period in labour market policy. Most actions were initiated at the central level, but some of them were carried out at local levels. The most important measures were common for all East European countries except the USSR and included the following: - Attempts to reduce the demand for labour. The governments froze employment levels in administrative and clerical jobs. County councils were instructed to determine the permissible level of each enterprise operating in their area. Methods of recruiting labour were restricted: limitations were placed on advertising vacancies and on recruiting activities in regions other than the  71 location of the firm. -Attempts to strengthen labour discipline within enterprises. For example, in order to decrease absenteeism, the practice of giving out sick pay was restricted. High labour turnover was countered by wage sanctions against "professional" job leavers. In Hungary, for example, workers changing their employer more than twice a year became subject to bureaucratic procedures, such as compulsory registration at local labour exchange offices. Legal provisions setting an upper limit for overtime were relaxed. Finally, efforts were made to promote in-plant training programmes. - Attempts to influence the allocation of labour among enterprises by administrative  means. Measures included "organized redistribution", the  strengthening of "professional orientation work" at schools and colleges,and the compulsory direction of new graduates to certain jobs (often in different regions of the countries) for a stipulated period of time (up to several years). In Russia these activities were taken to extremes. Thus, every Institute and University graduate was supposed to work a minimum of three years at a designated workplace in a specified region. This measure was seen as a necessary one, allowing each graduate to pay debts accumulated during the period of free education back to the state. In addition active measures were implemented to attract young workers to outlying regions of the country - Eastern and Western Siberia, the North and the Far East. One of the main tools used to attract volunteers was a significant inter-regional wage differential introduced in the  72 1970s. Although some measures were aimed at limiting wage competition (e.g. a new wage rate system was introduced which narrowed the wage differentials of workers doing the same kind of job), the regional differentials remained more than significant, even within the same occupation. - A campaign against the second economy and small-scale enterprises (particularly in Hungary and Poland). Local authorities increased the level of taxes imposed on small-scale farming activities, the number of licences for private industrial activities was restricted, and official declarations were issued against "unjustified profits" and "non-labour incomes" targeted at the private sector and cooperatives. The results of direct state intervention in labour markets in Eastern Europe did not meet expectations. A few measures, however, proved to be relatively successful, including the allocation of certain groups of new graduates to jobs  19  and attempts to provoke inter-regional flows of labour by increasing  the spatial wage gradient. But most of the measures listed above either failed to work properly or their labour market benefits were overwhelmed by negative side-effects. These failures can be explained by several main factors. General attempts to direct labour towards selected jobs and/or areas could not work in full because workers retained their rights to change jobs. Employees could quit their jobs at any time and were not obliged to accept the workplace suggested. However, there were some exceptions to this case, particularly in the USSR  73 where the attempts to recruit acquired the features of national campaigns supplemented by various forms of social and political pressure. For example in the case of "national youth construction projects" like the Baikal-Amur Railway and KAMAZ truck plant, all regions of the USSR were obliged to contribute labour to the programmes. Some measures were evaded by enterprises. Thus, they defended themselves against the freeze on administrative and clerical staff by reclassifying their non-manual labour as manual workers. Some of the measures, although designed to meet the interests of large state companies, proved detrimental to them. For example, the directives to limit overtime in industrial work and to impose an administrative ceiling on monthly earnings actually cut off one of the most important management tools to mobilize workers to work more hours and harder in order to meet quarterly or annual production targets. It is not surprising that the period 1975-1980 was one of continuous and very often indecisive struggle between  central economic management and particular  enterprises exerting pressure either to revoke such measures or at least to impose new ones aimed at eliminating negative side-effects. Accordingly, large enterprises applauded the measures taken to restrict the legal private (in Eastern Europe) and collective (in the USSR) sectors. But at the same time those large East European enterprises which had cooperative relations with this sector tried to exempt their partners from the restrictions and were generally successful in so doing (Sziraczki, 1990). In the USSR, especially in the periphery of the  74 country in so-called factory-towns (settlements with the majority of population working at one large factory) workers were scarcely able to quit their jobs because the factory-employer owned and/or controlled housing, social and medical services as well as personal subsidiary land holdings. When a worker quit the job for any reason, he/she lost everything and faced quite serious difficulties, since administrative restrictions on residential mobility did not allow a person to live in a different place without already having obtained a job. In turn, the job could not be obtained until the official residential permit was issued. In spite of large East European enterprises trying to protect the private sector, production in small-scale farming and industry decreased sharply, partly leading to a "food problem" and to shortages of services. These were already being felt in the mid-1970s (Charap and Dyba, 1991). Similarly, the campaign to extend the use of piece-work in factories, and the introduction of a new wage-rate, system were significantly counterproductive. They undermined accepted practices by instituting wage differentials, causing conflicts between labour and management as well as between different groups of workers. Besides that, new central policies in terms of labour markets gave way to a new trend in Eastern Europe, though not in the USSR, namely - to mergers. In the early 1970s (and more actively from 1975 on) more small- and medium-sized companies were absorbed into large ones struggling with labour shortages. As a consequence, the share of the small- and medium-sized sector  75 contracted considerably. For example, in Hungary between 1970 and 1978, in state-owned industry, the proportion of enterprises with fewer than 300 manual workers fell from 34.1 to 20.6%.Changes in the industrial cooperative sector were even more dramatic: over the same period the share of cooperatives with fewer than 100 manual workers dropped from 46.5 to 13.8% (Sziraczki, 1990). Enterprises recognized the weak impact of central and local actions to lessen the excess demand for labour, whilst mergers gave only a temporary respite. Therefore, more firms adopted the strategy of setting up their own internal labour markets by devising new rules to tie workers more securely to the firm. Reduced or even non-existent recruitment standards raised the problem of how to attract newcomers and not to break the rules of the administrative system. As a result, a compromise was reached, and county councils received the right to determine the permissible level of new annual employment for each enterprise operating in their area. Giant enterprises, of course, received a great advantage over smaller ones because of the significant degree of local power the former could wield. The examples of the automobile factories AZLK and ZIL in Moscow, the Polish iron-and-steel works in Krakow and Katowice, and the Czech truck-and-bus plant LIAZ in Liberec which used to attract thousands of newcomers per year are quite illustrative. So, to cope with labour shortages firms developed various policies which contributed to the spread of internal labour markets. Multiple job holding (often in the same enterprise)  was permitted, although some restrictions  76 concerned with maximum earning were still preserved. In order to keep the loyalty and cooperation of strategic workers, managers offered different financial bonuses and non-financial benefits, and, as a result, informal bargaining processes developed between these workers and management. Another form of motivation of workers was the promotion of them to different within-unit organizations like "Committees of Workers Self-Management" with extra rewards. At the same time enterprises tried to restrict labour demand to the lower levels of the job hierarchy and to build up a variety of promotion ladders. These internal labour market developments proceeded in non-bureaucratic ways, with different arrangements from one plant to another. Internal labour markets provided a degree of security against competition from other enterprises and against unpredictable changes in the employment and wage policies of central governments. Some additional points should be made with particular reference to the statements earlier in this paper on the educational characteristics of East European labour. In particular, it is worth mentioning that in a further response to labour shortages, firms stepped up their efforts to train their employees. Training courses for semi-skilled jobs decreased, reflecting a decline in demand for such labour, whilst vocational training courses were gradually taken over by the public education system. This is confirmed by the data in Table 17 . At the same time special-purpose, further-training courses were rapidly gaining ground. Of these around 80% were organized by enterprises for their own workers.  77 Moreover, it was usual for enterprises to make the training as specific as possible in order to decrease the probability that workers would use their knowledge elsewhere. In this way the spread of further-training courses diverted the acquisition of firm-specific skills from informal to bureaucratic channels and contributed to the stabilization of the labour market internal to the enterprise. As some authors mention (Sziraczki, 1990; Kolodko, 1991) a further important influence on the development of enterprise training was the need to adjust to cyclical variations in economic activity (those mentioned in Part I, Chapter 3) once the external labour reserves had been exhausted. From the early 1970s onwards, adjustment to cyclical changes in production came to depend increasingly on the ability of the enterprise to vary the intensity of the use of its own workforce. One of the most important means of achieving this was the mobilization of the internal slack through training courses. Summing up, the state followed events in the labour market rather than dictated them, and was relatively powerless to overcome the macro-scale contradiction between a relatively free labour market and controlled resource allocation. In addition, state actions led to doubtful results; the majority of implemented measures had unforeseen side-effects. This situation remained almost unchanged throughout the 1980s, although the gradually worsening general economic climate (accumulation of foreign debt, foreign trade imbalances, etc.) forced governments in Czechoslovakia, Hungary and Poland to revise their economic objectives and policies (Sziraczki, 1990). Romania ran  78 ahead of other East European countries in terms of attempts to reduce foreign debt without undermining the principles of full employment. To achieve this goal the central governments in the named countries restricted imports, investment and domestic demand whilst strongly stimulating exports. Some studies (Berg and Sachs, 1992; Kolodko, 1992),although operating with Sovietstyle statistical data, indicate the 1979-1983as the period of peak export activity reflected in a number of trade indicators. Such economic policy was implemented by means of non-market instruments. Some sources (Cukor and Kovari, 1991; Dornbush, 1990;Sziraczki, 1990) mention that, in fact, the transformation of control towards even more rigid central management and the replacement of the established legislative control by prompt state intervention in urgent cases proved to be an important factor enabling the countries in question to avoid insolvency in the early 1980s. Economic growth fell back and living standards stagnated whilst the countries achieved a trade surplus and slightly reduced external debt, although these processes had different shades within the region. For example, in Hungary the government also took various actions to try to reduce the isolation of the economy from the world market and to stimulate market adaptation by enterprises. Initial steps were taken to introduce price and currency reforms: a uniform exchange rate was introduced and foreign trade prices were adjusted to world market prices. In all the countries mentioned, attempts were made to modify the monopolistic, centralized structure of the economy by breaking up  79 some large-scale enterprises into small firms. So, for example, in Hungary between 1980 and 1983, 25 large-scale enterprises of the state sector were broken up into 332 smaller units (Gyurak, 1984), while in Czechoslovakia 15 big companies were reorganized into 234 smaller ones over the same period of time (Novoye Russkoye Slovo, 1993). Similar developments, although on a smaller scale, were noticed in Romania, Bulgaria and Poland (Hughes and Hare, 1992). In spite of these changes, no improvement was observed in the ability of enterprises for rapid adaptation to the market (Portes, 1992). The economic policy did little to increase the autonomy of enterprises: they remained heavily dependent on the state and this impeded the policy designed to impose flexibility. During 1980-1982, a significant divergence in economic policy took root between the USSR on the one hand, and Hungary and Poland on the other hand. The rise in political influence of Polish unions, and government actions in Hungary, caused employment policy to develop along liberal lines (as a continuation  of general economic liberalization), while in the USSR,  Czechoslovakia, Romania, Bulgaria and the GDR,  labour markets retained the  basic features of the 1970s. Some of the new developments will be described more fully, since they are what determined Hungary's and Poland's push ahead of other Eastern European countries as far as labour market transitions are concerned . 20  Although the USSR, Romania, Bulgaria and the GDR inherited their  80 labour market policies from the 1970s, in Poland and Hungary the demand for labour fell slightly owing to economic stagnation and investment restrictions. The governments recognized that measures implemented in the mid-1970s, aimed at direct state intervention in the labour market, had not produced the expected results, and they were gradually eliminated between 1979 and 1981. In addition, fearing that economic stagnation in the state sector could result in unemployment,  the central economic authorities in Hungary began to  reinterpret the notions of "full employment" and "job security". To avoid any increase in social tension the government introduced retraining, early retirement and job creation schemes (Sziraczki, 1990). Since Hungary was a pioneer in terms of measures which are under way in many East European countries now, its experience deserves a more detailed description. Programmes to assist in the retraining of workers were introduced in 1983 and were originally designed to retrain workers made redundant in a government-initiated above).  programme of enterprise rationalization (mentioned  After 1985 retraining was made available to workers scheduled to  change jobs because of individual company shake-ups and to people willing to change jobs for personal reasons. In 1987 retraining assistance was offered to the unemployed as well (Hars et al., 1991). Unemployment  benefits,  introduced in 1986 in Hungary, as  "unemployment aid", were first available only to workers who lost their jobs  81 Table 16. Vacancies, Job Seekers and Unemployment in Hungary in 19841989. Year  Quarter  Registered Vacancies  Registered Job-Seekers  Registered Unemployed  1984  1  74,349  1,890  1984  2  86,020  1,994  1984  3  75,950  2,253  1984  4  64,559  1.893  1985  1  60.800  2,460  1985  2  63,276  2.692  1985  3  55.277  3.178  1985  4  65.051  2.669  1986  1  61.468  3.143  1986  2  75,434  4,613  1986  3  81,654  7,595  1986  4  63,888  6,387  1987  1  56,652  10,381  1987  2  60,887  9,188  1987  3  55,473  10,465  1987  4  46,800  10,809  1988  1  43,762  16,845  1988  2  54,127  11,462  1988  3  70,129  15.254  1988  4  72,534  14,163  1989  1  68,131  22,678  1989  2  66,748  23,646  1989  3  20.236  22.146  1989  4  36,765  23,354  1.442  Notes: Data of 1984 and 1985 refer to 109 towns, and of 1986 and 1987 - to 125. Job-seekers are employed, unemployed and pensioners. Source: Sziraczki, 1990; Employment Information Centre, Budapest.  82 in company lay-offs affecting ten or more people. This eligibility condition was lifted in 1988, although a worker still could not receive benefits unless formally dismissed by an employer. In 1989 unemployed persons became eligible for an unemployment allowance paid for a further year and equal to 75% of the unemployment benefit (Hars et al., 1991). Table 16 below illustrates the development of unemployment in Hungary in the pre-transitional period. Hungary was the first Eastern European country which recognized the existence of open unemployment and began registration of the unemployed in 1986. In 1987 the Hungarian government launched a programme to facilitate early retirement. To be eligible for early retirement the employee was required to be within five years of retirement age (60 for men and 55 for women) and must have been employed for at least 25 years (30 years for men) and have worked for the present employer for the past five years. Early retirement pensions (normally paid by the employer) were offered to people who lost their jobs as a result of a company closure or reorganization involving staff reduction (Hars and Nagy, 1991). In 1988 the government introduced measures to encourage job creation in the least developed regions of the country. To promote investment the state offered enterprises preferential loans, or, in some cases,grants. Substantial funds were earmarked for this purpose by the Employment Fund. However, the annual average increase in jobs as a result of this programme over the 1990-1992 period was believed to be only around 3,000 (Hars and Nagy, 1991).  83 A programme to employ the unemployed on community work has been operating since 1987. In 1990 an average of about 4,000 people were employed on community work programmes each month (Hars et al., 1991). In 1989 the government began to give the unemployed start-up loans to establish their own enterprises or to engage in other self-employment activities. For the first four years of the loan, interest was to be paid by the Employment Fund. Although the loan conditions were quite liberal, the programme has since been discontinued. During the 18 months of its existence, about 43,000people received such a loan (Hars, 1989). In 1989 the government also launched a programme of subsidized employment designed to help university graduates find jobs. Under this scheme employers who hire graduates registered as unemployed at an employment office received the first 18 months of wage costs from the Employment Fund. By 1992, however, only slightly more than 300 university graduates had been placed in such subsidized work. Another major focus of employment policy since the mid-1980s has been providing special assistance to regions facing the greatest unemployment problems (see the next Part for details). Affected workers in the regions identified as needing special assistance became eligible for a wide range of additional financial support: they could be offered severance pay from the State, a resettlement loan, or an entrepreneurial loan, while employers hiring such workers had the corresponding social security payments - 43 % of wage costs -  84 reimbursed by the Employment Fund for up to 12 months (Hars et al., 1991). Table 17 below illustrates some general results of these measures. Table 17. Job Leavers by Reasons in Hungarian State and Cooperative Sectors, %, 1976 and 1985. Indicators, Years  1976  1985  Retirement, death  9.9  11.5  Maternity leave, military service  12.9  12.7  Voluntary leaving  56.7  44.0  Transfer initiated by employer  ••  5.7  Redundancy  ••  1.3  Dismissal for disciplinary reason  1.1  3.6  Termination of contract  13.2  13.2  Other  6.2  8.0  Total leavers  100.0  100.0  Total leavers in thousands  1,094.0  1,010.1  Labour turnover  17.2  13/3  Source: Sziraczki, 1990; Foglalkoztatottsag eskeresetiaranyok (Budapest, 1979, 1987). Poland was the only other East European country where active measures of labour market policy were implemented before 1990 (Tomes, 1991). Although the scale of pre-transitional measures implemented in Poland can not be compared to that in Hungary, several points are worth mentioning. - employment information centres were organized. Their tasks were to bring  85 employees and enterprises together and to ensure a satisfactory match between the supply and demand for labour. In addition to a lack of equipment and facilities, however, the offices did not have the experience needed to organize an effective placement service. Also there were, and still are,too few centres, as well as more candidates than vacancies. So, in 1990 there were 390 Labour Offices in Poland with a total staff of about 4,000people. Of these offices, 341 were regional establishments dealing directly with the unemployed. Each Regional Office served an average of 3,300 registrants, i.e. there was one staff member per 280 unemployed people (Gora, 1991). -The Polish Act "On Employment" of December 29, 1989for the first time recognized the status of the unemployed and provided for standard passive measures such as the payment of unemployment benefits. This was formulated, however, in such a way as to imply that those measures were to be a last resort form of assistance, which should be applied only in case the anticipated set of active measures of manpower policy could not be used (Dziennik Ustaw, 1990, art. 15). In reality the inadequacy of the administrative means available to the Ministry of Labour and the understaffing of the labour offices, mentioned above, have led to an overwhelming predominance of the straight payment of unemployment benefits as a means to deal with the unemployment problem (Chilosi, 1993). It is not surprising that countries like Bulgaria, Czechoslovakia, the USSR, and Romania entered the transition period unprepared to deal with  86 labour market changes, since their pre-transitional employment measures did not create the infrastructure necessary to handle rising unemployment. But in Hungary and Poland - countries running ahead of the rest of Eastern Europe in terms of labour market policies in the 1980s - neither the specific measures introduced nor the countries' labour market institutions have been able to slow the growth of unemployment to any great degree either (see Part IV below). We can see several main reasons for this. First, as was shown in Part I, the Eastern European economies suffered (and are suffering) a prolonged recession. Negative economic growth under transition results in a situation where the economy is unable to respond to the Table 18. Job Losses by Sector in Several Eastern European Countries as a % change between 1989 and 1991. Industry, Country  BL  CZ  HG  PL  RN  Agriculture  - 16.6  -21.2  -46.0  -6.6  +0.3  Industry  -23.3  - 11.0  -28.0  -13.0  -9.0  Construction  -31.4  -4.4  -48.9  - 15.5  -54.2  Services  - 11.5  -2.4  ...  -3.6  -3.3  Total  - 19.8  -7.8  -34.1  -8.9  -10.4  Note: BL - Bulgaria; CZ - Czechoslovakia; HG - Hungary; PL - Poland, RN Romania. Source: OECD, Employment Outlook, 1992.  87 drive to restructure. The decline in absolute employment, although low in relation to the decline in output, took the authorities by surprise. They had been lulled by the tone of forecasts. Table 18 gives a general idea about the scale of the problems. Second, as we shall see, in its present form the employment office network has proved unable to tackle the full range of employment problems effectively. In addition to the lack of technical and information resources, a major problem has been that many employment office staff do not have the necessary experience to deal with the rapidly worsening conditions. The majority of the retraining programmes have been enterprise-based and have consequently not been devoted to preventing unemployment. Third, local authorities readily accept community work programmes because much of the cost is borne by the state and their communities directly benefit. However, the unemployed recruited for local communities are mainly involved in very low-grade manual work and do not acquire significant skills. Also some unemployment-targeted measures in Hungary have collapsed because the state simply could no longer cope with the rising expenditures caused by the increase in unemployment. All East European countries are facing an urgent necessity to develop an employment policy truly appropriate to the labour markets in their present condition and to solve two main general problems of imperfect labour market performance: structural misallocation of the labour force and weak cost-sensitivity of enterprises.  88 §2.1.1. Patterns and Interim Conclusions.  The present situation in East European labour markets is partly a consequence of their pre-transitional heritage. The main feature of this heritage is rooted in a compulsory socialist paradigm in labour economics, which stipulated that employment was guaranteed for all. Put in practice, this paradigm resulted in economic growth generated by extensive use of inputs, in our case evident in extremely high rates of labour force participation. Questions concerning the rational rate and structure of employment were either completely ignored or treated as inseparable from the full employment assumptions. One of the consequences of this policy with respect to labour markets was a stable and declining number of people able to work, resulting in severe labour supply problems (worsened by regional diversities in labour supply), huge labour shortages and excess jobs. Incentives and the stimulation of labour either lost their significance as an instrument of labour market policies or became a marginal tool of government labour management. Generally speaking, the labour market under central planning was a resource-constrained labour market ruled by policies of centrally controlled wages and labour allocation (both horizontal and vertical). The role of labour hoarding, which emerged as a consequence of a resource-constrained labour market, was extremely significant in three respects. First, it predetermined a significant rate of overexpenditure of the state budget. Second, it contributed to a low level of marginal productivity of labour. And third, it supported overgrown party-state  89 and managerial employment sectors. Problems of severely constrained labour supply resulted from forgotten principles of economic rationality and lack of competitive relations on the demand side. We therefore bring to the fore one of the main conclusions, resulting from the discussion in Parts I and II and serving a link between the foregoing and the following parts of the thesis. restructuring and reallocation of  labour  on two main factors: the degree of  The ease with which  occurs under transition depends deformation  of  the occupational  structure of employment on the macrolevel and the qualifications of the labour force.  Labour market restructuring in most of the countries has been considerable, yet by late 1991 it had mainly taken the form of "clearing the desks", that is a process of disemployment from state enterprises. The growth of employment in the private sector of the economy has been continuously accelerating since 1991 (see part IV for evidence). There were huge declines in most of the aggregate economic indicators, and especially in GNP, that would have created an even greater labour market depression had they occurred in Western Europe. Yet beginning with 1991 labour hoarding began to shrink, indicating the pervasive labour rigidities, in which individual lay-offs had only recently been overtaken by large group lay-offs. Employment has been falling while unemployment has been rising since 1991. The rate of employment decline was lagging behind the rate of decline in GNP. Below we shall try,  90 therefore, to explain why employment declines at a slower rate than GNP. During 1990 and 1991 serious disequilibria in Eastern Europe had been deepening in all the countries, although there appeared some signs of recovery. Labour force fragmentation threatened to marginalize ethnic minorities and women much more than men. Thus 70% of all redundancies in the ex-USSR in 1991 were women, as were some 61% of the unemployed in Bulgaria (Samorodov, 1993; Beleva, Bobeva & others, 1993). Privatization has been taking many forms, with different but as yet unclear implications for the speed and extent of job shedding, the restructuring of employment and labour relations. According to some sources, there was a cut in employment across all sectors of the economy in all countries, but not a growth in services (Standing, 1993). Below we will attempt to test this hypothesis and, if true, analyze the causes of the occurrence. Active employment measures in all East European countries were located at the margins of macroeconomic policy. These measures fit no model of labour market development known in the literature. The individual character of the East European transition, resulting from both historical factors and the present economic policies of reformist governments, challenges the temptation to look for analogies and similarities in other regions of the world. From the discussion one might derive certain preliminary conclusions about the role of labour market policies in the economic restructuring process. The following few conclusions to be tested in Parts III and IV seem consistent  91 with the discussion above. • Labour market policies can significantly contribute to the speed of transition, mainly by raising productivity and changing structural patterns of labour allocation. • Although the definition of unemployment in transitioning countries is close to that accepted in the West, the pre-transitional heritage of East European labour markets and the current economic situation in the countries presumably create an unemployment of different origin. The problem of origins and causes of East European unemployment should be scrutinized and taken into account when elaborating on labour market policy under transition. From this point of view labour market policy must be an integral part of macroeconomic policy. • Goals of labour market policy under transition should be analyzed on the basis of the pre-transitional heritage projected to the transition phase, the current situation, and the goals of macroeconomic policy of the government. Without labour policies designed to help in managing the restructuring of labour markets, the risk of social instability, and of the segmentation and marginalization of labour markets are very high. regulations  and  institution-building  In particular, labour  assistance can help  in limiting  unemployment and the dislocation costs associated with economic restructuring (Standing, 1993). • The objectives of any labour market policy should be proven, clear  92 and explicit, as should be the criteria for monitoring them. Are they simple enough to administer and understand? Are they transparent enough so as to be legitimized? •  As some authors note, any policy measures should have their  alternatives. At present many policy recommendations are based on assertions backed by repetition, which may or may not be true. Thus, for example, it is very commonly claimed that labour market training is needed on a massive scale because marketable skills are lacking. How is that known? Similarly, it is widely claimed that employment services are desperately needed to reduce frictional employment. How is it possible when the question of the origins of East European unemployment has not yet been decided? Perhaps other equally effective networks do exist (Gora, 1993;Nesporova, 1993; Standing, 1993). This is not merely to state that such networks do exist or that less training is desirable, merely that such claims are usually a matter of commonly accepted opinion which has not been proved yet. • There is a clear and urgent need for systematic labour market data, which would serve policy development. This statement is closely linked to the previous one. •  Priorities, goals and mechanisms of labour market restructuring  should be linked together to create a statement of policy intentions. This coupled with considerations on the financing of policy intentions creates a framework for labour market policy.  93  PART III. STRUCTURAL PROBLEMS OF EUROPEAN LABOUR MARKET TRANSITIONS.  EAST  As we have seen, in 1990-1992 the East European economies were hit by a cruel combination of economic factors that made the goals of transition, however conceived, much harder to achieve. Yet despite the enormous declines in aggregate economic output, the overall decline in employment in 1990-92 was relatively modest (see details below). The general question about the relationship(s) between level of economic development (measured by GNP) and employment brings us to the problem of structural adjustment, which seems to be one of the most important on the transition agenda of any country. In terms of labour market transitions this problem can be narrowed down to issues of macrostructure of employment (occupational structure). These are examined in the present Part, which focuses on two main sets of questions: 1. To what extent are the problems of macroemployment restructuring in Eastern Europe shaped by the countries' historical development? The present chapter looks at some structural peculiarities inherited from pre-communist times. 2. What is the general tradeoff (if any) between the attained level of economic development and the employment structure? The character of the tradeoff is supposedly among the factors that determined the time lag in the decline of employment.  94  CHAPTER 1. OCCUPATIONAL STRUCTURE OF EMPLOYMENT: A PRELIMINARY ANALYSIS. The  present chapter attempts a brief historical review of the  macrostructural development of employment in the East European countries during the XX century, based on a number of comparisons of structural indicators. However the research priorities for the chapter and Part III in general posed above make it necessary to devote some space to the description of reservations and limitations of the approach, setting the research framework at the beginning of the analysis.  95  §3.1.1. Scope of the Approach. One of the first questions a reader will raise following our analysis below is whether one should expect individual countries to follow a common path of evolution of sectoral employment structures - or whether this should be expected, if at all, only of economic systems, which might be groups of countries. In other words, if we are looking for some sort of "common" or "general path", it means that universal regularities of development are assumed, at least in relation to the groups of countries. Given that any classification is somewhat arbitrary and incomplete as it leaves certain events unclassified, the question still stands: isn't this an assumption which can undermine our results whatever they are? International comparisons suggested below take place at a level, sufficiently aggregated to allow challenges to conclusions based on steadystate assumptions. Should we commit to universality in addition to this? We shall however note in response, that the Universal Development point of view is not an invention of the author. As will be shown in Part IV, this view is quite well rooted in the history of development thought. It originates in neoclassical postulates, and maybe even earlier, and received thorough intellectual and empirical testing later in the 1950s. Although some of its components did not survive the verification, some others are alive and practiceable, as our analysis below demonstrates. The author, however, does not associate himself with strict adherents of the neoclassical school of development with all its underlying assumptions. Our analysis based on  original  methodology is intended to make further inquires about the validity of the universal approach and to find its applicability to labour market analyses. Paragraph 3.2.1. below suggests our view on economic implications of the universal approach and the limitations it imposes on economic analysis. There are some definite assets, however, which also receive special attention. The methods used in Chapters 1 and 2 of the present Part pertain to behavioural modelling and require us to make a commitment to the abovementioned rigid behaviourial assumptions. Of course, the data available to us from the past may be uninformative about behaviour that might occur in the future. Moreover, although one can experiment with alternative models, the complexity of the modelling process makes it difficult to explore the full range of possible responses to a policy change. At the other extreme is an approach that simply eschews behaviourial analysis and focuses on direct, first-round effects of policy changes and ignores long-term indirect effects. The advantage of the approach employed by us seems to be primarily its flexibility. The calculations involved are not technically advanced, and one can explore alternative views of the future without great difficulty. Nevertheless, we are fully aware not only of the abovementioned shortcomings of the approach, but of some others as well. We are referring first of all to that part, where the situation in Western Europe was used as a template for the evolving patterns of the employment transition in Eastern Europe, although elsewhere in this paper the author warned of the danger of such analogies. The argument  97 against such an approach could be, of course, that all countries are unique and there is no obvious place to look for a parallel. One might choose the poorer countries of the EEC such as Portugal and Greece, or more developed countries during periods of economic contraction such as Austria, or use other approaches to comparisons  which are  devoted to historical peculiarities of the  development of employment structures (Chapter 1). There are two strong rejoinders to such an argument. The first refers to the need for Eastern Europe to compete in a world region currently dominated by Western Europe. We would therefore be particularly interested in international comparisons which use the Western European situation as a template. The second point is that we were looking for  general regularities determining the path of structural  development of employment beyond national and/or regional specificity while testing the suitability of the Universal approach. In our opinion any regional uniqueness is what determines the magnitude and directions of deviations from the general trend. The fact that a collection of countries with different backgrounds was used in our search for an integral parameter of occupational structure, and that each of them demonstrated "goodness of fit" in terms of the methodological approach employed (see below, Figures 7-8) witnesses in favour of the argument. Certainly, the idea that all these countries display as many differences as similarities remains persuasive. In our opinion it results in the following very important conclusion: one must be careful not to interpret a simulation on this  98  basis as a forecast. This is the reason the next Part pays more attention to a country-by-country, less-aggregated analysis of labour market development in Eastern Europe. Another important point is related to the statistical data used for the analysis. As we demonstrate below, discrepancies among sets of monetary indicators originating in different sources are very large (e.g. GNP). Different estimates of GNP per capita differ not only in magnitude, but also in accuracy. Two main criteria for preferring one source over another were: a) the frequency this particular source is referred to in the literature and how this literature estimates its accuracy; b) the ability of the source to provide a dynamic homogeneous set of indicators for multiple years. We are fully aware that these criteria do not guarantee that we have the most full, accurate and precise set of data currently available. We shall note, however, that one of the main features of the statistical methodology employed is that absolute magnitudes of GNP indicators are of less significance for us. What is important is that the data have been borrowed from the same source and for the whole period of study. This allows us to achieve the necessary homogeneity of the GNP data base. The reader therefore should be particularly careful about our conclusions involving absolute monetary indicators. Their value is constrained by the particular source of data used. In terms of overall significance those inferences are illustrative in character rather than being suitable for vast generalizations.  99 §3.1.2. Occupational Descriptive Analysis.  Structure of Employment:  an Overview  and  The term "occupational structure" and general ideas on its relation to the level of economic development were introduced by Clark (1957) and Fisher (1935; 1952). The essence of these ideas narrows down to the following: as an economy develops the share of employed in its Sector I (primary economic activities, i.e. agriculture, forestry, mining, fishing, etc.) decreases first mostly in favour of Sector II (secondary activities, i.e. manufacturing industries, construction, etc.) and then - along with the decrease or relative stabilization in secondary sector employment - in favour of Sector III (tertiary activities, which are widely understood to comprise services). In its general form the foregoing relationship, having been confirmed by a number of later studies for different regions of the world (Daniels, 1982; Freeman et al., 1982; Ivanov & Ushkalov, 1991;Bulantsev, 1993),does not cause any objections. It is important to ascertain, however, whether it bears a universal character. (For example, the case of New Zealand, which attained a high level of economic development under the domination of the first and third sectors, has been closely examined in the literature).  21  Up to the second quarter of the XX century the economic development of Eastern Europe was progressing in accordance with the general framework of the post-Versailles economic transformation in Europe. Germany and Czechoslovakia stepped up the rate of manufacturing, while Romania, Hungary, Latvia, Estonia, Lithuania, Bulgaria and Poland, with their agricultural  100 specialization, were promoting mining and allied industries. Poland, Lithuania, Latvia, Estonia and Finland at the threshold of the XX century were outlying areas of the Russian Empire, and the industrial spurt of Russia in 1880-1913 did not affect these borderlands. Table 19 illustrates the macrostructural employment differences pertaining to that period: Germany and Czechoslovakia were much more industrialized than the rest of the countries if measured by structural labour indicators. This status quo remained relatively unchanged throughout the first half of the XX century.  22  As  is apparent  from the table, along with  a comparatively  underdeveloped industrial sector, more than 65% of the labour force in Lithuania and Bulgaria was involved in primary activities, although agriculturerelated industries were relatively well developed. Hungary and Czechoslovakia were the more agricultural and industrial areas of the Austro-Hungarian Empire. In Germany, unified at the beginning of the XX century, two systems of agricultural production existed: the so-called Prussian system in the "Eastern Lands" and that based on private property in the "Western Lands". Macro proportions of employment within Germany remained unchanged until the post-war division of the German Reich. The occupational structures of Romania, Poland, Latvia, Estonia and Hungary were very similar to each other. Extrapolating the data from Table 19 and comparing the average weighted indicators (Average East European Country - AEEC) with the results  101 Table 19. Sectoral Employment in selected countries, %, various years. Countries  Year  Secto  Secto  Sector  Poland  1931  53.0  25.5  17.5  Latvia  1925  55.4  23.2  21.4  Lithuania  1923  67.4  12.8  19.8  Estonia  1924  52.4  26.9  20.1  Finland  1930  56.1  27.2  16.7  Bulgaria  1920  70.1  16.1  13.8  Hungary  1920  52.5  27.2  20.3  Romania  1925  53.0  25.5  17.5  Czechoslovakia  1921  34.5  44.4  20.2  Germany Eastern Lands Western Lands  1925 1925  23.6 18.1  50.1 52.7  24.7 31.0  Source: Clark, 1957. obtained by Fourastier (Bairoch, 1971) for a Hypothetical European Country (HEC), one can see a 25-year lag behind Western Europe (Table 20). Data are presented for years when the two sets of macrostructural indicators were closest. Since the average indicators, although being illustrative, do not reflect details, it is appropriate to draw attention to some particulars. East European countries can be classified according to the following scheme: A - countries of agricultural specialization with a comparatively welldeveloped industrial sector - Poland, Romania, Hungary, Estonia and Latvia;  102 Table 20. Sectoral Employment Structures for average countries, %.  Coutry, year, indicator  Year  Sector I  Sector II  Sector III  Average East European Country (AEEC)  1925  50.5  27.8  21.3  Hypothetical European Country (HEC)  1900  48.0  29.0  23.0  Note: AEEC - simple weighted average indicator calculated for Eastern Europe. HEC - simple weighted average indicator calculated for Western European countries. Source: estimated on the basis of Clark, 1957; Bairoch, 1971. B - countries of agricultural specialization with a comparatively underdeveloped industrial sector - Lithuania and Bulgaria; C - industrial countries - German Eastern Lands and Czechoslovakia. Comparing average indicators for the groups A and B with indicators for the HEC  we  attempt  to  determine  the  time  lags accompanying  their  macrostructural development. Table 21 is constructed according to the same principle as the one above: it illustrates those years when the indicators were nearly the same. As one can see, the A-HEC time lag approximates 45 years, the B-HEC lag - 95 years, and the average - 65 years.  103 Table 21. Sectoral Employment Structures for selected groups of countries, %, various years.  Countries  Years  Sector I  Sector II  Sector III  A  1925  55.5  24.3  19.8  HEC  1880  56.0  24.0  20.0  B  1925  68.6  15.4  16.0  HEC  1830  66.0  20.0  14.0  A-B  1925  59.3  22.0  18.0  HEC  1860  59.0  24.0  17.0  Source: estimated on the basis of Clark, 1957; Bairoch, 1971. Table 22 compares the countries in question with the Southern European countries: Table 22. Sectoral Employment in selected countries and selected groups of countries, %, various years.  Years  Sector I  Sector II  Sector III  Spain  1925  52.9  29.0  18.1  Greece  1924  53.3  26.1  20.6  Portugal  1927  56.2  24.5  17.6  Average  1925  54.1  26.5  18.7  A  1925  55.5  24.3  19.8  HECAS  1925  54.8  25.4  19.2  Source: estimated on the basis of Clark, 1957. As one can see, in 1925 there were no tangible differences between  104 Southern Europe, and group A in terms of employment structure. The average parameters for the Hypothetical  European Country with Agricultural  Specialization (with East European countries excluded) - HECAS, borrowed from Bairoch, are as stated in the last row of Table 22. In case of Lithuania and Bulgaria the abovementioned lag in relation even to the HECAS was around 35 years. The occupational structure in these countries was similar to that in Yugoslavia, included in the HECAS . The 23  German Western Lands and Czechoslovakia differed in their employment structures. Allied Tables 23 and 24 offer comparisons of these countries with the HECAS, HEC and Hypothetical European Industrial Country (HEIC) : 24  Table 23. Sectoral Employment in selected countries and selected groups of countries, %, various years.  Countries  Years  Sector I  Sector II  Sector III  Lithuania  1925  67.2  13.0  19.8  Bulgaria  1925  69.8  16.2  14.0  HECAS  1890  64.3  20.0  15.7  Yugoslavia  1930  71.4  16.0  12.6  Source: estimated on the basis of Clark, 1957.  105 As follows from the tables, Czechoslovakia and German Eastern Lands surpassed the HEC in terms of Sectors' I-II relation. The German Western Lands were running far ahead of advanced European countries at that time. The situation in Czechoslovakia was similar to that in Sweden and Norway (refer to Table 25). Table 24. Sectoral Employment in selected countries and groups of countries, %, 1925.  Countries  Sector I  Sector II  Sector III  Eastern Lands  23.6  50.1  24.4  Czechoslovakia  32.0  45.4  22.2  HEC  39.6  31.0  29.5  HEIC  27.6  44.3  27.4  Source: estimated on the basis of Clark, 1957; Bairoch 1971.  The steady tendency of labour outflow from Sector I into Sector II, common for the majority of European countries, originated in the middle of the nineteenth century. A hundred years later a new trend developed. In Great Britain,  Switzerland, Sweden, Netherlands  and Germany  the share of  employment in Sector II began to fall steadily after the level of "saturation" had been reached (45-50% employed ). A similar tendency was observed later in 25  other European countries. According to Clark (1957) the level of "saturation" is associated with a certain level of economic development (this hypothesis is subject to detailed empirical testing in Chapters 2 and 3 below).  As the  106 evidence shows,  in the majority of European countries the point of  "saturation" of SectorII and the beginning of labour outflow is timed to the Table 25 Sectoral Employment Structures for selected countries, 1925, %.  Countries Eastern Lands  Sector 1 23.6  Sector II 50.0  Sector III 24.4  Germany  70 9  51 4  7.7.7  Switzerland  77 1  51 5  7fi 4  Belgium  ?.14  50.3  28.4  Czechoslov.  37. 0  45 4  7.7.7  Sweden  33 8  41 n  75 7  Norway  35.7  39.1  25.2  Source: various sources.  approximate GNP per capita level of 3000-3500 dollars . When the 40% 26  employment level is achieved in Sector III, the GNP per capita level averages around 5500-6000dollars. The sample of 12 European countries studied includes both industrial pioneers like the Netherlands, and Great Britain and  countries of "late  industrialization" (Finland, Spain, Portugal). The latter had much in common with the East European countries classified as group "A". The above analysis favours the idea that one might expect groups of countries to follow a common path of evolution of sectoral employment  107 structures. We can see from the data analysis that even different groups of countries followed a general path of occupational macrostructure to a certain extent. This interim conclusion does not go any further, in particular it does not say whether all individual countries should follow the same path and what that path is. Chapter 2 below is intended to shed some light on these issues. Following the initial hypothesis on the interdependence between the occupational structure and level of GNP per capita (Clark, 1957), one  can  Table 26. Per capita GNP in some European countries, 1950, international dollars. Countries  GNP per capita in 1980 prices  Poland  2170  Hungary  2208  Finland  2758  Romania  1069  Spain  1640  Portugal  937  Czechoslovakia  3124  France  3125  FRG  2713  GDR  2119  Austria  2318  Source: Ivanov & Ushkalov, 1991.  108 assume that in the middle of the twentieth century these indicators should have followed  a similar track of development  in the countries  of "late  industrialization" and group A. However, the GNP per capita statistics do not prove this assumption (see Table 26. Data are not available for the Baltic states). There are some similarities among the the per capita GNP indicators in the "group-C" countries and European industrial countries. The problem is that the estimates of the GNP per capita for the former "Eastern Block" differ significantly depending on the source (Table 27). Estimates made by East European organizations do not deserve much trust because official prices used for calculations did not accurately reflect the scarcity value of goods. Moreover, East European countries used to calculate the socalled "Net Material Product". The methodology of its estimation differs Table 27. Alternative Estimates of per capita GNP in East European Countries, USD, various years. Country  HestonSummers 1985  PlanEcon, 1988  CIA, 1989  Bulgaria  5113  5630  5690  CSFR  7424  7600  7900  GDR  8740  9360  9670  Hungary  5765  6490  6090  2460  Poland  4913  5450  4560  1860  Romania  4273  4120  3440  USSR  6266  5550  9230  Source: Bulantsev, 1993  World Bank, 1988  1735  109 significantly from that for the GNP: the latter includes the output of service industries. In order to be able to work with extensive time-series data on GNP in Eastern Europe, especially in a comparative perspective, one should not mix statistics from different sources. For the kind of comparison where absolute magnitudes are of less significance than relative indicators (e.g. ranking countries) or the dynamics of an indicator (e.g. annual rates of increase) international comparisons can bring more robust results.  27  We shall note in conclusion that the points on the statistical bias made in this chapter should be related to the monetary-based indicators mostly. Their main aim was to demonstrate the uselessness of the approach, which would combine data from different sources. If it can work for the Western countries, it certainly does not work for Eastern Europe.  110  CHAPTER AND  2.  LEVEL  OCCUPATIONAL OF  ECONOMIC  STRUCTURE  OF  DEVELOPMENT  EMPLOYMENT IN  EASTERN  EUROPE.  Having discussed the historical aspects of macrostructural development of labour in Eastern Europe, we are still facing a number of questions relating to the structural characteristics of employment, which cannot be resolved by means of descriptive analysis only. Among them are the following: • does an interrelationship between the employment structure and the level of economic development exist? •  At what stage of the development of employment structures are the  former USSR and countries of Eastern Europe? What is the statistically defined magnitude of their relative (in respect to the level of economic development) and absolute lagging-behind in employment structures compared to the West? •  What is the methodology on which such an estimation might be  •  What are the general directions of the development of employment  based?  structures, and, consequently, their role in labour market transitions in Eastern Europe? Unfortunately, as was mentioned elsewhere, it seems difficult to find an answer to these questions using the type of statistical indicators that are traditionally utilized for analysing labour market performance in the West. We see two main reasons for that. First, the questions above imply the need for international comparative study, which is hardly achievable if one uses a  Ill traditional set of labour market indicators. They are often designed for the needs of individual countries or groups of countries, and could be misleading if used for the purposes of international comparisons. In the case of traditional indicators the tradeoff between international comparability and relevance to specific countries and policy needs should not be overestimated. Also, concepts of employment and unemployment in a phase of structural transition need to be analyzed in their multidimensional aspects. No single indicator or set of statistics will do justice to the wide ranging policy issues that they evoke. In this case analytical progress and statistical refinement will necessarily be mutually reinforcing. The present chapter presents an original statistical indicator and new methodology to attempt to answer the research questions posed above. It is hoped that the methodology will contribute to solving the problems of crosscountry comparability of labour statistics. The following paragraph is intended to offer some new insights into the limitations and assets of the approach, while §3.2.2. is devoted to the characteristics of the adjustments of labour markets in transition countries. An intended innovation of the methodology is supposed to be its close attention to labour market flows as opposed to mere changes in overall numbers of employed and unemployed. It is intended to identify those features of labour market adjustments that may call for the refinement of policies implemented thus far in East European countries.  112 §3.2.1. Scope, Limitations and Main Assets of the Approach. Before presenting a statistical approach which attempts to test the hypothesis  about  a  "common  path  of  employment  macrostructure  development", we shall once again note that to some extent it incorporates the ideas of universal development. It seems appropriate therefore to state some reservations concerning the approach, its limitations and main assets. The statistical analysis presented below will be conducted mainly with the aid of basic econometric and computer techniques and by making use of the maximum number of observations  available, with the aim of reaching  conclusions which can be generalized (which is prescribed by our objectives). In Part V certain problems which arose in the previous parts will be examined on a less aggregated level and in more depth on a case-study basis, selecting those countries for which more detailed data are available. One of the main tasks here is to convince the reader that we are fully aware of what the incorporation of the universal approach actually means from an economic point of view. Incorporating the view of universality means that all products would have to be produced with identical production functions throughout the countries used in the sample. In addition, presumably these would all have to be fixed-coefficient production functions (fixed predetermined rate of change of the capital-to-labour ratios, fixed occupational coefficients, fixed educational coefficients, etc.).  113 All products would have to be consumed with identical consumption functions throughout the sample, and the aggregate consumption function would have to correspond to identical micro-patterns of consumption demand, otherwise variations in the composition of total output would reimpose the possibilities of substitution in production. In other words, we disregard the factor of economic openness and international exchange which influences the employment structures. Alternatively, it seems that the Universal regularities assume that all products are produced with identical variable-coefficient production functions throughout the sample, but identical relative factor endowments and hence identical relative factor prices will occur for different countries in different periods of time, i.e. at the same levels of output achieved at different times. Identical relative factor prices would mean the countries having  identical  capital-labour ratios and identical occupational and educational coefficients. This case seems to require additional conditions, however: since a production function relates the maximum quantity of some exactly specified output to the quantity employed of some set of exactly specified inputs in a given period, we must be sure that the sample countries are operating on the boundaries of their production functions; in other words the condition is that all entrepreneurs maximize profits and there are no differences in the quality of entrepreneurship in the countries in question. One of the biggest conceptual difficulties with the model is that it uses  114  a linear and homogeneous production function (not quite Cobb-Douglas however) as a basic relationship between the structural indicator and the output. On the one hand, the structural nature of the independent variable may require future elaboration on the shape of the production function: in this case it should not necessarily be akin to the Cobb-Douglas relationship. The other point is concerned with the nature of the current stage of economic development in Eastern Europe: there is no evidence in the literature yet that the traditional linear forms demonstrate sufficient "goodness of fit" for the East European situation. The East European production function may be somewhere in between the Cobb-Douglas production function with an elasticity of capitalfor-labour substitution equal to unity and the Leontief relationship with an elasticity of substitution equal to zero. Indeed the East European elasticity equals to some  n with the shape of the production function yet to be  determined. Furthermore, we require unit-income and unit-price elasticities of demand for all products, and identical public spending propensities. Clearly, we will never observe the extreme case described above in the real world. Nevertheless, the question is: how close do we actually get to it? The answer is an empirical one, but since it runs in terms of probabilities, judgement will be required in the final analysis to tell whether we have come close enough. Close enough for what? It all depends on the purpose of the inquiry. If we are trying to explain how the occupational structure changes in  115  the process of economic development, any answer will do: coming close means an end to our investigation; not coming very close means a further search for new variables with greater explanatory power. However, if we are trying to reveal whether changes in occupational structure cause economic development, not coming very close may well mean that the whole notion of basing the analysis on international comparisons must be dismissed as a blind alley. If the feasible development paths of  occupational structure are very wide, the  question of the correspondence between the development level and a certain occupational structure must be faced anew, perhaps with different tools and modes of thinking than have so far characterized the international comparative approach. In other words, if given levels of economic development are not associated with an unique set of occupational coefficients, all the single-equation estimates of this study and others like it are subject to simultaneous equation bias: we are testing the reduced forms of what is in effect a series of demand and supply equations. In case the above does not make the point clearly enough, we will spell it out as follows. The model suggested below  is not aimed at revealing  whether economic development is caused by changes in occupational structure of employment. To do this much more advanced econometric tools and much more diverse homogeneous data would be required. Our model is aimed at an understanding of what may be the first part of the causative mechanism, specifically whether levels of economic development are associatedwith a certain  116 occupational structure. To do so we would need to express the three-category educational structure in a single indicator, empirically relate those indicators to per-capita indicators of economic development (say, GNP), and then - try to understand if there is any formalizeable dynamic relationship between the two. If we come close - further research is required. Besides the abovementioned limitations and shortcomings, our model has a number of conceptual advantages and merits, which seem to be significant enough to distinguish the model from its neoclassical roots. First of all, as was noted, it suggests a way to formalize a three-category employment structure into a single indicator. This seems to be a conceptual innovation; at least the author has not seen anything like this in the literature so far. The additional difficulty here is that percentages of employed in the three sectors of the economy are interrelated relative indicators, i.e. they sum to a constant. Unfortunately, the well-known statistical technique of "Mixture Models" (see Pindyk and Rozenfield, 1990; Cornell, 1985; Marquardt, 1974), developed to deal with the situation where independent variables in a regression context sum to a constant, pursued a number of times, proved useless in helping to solve this particular task. The original methodology was therefore needed. Secondly, our model uses a dynamic chain of time-series data, without which most of the projections resulting from the analysis of manpower are rather devoid of economic meaning. Thirdly, the model attempts to use the integral structural indicator as an  117 independent variable, i.e. to analyze the employment structure as one of the factors of production, placing it as an independent variable - an attempt never pursued in econometric sources in this context. Fourthly, the model takes into account outflows of employed from one sector of the economy into another, i.e. it deals with vertical mobility of the labour force. For doing so it uses a set of adjusted data on the employment structure, using the original methodology of adjustment (seebelow). This seems to be an important methodological contribution to the theory of manpower modelling. A few technical economic details which distinguish our model from the once-popular neoclassical cases are as follows. Since we use a dynamic integral indicator of the occupational structure, its magnitude changes as the percentage of employed in at least one sector of economy changes (seebelow). This means, first of all, that the ratios of different inputs in our model, and, correspondingly, the ratios of different products that enter into consumption do not stay unaffected and are changeable in the model. This seems to be a huge deviation from the Universal approach. It means that in our model  technical progress in both production and consumption is not entirely "neutral" as has already been mentioned. Even more, if technical progress is not neutral, it reflects its acquirability at different cost in different countries of the sample. What is of equal importance is that the model takes into account different labour force participation rates in the countries of the sample, different  118 rates of entrance and exit of population into and out of the labour force, and, indirectly, different population age structures, etc. This, beyond any shadow of doubt, is not compatible with the Universal approach. We shall note in conclusion, that not only our hypothesis, but also those evident in some sources (e.g. OECD, 1970) relate occupational structure of employment not only to the level of output, but often to the rate of change in output. Our model offers some insights in this issue as well.  119 §3.2.2. Occupational Structure of Employment and Level of Economic Development: General Tradeoff.  Until recently Eastern European sources paid little attention to the occupational structure of employment.  28  There has been elaborated one main  approach, which become nearly classic. Typically, the approach to analyzing the employment  structure of population  factorial forecasting based on employment  structure  narrows down to a version of  understanding  and different  the relationship between the  indicators  of economic  efficiency  (Anchishkin, 1973).The relationship of the employment structure with the level of economic development, measured by different parameters, was, consequently, unambiguously  derived from the function of demand for labour. The  unidirectional character of this approach is obvious since it does not take into consideration the widely recognized  factor of scientific and technological  progress - a qualitative variable, showing the nature of growth of efficiency of assets and production of labour. The causes of such a simplistic understanding of manpower, structural dynamics were described in the first chapter of Part II, devoted to performance of labour markets under central planning. The procedure suggested in this chapter is based on studying the feedback  between  employment  structure  and the  level of  economic  development. Questions may be asked about the expediency of using the three-category employment macrostructure while the character of labour activity is steadily  120 diversifying. In our opinion, three- or four-sectoral (categorial) divisions of the employment structure (quaternary activities, specifically, financial services, information processing and information supply activities, do not have an adequate statistical representation in Eastern Europe and are included in the tertiary sector) reflect not only the classification of manpower by the character of the final product, but the character of the employee's activity too, although classification problems do exist. In this context the occupational structure of 29  employment has a deeper meaning than that of just a complex economic indicator. In addition we have a pure methodological consideration in favour of the three-category classification. Further diversification of the employment structure inevitably complicates the problem of international comparisons, since more detailed divisions are specific to individual countries and cannot be used to reflect general macro tendencies in the development of the structure of manpower.  30  The present chapter tests the hypothesis on correspondence  between the level of per capita GNP with the three-category employment macrostructure (here calculated according to the UN methodology).  The  hypothesis can be posed more precisely as follows: one and only one numerical value of the share of employed in each of the three sectors of the economy corresponds to a given level of GNP per head.  Two data masses constituted the statistical base of the analysis: threecategory structure of employment in percent for 26 countries over 30 recent years and per capita GNP in international dollars for the same countries over  121 the corresponding period.  31  The results of the statistical analysis refer to three representative years (1970, 1980, 1987). Figures 1-3 extrapolate data for real West European countries, showing trends of interrelationship between GNP per capita and the proportion employed in three sectors of the economy.  32  Figures 4-6 illustrate  the sectoral development of an average weighted employment structure in relation to GNP per capita for a hypothetical (average)West European country. Employment structure for a hypothetical West European country was calculated as the weighted average of annual employment structures of all West European countries in a sample for each of the sectors over the period 1960-1987. Both empirical series confirm the hypothesis by Clark and Fisher. The share of employment in primary activities decreases across time (Figure 4). A certain same level of economic development is in average attained under the higher share of the I sector. The whole picture mirrors the acceleration of development of former agricultural European countries. The parallelism of the curves might denote the existence of some sort of general relationship. An average dynamics of Sector II is far less clear.  33  Nevertheless it is  apparent from Figure 5 that over the period 1970-1987 the range of shares of secondary employment dropped from 37-60 to 30-50%. This picture has been followed by a drop in the corresponding level of per capita GNP from 7,00010,000dollars to 6,000-9,000 dollars.  34  At the same time one of the characteristics of development of secondary  122 Figures 1-3. Sectoral Dynamics of Employment Macrostructure in Western European Countries: 1970, 1980, 1987, %. Share, %  ' See endnote 33. " In fixed dollar value of 1980.  GNP per head, thous. $**  Source: author's calculations  123 employment is a scattered distribution of the employment shares corresponding to a certain level of GNP per capita. The magnitude of the gap does not change over the period and equals 15 points. Although the overall tendency of decreasing the share of secondary employment is quite clear, the range is still significant and depends on the initial level of economic development. The development of tertiary activity in Western Europe is generally related to per capita GNP growth. As the economy develops, the share of tertiary employment tends to increase, most significantly within the interval 3,000- 6,000dollars (up to 35%). The rate of increase decreases after this level has been achieved, but remains relatively high. Beginning with 8,000 - 9,000 dollars (with approximately 55% employed in the tertiary sector) there is a "saturation level" after which the insignificant increase in the share of employed corresponds to the increase in GNP per head. This tendency did not exist in 1970, was noticeable in 1980 and was clear-cut in 1987. The data even shows a reverse dynamics (up to 10 points) of the tertiary employment within the 9,000- 10,000 thousand interval (Figure 3). What is revealed by this preliminary analysis is the fact that different shares of Sectors II and III correspond to a given level of GNP per capita. This analysis however does not yet permit any conclusion about the general relationship of the occupational structure and the level of economic development.  124 Figures 4-6. Sectoral Dynamics of Employment Macrostructure for Hypothetical Western European Country: 1970, 1980, 1987, %. Share, %  A -1910 B -1980  c -1987 -1 Sector  4  J  1  I  I  AT 50-  5 - II Sector  J  i  L  / 6 - III Sector  J  6  8  .10 12 U  GNP per head, thous. $  Source: author's calculations  125 §3.2.3 Maturity of Occupational Structure. If one formalizes a three-category occupational structure indicator, for example 10%, 40%, and 50%, as a point in three-dimensional space with the corresponding coordinates along each axis, the question stands: what research tools should be used to detect a tradeoff between GNP and occupational structure? Since we deal with relative indicators of structure, different numerical values of each of the three categories could correspond to the same value of GNP. Let us assume that two different employment structures correspond to the same 6,000 dollars GNP per capita level in two different countries: 30%, 40%, 30% and 40%, 40%, 20%. The first question would be: is it generally possible? Our initial working hypothesis rejects such a possibility, but the hypothesis needs testing. The second question would be the following: GNP per capita is represented by a single numerical value. Is it possible to formalize an indicator of occupational structure in such a way that it would be represented by one numerical value also? This would facilitate the solution of the task tremendously. In other words, is it possible to logically move from three coordinates in three-dimensional space to one coordinate only and then to compare the development of GNP and employment structure? The general methodological problem has been therefore posed as follows: is it possible  to formalize a  general relationship  between occupational  employment structure (as a single, not three-category indicator) and GNP per capita?If such a relationship does exist, we would be able to talk about the  126 "hypothetical" employment structure of a country. The hypothetical structure would be that which should theoretically correspond to the attained level of GNP. Comparing hypothetical and real employment structures for the same country, we would be able to determine the degree of deformation of real employment structure measured by its divergence from the hypothetical one. The solution of this problem would allow us to solve one of the crucial tasks of our study, namely to estimate the "maturity" (or the degree of "immaturity") of employment structures in East European countries, as measured by their divergence from a hypothetical sectoral structure or, in other words, would allow us to answer the question: what should the occupational structure of employment be in a country with a given level of economic development? Does real occupational structure correspond  to the hypothetical  owe? (The mature structure is taken to be that coinciding with the hypothetical one for a given level of GNP per capita). Coefficients of a linear correlation between each element of the threecategory employment structure indicators and level of per capita GNP have been calculated. They merely prove the regularity known from the literature, specifically, anticorrelation upon increasing the share of employed in Sector I, close relationship under a similar process in Sector III and an implicit relationship with the dynamics of the characteristics in Sector II. The only conclusion which might be made from the correlation parameters calculated is that of deterministic character of share of employed in the first and third  127 sectors. The inference on the transitional character of the processes in the second sector is supported by indicators of elasticity of per capita GNP in West European countries in all three sectors of the economy. As is seen from Figures 1-3, a decreasing share of the number of employed in the first and second sectors and an increasing share in the third one are associated with growth of the GNP, but from the standpoint of different counterparts (I, II and III sectors) the employment structure does not always change in a similar manner. There should be a novel, integral indicator which would  enable evaluation of the  behaviour of the object in general, not only from the standpoint of its individual qualities (Ivanov and Ushkalov, 1992). Such an integral parameter is a synthetic characteristic of the employment structure synthesized on the basis of parameters in the three sectors. This characteristic would integrate the dynamism-in-time of the three sectors in their interrelationship and bear a forecasting load. The problem of revealing the parameter was solved on the basis of the superposition method (method of inputs) known in econometrics. Computer programming tools were used for the analysis of the degree of substantiality of interrelationships between the sectors on the basis of characteristics of elasticity calculated for each country and every year and revealing the directions of flows of the labour force. In other words, the revealed interrelations were used to synthesize an empirical indicator suitable for international comparisons, i.e. to transit from the relationship:  128 Y=f[E;(E ;E ;E )]^xt 1  2  3  (1)  Y - GNP per capita; E - employment in economy in absolute terms; E E - sectoral shares of the number of employed; / - function r  3  to the relationship: Y=m,  where  (2)  \p - integral parameter of the employment structure; / - function. Functional relationships (1) and (2) have much in common with so-called single-factored models of economic growth which establish quantitative links between the production on the one side and one of the major inputs on the other side. The main innovative and distinguishing feature of (l)and (2) is that they formalize the links between absolute indicators Y on the left-hand side and structural relative characteristics on the right-hand side. Methodologically this means an attempt to determine every ^ corresponding to each given level of economic development, so to say "pure structural indicator of the labour resource"or "structural effect". This crucial moment has not received much reflection in econometric literature so far. Since the relationship (2) is a general form of a model built on "production inputs of the second order" (here employment structure), absolute-termed characteristics can be excluded from (2) without any significant detriment to the results of the analysis (Ivanov and Ushkalov, 1992). The computer investigation into time-series and cross-section data bases  129 revealed certain tradeoffs between the sectors of employment, which are represented by Figures 7 and 8. Such a close approximation of the results to the hyperbolic-shaped  curve (Figure 7) has permitted formalization of the  interrelationships between E /E and E /E - two essential mutual dependencies 3  2  1  2  of the employment structure. Figure 8 illustrates the situation after the system of coordinates has been changed for convenience. The function depicted by Figure 8 has the following general form: (E/Ez-Ej/EJ  +2^s/[(E{E -E /E^ 2  3  +2]+a'+y.  (3)  Or ZWtf  +a '+y, a «0.4 7 ~0.1 2  where  (4)  35  As it can be seen from Figure 8, two magnitudes of \{/ correspond to one magnitude of Y, but only one magnitude of Y corresponds to each \p. Hence, in first approximation i£ is an integral parameter of the employment structure. If one knows the magnitude of ^, he/she can restore the magnitudes of the shares of employed for each of the sectors. And, equally important, the parameter 36  takes into consideration the time factor, i.e. outflows of employed from one sector into another. This was done by analyzing the dynamics of the following indicators for all countries in the sample (where data were available): a) sectoral employment in absolute terms; b) number and sectoral structure of retirees; c) number and sectoral structure of new entrants in the labour force. Making adjustments for these indicators while examining changes in proportions of  130  131 employed in three sectors, it was possible to conclude what portion of the change is attributed to "pure" inter-sectoral flows of the labour force. After 37  that the most frequent directions of sectoral interflows were revealed (i.e. situations when change in the proportion of employed really meant change in absolute sectoral employment due to labour in/outflow). The most frequently occurring sectoral flows were those from primary to secondary and tertiary activities, denoted correspondingly as E/E and 2  E /E 3  }  [Figures 7-8 and (3)].  These were called "two essential mutual dependencies of the employment structure". The coefficient of linear correlation calculated for the time-series and cross-section data for Y and \[/ in the relationship (4) varies within the interval 0.75-0.95 for different countries during the study period. The integral indicator of the employment structure \p (hereafter - IIES) can be both positive and negative with absolute magnitudes changing within the open interval ]-50;50[. When shares of employed in an economy are split equally among the three sectors, the IIES approximates zero. In general, the increase in the share of the number of employed in the III sector and correspondingly the decrease of the percentile in I and II sectors, are associated with the increase of negative magnitude of the IIES (Table 28). 38  The calculations show that the average pre-transitional rate of the IIES's change in Eastern Europe approximated 0.1  annually. Over the 1960-1987  period the average IIES for Eastern Europe (excluding the former USSR)  132 Table 28. Integral Indicator of the Employment Structure (IIEC) for the Hypothetical Western European Country (HWEC), USA and East European countries in 1960-1987.  1960  1970  1977  1978  1979  1980  1  -3.998  -5.494  -6.539  -6.744  -6.661  -7.444  2  -0.920  -2.020  -2.891  -3.030  -3.125  -3.291  3  1.512  0.162  -0.099  -0.153  -0.216  -0.333  4  0.298  -0.101  -0.048  -0.104  -0.124  -0.180  5  -0.076  -0.248  -0.512  -0.582  -0.617  -0.633  6  0.588  0.228  0.075  0.085  0.003  +0.0  7  0.344  -0.109  -0.370  -0.363  -0.365  -0.494  8  -0.338  -0.692  -0.694  -0.754  -0.817  -0.917  9  3.734  0.579  -0.292  -0.306  -0.295  -0.292  10  4.821  1.213  1.201  1.001  0.795  0.338  Note: 1 - USA; 2 - HWEC; 3 - Eastern Europe without USSR; 4 - USSR; 5 Czechoslovakia; 6 - Poland; 7 - Hungary; 8 - GDR; 9 - Bulgaria; 10 - Romania. Source: author's calculations. changed from +1.512to -0.691.The "zero level" appears in 1975. The "negative magnitude-gaining" trend was quite stable all across the period, but the gap between the average Eastern European country and the hypothetical one was ever widening and increased from 3 to 7 times in the mid-1980s.  133 Even the indicator for the hypothetical Western European country (HWEC) was not comparable with the one for the USA: the indicator of-4.133 characteristic of the HWEC in 1987 characterized the American employment structure of the mid-1960s. It is obvious that this gap (in relation to the USA) is much deeper in case of the former USSR and Eastern Europe and reaches its Table 28 (continuation).  1981  1982  1983  1984  1985  1986  1987  1  -6.458  -6.554  -6.774  -7.127  -7.554  -7.815  -8.361  2  -3.200  -3.275  -3.461  -3.714  -3.875  -4.040  -4.133  3  -0.371  -0.422  -0.475  -0.525  -0.571  -0.609  -0.691  4  -0.174  -0.176  -0.184  -0.184  -0.188  -0.213  -0.217  5  -0.801  -1.005  -1.271  -1.382  -1.487  -1.513  -1.806  6  -0.126  -0.128  -0.060  -0.150  -0.185  -0.222  -0.260  7  -0.369  -0.373  -0.360  -0.353  -0.365  -0.419  -0.477  8  -0.903  -0.962  -1.013  -1.073  -1.135  -1.201  -1.269  9  -0.316  -0.337  -0.347  -0.363  -0.395  -0.422  -0.439  10  0.291  0.273  0.200  0.174  0.143  0.122  0.103  Note: 1 - USA; 2 - HWEC; 3 - Eastern Europe without USSR; 4 - USSR; 5 Czechoslovakia; 6 - Poland; 7 - Hungary; 8 - GDR; 9 - Bulgaria; 10 - Romania. Source: author's calculations.  134 maximum in the case of Romania (where the indicator is positive). Since "zero level" means an approximately equal split of employment among the three sectors of an economy, the following statement seems to be appropriate: in Romania the "saturation level"for the second sector has not been yet attained, i.e.  the first stage of employment structure formation has not been completed (that one characterized by the outflow of the labour force from the first into the second and partly into the third sectors).  39  The same argument can be related to both the former USSR and Poland. Here we refer not only to high amplitudes of the IIES, but rather to its low changeability over time. (As one can see from Table 30, the magnitudes of the IIES for the USSR remained at approximately the same level for at least 4-5 pre-transitional years). This could be interpreted as a result of a deliberate "conservation" of the employment structure over the totalitarian years (e.g. administrative restrictions on rural-urban migrations). The number of employed in Sector I in the USSR (even according to official Soviet statistics) remained unchanged in 1965-1987,although significant regional contrasts did exist. IIES of Czechoslovakia and GDR stood out from the average Eastern European picture. Although neither country exists as a political unit any longer, their example is quite illustrative because of the (relatively) high level attained during the pre-transitional years. This level, along with Western European experience, allows us to make some forecasting assessments. What is of great importance is not the IIES by itself, but its dynamics  135 in relation to GNP per capita. Figure 9 allows us to make qualitative estimations of the "maturity" of pre-transitional East European employment structures. Maturity (or rather degree of immaturity) is understood as HES's deviation from the absolutely GNP-elastic parameter under a given GNP for the HWEC. The location of curves for the HWEC and real countries in relation to each other is a graphical illustration of the structures' maturity. It is worth emphasising that "real" Eastern European curves lie below the HWEC's curve, i.e. their employment structure is less developed than would follow from the attained level of economic development. If in 1960 the maximum  magnitude of this gap (lag of structure behind per capita GNP level) was observed in Romania, in 1987- in the GDR (which reached a level of economic development requiring a significant acceleration of tertiary growth), the USSR, Czechoslovakia, Poland and Romania. The gap has widened in all these countries except Romania and has stayed at the same level in all other East European countries. The smoothest development of the employment structures in terms of their relation to the hypothetical one was observed in Czechoslovakia, the GDR, and Romania. In the other countries, and especially in Hungary and Poland, significant deviations and even elements of negative development occurred within the interval 4,000- 6,000dollars of per capita GNP. Hungary 40  and Poland were two countries with the most clearly seen economic recessions under communist rule.  136  Figure 9. Elasticity of Integral Indicator of Employment Structure (IIES) on the GNP per Head for Hypothetical Western European Country, USSR and Eastern Europe, 1960 - 1987.  Source: author's calculations  137  138 One can distinguish several main phases in development of the hypothetical employment structure. 1) Up to a GNP level of approximately 6,500 dollars, the employment structure develops relatively slowly. In 1989 all East European countries except the GDR and Czechoslovakia were going through this stage. One of the main characteristics of this phase can be formulated in different terms: at a share of the number of employed in Sector III not exceeding 40%, the indicator of per capita income does not surpass 5,000 - 6,000 dollars. If the share of employed in Sector II continues to increase at this stage, the per capita GNP starts showing a tendency towards decline. 2) On achieving an income level of approximately 6,500 dollars there is observed a decline in the share of employed in Sector I, accompanied by an increase in Sectors II and III. In this case there occurs a relative stabilization of the level of per capita GNP over a certain period of time. Structural change accelerates when GNP is between 6,500 - 8,000 dollars: it occurred in Czechoslovakia. It seems interesting that GDR has followed the structural crisis of the HWEC at the same level of 8,000 - 9,000 dollars of GNP (circles in Figure 9). 3) . Structural crisis occurs when GNP reaches 8,000 - 9,000 dollars. Growth in per capita GNP is followed by the decrease of negative magnitudes in IIES. We can suppose that this is connected with the transition to the postindustrial society (Bell, 1974). This can be viewed also as a crisis of the threecategory occupational structure in that it becomes acutely necessary to  139 distinguish quaternary activities. This is reflected by Figures 10 -12 illustrating the dynamics of sectoral employment structure in relation to GNP level. None of the East European countries has entered this phase so far. Under  these conditions  the issue of inter-sectoral mobility of  manpower becomes of great importance. The reduction of the number of employed in Sector I in Western European countries was noted even during the pre-war period. At the level of per capita GNP of 4,000 dollars the outflows from Sectors I and II into Sector III start to dominate. The process of releasing employees from Sector II occurs until the level of income of 6,500 - 7,000 dollars is reached, when the direction of  flow changes once more. The  prevailing role is played by the outflow of employed from the second sector into the third and, partly, back into the first. Starting with the level of income of 7,000 dollars a strong instability of the share of employed in Sector II is observed: multi-directional fluctuations may be as high as 3-5% a year. At this stage of the development of employment structures the dominant role is played by two outflows: from Sector II into Sector III and from Sector I into Sector III. On reaching a per capita GNP of 10,000 - 12,000 dollars the direction changes again: Sector III starts to decrease and Sector II to increase slightly. This is definitely connected with the methodological problem of needing to distinguish between Sectors II and III on the one hand and Sector IV on the other. Such are the general regularities of the formation of the employment  140 macrostructure at different stages of its development. There are revealed at least 4 levels of labour saturation of individual sectors. These levels define changes in directions of exchange of labour among individual sectors of the economy. It should be noted that with the growth of income the weight of a unit change of the share of employed in sectors is changed. Thus, while in earlier stages of development more than 50% of the changes in per capita GNP took place due to the dynamics of Sector I, while at the present time only about 15% do so. The weight of the unit change in Sector III grew up to 65 %. We have already mentioned the lagging of changes of the parameters of the employment macrostructure in relation to the parameter of per capita GNP. This phenomenon can be named a time lag. The processes causing it do not strictly follow one another, but overlap one another in various combinations to form a hierarchical structure. Depending on the level of saturation of sectors with labour the average duration of the lag is reduced. This can be explained in two ways: (a) a greater portion of the population involved for the first time in the labour activity is attracted into the  sector III and (b) starting with an  income level of approximately 5,000 dollars changes in Sectors II and III coincide both in time and direction, thus ensuring a specific cumulative (resonance) effect. Due to this phenomenon the time lags show a downward tendency even at  declining rates of growth of per capita GNP. This is a  preliminary hypothesis which is far from being fully developed and needs further study.  141 On the basis of the methodology described above and the regularities it has revealed let us consider the situation in the former USSR and the countries of Eastern Europe, which, though varying among themselves, have some common features in terms of labour market development. What should be briefly  mentioned  here  41  is the formerly popular concept  of rational  employment of population or effective full employment. As was mentioned elsewhere, in the countries of Eastern Europe there always existed the problem of imbalance between existing and new jobs and labour resources. Despite a high proportion of labour-active population employed in public production, the number of jobs considerably exceeded the number employed. At the same time, enterprises were interested in maintaining considerable labour reserves, which increased the number of employed. Measures to entice manpower made it difficult to strengthen discipline and increase labour productivity principally due to the fact that labour in the countries of Eastern Europe and the USSR was the cheapest production factor. This resulted in the situation where in 1989 in Soviet industry every tenth job was vacant while for newly created and modernized jobs this proportion was even worse (Ivanov and Ushkalov, 1991). In the industry of Russia in 1992 every sixth new job was vacant and every third among modernized jobs (Economy of Russia...).The consequences of this imbalance are well known: according to the poll data if there are vacant jobs at enterprises, the production volume is reduced by 1.7% for each percent of vacant jobs, and the productivity of labour - by 0.7%. The pre-transitional  142 situation in countries of Eastern Europe was similar. Furthermore, there existed a disequilibrium between the growth of the basic production assets and that of the labour force: the rate of withdrawal of obsolete basic assets from production was far from optimal, which resulted not only in excessive jobs, but also in an irrational employment structure. There existed a "black market" in labour and large scale hidden unemployment (hoarded labour). These processes have resulted in the situation where the employment structure is not only  underdeveloped in comparison with, say, Western  Europe, but also in comparison with a hypothetical structure which could have been reached at a given level of development. The hypothetical  proportions of employment  in countries of pre-  transitional Eastern Europe in the mid 1980s, forecasted on the basis of the integral parameter, mainly reflected the level of Western European countries of the  mid-1960s. The hypothetical  employment  structure  of the  USSR  corresponded to the US structure of the 1940s, while the actual structure corresponded to that in the period of the Great Depression. At the same time, as regards the level of per capita GNP, countries of Eastern Europe were at the level of Western Europe of the early 1970s. The maximum gap between the actual and hypothetical employment structures in 1988 was observed in Poland, the USSR and Romania, where differences in the parameters in Sector I were equal to 10-15%. In Sector III the  143 maximum gap occurred in the GDR and Czechoslovakia. They lagged behind the hypothetical structure by about 10%, while Hungary lagged by 8%. As regards the integral parameter, the smallest gap was observed in Hungary. According to the gap parameters Bulgaria ranked intermediate between Hungary and Czechoslovakia. The question of time lags is very important, which has been confirmed by the lagging of employment decrease rates behind the GNP decrease rates. One of the possible explanations was offered above. There are at least three more which seem to clarify the picture. They have been mentioned elsewhere in the paper and here we just mention them briefly: -job reallocation from the state to private sector has occurred to a large extent without flows in and out of unemployment. - The adjustment of employment to decreases in output has involved, especially during the early stages of transition, more reductions in hours worked than in number of employed persons. - A significant amount of labour shedding has been accomplished by pushing workers out of the labour force. It appears that lay-offs to date have been fairly selective, often involving persons close to retirement age. These conclusions are subject to more detailed examinations in the next chapters, as well as in Part V of the thesis.  144 C H A P T E R 3.  CONCLUDING  REMARKS.  Labour market transition to a market economy in  Eastern Europe  makes it necessary to rethink some important theoretical issues. Among these the problem of rational employment structure is perhaps one of the central ones since it is employment  which mediates between production and  consumption, exerting a powerful influence upon all economic, social and political processes. In their turn, employment structures in the region are both the result of prior development and the basis for new economic patterns. They also serve as a factor limiting the rate of structural adjustment under transition. For all the variety of patterns on the way to a market-based economy, all the countries of the region are now facing grave employment problems. The unemployment  problem  (or  imbalance  between  labour  resources and  opportunities for their effective use under new conditions), reflects structural economic disproportions inherited from earlier development. One of these is the disproportion between the achieved level of economic development and occupational structure. Using the so-called Clark-Fisher-Galenson relationship, we have analyzed the employment macrostructure in relation to the achieved GNP per capita level. The analysis was based upon extensive arrays of statistical data and a modified one-factor structural model of economic growth. The analysis showed the relationship between level of economic  development  (measured as per capita GNP) and employment macrostructure. We have also composed an integral parameter of occupational structure absolutely elastic in  145 terms of per capita GNP in countries of Western Europe. The pattern for Eastern Europe was the reverse. Thus it became possible to compare the evolution of employment  structures at the macro level as conditioned by  economic growth in Eastern and Western Europe, i.e. under planned and market economies, the period under study being 1950-1989. It may be well to point out here again that our aim was to test what isolated comparisons,  some existing development  theories and ordinary  commonsense seem to suggest - that there is a certain positive relationship between the occupational structural inputs and certain economic variables indicating a given level of development. While the correlation analysis enabled us to confirm the existence of the correlation, it cannot throw much light on the causal relationship: whatever indicator may be used to measure the level of development, does it result from a certain occupational structure of the working population, or is the latter a consequence of the development level attained? May the dependence be reciprocal? The causal relationship however, was not among our objectives, as was mentioned elsewhere. The fact that no clear line has been drawn between cause and effect should not obscure the importance of clarifying the relationship, as well as some definite methodological achievents. It also should not preclude us from making certain inferences. The employment macrostructure of the average West European country was regarded as the maximum of maturity. So, each country of pre-transitional Eastern Europe (including the former USSR) may be characterised by two  146 employment macrostructures - one real and one hypothetical, i.e. conforming to the achieved national level of per capita GNP. Comparison between real and hypothetical employment structures made it possible: - to measure the temporal lag between real and hypothetical employment structures in every sector (Figures 10 - 12); - to assess the gap between those structures in every sector (Figures 10 - 12). The figures mentioned and Figure 9 indicate that as early as 1960 the USSR and all the countries of Eastern Europe lagged behind the hypothetical value of the integral parameter of employment structure and that the lag was over 40 years. It should be noted that the lag in per capita GNP was far less significant, which means that occupational structure in these countries was much less mature than the one which would have been likely to correspond to the achieved level of per capita GNP under Western Europe-type market conditions. Without going any deeper into details of the gap behaviour in different East European countries, discussed elsewhere in this paper, we should mention that the deformation of the employment macrostructures in the countries of the region is made particularly visible through an analysis of disproportionate development of the three sectors (Figures 10 - 12). The employment share of Sector I in all the countries in the pretransitional period was much higher than the proportion expected on the basis of level of development. In 1987 the gap was biggest in Romania and the USSR  147 while in Czechoslovakia it was the smallest, bringing the country close to the world standard. The employment trend in Sector III was the reverse of that in Sector I and was underdeveloped in all the countries of the region, the gap being about 8-10 points everywhere except for Poland (some 2 points). The situation in Sector II was far from uniform. Hungary was unique in having achieved the hypothetical level. Our analysis of occupational structures leads us to conclude that not only were they deformed to an extent unprecedented in postwar Europe, but some features had been inherited from earlier periods. Nevertheless, in 19601989 the scope of the deformation was constantly  growing  (except for  Czechoslovakia). Short-term positive developments, observed in this or that sector, were unsteady and failed to improve the overall employment structure. This situation arose because of a simplified understanding of the universality of labour in the Soviet period as related to the distribution of labour. It was assumed that universal employment not only alters the pattern of social division of labour, but also requires specific proportions in labour distribution between major fields and sectors of production. Universality of labour was treated as a direct implication of public ownership of the means of production in its highest, most developed form - state property. This mechanical interpretation of the planned distribution of labour force by sectors tended to reduce economic growth to increasing just National Income (Net Material  148 Product) created in the field of material (tangible) production. This approach divided employment structure (and the entire economy) into the categories of material and non-material production, subdivided into smaller sectors. The latter economic field was regarded as subordinate, and its output had no direct effect on the calculated growth of national income. The result was a narrowed list of socioeconomic priorities. Objective regularities in the development of the employment structures were practically ignored, leading ultimately to a backward pattern of employment. This situation resulted in the permanent reproduction of inefficient (immature) employment structures during the pre-transition period, both at the micro- and the macro levels. As indicated above, during the last pre-transitional decade the integral maturity index of occupational structure was improving very slowly. The lag from an adequate level was considerable. Thus, in 1960-1989 this gap increased from 2 points to 5-7 points in Poland, Romania, and the USSR, and from 1-1.5 points to 5-7 points in Czechoslovakia and Hungary. Even without knowing the short-term results of the first years of transition described elsewhere in the thesis, it is possible to conclude on the basis of our study that deep structural change in employment was likely to proceed even without any economic change whatsoever. Since output is falling (the causes of which were examined in Part I) the countries are facing largescale long-term intersectoral shifts of labour, which implies unemployment of a structural cyclical character.  149 Perhaps it should be noted that we have tried  multiple regression  analysis in attempt to uncover the role (weight) of employment change in each sector of the economy in the unit change of output. Trying this analysis on the basis of both "classical" multiple regression and "mixture models" techniques (elaborated to deal with the situation where the independent variables in a regression context sum to a constant, seePindyk and Rubinfield, 1990; Cornell, 1985) we have not arrived at significant magnitudes of the regression coefficients. This proves the existing techniques inadequate for the purposes of our analysis and calls for new research tools and/or research conducted on a less aggregate level to minimize the role played by classification and definition problems. Our attempts to introduce additional explanatory variables into the regression equations for each country of the sample (e.g. absolute number of employed in the economy, etc.), although frequently improving the correlation coefficients, made the estimate highly indeterminate because of the high standard deviations, caused in turn by the high intercorrelation between most of the independent variables used. Although there were few exceptions (i.e. statistically significant results for some countries in the sample), they should be regarded as exceptions. We would conclude therefore that at the present level of the study the results of the multiple regressions do not permit the construction of a hypothesis on the role played by each sector in the unit change of the output. The best overall results have been obtained for the whole economy in  150 terms both of correlation coefficients and of the character of tradeoff between the output  and the integral indicator of the occupational  structure of  employment. The latter has been formalized. Our results can further be improved by a careful correction for  "deviant" countries. At the least,  therefore, our international comparison approach could serve as a useful macroeconomic checking device for national manpower estimates. All of the above is intended to contribute to the following statement. Having revealed the general dynamic tradeoff between employment structure and per capita GNP, one can ask: in what way are the problems of unemployment in Eastern Europe caused by the situation described? Does this unemployment have a cyclical structural character or does it originate from some other source or sources? Our analysis does not allow us to answer this question fully. There is a  link between the collapse of GNP, structural  imbalance in employment and the unemployment issue. And this link is a very important characteristic of the macroeconomic situation under the transition, which, though differing in extent in individual countries, in our opinion is common for the region as a whole. Unfortunately at the present stage we are unable to say whether this link has a deterministic character. It is interesting to note in the light of the above discussion and in conclusion of the present chapter that so far East European unemployment has been studied in a rudimentary way. The prevailing views on this matter, which do not link unemployment to the fall in output and do not assume it has a  151 structural character, are summarized by Chilosi (1993): "...the rise of unemployment has remained remarkably behind the fall in production... [I]mportant factors increasing unemployment have not yet had the time to work themselves fully out. ...[U]nemployment seems to derive mostly from the contraction of demand, rather than from large scale restructuring which is still to come". 42  As was mentioned, the temptation to forecast unemployment in Eastern Europe in the long run as the difference between the actual and hypothetical employment structures should be overcome. The only thing such a calculation could demonstrate would be the maximum possible unemployment, but some methodological stipulations are necessary here. It is very difficult to make a forecast since, even if one knows the possible outflows of manpower from Sectors I and II, the ability of the tertiary activities to absorb labour is unknown and to a large extent unpredictable. This ability is connected with the pace of privatization, the rate of job increase in Sector III, the vacancy rate, the pace of development of new private enterprises, and the success of government stabilization policies and policies towards employment, as well as with a mass of other factors, the majority of which are beyond the scope of the present thesis. Before we move on to some detailed discussion on labour market developments  under transition, it would be appropriate to make some  concluding remarks of an overall character related to the general logic behind our empirical study in the present Part. There is some evidence that East European unemployment is two-fold.  152  It is associated on the one hand with the compression of real demand deriving from stabilization policies and on the other hand - with structural dislocations of economic activities. Unlike the majority of relevant sources, we would like to lay the emphasis on the interconnections between these two causes. The general logic here is as follows.  43  Open inflation ensues from price liberalization and the cost push caused by the phasing out of subsidies. Price liberalization is a measure of economic stabilization, because it leads to the elimination of repressed inflation, as well as to institutional change, because it is the prerequisite for the establishment of functioning markets. The phasing out of subsidies responds to the same logic: it eliminates a source of public expenditure and increases costs to enterprises; at the same time it allows market calculations to be based on a price structure in which some degree of confidence can be placed, in the sense that it is not too far from the structure of social opportunity costs. Price liberalization and the phasing out of subsidies lead to a fundamental change in the profitability of different kinds of economic activity. Hard budget constraints imply that unprofitable activities should in principle be suppressed, and the workers performing them be laid off. These arguments being tied to the discussion on the causes of output collapse, form the general logistic framework of the discussion on structural unemployment. Quite clearly, at the level of aggregation of the present study and with the kind of data used, any further interpretation has to be made with care.  153 Further analysis will have to be pursued on a case study basis by taking, for example, the countries with different occupational values and which are at analogous levels of development, and by bringing in a host of additional explanatory variables in order to determine which are the most important factors behind those variations.  154  PART IV. LABOUR ISSUES IN GROWTH DEVELOPMENT THOUGHT: AN OVERVIEW.  AND  What follows is a survey of the main growth and development theories with structural aspects of labour in focus. Although relatively brief, it aims at rooting the model submitted in the foregoing chapter in the existing theories of economic progress. In the present chapter, however, these theories may seem to be taken from the original context of their source. This can be justified. We are attempting neither a general survey nor criticism of the theories, our primary goal being an outline of the theoretical evolution of labour problems. The model we have presented in the above chapters has a few very important assumptions.  First of all, it assumes universality in economic  development, an assumption which may seem anti-geographical and significant enough to question the results. As we will see, however, the relative validity of nationally-specific paths of economic development and paradigms of general regularities in development is far from settled in the literature. Building a model as we did does not mean taking sides and is suggested as one of the approaches appropriate to analyze the problem. Other assumptions of the model, such as the one that the forces of competition within the economies are sufficiently strong and well-articulated, that employers are sufficiently sensitive to price changes to respond  by  changing their production techniques, and that the single-factored model does not bear significant causative errors, etc. are indeed open to questions, especially  155 if applied to small and poor countries. This is a situation where we face the trade-off between the analytic simplicity of a model and the seemingly greater complications associated with the facts of growth to be explained. In our opinion, one of the most significant assumptions was that related to a closed economy. World trade has a large influence over the sectoral allocation of labour, especially in manufacturing. Trying to take the factor of "openness" into consideration while building the model would however involve much more complicated techniques and itself can be a topic for a dissertation. It is well-known that the terms "growth" and "development" have different meanings. Without going deeply into terminological differences, we 44  would claim that the model presented in the earlier chapters and linking change in output to changes in relative (structural) indicators of employment is neither development nor growth theory in their accepted classic form. On the one hand, it features output as a dependent variable, is highly aggregated and uses a limited number of inputs. These characteristics, as well as quantitative relations linking output and input, make our model very close to growth models. On the other hand, using relative and interconnected variables as inputs, which, given the dynamic character of the model, reflects differences 45  in sectoral responses (usually neglected or ignored by growth models), our model approximates development models. These however are less amenable to mathematical formulation, graphic representation and statistical testing. In our personal opinion, models of economic development are at once less purely  156 scientific than models of economic growth, and, because of their greater detail, considerably more suggestive. The possibility of including variables not easily quantified implies an added diversity at the expense of lower scientific content. Trying to identify the role played by occupational  indicators of  employment in various economic theories, we will make references to both growth and development models. It is worth mentioning, that for more than a century after Ricardo, Malthus, and Marx developed their models, economic theories were largely concerned with problems other than growth. Growth  was intellectually  neglected in part because it proceeded so effectively in Western Europe and North America despite the spectre of Ricardian diminishing returns. Growth resulted from both changes in technology  introduced by entrepreneurs,  described in Joseph Schumpeter's development  theory  (1911) and from  economies of scale that cheapened output as volume increased, first discussed thoroughly by Alfred Marshall. Growth was rapid and fairly regular and was therefore displaced from attention by other, more pressing policy problems. Not until the Depression of the 1930s and the period following World War II did the question of economic growth arise again, first as a by-product of the Keynesian revolution in income theory and later as scholars became aware of the international incidence of poverty outside Europe and North America. In a model that focused more on deficient demand than on gaps in supply, Keynes pointed out the possibility of economic equilibrium at less than  157 full employment, which is governed by the level of national output. His concern emphasized economic stability rather than growth as such, and his model of income determination dealt with the very short run. There is no need to rehash the importance of Keynesianism and its role in determining the role of government in capitalist society. For our study two points are of importance. First, it was Keynes who set up the general direction of economic study for years to come: the emphasis was laid on aggregation at the national level with total neglect of internal economic structures and regional disparities. Second, employment was viewed as a derivative of a number of factors influencing the change in output, specifically, consumption level, investment and international exchange. For years thereafter employment  was forgotten as a factor of  production and there was no feedback between employment and the output level examined. The Keynesian spirit gave root to the Harrod-Domar model (Harrod, 1939,1948;Domar, 1957). Harrod observed that the national income sufficient to provide full employment  (at a given wage) in one period would be  insufficient in the next because of the additional production capacity created in the first period that would be available in the second. In brief, the additional spending needed for full employment in the second period could be calculated using the relationship between the capital stock and its output, i.e. by the capital/output ratio. In its main features the Harrod-Domar model shows the dynamic relationship between a single scarce factor of production (capital stock)  158 and a single homogeneous output, which, generally, has much in common with our model. In the Harrod-Domar views, as well as in Keynesian, labour was placed as a dependent variable. Of course, labour can be introduced in the Harrod-Domar system, but there must be a constant ratio of capital to labour. Such a ratio will be constant only if labour grows at the same rate as capital stock - a coincidence at best - or if labour is redundant regardless of the rate of capital expansion, implying that labour is available at a constant wage rate. If labour and capital grow at different rates, then under the model either labour or capital must be underutilized. The Harrod-Domar theory did not survive empirical testing. One of the main reasons was the growth observed in the 1950s. It has been faster than can be accounted for by the rate of physical capital formation and a fixed capital/output ratio. The theory could be "saved"by allowing the labour/capital ratio to change, but then it ceases to be a theory and lapses into the category of tautology.  46  These rigidities of the Harrod-Domar model led scholars to explore theories that permitted greater flexibility. When the prices of capital and labour changed and when correspondingly appropriate substitutions  were made  between the factors of production, the outcome in terms of growth of output held considerable interest. One of the first scientists who started reasoning along this lines was Robert Solow. He started combining variable proportions of the factors and using flexible factor prices (Solow, 1970). Solow and many others  159 have used a particular mathematical foundation in their theoretical and empirical research: the Cobb-Douglas constant returns to scale production function, in which capital and labour can be substituted for each other as production  factors. These researches  (later called neoclassical theories of  economic growth) have arrived at one crucially important conclusion, specifically  a difference between constant returns to scale and the law of  diminishing marginal returns to a single factor of production. The difference implied that the "returns" in constant returns are returns to all the factors of production taken together, while the law of diminishing returns refers to the returns to factors taken one at a time, holding other things constant. The conclusion was that marginal returns to either (or any) of the factors taken individually may diminish, ceteris paribus, at the same time that equal proportionate increases in all the factors lead to an equal increase in output. This conclusion, extended and generalized by James Meade from Cambridge University (Meade, 1962) gave rise to emphasising structural aspects while studying production factors in their relation to final homogeneous output and served as a foundation for Colin Clark's theory. The use of changing relative factor prices, innovated by the neoclassical theories, and productivities to change the proportions in which inputs are combined in the production process was one of the main theoretical peaks achieved by the 1960s. Behind those possibilities for factor substitution was the implicit assumption that, as we mentioned earlier, the forces of competition within the economy are sufficiently  160 strong and well-articulated, that changes in relative scarcity are reflected in relative price changes, and that employers are sufficiently sensitive to these price changes to respond by changing their production techniques. These assumptions are very similar to those for our model. The inherent instability of the Harrod-Domar model was intellectually challenging. So was the implication of instant and complete adjustment to factor price changes through factor substitution featured in the neoclassical model. Nicholas Kaldor has tried to overcome both problems (Kaldor, 1957). His objection to ready substitution between capital and labour lies in the rigidity of the technology already embodied in existing machines. For Kaldor, all technological change is embodied in physical capital. No technical progress can occur without accompanying investment. In the neoclassical model all technical change is "disembodied" in the sense that it proceeds in time with or without the accompaniment of investment. The escape from instability, in Kaldor's view, is tied to the relations uniting technical progress and the capital/output ratio. If technical progress was to grow more swiftly than the capital stock, the marginal productivity of capital would increase leading to more investment. And conversely, if capital investment were to grow faster than technical change, the marginal product of capital would fall, discouraging such a rapid rate of investment. Economic growth, proceeding according to this scheme, tends to work along an equilibrium path in which the growth rates for the capital stock, total output, and labour productivity are all equal. One can note that in this  161 form the model is nothing breath-taking even without deep analysis of its shortcomings and assumptions. In our opinion, however, this model was very important, because it was unsuccessfully criticizing one of the main postulates of the neoclassical theories. The neoclassical notion of disembodied technical progress comes from the opportunity to increase growth by means of structural re-arranging the inputs without any absolute increase in the stock of inputs. This idea was then drawn upon by Clark and Fischer in their structural theory of labour. If we consider all the above contributions  carefully, their most  remarkable common characteristic is that they have not come from conventional marginal economic analysis, which was the major direction in economics before the 1930s. On the contrary, they have mainly come out in polemic with it. Moreover, they all deal mainly with problems of production. None of them touches upon problems of optimum allocation of scarce goods/factors. All the contributors just mentioned stem from what sometimes is called "the production" or "industry approach" to economic growth (Pasinetti, 1981) as opposed to marginal economics and the optimum allocation of scarce resources - a direction developed much later, in the 1960s with its highly innovative mathematical tools of optimization under constraints. The following main conclusions coming from growth theories and extensively confirmed by empirical studies, even in simplified presentation below, seem to summarize almost 150 years of economic modelling (till  162 approximately the 1960s): - Economic growth as a process implies that the capital stock grows more rapidly than the labour force. The capital/labour ratio increases. - The increasing amount of capital combined with complementary labour implies that labour productivity, measured simply as the amount of output in a period divided by the labour inputs in the same period, also rises. - The growth process maintains roughly constant the relative shares of labour and capital in national income. - The return to capital is constant, or at least shows no definite trend over time. - The capital/output ratio is constant, or at least shows no definite trend over time. As one can see, these conclusions were far too simplistic to allow further insights in the factors of economic growth. In particular, there was a necessity to take the analysis in a number of different directions, for example into a number of sectors, the allocation process between sectors, the relations among inputs in the production process, etc. Those scholars who understood remarkably well the requirements of production, did not go into the dynamics of it, which is indeed the aspect which gives it full relevance. The others mainly concentrated on the exploration of particular, though important, aspects, or only on the macroeconomic aspects of the process of growth in a modern society. So, the 1960s was a period when it became obvious that existing growth models are not omnipotent, a period when these models started to be used as building blocks for more complex development economics. Dual economic models were one of the first steps in this direction. To mention just a few, these theories, dealing with interactions of economic sectors, stem from the works by Arthur Lewis, John Feiand Gustav Ranis (Lewis, 1954; Fei & Ranis, 1964). The heart of the Lewis-Fei-Ranis view centers on  163 agricultural labour as initially unlimited in supply owing to the size of the agricultural labour force and initially low productivity in the sector. This view was the first one dealing with intersectoral flows of labour, and even more with the interaction of different sectors of the economy. The idea of growth  as an unlinear process which would continue  permanently once momentum is gained was a leitmotif in the development literature in the 1940s and 1950s. It became one of the tenets of the neoclassical school. This idea has been summarized by Lewis: Once a snowball starts to move downhill, it will move of its own momentum, and will get bigger and bigger as it goes along... You have, as it were, to begin by rolling your snowball up the mountain. Once you get there, the rest is easy, but you cannot get it there without making an initial effort, (cited in Lin, 1994.). It would be unfair to say that the neoclassical school did not contribute studies to structural issues of economic growth. Suffice it to recall Francois Perroux, Gunnar Myrdal and Albert Hirschmann. The notion of vertically polarized economic growth was first introduced by Perroux ("growth pole") and then translated into spatial concepts of economic growth by Boudeville ("growth center"), Hirschman ("spatial polarization") and Myrdal ("backwash effect"). Although the detailed analysis of these theories is well beyond the scope of the present thesis, some features, related to our topic, are worth noting. Structural studies within the neoclassical school attacked the accepted equilibrium concepts. First, the concept was extended in the regional discourse of economic development. Second, new studies challenged the notion of static  164 economic equilibrium. Both innovations significantly enlarged the analyzed spectrum of factors of economic growth, and especially emphasised its structural sides. One of the first insights into dynamic structural patterns of production factors was suggested by A. Fischer (1952)and Colin Clark (1957). The alreadymentioned  Fischer-Clark hypothesis specified a sequence in labour force use  and was the first to touch upon quantitative definition of underdevelopment (on the basis of per capita statistics, which is the way we still mainly define it). High proportions in agriculture would diminish as development proceeded, and would be replaced by larger numbers in unsophisticated industry, reflecting high income elasticities of demand and low productivity. As incomes continue to rise, income elasticities of demand for services of many types would gradually predominate. Productivity in services may rise less rapidly than elsewhere. As a result, labour force allocation in services would rise, as a proportion of the total, owing to improvements in productivity in industry unmatched by similar advances in the service sector. The lowest-income countries would therefore be characterized by the highest concentrations of workers in agriculture; middleincome countries would feature high proportions in industry, while the highest proportions in services would be found in the highest-income countries. Numerous attempts have been made in the literature to test and verify this concept. The evidence, however, is mixed (e.g. Daniels, 1982; Freeman et. al.,1982;Ivanov & Ushkalov, 1991;Bulantsev, 1993;Ivanov & Bulantsev, 1991;  165 Herrick & Kindleberger, 1983;Pasinetti, 1981;etc.). Summing up the inferences from the literature, the following statement seems to be true: as a general concept, the Clark-Fischer hypothesis holds water, but there are numerous factors influencing the allocation of labour and allowing exceptions from the rule. Our study, although deeply rooted in the Clark-Fischer view, did not aim at testing it, but shifted the emphasis towards questions of intersectoral mobility of the labour force. Numerous variations in labour force deployment summarized in the literature, by sector, suggest a further question relevant to our study. A neoclassical framework implies that labour productivity would tend to become equalized as the development process proceeded. The Lewis-Fei-Ranis model, already mentioned, reminds us that low agricultural productivity and high industrial efficiency would tend to converge as incomes rose and as intersectoral movement of labour and capital had their equalizing influence. Empirical observations  illuminate (Herrick & Kindleberger, 1983; Pasinetti, 1981;  Burmeister & Dobell, 1970; OECD, 1970). Sectoral productivities do not converge quickly or monotonically to the national average, thus offering some additional support for Clark-Fischer reasoning and taking some grounds away from neoclassicists. We would therefore stress the above issues of production factors structure, productivity evolution and resource allocation as the important theoretical departure points from  the neoclassical approach to  growth in the early and middle 1960s. Besides the works of Fischer, Clark and  166 Kuznetz (1971), the research of Hollis Chenery and his associates is the most directly related to these transformational shifts (Chenery & Syrquin, 1975). Mostly due to their works, by the mid-late-1960s there was a growing awareness among theorists and practioners that explanations of and prescriptions for growth and development must cover many non-economic and non-quantitative variables if they are to claim success. Looking back in 1984, Clark considered that the three most important merits of his work were to have emphasised the significance of structural issues in employment, the significance of tertiary sector growth as an indication of development and to have pointed out that capital alone does not produce growth (Clark, 1984). Colin Clark's works have contributed to a wave of research on international development and inequalities of a different sort, sometimes unified under the headline of "Underdevelopment Theories". In this chapter we have to mention the Dutch scholar J. Boeke, who created a logical transition chain from Clark's hypotheses to numerous Theories of Underdevelopment. In moving from C. Clark to J. Boeke, one undergoes a radical change in perspective, characteristic of that period. Boeke's The Structure of the Netherlands Indian Economy was published in 1942 by the Institute of Pacific Relations in New York. It was part of the concern in both America and Britain with the postwar order and in particular with the less-developed, larger part of the planet. If C.Clark's studies were statistical and global, Boeke's was regional and  167 descriptive: it was what you would get by applying a magnifying glass to C.Clark's statistical items. In our opinion, Boeke's work is significant at least in two respects: in the thorough descriptive exposition of a dual economy (significantly expanded later by geographers, such as Santos and McGee) and in the presentation of the effects of culture on economic activity. It would be wrong to say that Boeke was first to notice all this. Dual economy and cultural aspects were widely recognized by Marx and developed by Hilferding and Lenin. Boeke, however, was the first to present a non-Marxian expression of dualism, where different modes of production were not mutually exclusive, but could overlap and coexist. Boeke's dual economy framework  gave rise to  numerous growth models and strategies that were predicated on regional contrasts within a country and micro-level regional studies. Boeke's works were taking root in case studies of rural-to-urban migrations. Of course, the notion that a labour surplus existed in backward countries owing to the inefficiency of their agriculture was fairly conventional and apparently solidly established during the post-Clark period. In our opinion, what Boeke contributed was a social model that included different, even antagonistic operative modes, such as that the profit-incentive did not function in the villages in the same way as in the cities. In this respect, that is, in the weight which he attributed to culture, he was a forerunner of what became known as the modernization school of American sociology, which was the principal alternative to development economics as a source of Western strategies for progress in the Third World  168 (Faria, 1991). It could be seeing too much in the short period covered, but the temptation is irresistible to depict an expanding progression from C.Clark, who was among the first to emphasise structural issues of employment, to J.Boeke, who dealt with the problem of an available underused work force, to Rosenstein-Rodan (1948), who first proposed a strategy to absorb it via fast industrialization, and finally to Mandelbaum (1961),who described the means to  make  that  strategy  workable,  and only  then  - to  Theories of  Underdevelopment. It may seem to sound too pat, because, for one thing, these authors were not discovering anything entirely new that was not known before them. However, they were among the first to challenge major neoclassical conclusions and try to answer the questions posed by a relatively finished set of hypotheses conducted in the earlier periods. Despite the variety of circumstances present in Third World countries and the obstacles for economic growth and development there, the economic research of the 1930s-1970s has uncovered some regularities in the process of economic growth and structural change, concerned, first of all, with changes in the structure of output and production factors, and shifts in the structure of domestic demand, consumption and investment tendencies. Even from the present-day positions  calling for complex theories  of economic growth  incorporating interdisciplinary approaches, one should not neglect general tendencies. Eastern Europe is one of the best examples of possible consequences  169 of such a neglect. Surely, each country has unique characteristics. In our enthusiasm for uncovering features common to the development process in many countries, one should not lose sight of local differences. The general methodological question however still stands: "Do local differences determine the general path of development or do they influence the degree of deviation from somewhat general tendencies?" The approach to debating this question is mainly a matter of disciplinary discourse. As usual, the truth is somewhere in between. There is no doubt that one should avoid slavish adherence to already established general patterns of development, as well as expressions of overwhelming concern when a country diverges from these patterns. As Herrick and Kindleberger (1983) put it with explicit reference to Eastern Europe: Economic history, recent and not so recent, has demonstrated inescapably that giant hydroelectric installations, impressive steel mills or automated oil refineries, while symbolic of development, are neither necessary nor sufficient to cause it. Toward the end of the 1960s it was found that despite years of impressive development efforts and substantial investments in many Third World countries, the gap between the developed and developing economies had been widening judging by numerous indicators. Such problems as regional disparities in growth, wide-spread poverty, income diversification,  and  abundant social ills became more than noticeable on a seemingly problem-free background of aggregate economic growth. It was the period when Chenery's  170 famous statement on "growth without development" appeared. One can see little succession between Underdevelopment Theories and early orthodoxies. We would pin-point two major aspects of differences which seem relevant to the objectives of the present chapter. First, the new generation of theories was less explanatory (less scientific?) and more normative, aimed at improving the applied planning process. Secondly, these theories significantly expanded the disciplinary context for modelling, creating new objectives and methods for both theoretical and applied research. One particular type of theory should be mentioned here, namely theory emphasising issues of labour force mobility (mostly horizontal). We refer to the famous "core-periphery" models. Sometimes they are mistakenly deemed to be pioneered by John Friedmann (also mistakenly linked to the neoclassical branch of development thought: Lin, 1994). Without denying Friedmann's merit of creating a coherent and elegant theory, we would point out the role played by R. Prebisch, who had been very influential in shaping economic policy in a rigid neoclassical direction in his native Argentina before the war, and then extended his views. The labour mobility idea in its core-periphery sense was first employed in his analysis (1950), although not with the ideological connotation that it would acquire in the hands of dependency theorists, but as a conventional designation for a geographical arrangement underpinned by historic-economic circumstances that evolved in the "natural course" of events. Although somewhat simplistic, Prebisch's idea narrowed down to the view that  171 primary production for international exchange had nearly reached the end of its possibilities for generating internal growth due to various factors. Among them were: higher productivity and technological advantages in the core (read Western) countries, and deterioration of prices for primary commodities in relation to secondary-sector goods. We can see here how these ideas overlap with the Clark-Fischer hypotheses and our model. What is important for us is that this kind of denial of international specialization and division of labour principles dating back to Adam Smith, signifies substantial departure from neoclassic principles. Particularly important is that the core-periphery order of things went against a Pareto equilibrium in the international factors market. In our survey of development thought it is important what the later development theories (1960s-1980s)inherit from the earlier works. The above comments on Prebisch's ideas prove our earlier thesis that there was little succession between classic and neoclassic stages on the one hand and Underdevelopment and Post-Dependency Theories on the other. There was lively debate in the literature, especially in the 1970s, discussing how classical and neoclassical  theories  contributed  to  development  thought.  The  overwhelming consensus is that Keynes' "General Theory" was inapplicable to underdeveloped economies. This conclusion has some applicability towards Eastern Europe, and we therefore will devote some space to detail. The main line of reasoning (e.g. Rao, 1952; Seers, 1979; Kurihara, 1959; Thirl wall, 1982)  172 was as follows. While in a developed economy the marginal propensity to consume decreases with an increase in income and brings down with it the value of the multiplier, thus requiring easier money and higher investment to restore equilibrium with full employment, in an underdeveloped economy "the marginal propensity to consume is likely to be not far short from unity", which produces a higher multiplier "with  the result that comparatively small  increments are likely to bring about full employment". This reading of Keynes led to the inferences, which may be called brave at the very least: The paradoxical situation, therefore, arises that the poorer the community the greater the ease of obtaining for it a condition of full employment caused by changes in its net investments (Rao, 1952). This did not tally with the fact that "deficit financing and created money" in backward agricultural countries had a minimal multiplier effect on output. Without going into the discussion, we will note however some consistency between this conclusion and our inferences on "immature" structures of employment. We are leaving aside numerous complex development theories evolved through the 1980s, such as Post-Dependency Theories (World System Theories, Structuralist Theories, Neoclassical Marxism, etc.), as well as Post-Marxist Theories (French Regulation School, Relativism, Advanced Structuralism, etc.) suggesting innovative, complex and elegant models for development. This is justified by a constrained volume of the present chapter. The 1980s were the decade when economic modelling returned to  173 growth theory, turning it to development. This was partly a result of reality proving fractious to the implementation of Western developmental techniques and strategies in the Third World. Disillusionment began by the mid - late 1970s and became obvious in the 1980s.This disillusionment can be detected in several directions, such as the persistence of poverty, high unemployment and other types of social inequality. The bewildered state of development thought in the 1980s was well summarized by Faria: Rich countries are getting rich and poor countries are getting poorer. How could it be that after a quarter century of development policies, development was not taking strong root in the Third World, whereas it had sufficed less than a decade of the Marshall Plan to get Europe on its feet and on the road to prosperity? ...Whatever per capita figures expressed, there could be no development with growing poverty, unemployment or inequality (Faria, 1991). A well-known monograph by M. Todaro published at the end of the 1970s left very little room for illusions about the achievements of development thought. And this despite an exhaustive analysis of its theories and its policies that gave it all conceivable intellectual benefits, even to disciplinary respectability in the face of orthodox economics. Todaro admitted that, given certain distortions easy to maintain and prolong (artificially low agricultural prices for instance) one can have equilibrium with mass unemployment. Theoretically, underdevelopment could last forever (Todaro, 1977). In our opinion this was one of the first signs of the retreat of development thought to the laissez faire camp. The situation was dismaying. On the one hand, as we tried to show, classical and neoclassical ideas could hardly be applied to developing countries,  174 although they have proved some general regularities of economic development. On the other hand, almost 30 years of theorizing on Third World development (since the rise of the first Underdevelopment Theories by Frank, Baran, Amin, Geertz and others) and attempts to implement the results of these theories did not bring any significant results. This situation in the development thought was confirmed by a number of texts, some of which became nearly classics of the decade (refer, for example, to Stewart, 1977;Lipton, 1977; Schumacher, 1973). These texts did not take underdevelopment as one large problem requiring one overall solution. They just went into the question and dismantled from all possible angles the dogmas created by numerous development theories. One of these dogmas was the belief that development requires urbanization and the transformation of villages into farm factories. As we can put it, it was not just a question of restoring a balance between rural and urban growth, or of articulating better the two sectors, as some sources stated (Islam, 1974), but of actually taking things as they were and improving them, of respecting the predominantly rural order in most underdeveloped societies, protecting and stabilizing it through the conservation of skills, the alleviation of fiscal impositions, the stimulation of production and the rational adjustment of prices mainly through market forces. These ideas were taken over by many sources during the decade and initiated another crucial turn in  development thought, at least in the  economics-oriented part of it. The new research angle taken by the economic  175 development literature in the 1980s primarily involved heterogeneity of factor markets and questions of their internal structure and dynamic interrelationships, as well as issues of resource allocation, multisector models and (social) efficiency of growth in constrained and infinite time horizons. Although this can hardly be labelled as a return to classic ideas, there was a definite shift in research emphasis toward growth again.  47  Among labour-related theories, one should first of all mention studies conducted within the framework of Human Capital theory which emphasised the role of labour in the development process (Becker, 1975). Models studying issues of time allocation within a life span have been developed taking into account cultural peculiarities and characteristics of the pace of life in different countries (Linder, 1970; Kerr, 1972). New studies in international economics opened up horizons for conventional economic analysis (like the backwardbending supply curve for labour, resulting from case studies in Third World countries). New discoveries were related to other behaviour, sometimes noneconomic or nonrational. This involved the incorporation of qualitative variables in conventional economic analysis, and explains what we mean when we say that in the 1980s economics returned to growth theories at the same time as it turned to development. One of the most significant achievements of this decade was the development of a theory of Social Dualism, which splits a society into two parts, each characterized by a set of values. Although an early expression of this view can be found in the Dual (Lewis-Fei-Ranis) and Stage  176 (Rostow) Theories, already mentioned, it was the end of 1970s - beginning of the 1980s when this idea was incorporated in traditional economic analysis, bringing developmental questions to the fore. Two questions related to the role labour issues play in development and relevant to the present thesis, specifically unemployment and the employmentoutput relationship, were in focus during the 1980s. The idea that reallocation of resources can be a major source of employment dynamics does not belong to the 1980s. Back in 1817 David Ricardo wrote: A great manufacturing country is peculiarly exposed to temporary reverses and contingencies, produced by the removal of capital from one employment to another... [Considerable distress, and no doubt some loss, will be experienced by those who are engaged in the manufacture of such [adversly affected] commodities; and it will be felt not only at the time of change, but through the whole interval during which they are removing their capitals, and the labour which they command, from one employment to another. 48  In the 1960s this idea re-surfaced in Friedman's famous presidential address in the context of the "natural rate" of unemployment. According to Friedman, the natural level is ...The level that would be ground out by the Walrasian system of general equilibrium equations, provided there is embedded in them the actual structural characteristics of the labour and commodities markets, including market imperfection, stochastic variability in demands and supplies, the cost of gathering information about vacancies and labour availabilities, the costs of mobility and so on (Friedman, 1968). Different theories have emerged since that 1968 attempt to formalize the natural rate of unemployment and understand the implications of the idea of  177 sectoral reallocation. Broadly speaking, these theories can be separated into three main groups: sectoral reallocation, where workers move (with some cost) across sectors due to changes in the wages in different industries (Lucas & Prescott, 1974); mismatch, where individual workers and firms separate after observing they are not matching well (Pissarides, 1990); and search, where workers and firms engage in costly search activities once separated (Burda, 1993). These are best seen as complementary rather than competing theories; many sources combine various elements of all three. The last ten years have been marked with a lively debate about the relative importance of reallocation shocks and aggregate shocks as causes of changes in the unemployment rate. On the one side of the debate, it is argued that sectoral level shocks are the most important determinant of unemployment (Pissarides, 1978). On the other, it is argued that aggregate shocks may be the driving force, with different firms responding differently to the same shock (Shapiro & Stiglitz, 1984). In conclusion and in relation to our findings submitted in the earlier chapters, we would note that there are two obvious interpretations of the pretransition East European data. The first is that East European economies found their own peculiar way toward economic development that did not require the same shifts in employment structure as in the Western countries. The second is that there is a universal relationship between employment structure and output and their relatively high level of development with a given economic  178 structure either could not have happened or was a result of "forced" development. Although the analysis of the economic development thoughts helps to justify some views, it also suggests insights preventing us from taking sides. It is our belief, however, that taking sides is not and should not be the result of research. Since those sides themselves are a relative proposition, it would seem then that too much can be made of this issue.  179  PART V. EAST EUROPEAN ECONOMIC AND LABOUR MARKETS.  TRANSITION  CHAPTER 1. MAIN CHARACTERISTICS OF THE PERFORMANCE OF LABOUR MARKETS UNDER TRANSITION. What makes the East European case peculiar is the functioning of CPEmodel labour markets under entirely different macroeconomic  conditions  resulting from reforms currently under way. These conditions  include an  unprecedented  scale of restructuring (including property privatization), an  increasing inflationary pressure, the overall contraction of industrial production and the collapse of output, and a severe foreign trade deficit. Some of these changes were the focus of attention in Chapter 3, Part I, some of them are summarized in Table 29. Table 29. Some Macroeconomic Indicators for Eastern Europe in 1992. Country, indicator  1991 % Change in RGNP IP E  1992 UCh in  thousands  % of labor force  Bulgaria  -17.0  -27.8  -14.5  +210  11.8  Czechoslovakia  -16.0  -21.2  -7.4  + 154  5.6  Hungary  -10  -19.0  -6.0  +207  10.1  Poland  -9  -11.9  -5.5  +693  12.3  Romania  -14  -18.7  -11.6  +563  6.2  Note: RGNP - real GNP; IP - industrial production, E - employment, UCh change in unemployment. Source: Burda, 1993; OECD Employment Outlook, 1991; 1992.  180 The time lag resulting from  low adjustability of labour markets to new  conditions at both micro and macro levels, as well as specific features of transitioning economies and the infancy of market-based clearance forces result in frequent market failures and a necessity for government intervention. Before we proceed further with the scope for government  policies towards  employment, some conclusions should be drawn concerning the performance of labour markets under transition both theoretically and empirically. The present chapter attempts such conclusions, addressing the mediating role of labour markets and unemployment in economies in transition. In order to offer policy directions, one must identify the mechanism through which the policy might work and empirically examine its strength.  181 §5.1.1. Recent Labour Market Developments Introduction.  in Eastern Europe: an  As was mentioned in Part I, in 1989-93East European economies were affected by a potent combination of economic and political factors. The influence of  some of them was hardly avoidable, and some were clearly  provoked by either underestimations of serious problems or by excess optimism with regard to the goals of transition. In addition, as noted elsewhere in the thesis, until approximately 1992 the situation was much less scrutinized than it is now and goals of transitional policy were either vaguely formulated or completely absent. With  labour market  measures  being  a "forlorn"  periphery of  macroeconomic policies in the majority of East European countries, the period of active economic transformation has recently brought some new developments to this sphere. These new developments owe their existence to the combination of two main factors of a general character. First is the noted underdevelopment of labour market policies in Eastern Europe. The second factor is the pretransitional heritage of East European labour markets. This heritage, combined with new macroeconomic conditions, determines the uniqueness of the East European labour situation. Before we proceed with a detailed analysis of the pre-transitional heritage of East European labour markets, it seems appropriate to outline the main recent developments brought about by macroeconomic liberalization. Since the late 1980s, structural changes in the East European economies  182 have taken several forms, each of which had accelerated by 1991-92.The state sector, although still the dominant source of production, income and employment, had shrunk, while the private economy had been growing. Tables 30 and 31 provide estimates of the evolution of state and private employment between 1990 and 1992. In all East European countries except Russia the employment decline since the beginning of transition exceeded 20%. But the employment decline had lagged behind the output decline. Since the reasons for that are analyzed elsewhere in the thesis, we shall note briefly here that at the start of the reforms, firms commonly reaped windfall profits and this occurrence, coupled with short horizons, led firms to react to the sharp initial drop in output with only gradual reductions in employment. Bulgaria stands as one exception, perhaps because of the size of the output shock and the inability of firms to protect employment. Russia represents the most extreme case: some sectors indeed reported increases in employment in 1992,probably reflecting an elasticity close to unity of subsidies with respect to the firm's employment. Once transition had begun, productivity declined in all the countries. It is now stable or going up. Productivity. in Poland stands at 85 % of its value at the start of transition, up from 70% at the trough. In former Czechoslovakia it is 100%, up from 85%. In Hungary it stands at 110%, up from 85%. The outliers are Bulgaria, where the employment decline has followed closely the decline in output, and Russia where productivity is still declining. Privatization initially occurred more as a result of internal restructuring of industrial enterprises and  183  CNOO  ODCN CON  >e  vocn  in-*  6  OO  r-cn  '-'ON  O N O  S2i  r-cN  TJ-NO  ON,  rtON  NOcn NOcn  O N oo  CN  _ O N  ON  O '  ON^H  00>-H  OOj-H  2 ON  Oq mr-~!  SOrt  NOcn  ON^H  <NoO  o©  i—i O N  ooi_;  r^cN  ^ O N  r^cn i-Hod  O N O  OO-H  OO^H  O  ON  ON  a> J3  too ON ON  on  t-l  r-q  o o  in"*  ON  OO^H OO^H  CD  C  O o 6 o PU >> X) on CD  'C -t-> c o U  CN ON ON  cN^t-r^ "^•CN-H  -a a CM  3tocn  on  cnoo^t ^•CN—i  ONOOO r^^cN cninoo inoovo  o  ^cn^  ONCNr^  vogJcN  OOo  OcnoN NOONr-~ -;r-rn  oo oo  —IONCN OOOOON CNONCN CN-H  cncno --ivo Tien CN  Sot  O O o O O o  O00CN  cnvOrC cNcnoo  CNONCN  Ooo^  OOo  NOO^O  iricNcn mm  0 0 5  oo  oocnto r^NO^H  C2 cM CD  & =3  W on cM  PQ  o ON ON  ON ON  voo£  NO  cnON^'  OO  CNCN  oo  o oo o  ON ON  0  O O o  •^tONfQ  t^-NO  e OJ  6 >> o  t  o  o  2 <u  d cn  2 <L>  J D  on  ^  5  £>g u o  1^22  Jf i t  O  2s  on  _  2 <u  2 <u  on W  SnB  2 CD w dl  M  2^  on O H  on O H  22 ^  on O H  ^  on  on  ^  OH  on O H  Sf2— &2  on O-  >NH C  2 <u  E <D <D  3_22  i i 2*0-0 ^  to >>>>  ^ 2 2  OHOH N o 66 o o E E UHPQPQ ftnHWPQ «-g  B*0 T3 CD CD E CDrt-k  eS g CD ai - 2 n 3 T 3 CD > N C , >>>N « PQ n * 2 2 22  > I5 IO « C  C C H P£ HW 6  • ooE E rtHPQW  g <U (D  .  _ (11 2  0  0  lilt PiHP-iPLi  184  NO  CM  o  PS m  O  co ON  in  d  oo  o  co d  Tt  co  ID  .S >>o  o P ITtvo  CM S £P > >> fl .5 O . cM — U 3 PH P u  U  e  q co  q  co  NO  CN  NO_ ON  CM  oo  NO  Tt-  oo co  6  B >>  co d CN  • cM  co  Tt  CO  q  CN CN  NO  loo CN  ON  NO  r-'  G  ON ON  CD  o  't-< NO  CO  CN*  CN  in  CN'  ON  |ON  CO CN  in  o  u c  E m  CM  E  '6  o  U  "o  •S3 N  U  a  Pi  cM  cM  "3  "S  _cM GO  B  GO  03  CM  P5  X5  B cM  2i> o i—  =3  o  185 the emergence of small-scale enterprises, co-operatives or joint ventures than by explicit privatization of state enterprises through direct legislation (Fischer and Standing, 1993). The latter form accelerated in 1991, as evidenced by legislation in Romania, Czechoslovakia and elsewhere. In 1993-94 Russia definitely outpaced the rest of East European countries in terms of privatization and partly even in terms of development of privatization-related legislation.  49  However, the shift from state to private sector production and employment during this period came primarily from new sources of jobs outside the economically stagnant or sharply declining state sectors. The employment cuts in the latter sphere  greatly outweighed the absolute growth of jobs in the  former. The evidence allows us to draw a more detailed picture, one in which in all countries becoming unemployed during 1989-93 originally was voluntary and has become increasingly involuntary. The approach of firms was initially to use shorter working hours and involuntary holidays, as well as to use those government programmes which allowed for transfers to workers who left. In 50  Poland in 1990 at least 70% of the employment decline was through early retirements. Although elsewhere the numbers were smaller, as in Hungary with 20% of the decline between 1990 and 1992 conducted in this way, the response was quite common. Further, involuntary separations were concentrated on ancillary workers, including administrative and clerical staff. This fact may partly explain the high initial proportion of women in total unemployment. It is revealing that only in Hungary, where more significant structural change  186 occurred by 1992, does male unemployment dominate female unemployment (Blanchard, Commander & Coricelli, 1994). Privatization has also taken the form of conversion of state enterprises into joint ventures with foreign firms. It has also been seen in the growth of small-scale firms, mostly in the retail trade sector. One of the biggest questions in 1992 was: "What should be done with the large-scale enterprises that dominated the industrial landscape of Eastern Europe over the last decades?" Clearly, alternative approaches have been pursued in different countries of the region. For example, in Hungary many of these enterprises have been transferred into joint ventures with foreign capital. In Czechoslovakia, Poland and Romania ownership rights are to a greater extent being distributed to the population through Employee Share Ownership Plans. In Russia privatization has taken a form intermediate between the two: at the first stage ownership rights have been transferred to the population through a voucher scheme, and at the second stage enterprises are allowed to be sold at auctions. This undoubtedly will speed up the formation of the stock market in the country. Without making any further judgements on the respective merits of the various forms of privatization, which is well beyond the scope of the thesis, it is important to consider whether the choice will have different labour market implications, or require distinctive labour policies. There are other aspects of the economic restructuring shaping labour market developments.  We can  predict a mass outflow  of labour from  187 agriculture lasting for some years. Agriculture was a labour reservoir for decades, absorbing much larger proportions of the labour force than was typical in industrialized Western economies (refer to Part III for details). There has also been a start made in the considerable shift needed from heavy industries "material production" -to consumer goods and services. There has also been the beginning of a move within manufacturing away from the "military industrial complex" towards high-tech light industry. The emergence of open unemployment is one of the major recent developments in East European labour markets. In 1991-93 dis-employment became the dominant factor of unemployment growth. Without going any deeper into questions of the origins and scale of East European unemployment here (see Part III for a detailed analysis), we should note that with respect to Russia Alexander Shokhin, the Russian Minister of Labour appointed in August 1991, forecast in November 1991 that in the ex-USSR about 30 min people would lose their jobs in the next two or three years and that up to 15 min people would become long-term unemployed. This forecast indicates the unemployment levels to be expected in Eastern Europe.  51  188 §5.1.2. Transitional Macroeconomics and East European Labour Markets: General Implications.  The majority of sources agree that unemployment, which is one of the major concerns in present-day Eastern Europe, is not merely a by-product of transformation, but is essential to change (e.g. Burda, 1993; Paunov, 1993; Charap and Dyba, 1991; Lindbeck, 1992; Aghion and Howitt, 1991, etc.).  52  Usually authors offer several points to support the view. First, emerging unemployment will offset a growing imbalance in the bargaining power of workers over managers in the aftermath of central planning. Unemployment provides the "disciplining device" (Shapiro and Stiglitz, 1984; Burda, 1993) to raise productivity. Second, unemployment growth  may be necessary to check the  of real wages (Blanchard, 1991; Burda,  1993). Indeed, open  unemployment may be the only effective means to this end in the absence of collective bargaining. A third reason is offered by the new macroeconomics literature on the matching or flow approach to labour markets, in which job creation is seen as a stable function of the stock of unemployed workers and open vacancies (Pissarides, 1990; Blanchard and Diamond, 1992; Burda and Wyplosh, 1990;Burda, 1993).Unemployment  is needed to allow the emergence  of a new, more highly productive private sector and to force the acquisition of new skills by workers since it is a higher rate of productivity growth (as opposed to the level of productivity) which implies a higher present value to a labour-market match (Ibid.). More restructuring requires new job matches, 53  189 reducing unemployment by raising the job-finding rate of unemployed workers. This in turn requires more unemployment.  At the same time, more  unemployment at a given level of vacancies makes them "too efficient", causes labour market congestion, and leads to long-term unemployment (Ibid). To the extent the firing decision is in public hands, the state can influence this process greatly. Postponing layoffs reduces congestion, but at the cost of  less job  creation in the higher-productivity private sector. Is there an optimal speed at which to release workers into unemployment? Although the point that "unemployment is needed" may seem dubious, we will try to demonstrate that it holds water. Since the flow approach appears to be one of the crucial chains in our reasoning, it seems appropriate to devote some space to citing its main theoretical considerations and applying them to the current East European situation. In our opinion, the flow approach summarizes labour market problems induced by the current government macroeconomic policy. The description of the flow approach and its application to East European labour market problems (principally unemployment) would therefore serve as a guiding logic for this chapter. Before the flow approach emerged, issues of unemployment had been focused upon by the historical literature which predominantly studied the effect of technical change on the marginal product of labour, and was principally concerned  with  the  tradeoffs  between the  level of productivity  and  unemployment. The flow approach, associated with C. Pissarides (1990), is  190 highly innovative since unlike the traditional Keynesian theory it can explain long-run fluctuations in unemployment, and unlike the Real Business Cycle theory it contains the frictional elements required for unemployment to be compatible with equilibrium (Aghion and Howitt, 1991). However, in our opinion, this approach, when applied to Eastern Europe, does not take into account some of the important reasons proposed in the historical literature for why mismatch can occur or match can be delayed in time, for example, skill obsolescence and type of training (Burda, 1993).  Below we offer some  theoretical insights into this problem and then turn to empirical evidence. The dynamic view of labour markets in general form stresses the role of unemployment in job creation: with higher unemployment or more vacancies, more job matches occur. Since most flows into unemployment result from involuntary separation, for new jobs to be created others must have been destroyed (Burda, 1993). In 1992 the percentage of school-leavers and other new entrants in the labour market was 18% in Czechoslovakia, 4% in Hungary, 15% in Poland, 12% in Romania, 16% in Russia and 10% in Bulgaria. According to Pissarides (1990), workers find searches less costly when unemployed. ...both unemployment and vacancies matter in hiring. The rate of hiring appears to be determined by both sides of the labour market, not only by the demand side, as is often assumed in macroeconomic models (Blanchard and Diamond, 1989).  The term "matching function" describes the process by which jobs and workers  191 come into contact to form employment relationships. The number of matches (hires) x depends both on the unemployment rate u and the vacancy rate v. The matching function: x=f(u,v); f(u;0), f(0;v)=0;f , f > 0 v  u  (A)  captures spatial aspects of the search process, imperfect information, and all types of occupational and industrial mismatch between jobs and workers (Pissarides, 1990; Burda, 1993; Aghion and Howitt, 1991). It also subsumes potentially endogenous search activities of workers and firms. In conventional analysis, x grows with increases in both u and v, with a diminishing marginal product for each, capturing the notion of congestion in labour markets. More controversially, it is often assumed to have constant returns to scale: doubling unemployment and vacancies doubles the flow of job finds. Some available evidence supports the assumption of constant returns in matching for industrial countries, but some points to increasing returns (Howitt, 1985). Under constant returns to scale the vacancy - unemployment  ratio v/u is a sufficient statistic  for the tightness of the labour market, which determines both the hiring rate h=x/v and the job finding rate f=x/u. It is obvious from here how the matching function is subject to changes in government policies: changes in unemployment benefits, income and indirect taxes can all affect its properties by influencing the activity of job searches.  54  It is apparent from the form of the matching function, and it has already been mentioned,  that  it recognizes  that  the job-matching process is  192 uncoordinated, time-consuming and costly for both workers and firms. Under constant returns to scale the functional form generally used for the empirical estimation of the matching function is the Cobb-Douglas: X =A U*y-\ t  (B)  t  where A denotes "technological progress" in job matching, and capital letters are used to denote discrete time data (Boeri, 1994). Since the data on hiring is very limited in Eastern Europe, new hires could be proxied by outflows from unemployment to employment. By taking the logarithms of equation (B) and substituting outflows to jobs for new hires, the regression model is obtained: ln(0„)=ln(A,)+ ln(C7,)+61n( V,,)+u a  it  (C)  Empirical estimates are submitted in Table 32 against monthly data on different regions (denoted by the subscripts 0 in each country. As one can see from the table, the dependent variable is monthly outflows to jobs, and the explanatory variables are the stock of unemployment and vacancies. The coefficients of both explanatory variables are positive and statistically significant (except for the case of vacancies in the Slovak Republic) and vary insignificantly over different estimation procedures. Unlike in West European countries, USA and Canada, the estimated matching functions demonstrate decreasing returns to scale: the coefficient for the stock of vacancies is small when compared to the coefficient for the stock of unemployed (Pissarides, 1986; Blanchard and Diamond, 1990). In other words, changes in the number of unfilled vacancies in Eastern Europe seem to have little influence  193 Table 32. The Matching Function in Eastern Europe, selected countries, various years.  Estimated equation: log (Oit)=log(Ait)+b log (Uit-l)+c log(Vit-l)+uit Country, indicators  Variables U V  Poland Jan 92 - Mar 93 49 regions 686 observations  0.60 0.24 Yes" (9.28) (7.34) 0.65 0.14 Yes (9.16) (7.00)  Pooled regression OLS Random effects GLS  Hungary Apr 91 - Apr 93 20 regions 480 observations  0.73 0.07 Yes (16.71) (2.00) 0.93 0.06 Yes (15.67)(1.40)  Pooled regression OLS Random effects GLS  Czech Republic Jan 91 - May 93 8 regions 232 observations  0.75 0.20 Yes (33.93)(4.91) 0.77 0.15 Yes (18.66)(1.95)  Pooled regression OLS Random effects GLS  Slovak Republic Dec 90 - Jun 93 4 regions 124 observations  0.77 -0.02 Yes (16.46)(0.21) 0.82 -0.03 Yes (10.20) (0.46)  Pooled regression OLS Random effects GLS  TD  d  d  d  Estimation method  Test statistics R CRS (b) H (c)  0.72 -5.48* 0.44 -2.71*16.42  0.84 -2.98* 0.85 -2.10**18.5*  0.96 -7.37* 0.94 -2.92* 2.26  0.92 -0.31 0.90 -0.46  4.19  Notes: Robust (to heteroskedasticity and serial correlation) t-statistics in parentheses. Ait - constanta of technological progress. Oit - outflows to jobs in region i at time t. Uit - unemployment stock. Vit - vacancy stock. T-test of the hypothesis of constant returns to scale (b+c=l).One asterisk denotes significance at 1% level, two asterisks - at 5%. Hausman test of fixed effects against random effects. Asterisks denote significance at 5%. d - for dummy coefficients please refer to Boeri, 1994. Source: Boeri, 1994.  194 on outflows from unemployment to jobs.If both the number of unemployed and the number of vacancies double, it would result in  less than proportional  outflows to jobs. The question is: why does this happen? As some authors mention, one of the reasons is limited coverage of vacancy data (Boeri, 1994; Burda, 1993). But problems with vacancy data are also common in OECD countries, where, generally, vacancies reported to labour offices are a minor share of the total number of job openings. Therefore, measured job finds should be responsive especially to measured vacancies. A more convincing explanation seems to be that most job openings are filled by those who are searching for jobs while still employed. This seems to be likely since most vacancies are indeed filled by private employers who recruit workers directly from the state sector. One more explanation for the weak effect of vacancies on job finds can lie in the mismatch between the regional distribution of job-seekers and that of the unemployed. This hypothesis is subject to testing in part 5.1.3. The  problem  of  time  lags is  another  important  theoretical  macroeconomic consideration relevant to labour market transitions. Wages react slowly to labour market conditions. Wages also react to prices and prices to costs with lags (Blanchard and Muet, 1993). There are further lags in the response of exports and imports to changes in the terms of trade, in the response of output to changes in demand and, perhaps, in the response of unemployment to output. Two points following from Table 33, both of a  195 diagnostic character, are worthy of special attention. First, there is no evidence that employment in Hungary and Poland has become significantly more responsive to output changes than in the past. Similarly, there is no evidence of significant changes in adjustment speeds. Second, there is a large gap between OECD countries and East European countries (represented here by Hungary and Poland) in terms of short-run and long-run employmentTable 33. Estimates of Employment Responsiveness to Changes in Output in 1980-1985and 1986-1991,various countries. Short-run 1980-85 1986-91  Long-run 1980-85 1986-91  Mean lag 1980-85 1986-91  PL  0.04 (1.4)  0.04 (1.7)  0.17  0.18  3.35  3.55  HG  0.02 (0.8)  0.05 (1.9)  0.11  0.26  4.26  4.26  AG  0.03  0.05  0.14  0.22  3.76  3.88  SW  0.18 (2.4)  0.27 (4.6)  0.75  1.04  3.17  2.85  SP  0.32 (2.1)  0.29 (3.4)  0.70  2.07  1.17  6.14  AG  0.25  0.28  0.72  1.55  1.86  4.00  Note: Poland and Hungary are the only two East European countries for which quarterly series on industrial production and employment were available over a sufficiently long time-span, "t-statistics in parentheses. PL - Poland; HG - Hungary; A G - average; SW - Sweden; SP - Spain. Source: Boeri and Keese, 1992; OECD Main Economic Indicators, various years.  196 output elasticities, and this gap has widened in the course of the 1980s. If in Spain and Sweden (countries from two "poles" of the OECD) a unit fall in output leads to a decline in employment of about 0.3% within a quarter, the corresponding fall in Hungary and Poland was only 0.05%. The issues of time lags has not only medium- and long-run implications (the ones following from Table 33), but also is related to short-term issues of unemployment. When government intervenes to subsidize unviable enterprises in order to keep unemployment low, a quite popular measure in all East European countries in the 1990s, it artificially extends the time lag in the employment-to-output adjustment process. Evidence from Russia in 1993-1994 shows that in those regions where the rate of increase in government subsidies (so-called loans-generated inflation) outpaced the rate of increase in the consumer price index, the rate of decrease in absolute employment was very low (Tyumen, Lipetsk, Tambov, Kaliningrad, Volgograd, etc. oblasts,total - 50 subjects of the Russian Federation). In the areas, where the situation with subsidies was opposite (Moscow, Sankt-Petersburg, Komi Republic, Vladimir, Ivanovo, Perm, Penza, etc. oblasts) the decrease in absolute employment was much more noticeable. Relatively low employment-output  elasticities even after the start of  transformation have many other explanations in the literature (Boeri and Keese, 1992; Flanagan, 1992; Paunov, 1993; Blanchard and Muet, 1993), and some of them were submitted in the end of Part III. Other have to do with the political  197 economy of the liberalization process, namely the consensus-seeking patterns followed by governments in a context where employment is still dominated by state sector jobs and the right to a job was previously guaranteed. Other explanations such as the control exerted by workers on management, labour hoarding and the absence of mechanisms enforcing the exit of firms, refer to the microeconomic legacies of the past system. As was shown elsewhere in this thesis, employment has still to adjust to changes in output. The inevitability of adjustment is not in doubt. We tried to demonstrate that it is merely a matter of time. In terms of our discussion, therefore, two distinctions turn out to be conceptually important and empirically relevant. The first has to do with the initial employment structure and the degree of inflation at the start of transition. As showed above, the degree of inflation has a great significance in determining the rate of employment-to-output adjustment and the speed of releasing workers to unemployment. Figure 13 is illustrative while generalizing on the matter. It graphs the situation when competitiveness, measured by a set of economic indicators (see Blanchard and Muet, 1993 for details) is low. This is true in relation to the majority of East European countries given the collapse of COMECON and the small share of exports which went to the West. Initially low (point B in Figure 13), then improving, but still somewhat insufficient competitiveness may initially not be sufficient to prevent a further increase in unemployment. This process is approximated  by  the  BCE schedule,  showing  steadily  improving  198  Figure 13. Logic of Adjustment to Equilibrium Employment under High Unemployment, Low Inflation and Fixed Nominal Exchange Rate.  The SS schedule describes the supply side. It says that higher unemployment leads to an improvement in competitiveness. As unemployment rises real wages tend to decline, which lowers the domestic cost of production and therefore the prices of domestic goods. The negative relationship between competitiveness and unemployment along the DD schedule corresponds to the demand side. A gain in competitiveness raises world demand for domestic goods and therefore lowers unemployment. If competitiveness is too low to start with, such as at point B, improving but still insufficient competitiveness may initially not be sufficient to prevent a further increase in unemployment and thus will result in a BCE path.  Source: Blanchard & Muet, 1993  199  Figure 14. Logic of Adjustment to Equilibrium Employment under Positive Inflation Differential, Low Competitiveness and High Unemployment.  — — (j Unemployment  If the economy starts transition from high inflation, pressure from unemployment will decrease inflation. But for some time inflation will remain higher than the foreign inflation, so that competitiveness will further deteriorate. Eventually the inflation will come down to the level of foreign inflation or lower, competitiveness will improve and the process will take the economy back to full employment (path ABCE).  Source: Blanchard & Muet, 1993  200 competitiveness but a transitory increase in unemployment. The development of Hungary and Czechoslovakia over the period 1989-1993 confirms the inferences above. These countries started with relatively high unemployment which was associated with low inflation. By fixing the nominal exchange rates the countries managed to improve competitiveness from the start. The AE schedule shows no transitory increase in unemployment when a country starts from a higher level of competitiveness. The other way around, if an economy starts transition with relatively high inflation (the situation of Russia, the Baltic states, Poland and Bulgaria), pressure from unemployment will also reduce inflation. But for some time domestic  inflation will remain higher than foreign inflation, so that  competitiveness will further deteriorate. When inflation becomes less than foreign inflation, competitiveness will improve and unemployment stabilize. Such a process of adjustment is depicted by Figure 14, by the ABCE path. For a certain time, namely from A to B, inflation declines, but both competitiveness and unemployment still worsen. From B on, competitiveness increases; from C on unemployment starts to decrease, theoretically returning the economy to full employment. How long it will take for the economy to eliminate the initial inflation differential and to start the period of decreasing unemployment depends very much on the nature of the lags and the character of government interventions. The important distinction here is between nominal rigidities  - lags in the  201 adjustment of nominal wages to prices, and of prices to wages - versus real rigidities, roughly all other lags, such as lags in the response of wages and prices to demand conditions or in the response of exports and imports to real exchange rates. Supporting the above discussion by some evidence, it would be interesting to mention that Blanchard and Muet (1993) show that the real wage, adjusted for productivity, adjusts slowly to unemployment. According to Western experience, the speed of adjustment averages around 10-12% of change in unemployment per year. The second interesting point is that the mean lag of wage inflation on price inflation is relatively short; the adjustment is usually complete within a year. Now it is clear why - aside from weaving a social security net - the other main government intervention for tackling the unemployment problem in all East European countries has been to keep real wages down. At the beginning the employment objective was not pursued at all, the main aim being the control of inflation as well as real earning in the state sector. Although the relevant data is not available for all countries, information on Poland serves as a good example of the intervention. In 1990 the average real wage was down 24.4%from 1989, in 1991-0.8% up from 1990, in 1992-down 2.8% in relation to 1991 (Chilosi, 1993). Later, specifically after 1991-92,when it became impossible to neglect the employment objective, some governments, especially those in Russia, Bulagaria and Romania, started using inflationary measures to keep unemployment down. Theoretically still trying to pursue the  202 objective to restrict price- and wage-generated inflation, the governments continued subsidising state enterprises heavily. These subsidies were increasing steadily because more enterprises needed them. This resulted not only in the workers receiving money they did not earn (industrial output had been falling constantly  since the end of the 1980s), but in provoking wage-generated  inflation too. Government policies therefore were keeping unemployment somewhat low, disturbing natural time lags in the process of employment-tooutput adjustment. One of the most important East European rigidities, relevant both to our theoretical viewpoints and to empirical evidence on the lack of adaptation of employment to the fall of production is the rate of expansion of the private sector. There are opposing opinions on this process. Although there are no direct data available, some indirect estimations indicate that the private sector has been expanding at a higher rate than the state sector has been contracting as shown by a variety of measures. Still, owing to the relatively small size of the private sector in all East European countries, the growth of its employment has been smaller than the number of those rendered unemployed by the contraction of the state sector. However, if the momentum of the growth of the private sector keeps its pace, its contribution to job creation can eventually compensate for the shedding of labour in the state sector. This already has taken place, for example, in Poland, where in some months in 1992-93 the increase in private employment exactly matched the decrease of employment in the state sector  203 (Rzeczpospolita, October 30, 1992). However, unemployment  continued to  increase. Figures of unemployment in all countries in question are still markedly lower than the combined sum of open and hidden unemployed. The reserves of excess employment accumulated by the state sector could present a grievous social problem if allowed to be transformed  too  quickly into open  unemployment, following privatization. However, as some authors point out (Chilosi, 1993; Hare and Revesz, 1992)  the growth of the private sector can add to the problem of  unemployment because of the change of management objectives in privatized enterprises. According to Chilosi, in the Polish private sector in 1992 average wages were more than 10% lower than in the state sector. By keeping up its employment, especially in the areas hardest hit by the unemployment problem (like Koszalin region in Poland with a 23.7% unemployment  rate, or  Yaroslavskaya and Vladimirskaya oblasts in Russia with 12%),even at the cost of increasing hidden unemployment, the state sector is buying time, allowing the booming private sector the time to absorb the labour force. This would undoubtedly imply higher loan emissions and open inflation, which would bring countries back to the situation formally depicted by Figure 14.  If  controlling inflation is still among the priorities for the restrictive economic policies of the government, this would imply higher unemployment as described by Figure 13 above. As was mentioned, the situation is different in Russia, Bulgaria and  204 Romania.  The fact that the level of unemployment  there did not meet  expectations can be explained by the moderate pace of privatization up to 1994 and much less rigid monetary policy. The government keeps subsidizing unprofitable enterprises, thus following the scenario of more gradual release of workers into unemployment.  55  We tried to demonstrate how current general macroeconomic policies in different East European countries influence labour market policies of the governments and how these, in turn, fit in the flow approach. Reasoning along these lines seems important in determining medium- and long-run priorities of employment policies.  205 §5.1.3. Empirical Illustration of Regional Dimensions. In very concise form the conclusions from the foregoing might be put as follows. Transformation has been very difficult so far and there are as many causes of those difficulties. Table 29 shows that GNP fell in 1992 in the majority of countries. It continued to do so in 1993 and probably 1994, for which there are no data available yet. Declines in industrial output were dramatic but independent of the pace of reform. The collapse had cyclical causes (see Chapter 3 Part I for details) and at the same time its depth was intensified by the end of COMECON and the shift of economic activity from a lattice of rigid trading relationships of central planning to a market environment of imperfect information, placing more value on the search for and brokering of business contacts. Conventional supply and demand analysis fails to capture this key aspect of economic transformation. This also applies to the labour market. Firms slashed output and, after some delay, began to shed labour. Despite considerable interfirm mobility under communism, labour markets were not ready for unemployment and became very congested. A crude indicator is the ratio of registered vacancies to unemployed shown in Table 34. Despite the common view that "there is no job creation in Eastern Europe", labour markets exhibit both job seekers and openings to an extent comparable with the West, and reported vacancies are only a fraction of available job openings. As we have described in detail, unemployment has structural aspects that  206 cannot and must not be ruled out. First, the recent sharp rise in unemployment Table 34. Officially Registered Vacancies and Unemployment in 1992. Unemployment (u), thousands  v/u, %  Country, indicator  Vacancies (v), thousands  CzR  85.0  142.0  SIR  13.2  282.3  4.7  PL  31.5  2,228.0  1.4  HG  25.0  546.7  4.6  BL  12.0  453.0  2.6  RN  3.3  675.0  0.5  UK  128.0  2,678.0  4.8  FR  59.0  2,753.0  2.1  WG  355.5  1,716.0  20.7  EG  30.2  1,149.1  2.6  59.4  Note: CzR - Czech Republic; SIR - Slovak Republic; PL - Poland, HG Hungary, BL - Bulgaria, RN - Romania; UK - United Kingdom, FR - France; WG - West Germany; EG - East Germany. Source: Burda, 1993; OECD Main Economic Indicators, 1992. is concentrated in sectors I and II of the economy. Second, the reorientation and expansion of international trade is currently under way, as well as price level adjustment. Third, the practice of labour hoarding is still alive from the shortage economy under central planning. Except in Hungary, Poland and ex-Yugoslavia,a new private sector must be built up from scratch. Siebert (1991)stresses this aspect of the transformation  207 Table 35. The Turnover of the Unemployment Pool in East European countries, 1992, percentage of the population of origin. Monthly inflows (a)  Monthly outflows (b)  Monthly outflows to jobs (c)  Unempl. rates (d)  Bulgaria  1.2  5.8  1.2  16.0  Czech Republic  0.5  24.8  17.4  3.2  Hungary  0.5  5.2  2.2  11.6  Poland  0.6  4.4  2.2  13.3  Romania  0.6  1.3  0.9  5.5  Slovak Republic  0.8  10.4  5.1  12.1  Former GDR  1.5  10.8  3.9  15.6  FRG  0.9  16.3  6.9  5.8  Austria  1.0  25.8  16.3  3.7  France (1991)  0.9  11.4  4.8  9.5  Norway (1991)  1.3  35.8  16.6  5.5  UK (1991)  1.0  13.5  ...  8.1  Notes: a) Average monthly inflows into unemployment divided by the workingage population. b) Average monthly outflows from unemployment divided by the average yearly number of unemployed. c) Average monthly number of vacancies filled via the mediation of the public employment service divided by the average yearly number of unemployed. d) Number of registered unemployed at labour offices as a percentage of the labour force. Source: Boeri, 1994.  208 Table 36. Characteristics of Long-term unemployment" in Eastern Europe, 1992. UIR  UR  (c)  su  SLTU  Bulgaria Women Youth (f) Unskilled Primary/low education  1.68 1.74 1.75  16.00 13.49 16.74  53.36 45.12  50.95 31.97  Czech Republic Women Youth (f) Unskilled Primary/low education  0.65  3.10 3.60 5.70  57.81 31.83  59.83  Slovak Republic Women Youth (f) Unskilled Primary/low education  1.04  Hungary Women Youth (f) Unskilled Primary/low education Poland Women Youth (f) Unskilled Primary/low education  LTUR  6.10 6.04 5.42  0.40  35.50 11.40 11.70 26.22  43.64 28.16 18.24  51.55 17.53 23.71  4.11 4.94 3.99  23.63  29.45  0.90 0.74 1.87 1.09 1.27  9.30 7.80 19.25 13.15 14.39  40.14 26.97 14.72  39.75 20.26 16.51  1.32 1.34 2.08 2.58  44.23  51.77  2.93  0.87 0.74 1.87 1.09 1.27  15.97 14.78 31.15 15.84  50.26 8.29 7.26  52.19 7.25 8.00  4.82 4.45 6.20 6.23  12.08  23.28  23.29  5.26  Notes: UIR - Unemployment Inflow Rate, (b) - inflows for each category as a percentage of a group-specific labour force. UR - Unemployment Rate, (c) - unemployment within each vulnerable group as a percentage of the group-specific labour force. SU - Share in Unemployment. SLTU - Share in Long-term Unemployment. LTUR - Long-term Unemployment Rate, (d) - long-term unemployed within each vulnerable group as a percentage of the group-specific labour force, (f) - The standard definitions of youth refers to those under the age of 25. Exceptions: under 30 in Bulgaria, between 15 and 19 in Poland, under 22 in the Slovak Republic, under 24 in Hungary. Source: Boeri, 1994.  209 for Eastern Germany, where the private sector, predominantly engaged in services and light manufacturing, is now. growing rapidly. One should note that a third of all employment in Hungary is the the second (i.e. entrepreneurial) economy (Burda, 1993). One of the most important features of unemployment in Eastern Europe is the low turnover of the unemployment pool (Table 35). In spite of large job losses in state enterprises, a relatively small fraction of the working-age population appears to enter the unemployment pool every month. It is not surprising that unemployment has built up rapidly throughout the region, partly owing to the small number of people leaving the unemployment pool every month (refer to Table 36 for details). In our opinion there are two crucial policy implications from the above. First, low outflows mean increasing duration of unemployment. Secondly, despite low inflows, unemployment may stabilize in some countries only at levels which are higher than those currently experienced, because further adjustment of employment to decline in output is expected. In other words, it is questionable whether measures aimed at reducing the flows from employment into  unemployment  unemployment.  can be  effective  in containing  the  growth  of  A shift of policy focus may be required that emphasizes  measures to foster outflows from unemployment. Table 37 provides some additional insights into the matter. In addition to all the above, it should be emphasised that East European  210 labour markets suffer significant geographic mismatch. For example, in Czechoslovakia in 1992 the registered unemployment rate was 5.8%, but in Prague it was only 0.4%. Assuming steady-state conditions, the average prospective duration of unemployment was 3.7 months. In Northern Moravia unemployment was 4.3%, the prospective duration of unemployment - 6.6 months. In Eastern Slovakia unemployment was 12.6% and prospective Table 37. Distribution of Inflows into Unemployment in Eastern Europe by Reasons for Separation, 1991,1992, selected countries, %.  Poland 1991  1992  Czech Republic 1991 1992  Slovak Rep. 1991 1992  Mutual Agreement  22.91  20.09  14.29  35.29  28.44  Quits  31.44  30.68 50.00  18.60  33.33  31.19  Individual layoffs  28.91  34.05 3.57  20.93  5.35  19.27  Group layoffs  9.64  6.18 14.29  3.88  4.63  4.59  Other  7.11  9.00 17.86  27.13  Total  100.00 100.00  Source: Boeri, 1994  29.46  100.00 100.00  21.39  16.61  100.00 100.00  211 Table 38. Regional Differences in Unemployment and Vacancies in Bulgaria and Czechoslovakia in 1992. Regions, indicators  Vacancies,, v, thousands  Unemployment «, tnousands  #'  v/  Bulgaria  1.3  50.2  2.8  Sofia  3.6  40.4  8.7  Burgas area  1.1  44.6  2.4  Varna area  1.0  40.9  2.5  Lovetch Mikhaijlovgrad  1.0 0.6  43.8 38.5  2.3 1.6  Plovdiv  1.2  89.4  1.3  Russe  0.5  39.7  1.4  Sofia area  1.6  60.2  2.7  Haskovo  1.4  55.0  2.5  Czechoslov.  6.5 9.8  46.1 21.1  54.4 101.3  17.8  3.7  485.0  C.Bohemia  8.8  16.9  52.3  W.Bohemia  8.2  9.7  84.7  S.Bohemia  4.9  10.3  47.9  N.Bohemia  9.8  21.2  46.3  E.Bohemia  9.7  19.2  50.7  N.Moravia  10.4  48.9  21.2  S.Moravia  8.6 3.2  39.2 71.0  22.1 7.4  Bratislava  3.8  19.2  19.6  W.Slovakia  2.3  94.2  2.5  C.Slovakia  2.8  83.8  3.3  E.Slovakia  3.7  86.7  4.2  RepubTic Prague  R^'puSfec  Note: N. - North; S. - South; E. - East; W. - West; C. - Central. Source: Burda, 1993  212 Table 39. Regional Unemployment Differentials and Regional Mismatch.  Country Bulgaria 1991 1992 1993 Czech Republic 1991 1992 1993 Slovak Republic 1992 1993  ARU, %  SD  CV  IM  2.6 14.0 16.5  0.7 2.6 3.1  0.2 0.2 0.2  0.3 1.5 1.7  2.1 2.7 2.4  0.5 1.1 1.1  0.2 0.4 0.5  0.3 0.6 0.7  10.6 11.0  2.4 3.9  0.2 0.4  0.7 1.0  4.0 10.0 14.1  1.9 2.4 4.0  0.5 0.2 0.3  0.5 0.9 1.2  6.5 13.5 15.6  2.5 3.9 4.5  0.4 0.3 0.3  0.8 1.8 1.9  9  8  4  Hungary 1991 1992 1993  20  Poland 1991 1992 1993  49  Notes: ft - number of regions; ARU - average rate of unemployment; SD standard deviation; CV - coefficient of variation; IM - index of mismatch. The index of mismatch (Boeri, 1994) is calculated as follows: IM=V2Z S [(U -V ) -(U-m where S denotes the labour force share of region r, f/ and V are, respectively, regional unemployment and vacancy rates, and U and V the the countries' average unemployment and vacancy rates. As all measures displayed in the table depend on the number of observations available, and regional breakdowns for some countries are finer than for the others, the submitted indexes are not comparable cross-sectionally, but simply over time. Source: Boeri, 1994 r  r  r  r  r  r  r  213 duration 22.1 months. Table 38 shows that mismatch also extends to the availability of jobs, and this is evident both where unemployment  is low  (Czechoslovakia) and high (Bulgaria). The within-country variance of the vacancy-unemployment ratio is as high as that between countries and it is thus unlikely that measurement error can account for this mismatch. We would emphasise therefore the key role of matching function in both general and regional analyses of present-day East European labour markets. Some comments are necessary with respect to the regional mismatch and to the spatial dimension of employment policy. Those comments follow directly from the empirical observations summarized in Tables 38 and 39. Under the conditions of transitioning countries, where there are housing shortages  and  other  obstacles  impending  horizontal  mobility  of  population,regional mismatch may be an important factor preventing increased demand for labour from having a significant unemployment  impact on flows from  to employment. Table 39 provides some measures of the  dispersion of unemployment rates across regions, and an aggregate index of mismatch in the regional distribution of unemployment and vacancies. The table suggests that the mismatch between unemployment and vacancies may have increased in all countries (last column). Conversely, there are no direct indications of an increasing dispersion in the distribution of unemployment across regions. There seem to be increasing discrepancies between the regional distribution of unemployment and vacancies, rather than increasing regional disparities in the incidence of unemployment. The most convincing explanation for this would be regional differences in the expansion of the private sector and  214 therefore differences in the growth of vacancies. There are various ways to deal with the unemployment issue. The easiest would be to leave the matter. to market forces and to shelter the resulting unemployed through the existing social security net. But there would be two main objections here. The first is the spatial variation in the intensity of unemployment  under the condition of restricted horizontal mobility of  population. The second is the general underdevelopment of the social security system, which is unlikely to improve in an era of budget deficits. For those reasons reliance on market-clearing forces would be a socially dangerous (especially in certain areas) and costly solution. The following conclusion seems to be fitting in the present chapter. Elsewhere we argued that unemployment  is essential to change in Eastern  Europe. Although we still do support this viewpoint, there can be obvious objections if one views unemployment as an effective device for inducing labour mobility  towards  more  productive  employment  assuming  steady-state  conditions. These objections are related to regional aspects of the East European labour market situation. The areas with the lowest v/u ratio (i.e. where increasing outflows from labour force to unemployment are not followed by the adequate growth of vacancies in the private sector to keep the ratio at least around 1) should be those of primary government importance with respect to labour market policies. Elsewhere below we offer scenarios of government activity in such areas.  215 The different types of intervention from the Ministries of Labour or their analogues should also be geared to the characteristics of the individual regional situation. In particular, and this follows directly from regional discrepancies in the v/u ratio, in the areas where unemployment  is mostly  structural, assistance to the unemployed should take mainly the form of professional education, training and retraining, in order to confer on them the capabilities and qualifications that are required by the market. This would be probably inappropriate for the areas where the availability of new jobs, whatever the specialization is, is slight, and where the intervention of active labour market policies should focus in particular on direct provision of subsidized jobs or job creation through the activation of public works. The alternative way of closing down unprofitable enterprises as a general universal radical approach to the reform cannot be recommended as a country is not a homogeneous entity.  216 §5.1.4. Unemployment  Compensation under Transition.  The discussion of unemployment compensation by economists in the West is typically concerned with its negative aspects. The main areas of concern are the extent to which unemployment  benefits raise wages, encourage  unnecessary quits or layoffs or excessively prolong spells of unemployment. On this view, unemployment compensation needs pruning, and this happened in many OECD countries during the 1980s (OECD, 1988). The current situation in Eastern Europe, however, presents a marked contrast. Unemployment compensation should be seen there as a natural complement to the reforms taking place. Setting aside radical alternatives such as basic income guarantees, the two principal ways of providing income support during unemployment unemployment  insurance  and unemployment  assistance. The  first  are is  contributory (with contributions coming from employers and workers) and depends on an individual's employment history. Benefits are usually set as a proportion of past earnings. They are always of limited duration and are not in general reduced if the claimant or his/her family have other sources of income. Unemployment  assistance, on the contrary, does not  depend on the  employment history and is of unlimited duration, but is subject to a test of family income and is not related to past earnings. Both benefits may be withheld from those who quit their jobs voluntarily, who fail to search for work, or who turn down job offers.  217 All but three OECD countries have unemployment insurance and many have in addition some form(s) of unemployment assistance for those who do not qualify for insurance benefit or who exhaust their entitlement (OECD, 1988). However, the details of both types of compensation scheme vary considerably from country to country (Ibid.). In the USA there is no real means-tested safety-net and the unemployed without unemployment insurance may receive nothing. Most European countries do have unemployment assistance nets, although there are notable exceptions as far as eligibility is concerned, for example Italy and Spain (Ibid.). Western analysts of unemployment often treat both unemployment insurance and assistance as having the same impact on the labour market, ignoring both the different level of income support the schemes may provide and the different conditions under which payments are made (Atkinson and Micklewright, 1990). According to some sources, unemployment insurance ranks above unemployment assistance in terms of social policy in that it provides benefit without the unemployed having to undergo special tests and verifications. Therefore unemployment assistance is to be preferred on economic grounds, the reduction of benefit via means-testing increasing the incentive to work (Burda, 1988). If a labour market is characterized by the frequent occurence of second jobs, as in Hungary, then, according to this view, meanstesting of unemployment compensation increases the incentive to find another primary job. The main conclusion would be that means-testing increases  218 outflows from unemployment and lowers inflows. Moreover, a means-tested benefit may be preferred on the basis of cost and targeted at those with the greatest need. Since a general discussion of advantages of either unemployment insurance or assistance is beyond the scope of the present section, we shall pinpoint only those aspects relevant to the current East European situation. There are a number of reasons why unemployment insurance may be preferred to unemployment assistance as an instrument of active labour market policies in the region. First, a state-organized unemployment insurance system represents an important demonstration of "solidarity"  for a workforce facing labour  market upheaval. As some authors mention, in Western labour markets this has been an important consideration in the past (Atkinson and Hills, 1989). In Eastern Europe any policy that might foster social solidarity seems to have much to recommend it. Contributory benefit implies redistribution from those in safe jobs to the unemployed. It  implies a benefit that is "definite and  claimable as of moral if not legal right" (Micklewright, 1991). The notion of benefit as "right", although being subject to certain conditions, should replace that of a right to work. Another reason is that benefit as a "right" for those with an employment record, provides an incentive to enter the labour force. This is the "entitlement effect" of unemployment insurance, mentioned by M.Friedman in his Nobel lecture (Friedman, 1977) and extremely important for present-day Eastern  219 Europe. The argument can be adapted also when the labour market is segmented. If there are two segments, a primary and a secondary sector, and unemployment  insurance is restricted to the primary sector (because the  primary sector is defined as representing the formal economy and the secondary sector as the informal economy), the total reward offered by the primary sector provides an incentive to take primary sector jobs. Thirdly, unemployment  insurance does not involve high implicit  marginal rates of tax on the earnings of the family of an unemployed person (Micklewright, 1991). The same is not true for unemployment assistance where the family means-test implies that the benefit will typically be reduced if other income is present. The implicit marginal rate of tax may be 100% for earning above some defined level, implying a strong disincentive to work. In our opinion this represents a very important neglected aspect of the difference between unemployment  insurance and assistance. When considering the  implications of means-testing for family labour supply in East European labour markets, the high incidence of two-earner couples should be noted. As was mentioned elsewhere in this thesis, participation rates for women of working age typically range from 75 to 85%. Any reform of social security in Eastern Europe which threatens these levels of participation cannot be seen as a step in the right direction. The distinction between unemployment insurance and unemployment assistance is therefore a substantial one. When looking at how Western  220 compensation schemes actually work, it is also very important to distinguish between the receipt of the two types of benefit. For example, Table 40 shows the proportions of the registered unemployed in Great Britain and West Germany receiving the two types of benefits. One point stands out: the proportions  of unemployed receiving  unemployment insurance are low. A situation like this is not an exception; on the contrary - it is typical for the West. For example, the same indicator for the USA in 1987 has been estimated to be below 30% (Blank and Card, 1989). The main reason for this situation is long-term unemployment. Table 40. Proportions of the Registered Unemployed Stock in Receipt of Unemployment Insurance and Unemployment Assistance Benefits, %, 1988. Country, benefit  Great Britain  West Germany  Unemployment insurance  26  40  Unemployment assistance  56  23  Neither unemployment insurance nor unemployment assistance.  18  37  Total out of  100  100  Source: Micklewright, 1991.  For example the British unemployment insurance system does not pay benefits for those unemployed over a year, and in Germany this system is also of limited duration. On the other hand, an insufficient employment record is also an important reason for non-coverage.  221 In other words, OECD compensation schemes were not designed to cope with the levels of unemployment experienced in the 1980s. And they indeed may not be appropriate models for the East European labour market upheaval. It is on grounds of inadequate coverage that western schemes can not be recommended to the East. This conclusion is very different from the argument which maintains that it is the disincentive effects of West European schemes which should not be tried in the East (e.g.Layard, 1990). Unfortunately  East European countries have not attempted many  innovations in the system of unemployment insurance and have adopted the OECD styles of unemployment benefits, funded by payroll taxation, based on insurance principles but not experience-rated and not means-tested (unlike follow-up social assistance, which is). The criterion of income replacement (as a % of previous net income) determines the amount of benefit, usually with minimum and maximum amounts. Initially, these programmes were quite generous, especially in the potential duration of benefit (OECD, 1992). For example, unemployment benefit in Poland was originally available without any time-limit. More recently, eligibility, duration and income replacement have been tightened after pressure from national finance ministries and international organizations like The World Bank and IMF. Some details for 1992 appear in Table 41 below.  222 Table 41. Unemployment Benefits in Some Eastern European Countries, 1992. Eligibility rules  EC  RR & PD  BI  Employed > 6 of last 12 months, job losers only, special programme for school leavers.  37%  60% of gross average wage in last 6-12 months (6 months at 80% of min. wage for school leavers.  671  2  Employed > 12 months in last 3 years, quitters and new entrants ineligible.  38%  3 months at 60%, then 3 at 50% (can be topped up to meanstested "social minimum").  522  3  All eligible according to contribution (360 days minimum), quitters and school leavers eligible after 90 days unemployment.  78%  Duration depends on employment in last 4 years (max 18 months, no min). First 2/3 of duration - 70% of average wage over last 4 years, then 50%.  3,388  4  Employed >180 days in last year, except school leavers, disabled, mass layoffs; moonlighting permitted.  73%  36% of last quarter's average wage for up to 1 year (recently extended)  1,240  5  School leavers ineligible until unemployed 5 months.  64%  Depends on service and education: workers 60%, graduates 70%, for up to 270 days (recently extended).  1,286  1  Note: 1 - Bulgaria; 2 - Czechoslovakia; 3 - Hungary; 4 - Poland; 5 - Romania. ER - eligibility rules; EC - effective coverage; RR & PD - replacement ratio and prospective duration; BI - Burda Index. Source: OECD, 1992; Burda, 1993.  223 The value of unemployment benefits to eligible individuals is hard to estimate: systems differ along many dimensions. The last column of the above table presents a crude measure developed by M.Burda (1988): it is "the present value of the benefits package for a full eligible claimant as a percentage of average weekly earnings, which is then multiplied by the coverage ratio, the ratio of insured unemployment to total. The measure ignores taxes (in the Visegrad countries benefit is taxed but at negligible rates); assumes all recipients of benefit earn the average wage; and assumes the coverage ratio is a good indicator of ex ante eligibility. For example, suppose the average worker in Transylvania eligible for benefit was entitled to draw 75% of salary for 26 weeks, and 50% of the unemployed were eligible. Discounting at 20% per annum (0.3512%per week), the index would be 930 (930% of weekly salary). Even apart from the index, the data in Table 41 above are revealing. The generosity of unemployment benefits varies significantly across countries, a variation caused more by differences in duration and coverage than in the rate of income replacement. Hungary is by far the most generous, which partly explains the huge pressure on the government budget and criticism by the ILO (1991). At the other end, the former Czechoslovakia and Bulgaria are very frugal: the former by policy design, the latter due to fiscal constraints. Poland and Romania are in the middle of the pack. Many Western analysts  of labour markets may see reform of  unemployment compensation as meaning the changing of benefit rules so as to  224  increase incentives to work. Some may wish to see "workfare" conditions for unemployment benefit, the payment of benefit in return for unskilled work (Micklewright, 1991). But the evidence on the supply side impact on unemployment benefits does not in general indicate effects that are both large and robustly determined. Nor can evidence from the OECD region be seen as necessarily representative of effects in Eastern Europe. As far as the demand side is concerned, the evidence, largely from the USA, is rather stronger on the likely impact of increased experience-rating of employer unemployment insurance contributions in countries where temporary layoff unemployment is prominent (Micklewright, 1991). On the other hand, experience-rating would seem to have little to recommend it for the foreseeable future in nascent unemployment insurance schemes in Eastern Europe, where one of the main aims of the labour market policies is to encourage reallocation of labour. The focus on disincentives obscures the fact that reform may also mean a strengthening of unemployment compensation schemes. The aim should be the provision of adequate protection to all those satisfying reasonable (rather than punitive) availability and job search conditions. For the reasons described earlier, the strengthening of insurance benefits should be seen as a priority. Moreover, the positive economic arguments for unemployment insurance show that an improvement of income support does not need to be at the expense of efficiency. In conclusion, two aspects of income support for the unemployed in  225 Eastern Europe are particularly worth mentioning. Firstly, the benefit coverage of people who are recurrently unemployed needs to be considered. As in the OECD  countries, there will be many people who move in and out of  unemployment; temporary jobs are a prominent feature of capitalist economies. Secondly, a proper safety-net is needed for those without unemployment insurance coverage. East European countries need to get this in place from the outset. The experience of Great Britain in the 1920s is illustrative: a succession of amending unemployment insurance laws failing to get to grips with the problem of high unemployment.  226 CHAPTER 2. SUMMARY IN EASTERN EUROPE.  O F ACTIVE LABOUR M A R K E T POLICIES  The present section summarizes the main active labour market measures implemented by East European governments in the beginning of the 1990s. The experience of Western Europe, especially the high equilibrium unemployment in France, Italy and Spain seemingly has not been lost on policymakers in Eastern Europe. On paper they understand that deep recessions with poorly planned and passive systems of support for the unemployed can lead to long-term unemployment and politically irrevocable commitments to support these individuals. The preferred approach is now "active employment measures" (AEM). In theory AEMs must reduce equilibrium unemployment. First... [they increase the supply of vacancies]. Public works, soft loans for job creation or selfemployment, wage subsidies, tax incentives for private firms, and unemployment benefit conditionality are [among the factors increasing the supply]. Second, AEMs enhance information exchange and monitoring of unemployed workers' prospects, while retraining programmes eliminate the mismatch between vacancies and unemployed. (Burda, 1993). Yet AEMs are now used in response to a situation that has already deteriorated: under the conditions of high and growing unemployment and often without centralized information processing, job matching is managed haphazardly. Benefit recipients often take moonlighting jobs (explicitly legal in Hungary and Poland) or even draw benefit in several districts, which is illegal. Conditionality of benefit on readiness to accept alternative job offers is rarely enforced except in the former Czechoslovakia. The rise in obligations for benefits (or passive  227 measures) inevitably crowds out funds available for active measures. In Bulgaria, for example, ambitious programmes encompassing 35% of the 1992 budget of the "Fund of Employment and Requalification" included a wage subsidy along the lines of the successful Czech and Slovak programme; yet the rapid rise in unemployment at the end of 1992 forced 90% of those funds to be paid out as unemployment benefits. In Romania only 3% of the employment budget in the first six months of 1992 was aimed at AEMs; a similar situation prevailed in Poland. In Hungary in 1991 a wide range of AEMs was cancelled for budgetary reasons. Table 42 below summarizes the situation with AEMs as it was in 1992. Only the Czech and Slovak republics have implemented an AEM policy in the Scandinavian mode (described in Calmfors and Nymoen, 1991), where spending on AEMs remains relatively constant at around 25 % of the budget in both republics  (OECD,  1992),  and conditionality  is important. Benefit  recipients who are retraining enjoy a replacement rate of 70% rather than 5060%. Considerable funds have also been spent on public works, on which local labour offices actively place those with expiring benefit claims. They also pay their wages for the first six months. In 1991  157,000 jobs were created in  short-term make-work projects run by local authorities and longer-term subsidised employment initiated by the private sector. Of the latter, roughly 50,000jobs were "Assistant" or "Praktikant" positions offered to school leavers, paid at minimum wage by the employment office for up to the first year. In Slovakia this programme began after some delay; only 24,000jobs were created  228 in 1991 (Burda, 1993). Table 42. Summary of Active Employment Measures in Eastern Europe.  Bulgaria:35% of Fund for Unemployment aimed at AEMs (unlikely to reach this figure). Public works on an experimental basis. Very limited support for entrepreneurs. Wage subsidy for school leavers envisaged. Local labour officies computerized, but not well manned (1 worker/290 unemployed). % of employment budget outlays in AEMs: 34% in 1991, 17% in 1992. CSFR: Local public works financed by local labour offices; apprenticeships for school leavers, at minimum wage for 1-year duration, sponsored by local labour offices; active computerized and coordinated employment offices running both job matching and unemployment benefits; incentives for retraining (70% replacement ratio rather than 5060%).Incentives for entrepreneurial support (12 months' unemployment benefit up front as "soft loan" + the same for hiring employees who are unemployed). % of employment budget outlays in AEMs: 23% in 1991, 30% in 1992. Hungary: Heavily limited by budget problems. Poorly coordinated employment offices, entrepreneurial programme (< 1,000 participants) abandoned in 1991. Regional development programme abandoned. Experimental short-time programme in 1991 had 35,700 participants. Roughly 100,000 in retraining programmes. % of employment budget outlays in AEMs: 30% in 1991,24% in 1992. Poland: Limited to some public works, entrepreneurs' programmes. Labour offices poorly staffed (over 300 unemployed per caseworker). In 1991 10,300 received training, 21,400 received startup loans. % of employment budget outlays in AEMs: 18% in 1991,18% in 1992. Romania: Wage subsidy programme for school leavers since 1991: for 9 months 60-70%of salary paid by Unemployment Fund. Early retirement programme imposes punitive "retraining charges" on the gross salary of older unemployed who refuse to retire. About 60,000 in a retraining scheme. For first 6 months of 1992 expenditures on retraining and wage subsidies were nearly 3% of total outlays. Short-time arrangements with unions allowed postponement of some layoffs.  Source: OECD (1992); Burda (1993); personal data.  229 As can be seen from Table 42, Czechoslovakia's set of AEMs had probably the greatest success among  those in Eastern Europe. Despite an  output collapse similar to that in other transforming economies, unemployment remains  low  and exits from  vacancy/unemployment  unemployment  into  work  high. The  ratio is the highest in Europe. Even the evidence for  Slovakia, where the collapse of heavy industry was steeper, is encouraging: unemployment has been falling and job creation picking up; job finding rates rose in 1992. In Hungary the shift in supply-demand conditions in the labour market has led to a growing structural imbalance in two areas. First, there have been changes in the structure of labour demand, with a sharp fall in the demand for unskilled, semi-skilled and non-manual labour relative to the demand for skilled labour (Table 43). Second, there has been an increasing differentiation in labour market conditions by regions. The ratio of vacancies to unemployment by counties varied from as high as 32:1 down to as low as 0.6:1 in 1992. Regions can be divided into three categories according to the local labour market situation. In Table 43. Vacancies and Unemployment by Skill Levels in Hungary, %, 1992. Skill levels  Vacancies  Unemployed  Skilled  48.5  14.6  Semi-skilled  25.2  16.5  Unskilled  17.4  50.1  Non-manual  8.9  18.8  Total  100.0  100.0  Source: Employment Information Centre, Budapest.  230 some areas acute labour shortages still persist. In others, though aggregate demand for labour exceeds supply by a substantial margin, demand for labour is tending to decline sharply, with structural mismatch between the skills required and those offered. In some regions considerable problems of unemployment have emerged. Such regions include traditionally underdeveloped areas, agricultural areas, regions dominated by crisis industries and areas with a high concentration of branch enterprises whose core is located elsewhere. As everywhere in Eastern Europe, unemployment in Hungary can be regarded as a regional phenomenon within the context of structural adjustment. Faced with redundancies, structural unemployment  and increased social tensions, the  government has assigned new responsibilities to the employment  offices,  extended the retraining scheme (although under severe budget constraint) and introduced a system of re-employment benefits. (Earlier the Employment Fund was set up to finance various programmes). In particular, the retraining scheme was extended to provide "make-up" money for employees enrolled in retraining courses organized by their firm, and assistance through direct subsidies and tax allowances for employers to cover part of the training costs. As was mentioned elsewhere in this thesis, reemployment benefits were introduced in 1986. The scheme provided financial security for redundant labour for one year. At any time within a year the reemployment benefit would be promptly terminated if the labour office could offer an appropriate job (requiring the same skill, with an earnings loss of less  231 than 10%, and a commuting time of less than two hours) to an unemployed person. The labour offices have been given new responsibilities, such as helping with the placement of employees involved in redundancy programmes and taking responsibility for hard-to-place labour. Furthermore, a "subsidized public work" scheme has been set up. It has been financed jointly by the government and the counties concerned, and implemented by the local employment offices. The scheme has provided hard-to-place labour with temporary work at a minimum wage level (drainage, afforestation, gardening, small maintenance projects in public places, etc.). In practice the shortcomings of the Hungarian AEMs are large and frequent, even given the severe budget problems, which influence the form and volume of some programmes (Table 43). First, until recently, the retraining scheme covered only those who were still employed while those who needed it most were not eligible (school leavers, the unemployed, those outside the labour force). Second, declaring redundancies and placing workers on the reemployment  benefit scheme was a long bureacratic procedure, so that  enterprises were inclined to shed labour by different, less cumbersome methods. Third, only those unemployed who had been placed on the scheme by their enterprise were entitled to draw re-employment benefits. Voluntary leavers were excluded. Fourth, the job creation programme suffers a severe shortage of funds and lacks identified "targets". The "subsidized public work" scheme is only a  232 temporary solution which, in addition, tends to reduce the future employment chances of its participants because of the poor reputation of labour (ethnic minorities like gypsies, unskilled, undisciplined) in the scheme. Turning  to the Bulgarian experience,  in 1992 the number of  unemployed there reached 600,000 people. In spite of the seemingly small absolute number of unemployed, the rate of unemployment in Bulgaria reached the highest level of any of the Eastern European countries (i.e. while the unemployment rate in Bulgaria was about 15% in 1992, it was about 12% in Poland, 6% in Czechoslovakia and 9% in Hungary). The rate of growth of unemployment in Bulgaria was also clearly higher than in these countries, rising 23-fold from 1990 to 1992. Explanations of this peculiarity can be linked with the legislative system in Bulgaria and with the unemployment measures. Some  sources  (Paunov,  1993)  mention  that  the  increment  in  unemployment consists mainly of displaced workers (75%). Nevertheless, since some restrictive legislative regulations were waived in 1991-92,the percentage of increment in unemployment  consisting of school-leavers and other new  entrants has been relatively larger in Bulgaria than in other East European countries. In 1992 the percentage of school-leavers and other new entrants in the labour market was about 18% in Czechoslovakia, 4% in Hungary, 15% in Poland and 12% in Romania. The structure of the unemployed in Bulgaria by different criteria can be seen in Table 44. When analysing these unemployment levels and structures, and these  233 Table 44. Structure of Unemployment Gender and Age in 1992, %.  in Eastern European Countries by  OLDER PU UR  YOUNG PU UR  MIDDLE PU UR  14  45  25  50  12  5  3  53  8  30  10  68  6  2  1  9  40  7  25  14  72  9  3  2  49  11  51  13  35  26  64  12  1  1  41  3  59  6  47  11  48  4  5  2  MALES PU UR  FEMALES PU UR  BL  45  11  55  CZ  47  6  HG  60  PL RN  Notes: PU - proportion of unemployed; UR - unemployment rates. BL - Bulgaria; CZ - Czechoslovakia; HG - Hungary; PL - Poland; RN Romania. Age categories: for Bulgaria - up to 30, 31-50, over 50; for Czechoslovakia - up to 25,26-50,over 50; for Hungary - up to 26, 27-55,over 55; for Poland - up to 25, 26-54, over 54; for Romania - up to 25, 26-50, over 50. Source: Employment Outlook (OECD, 1993);Paunov (1993);personal data. Table 45. Structure of Unemployment in East European Countries by Level of Education in 1992, %. OTHER PU UR  HIGHER PU UR  SECOND. PU UR  VOCAT. PU UR  BL  10  13  33  15  20  11  37  6  CZ  3  3  22  4  39  10  36  7  HG  2  1  6  2  44  14  48  9  PL  2  4  29  11  38  13  31  11  RN  1  1  51  8  39  5  9  2  Notes: BL - Bulgaria; CZ - Czechoslovakia; HG - Hungary; PL - Poland; RN Romania. PU - proportion of unemployed; UR - unemployment rates. For Bulgaria vocational includes secondary education and for Poland secondary education includes secondary vocational. Source: Employment Outlook (OECD, 1993);Paunov (1993)personal data.  234 shown in Table 45 above, it is worth noting that some of them are due to the specific measures used in the Bulgarian and other countries' statistics and legislative regulations. For example, in the age structure of unemployment in Bulgaria the "young" age group shows almost the highest levels, but it should be remembered that the Bulgarian administrative authorities include citizens up to 30 years old in this group, while in the other countries the upper limit for the "young" age group is 25. Moreover, new school leavers in Bulgaria are now eligible for registering and claiming unemployment benefits, while in most other countries they are not. When this is taken into account, the proportion of younger unemployed  in Bulgaria is in fact quite similar to that in  Czechoslovakia, Poland and Hungary. Nevertheless, unemployment rates for the "young" group in Bulgaria are still very high, which is a dangerous portent for future chronic problems. Unemployment structure by gender reflects quite clearly the fact that women's participation rates were very high before the transition started. In Bulgaria it reflects also the structure of job losses by sector, i.e. the sharp decrease of both output and employment  in the textile and other light  industries, where women account for the large majority of the employed. Some problems with the duration of female unemployment may be expected in all East European countries, insofar as there is evidence that the number of vacancies and the outflow from unemployment into the private sector are considerably lower for women than for men.  235 The duration of employment benefit payment in Bulgaria depends on the previous work experience and the age of the unemployed (see Table 46 for details).  56  Table 46. Duration of Unemployment Benefits in Bulgaria.  Work experience, years  Age, years  Duration of payments, months  Up to 5 Over 5 Over 5 Over 10 Over 20 Over 20 Over 25  for all up to 40 more than 40 more than 45 more than 51 - men more than 51women more than 56 - men  6 7 8 9 10 12 12  Source: Paunov (1993). Previous work experience of at least 6 months is needed. This does not apply to school and army leavers, who are eligible to claim unemployment benefits without the initial waiting period and without having previous work experience, and receive benefits equal to the minimum wage in the country for no longer than six months. This unemployment benefit scheme can be described as one of the toughest in Europe with the exception of the regulations concerning school leavers. This is the main reason for the fact that in 1992 less than 50% of the registered unemployed in Bulgaria received unemployment compensation, compared to around 80% in the other East European countries. Moreover, the  236 compensation calculated by means of the above formula is vulnerable to rapid change of inflation and to the partial and clumsy indexation. That is why the average unemployment benefit payment in Bulgaria in 1992 was just about the minimum wage in the country, which was quite low and allowed for covering about 70% of the optimistically calculated "social minimum" living cost. This ranks Bulgaria in the last place in Eastern Europe in terms of the replacement ratio (the extent to which unemployment benefits cover the loss of income from employment). No broad training and retraining programmes exist in Bulgaria, there are no start-up loans to encourage the growth of self-employment,  and no  subsidized employment schemes and wide-ranging public-works programmes have been adopted. That is why the structure of public expenditures on labour market policies in Bulgaria displays a low proportion of active measures (16.8% for training and 3% for other measures). And the total labour market expenditure as a percentage of GNP was less than 0.8% in 1992 in Bulgaria compared with 1.4% in Poland and 1.6% in Hungary.  237 CHAPTER 3. CONCLUSIONS. This chapter of the thesis offers conclusions for Part V, as well as some broader inferences which apply to the study in general. •  Both the relevant literature and this  study show high and rising  levels of unemployment in East European countries. Both unemployment rates and absolute figures of unemployment  vary among the countries. While  unemployment is rising steeply in Bulgaria, Romania and Hungary, its growth is slowing down in Poland, the Czech Republic and the Slovak Republic. Such discrepancies reflect the different ways in which labour markets have adjusted in different countries, as well as macroeconomic developments, a slower decline of output in the former Czechoslovakia, the scope of labour market policies, and the effects of recent changes in unemployment benefit regulations in most countries of the region. Russia represents a separate case with its soft budget constraint and, at the same time, a high pace of privatization characteristic for the last two years of the country's development. •  Unemployment inflows and outflows are very low by Western  standards, especially when compared to employment losses and the scale of the ongoing reallocation process. There is significant de-employment in the public sector occuring simultaneously with growth of the private sector, but the mechanism of labour reallocation between these two sectors of the economy does not necessarily involve the transition of a worker to the unemployment pool. The data show that most workers tend to move directly from the state to  238 the private sector. Actually, there is quite a significant flow from employment to  outside  the  labour force, but  not  from  employment  to  formal  unemployment. • The spread of long-term unemployment seems to be common to all countries of Eastern Europe. As T. Boeri (1994) points out, except for those in the Czech Republic, more than 30% of registered job-seekers have already been unemployed for at least 12 months, in spite of the relatively narrow scope and short duration of unemployment unemployment.  benefits and the short history of open  Given the small magnitude of inflows and outflows, the  duration of unemployment is likely to rise, even if the economy picks up and unemployment stabilizes at its current level. The role of unemployment inflows in shaping the distribution of long-term unemployment (there are "vulnerable" groups of employed population) may indicate that the structure of job losses by sector and occupation may have been a factor in spreading  the costs of  economic restructuring unevenly. The small role of outflows in terms of the characteristics of long-term unemployment show that once an individual has entered the unemployment pool, he or she either finds it difficult to exit or does not want to. •  The fact that most "vulnerable" groups in the economically active  population (low-skilled workers) have very similar shares of total and long-term unemployment, in our opinion shows that the high level of representation of these groups may be attributable more to the fact that they face a higher risk  239 of becoming unemployed than to lower exit probabilities. Only age indicators of the unemployed seem to have a significant effect on exit probabilities: as in West European countries, the proportion of long-term unemployed among young people is much lower than their share in total unemployment. Unemployment duration is therefore short for this group. •  For the reasons stated in Part III, the lag between output and labour  adjustments is quite long in the countries of the region. These countries will face major inflows into unemployment in the near future, although there are numerous factors which can influence this process. The experience of Poland in 1992 suggests that sharp employment declines may well continue after a recovery starts. Much would depend on active employment measures of the government with respect to privatized enterprises. The evidence indicates that many employers reacted to falling demands for their products by reducing the number of work hours and introducing "involuntary vacations" for the workers rather than by implementing large-scale lay-offs. Although there have been time lags between output and employment declines in all the countries, one must not get the impression that absolute reductions in employment were hardly noticeable. During 1989 - 1993 they averaged 12% over the region involving a net loss of around 6 min jobs. We therefore would insist that a relatively low rate of adjustability of employment to declines in output can not by itself account for the low magnitude of inflows into unemployment. There are other factors involved, and we tried to point  240 them out. •  As some sources put it, there are three main phases of reduction in  employment which all East European countries seem to have gone through. At first, employment decline was achieved mainly via attrition, a freeze on new hirings, and induced retirement. This approach was reflected in a rather moderate decline of employment (except in Poland, where employment has been rapidly declining from the very beginning of transition), and a small rise in the number of unemployed with previous work experience. Next, major layoffs of workers take place that disproportionately  affect people close to  retirement age. Finally, in the third phase, there is a closer correspondence between job losses and the rise of unemployment. •  Two main reasons might be offered for the small magnitude of  unemployment outflows, especially flows from unemployment to employment. First, in Eastern Europe on-the-job search plays a major role in labour mobility, which makes it even harder for the unemployed to find a job. Second, there are significant regional mismatches between the distribution of unemployment and vacancies. •  There is a need for a shift of policy focus from inflows into  unemployment  to promoting flows out of unemployment.  Labour market  policies by themselves can not be expected to foster the job creation process. Rather they seem to work better in periods of economic recovery than during downturns.  Nevertheless,  even by  simply  redistributing  employment  241 opportunities in such a way as to reduce the share of long-term unemployed, labour market policies may help to avoid  the marginalization of a large  segment of the labour force. By increasing competition for jobs and by reducing the number of long-term unemployed and associated skill losses, they could permit reductions in labour costs to be attained and hence could stimulate the operation of equilibrating mechanisms.  242  PART VI. RETROSPECTIVE VIEW AND AFTERWORD The purpose of this chapter is not to summarize the findings yielded by the preceding analyses. This has already been done at the end of each of the Parts of this study. It would appear more fruitful to attempt  a purposeful  review of the study, referring to the aims of the thesis and the intended methodology. This thesis arrives at a moment when the approach towards transition is going through a reappraisal. The emphasis is shifting away from the mere and rather naive quantification of "transition targets" (fix the exchange rate at a certain level, balance the budget within the next 500 days, etc.) towards a more sophisticated investigation into causal structures governing the interrelationships between various economic and social processes. Partly as a result of this particular timing, this thesis confronts a dilemma on two levels. On the one hand, everyone interested in East European issues is faced with a massive amount of literature on transition. Generally, most of the literature does not have a conceptual character, deals with narrow issues and becomes outdated relatively quickly. The result is widespread confusion and a tendency to reject or accept ideas somewhat indiscriminately. We tried to overcome this problem by analyzing and reviewing the current situation with labour markets on the basis of original methodology. On the other hand, given that the literature is diverse and often obscure, the author himself found the task of making sense of East European transition overwhelming. Thus, this  243 thesis is far from offering the "final truth"; it suggests views and concepts which sometimes may seem dubious and deserving much deeper investigation than allowed by the scope of the present study. Of course, the abovementioned difficulties influenced the objectives of the study. In particular, the study is confined to the macro level and does not descend to the level of the region or enterprise (p. 5). Although the time seems to be appropriate to start extensive case studies in Eastern Europe (looking, for example, at processes of regional or local economic transformation), this would add tremendous and probably unsolvable data problems preventing the researcher working in "office conditions" from making robust inferences. This task would have required field studies and research trips. In general, the study has reached its main aim, i.e.has suggested answers for what were called the "principal questions" of the thesis (p. 5). The role of labour markets in helping to ensure success for economic transformations has been emphasised (Parts I and V), the current problems threatening that success have been revealed (Parts I, III and V); and the goals and mechanisms of labour market transitions have been identified (Parts III and V). More generally, an attempt has been made to examine which factors, including other than economic variables, influence the occupational structure of the labour force (such as the heritage factor and issues of universality). Attempts have been made to evaluate the impact of each of those factors (Parts II and III). In this way, some light has been shed on the possibilities and the limitations of an  244 international comparison approach to the structural aspects of labour force (Part III). Unfortunately, as will be pointed out below in more detail, the information and data available made it very difficult to indicate with any precision to what extent international  comparisons  can be used with same expectation of  reliability, and to what extent their use can result in error. In spite of some limitations of our methodology for international comparisons offered in Part III, and the somewhat irregular way in which international comparisons may seem to be handled in the present thesis, the results of this study may be of importance. They should not be taken as unshakeable, but as a guide indicating possible directions for further research. Narrowing the study to the macro level and restricting the term "transition" to the ongoing economic transformations  (pp. 4-5), we have  suggested that general methodological approach which seemed to fit the research priorities and constraints the best. The transitional development of labour markets in Eastern Europe was viewed as a triply-influenced process (pp. 9-10). Specifically, labour markets are being influenced by broad economic policies pursued by the present governments, as well as by features inherited from Communist times and processes within the labour force itself induced by systemic changes. Although it was stipulated at the very beginning that the thesis pays less attention to the latter component  (p. 10), providing some  occasional insights into the development of labour market organizations in Eastern Europe, this can be regarded as one of the shortcomings of the study.  245 Now, when the work is completed, it seems that utilizing more of a structuralfunctional approach, i.e. focusing more on labour market institutions, labour market-related interest groups, unions and possibly on other less-defined entities and forces, and linking them to the macroeconomic policies of the governments, would have improved the study. In terms of the first two components, specifically broad macroeconomic policy and the heritage of East European labour markets, Chapter 3 of Part I and Part II respectively furnish detailed insights into these issues. Part I also links the discussion to broader theoretical issues, trying to identify the conceptual roots of macroeconomic policies in East European countries in existing macroeconomic frameworks. It attempted to answer the question how the East European situation fits existing theoretical considerations. In general, Parts I and II provide a certain logical continuity for the analysis, taking East European labour markets from the early 1950s to approximately 1992-1994, the last years for which the author had data (see Conclusions...p. 88). Using statistical data extensively, these parts of the thesis serve as an essential premise for the rest of the study, specifically for the more detailed  investigation  into  the  three  basic  concepts  characterization of the labour market: employment,  underlying any  unemployment, and  economic inactivity. The rest of the study was therefore focused mainly on two issues which were called "leitmotif problems". These were the degree of labour market restructuring to be expected, and the origins and scale of East European  246 unemployment (p. 13). Prolonged economic depression suffered by all East European countries, followed by huge declines in most of the aggregate economic indicators, as well as innovations  in the labour markets induced by systemic changes (e.g.  transformations of property ownership structure, processes of disemployment from state enterprises, etc.) brought the study to the issues of labour market restructuring under transition. One of the main questions here is the rate of employment decline, which was lagging behind the rate of decline in GNP. This, as well as some other questions related to the occupational structure of employment, were the focus of Part III (pp. 93 and 104). The model suggested in Part III to assist in revealing the dynamic tradeoff between output and the sectoral structure of employment has some conceptual and technical difficulties mentioned elsewhere in the thesis (e.g.p. 141). On the other hand it has a number of conceptual advantages. First of all, it suggests a way to formalize the three-category employment structure into a single indicator. This seems to be an innovation, at least we have not seen such attempts in the literature. Secondly, it uses a dynamic chain of time-series data, without which most of the projections resulting from the analysis of manpower are rather devoid of economic meaning. Thirdly, the model attempts to use the integral structural indicator as an independent variable, i.e. to analyze the employment structure as one of the factors of production; an attempt rarely present in econometric sources. And last but certainly not the least, the model  247 takes into account outflows of employed from one sector of the economy into another using a set of adjusted data (p. 116), i.e. it deals with vertical mobility of the labour force. This seems to be a contribution to the modelling of manpower. Though the approach suggested incorporates a series of successive approximations, one could very well imagine a simultaneous solution applied to it as a result of further elaboration. Chapter III, especially paragraph 3.2.2. offers some other arguments in favour of the model, as well as some criticisms. Here we would note that the model does the job as far as the objectives of the study are concerned. In particular, it answers questions about the existence of the general tradeoff between the attained level of economic development and the employment structure and some initial questions about the character of the tradeoff. It helps to understand at what stage of the development of the employment structures are the former USSR and countries of Eastern Europe and what is the statistically-defined magnitude of their relative and absolute lag in employment structures compared to the West. In general, the analysis helps to approximate the degree of labour market restructuring to be expected in Eastern Europe. This analysis is facilitated and complemented by a historical descriptive approach utilized in Chapter I of Part III. Although for the reasons stated in §3.2.2.the approach implying a universal law of development is hardly justifiable, we would insist that the thesis offers some insights validating the existence of "universal employment structure growth paths".  248  One point deserves special attention.  Elsewhere in the thesis we  expressed concern about the available data and methods of their collection (p. 6-9). Even though the data problem seems to have been overcome, two particulars deserve mentioning in conclusion. First, the data problem has been solved by combining data of different origin (i.e. centralized statistics and household-generated), comparing the data from different sources and evaluating their reliability. In the case of monetary economic indicators (e.g. GNP) two criteria have been decisive. Firstly, the source should be well-known and frequently referred to in the recent literature on transition. Secondly, the source should be able to provide extensive masses of time-series data calculated on the basis of the same methodology. In this way the necessity to extrapolate and approximate indicators was minimized and homogeneity  of the data was  maximized. The methodology applied for modelling helped here too: in our model not the absolute magnitude of indicators, but the rate of their change over time was the focus of attention. However, some conclusions resulting from the statistical analysis of the data may seem, akin to normative choices  imposed by the author. They  certainly appear so if taken out of context. However, Part IV, devoted to the role played by labour issues in development thought, roots our conclusions in the existing models of economic progress. It shows that the questions posed as our research priorities in Part III are not arbitrary, but result logically from the development of growth theories, being issues of both scholarly and practical  249 significance. Part V deals with the last of the "principal questions" of the thesis, namely with the performance of labour . markets under transition at a less aggregate level than the previous Parts. This includes empirical study on both national and regional levels, analysis of unemployment trends, and identification of goals and mechanisms for government labour market policies. Chapter 2 of Part V offers a summary of labour market policies in Eastern Europe, as well as some recommendations. Overall, Part V derives practical implications for government from our model. We would therefore conclude that we have mainly reached the goals posed at the outset of the thesis. This of course does not imply complete happiness with what has been done, but simply expresses our hope that these more than two hundred pages of text were educative, intellectually challenging and practically applicable. These are seen as the three main general goals of any scientific investigation.  250  ENDNOTES 1. Its idea is briefly as follows: as a part of the social compromise and in return for macrolevel employment and labour market security (i.e. near full employment and a system of employment-protection regulations) workers and unions in particular accept management's right to modernize, with the presumption that workers and employers share the benefits of productivity growth. The primary mode of regulation is wage-oriented rather than marketbased and involving shutdowns and labour force redundancies (Howell, 1992). 2. We omit numerous "frontier" models which are "neither...nor" and pay attention to basic concepts only. 3. This point is important for the study conducted in Part III. 4. Strictly speaking, in some cases the turn took place in 1989. For example, in Poland the industrial production, that is the basic macroeconomic aggregate determining the GNP level in Eastern Europe, was declining in absolute terms beginning with May 1989 (Rocznyk Statystyczny, 1990). It may be assumed therefore that the GNP was already falling in the second half of that year. It should be noted that 99% of national product of every Eastern European country was produced within its boundaries, there is no noticeable difference between GNP and GDP indicators, which can be used interchangeably for the sake of convenience. 5. There are authors who deny the suitability of the term "recession" to describe the nature of the phenomena in Eastern Europe, while stressing their specific character. Without going deeply into a theoretical dispute we must note that by recession in the region we understand a phase of reduced economic activity during which the absolute level of GNP declines. 6. The situation is similar in the mentioned cycles typical of the centrally planned economies. 7. The nature and origins of East European unemployment are analyzed and discussed elsewhere in this thesis. A very important point at the moment is that labour market performance at the macro level is closely connected with the specific features of East European recession. This is the crucial statement for further analysis. 7. On the other hand there is in fact much more economic activity in the countries than seems to be the case: it has simply shifted to the private sector, whose traditional methods of collecting statistics cannot detect it.  251 8. For example, it is now clear that the military industrial sector has been heavily mismanaged in Russia during the perestroikayears and later. Since this sector used to concentrate highly-skilled employees, intellectuals and scholars and since its product was quite competitive in the world market, it must have received a priority treatment from the government in order to save the sector from a "brain drain". The contemporary situation in the military industrial sector is pitiful with its needs heavily underfinanced and the majority of employees approaching the retirement age. In terms of the sequencing of economic reformist measures the debate continues on whether prices should have been released before significant progress in privatizing state property was made. Some experts claim that the sequence implemented had created the grounds for hyperinflation and actually slowed down the reformist efforts of Gaidar's government in Russia in 1992-93. 9. In 1989 the so-called Trzeciakowski Plan became popular in Poland. It assumed an inflow of 10 bin USD over three years, inclusive of the reduction of the foreign debt. At that time this proposal was not sufficiently seriously treated - either in Poland or abroad - and was viewed as unrealistic. Three years have passed, and the scale of the foreign aid has been even greater, but its effects have been comparably less than expected (Kolodko, 1993). 10. Granting aid to the former USSR, comparable to that obtained by the former GDR in 1991, would cost about 1 trl. USD. As the experience demonstrates, even that amount of external resources would not prevent production collapse. 11. It was declared impossible for the worker, as a co-owner of the means of production, to be an employer and employee at the same time. 12. The owner of the workforce and the owner of the means of production were expected to be becoming more and more identical with the "withering away of the state". 13. Sometimes the additional theoretical requirement was introduced: that the economically active population had to be used in the public sector. 14. One should keep in mind that this point of view, namely - equilibrium under full employment - does not originate in socialist doctrine. The idea of increasing demand for labour under constant/given labour supply was quite common in Western economic literature up to 1936-the year of publication of Keynes's famous General Theory on Employment, Interest and Money. It should be noted that Keynes wrote not on "frictional" or voluntary unemployment, recognized by both neoclassical and socialist economists, but on involuntary unemployment understood as a situation in which both aggregate labour supply under given wages and aggregate demand for labour under the same wage exceed  252 the real volume of employment. 15. A person who was fit to work, but not working, and with no other source of income, was liable to prosecution and imprisonment, according to Constitutions. 16. Labour hoarding can be defined as a situation where an establishment is paying for more worker-hours than are necessary to produce current levels of output. Indicators of labour hoarding can be calculated either in terms of hours of work or in terms of persons employed. 17. Country differences in the educational composition of the labour force may partly reflect differences in classifications. For example for Austria the category "vocational" in Table 7 includes a significant proportion of the workforce who have, in fact, received vocational training at the level of higher education. Comparisons of enrolment ratios can also be made by looking at enrolments at all levels of education by single years of age. For those East European countries where data is available the comparison with OECD countries is even more unfavourable than shown in Tables 8and 9. 18. The following example was borrowed from Boeri and Keese, 1992. A survey of Hungarian firms in 1991, including some firms in the private sector and joint ventures, revealed that many managers considered that the general education level of skilled and unskilled manual workers was very low and they were in general not satisfied with the quality of their workforces. 19. Usually these jobs were unattractive because of their geographical locations and were considered to be "dead-end jobs" with no professional career opportunities. According to some studies (Penkin, 1990) from 2/3 to 3/4 of "directed graduates" in the USSR in 1970-1980 changed their jobs after the compulsory three years and returned to their home regions, which contributed to high labour turnover. 20. The first unemployment compensation benefit in Eastern Europe was introduced in Hungary in 1986. 21. The terms "occupational structure of employment", "macrostructure of employment", "employment structure" and the like are used as synonyms hereafter unless otherwise specified. 22. Thus in 1950 - 1955 the share of secondary employment in Western Germany was 52%,in Czechoslovakia -50%.The data for some other European countries are the following: Austria - 45%; Belgium - 47%; Denmark - 42%; Finland - 37 %; France -41%; Holland - 45 %; UK - 48 %.  253 23. All estimates are given for countries in their post-WWII boundaries. 24. The mean sectoral distribution was calculated on the basis of the sample of 12 industrial European countries - France, Switzerland, Norway, Sweden, Austria, Belgium, Netherlands, Great Britain, Denmark, and Italy. 25. Empirically determined threshold. 26. Here and below measured in constant prices of 1990. 27. It is also worth mentioning here that the post-war economic isolation in the COMECON countries triggered an artificial spurt of the industrial sector. For example, Czechoslovakia was reported to have produced domestically some 65% of all categories of industrial goods at a five-digit SITC (Standard International Trade Classification) level - a higher proportion than Japan, despite, according to some estimates, Japanese GNP being at least 20 times as large (Kolodko, 1993). Similarly, East Germany produced a wider range of industrial products than did West Germany. 28. The monograph by M.Kabaj (1972) is worth mentioning here. The author, however, analyzed occupational structures of employment in a global historical context. 29. Most Western statistical sources distinguish between "employment" and "occupation". Up to recently East European statistics lacked such a distinction, which made East - West international comparisons in terms of labour statistics very difficult. The need for urgent reform of methods for data gathering has been mentioned elsewhere in the thesis. The point we are trying to make here is that the lack of an employmentoccupation statistical distinction in Eastern Europe represents another argument in favour of using three-category structural indicators for the purposes of international comparisons. 30. Here we assume universality of economic development, which some readers may find difficult to justify. Part III below offers some insights in this regard. 31. All statistical data on employment structure used in this Chapter have been borrowed from "Yearbook of Labour Statistics" (1961-1989) and Statistical Yearbook/Annuaire Statistique published by the United Nations. The data on GNP per capita refer to the IWTAB electronic library (1987-1988) and partly to Statistical Yearbook/Annuaire Statistique and Sirakov (1989). These data have been corrected/adjusted using the following publications: A System of National Accounts, Studies F, No.2, Rev.3, UN Publication, Sales NO.E.69XVH.3; and Basic Principles of the Balances of National Economy, Studies in Methods, Series F, No. 17, UN Publication, Sales No.E.71.XVII.4.  254 The statistical data consisted of that for Australia, Austria, Belgium, USA, Canada, Japan, New Zealand, Denmark, Finland, France, West Gerzaeny, Italy, The Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, Great Britain, Czechoslovakia, Poland, USSR, Romania, Hungary, East Germany and Bulgaria. Questions may be asked about using absolute GNP indicators for East European countries after the author has expressed strong doubts about this statistic in the previous chapter. We shall note that absolute magnitudes of GNP indicators are of less significance in this chapter. What is important is that the data have been borrowed from the same source and for the whole period of study. This allows us to achieve necessary homogenity of the GNP data base. 32. Line graph here shows dynamics of the indicators in time. For example, the proportion employed in primary activities is generally highest in countries with low incomes. As income rises, the proportion employed in agriculture decreases for both individual countries and an average country. 33. Closed dotted areas in Fugures 1-5 contours zones where actual data points for real countries are located. These areas are closed because there is no clear interrelationship between the proportion employed in secondary activities and income per capita. Therefore it is impossible to extrapolate the data and create the line graph. Growing income does not necessarily mean that proportion employed in secondary activitites changes in a certain way. The situation is much more clear for the I and III sectors of the economy. 34. In fixed dollar value of 1980. 35. The magnitudes of coefficients a and y have been established empirically. 36. Monosematicity of the parameter has some mathematical provisos which however do not play the determining role. 37. Some statistical inaccuracies do exist here. They are connected, first of all, with data shortage. Thus, sometimes it was impossible to receive data on sectoral structure of retirees or new entrants to the labour force. In these cases methods of extrapolation and/or indirect assessments were used. Besides, mentioned indicators used for analysing intersectoral flows do not take into consideration those people who, for example, did not retire but ceased being in the labour force, as well as those who stayed in the labour force, but remained unclassified (see Part I Chapter I on classification problems). These inaccuracies are however statistically insignificant. 38. For example, in 1978 in Hungary the IIES was equal to -0.363with 23.1, 49.3and 27.6%corresponding sectoral shares. In one year - 1979-the share of employed in the second sector due to labour outflow dropped by 0.4 points. 0.3 out of these 0.4 occurred due to the increase in the third sector; 0.1 - in the  255 first sector. The IIES has changed by -0.002. 39. Due to the methodology described above, the terms "in/outflow" used here to describe the change in the share of employed mean proportions corresponding to the absolute changes in sectoral employment. 40. Fixed dollar value as of 1980 everywhere in this thesis. 41. Refer to Chapter 1, Part III for a detailed analysis. 42. The latter point is stressed in particular for Poland by Kwiatkowski (1992). As an indicator of this is offered the fact that the share in total unemployment of workers freed through collective dismissals, mostly originating from the restructuring of large state firms, though increasing over time (Cilosi, 1993; italic mine hereafter - SI) was still only 24.6% in the first half of 1992 (Zycie Gospodarcze, 1992). This share had increased to a third at the end of November 1992, however it was only 18.8% as far as newly registered unemployed in November 1992 were concerned. It must be added (Chilosi, 1993) that collective dismissals ("dismissals originating in causes concerning the employer", according to Polish terminology) are regulated, because of their potentially greater socially disruptive effect, so as to provide better treatment for dismissed employees. Particular procedures are envisaged, as well as the payment of indemnities financed partly by the dismissing firm, partly by the Fund of Labour (Dziennik Ustaw, 1990). 43. We would like to stress the logical bonds between this comment and the content of Chapter 3 in Section 1. 44. In particular, economic growth means more output. Economic development implies not only more output, but also changes in technical and institutional arrangements by which output is produced and distributed. Growth may involve greater inputs measured per unity of output, or may result from greater efficiency. Development goes beyond this to include changes in the composition of output and in the relative sizes of contributions of the various inputs to the productive process. 45. The interdependence of explanatory variables made it impossible to run * multiple regression models in order the estimate the role/weight of each sector •in change of output. 46. Some later studies applied to Third World countries assumed that the Harrod-Domar theory was applicable to underdeveloped economies, although to a relatively simple model of such an economy. See, for example, Bruton, 1955.  256 47. We leave aside a significant block of development theories which continued to progress during the 1980s, such as: the dependency debate, post-Marxist reactions, revival of structuralism, intellectual relativism and others. Progressing within the general interdisciplinary realm of social sciences, they have slight relevance to our topic. 48. As quoted in King, 1993. 49. If activity of foreign investors could be considered as one of the indirect indicators to evidence the success in privatization, the following statistics are worth citing. Around 10% of the outstanding stock of privatized enterprises in Russia belongs to foreign companies and individuals. In 1993 20% of winners of all "investment competitions" in Russia were foreign firms and joint ventures. Total amount of foreign investment in Russia by the end of 1993 is estimated at approximately 1.5 bin USD and 35 min German marks. In 1994 this figure is estimated to approach 2-2.5 bin USD and "if all goes well" even 6-10 bin USD, according to Anatoly Chubais, President of the Federal Property Fund (Finansovyelzvestiya). In addition, in November 1994 Russia started selling state enterprises by auction. Foreigners have the same rights to participate in the auctions as Russian investors, with a few minor exceptions (Kommersant-DAILY, 1994). 50. The extreme case is Russia where by mid-1993 nearly 15% of the state workforce was on involuntary leave or short time (Blanchard, Commander & Coricelli, 1994). 51. According to the Finansovye Izvestiya {Financial News, May 04, 1995) unemployment in Russia in 1994 reached 5.3 min (7.1% of the economically active population). Out of these only 1.6 min were registered by central and local employment centres as people looking for jobs. The number of "partly employed" (people involuntarily working part-time) was4.8mln (6.4%).Every month in 1994 from 20% to 30% of the total number of employed in the Russian economy (69,6 min) "enjoyed" involuntary unpaid vacations initiated by the management of the enterprise. These people were not included in formal unemployment. 52. Analogously to the "Italian strike", the situation in East European labour markets may be called "Russian unemployment at the work place".This reflects the scale and depth of "hidden unemployment" or labour hoarding problem in the region. 53. This does not mean that unemployment is "good" in general and it would be beneficial to have more of it. The logic of the argument justifies a high open unemployment rate in East European countries during transition only and does not go any further. The structure of discussion maintains the argument that  257 unemployment is essential to change (providing a "disciplining device", a "bargaining power device", a check on the growth of real wages, an incentive for re-training, etc.).This does not allow the discussion to be turned around, to argue that it is not unemployment that is "desirable", but vacancies. In this particular East European situation vacancies are a function of unemployment, but not vice versa. This standpoint, which may seem "politically incorrect" from a traditional academic North American viewpoint, is supported by the flow approach to the labour market as well. 54. Some authors (e.g.Burda, 1993) mention that one of the theoretical limitations of the matching function is that it does not take into account job-tojob switching. However, this type of switching, being unrelated to unemployment issues, is not very relevant during the transformation. 55. It is worth mentioning that not all people temporarily dismissed from state enterprises people can be included in the unemployment pool. One of the most illustrative examples relates to several big industrial enterprises located in cities of Siberia and the Far East. In October 1994 such enterprises as Barnaul Tyre Plant, Altai Diesel,z.nd Khimvolokno (Artificial Fibres Factory) returned to fullscale operation, but faced an unexpected problem. Around 70% of temporarily dismissed workers did not return from their "involuntary vacations". According to the poll, they had found new jobs and are not willing to leave them now. This local episode has in our opinion a somewhat deeper significance. One the one hand it once more illustrates the structural character of unemployment in Eastern Europe, where people are looking not for any jobs, but for those which suit them the best. This is true even in such peripheral areas as the Russian Far East and Siberia, where the pace of economic change is significantly slower than in European Russia. On the other hand, the number of vacancies is high enough to supply jobs for people who are desperate to find a job. According to the poll, the number of unemployed in Russia who would take any job is approaching zero. The majority of the unemployed want to find a job providing better conditions than their last one (Jzvestiya, 1994). It may seem paradoxical for East European people, but the experience shows that should an unemployed person give up inordinate ambitions or even look actively for a job, he/she usually finds a better job within the next two months (Izvestiya, 1994). 56. The eligibility criteria, duration and coverage of the Bulgarian unemployment benefit scheme in force since 1991 provide that the amount of the monthly unemployment benefit payment is formed as follows: U=W +0.2(W -W ), where U is the monthly unemployment benefit payment, W is the average monthly wage the individual received during the past six months of employment, and W is the minimum wage in the country. min  av  min  av  258 BIBLIOGRAPHY. Adam J., ed. Employment Policies in the Soviet Union and Eastern Europe. Macmillan, 1987. Adam J. Employment & Wage Policies in Poland, Czechoslovakia and Hungary Since 1950. Macmillan, 1984. Agh A. Hungarian Experiences of the Democratic Transition to a Multiparty System. Unpublished paper. 1990. Aghion P., Howitt P. Unemployment: a Symptom of Stagnation or a SideEffect of Growth? European Economic Review, 34, 1991. Akerlof G.A., Rose A.K.,Yellen J.L.,Hessenius H. East Germany in from the Cold: the Economic Aftermath of Currency Union. Brookings Papers on Economic Activity, #1, 1991, pp. 1-87. Anderson P. London Review of Books, 10 May 1991. Andor L. The New Right in Eastern Europe: a Vision of Decline and Decay. Futures,N.24,No.2, 1992. Atkinson A.B. Institutional Features of Unemployment Insurance and the Working of the Labour Market. Welfare State Programm Discussion Paper #56, STICERD, London School of Economics, London, 1990. Atkinson A.B., Micklewright J. Unemployment Compensation and Labour Market Transitions: a Critical Review. ESRC TIDI Programm Discussion Paper 143,LSEand EUI working paper in Economics 90/9 EUI, Florence. Atkinson A.B.,Hills J. Social Security in Developed Countries: Are There Lessons for Developing Countries? Welfare State Programm Discussion Paper #38. STICERD, London School of Economics, London, 1989. Bairoch P. Le Tiers - Monde Dans L'Impasse. Le Demarage Economique Du XVIII Siecle. Paris, 1971. Bari T., Kees M. Labour Market and the Transition in Central and Eastern Europe. OECD Economic Studies, No. 18, Paris, Spring 1992. Barro R. Intertemporal Substitution and the Business Cycle. Journal ofMonetary Economics, Vol. 14,1981, pp.237-68.  259 Becker G. Human Capital. Columbia University Press, NY, 1975. Beleva I., Bobeva D., Dilova S., Mitchkovski A. Labour Market Trends and Policies: Bulgaria. In Structural Change in Central and Eastern Europe:Labour Market and Social Policy Implications. OECD, Paris, 1993. Bell D. The Coming of Post-Industrial Society: a Venture in Social Forecasting. Heinemann, London, 1974. BenceG. Social Theory in Transition. SocialResearch,\99Q, V.57,No.2, p.245. Berg A. and Sachs J. Structural Adjustments and International Trade in Eastern Europe: The case of Poland. Economic Policy, No. 14, April 1992, pp. 117-155. Bermeo N., ed. Liberalization and Democratization: Change in the Soviet Union and Eastern Europe. Baltimore, Johns Hopkins University Press, 1992. Blanchard O.J. Notes on the Speed of Transition, Unemployment and Growth in Poland. Mimeo, 1991. Blanchard O., Commander S.,Coricelli F. Unemployment and the Labour Market in Eastern Europe. In Unemployment in Transition Countries: Transient or Persistent?OECD, Paris, 1994. Blanchard O.J.,Diamond P. The Flow Approach to Labour Markets. American Economic Review, Papers and Proceedings,1992. Blanchard O.J.,Muet P. A. Competitiveness through Disinflation: an Assessment of the French Macroeconomic Strategy. Economic Policy, April 1993. Blank R., Card D. Recent Trends in Insured and Uninsured Employment: Is There an Explanation? NBER Working Paper #2871,1989. Boeke J.H. The Structures of the Netherlands Indian Economy. NY, 1942. Boeri T., ed. Unemployment in Transition Countries: Transient or Persistent? OECD, Paris, 1994. Boeri T.,Garonna P. Unemployment in Economies in Transition: Conceptual and Measurement Issues.OECD, Paris, 1993. Boeri T., Keese M . From Labour Shortage to Labour Shedding: Labour Markets in Central and Eastern Europe. Labour Market and Social Policy Occasional Papers, No.9, OECD, Paris, 1992.  260 Boeri T., Keese M . Labour Markets and the Transition in Central and Eastern Europe. OECD Economic Studies, No. 18, Spring 1992. Boeri T.,Sziraczki G. Labour Market Developments and Policies in Central and Eastern Europe: a Comparative Analysis. In Fischer G . , Standing G., eds. Structural Changein Central and Eastern Europe .-Labour Market and SocialPolicy Implications, OECD, Paris, 1993. Borensztein E. and Montiel P.J. Savings, Investment and Growth in Eastern Europe, in G.Winckler, ed., Central and Eastern Europe Roads to Growth, IMF and Austrian National Bank, Washington, D.C., 1991, pp. 153-187. Bova R. The Political Dynamics of the Post-Communist Transition: A Comparative Perspective. World Politics 44, 1991, No.1, p.117. Braudel F. Histoire et Sciences Sociales. La Longue Duree. Annals E.S.C.1958, No.4, pp.725-53. Braudel F. The Games of Exchange, Moscow, 1989. Bruno M . , DiTella G., Dornbusch R., Fischer S. Inflation Stabilization. The Experience of Israel, Argentina, Brazil, Bolivia and Mexico. Cambridge, MA, London, 1989. Bruton H.J. Growth Models and Underdeveloped Economies. Journal of Political Economy, August 1955. Bulantsev V.Yu. Dinamika Zanyatosti Naseleniyav Nekotorykh Stranakh YugoVostochnoi Azii i Dalnego Vostoka. (The Dynamics of Employment in Some Countries of South-East Asia and Far East). Vestnik MGU, Seriya Geografiyz. (Bulletin of Moscow State University, Geography Series), No.3, 1993. In Russian. Bulantsev W.Yu.MacrostructureofEmployment: Post-Warand Transition Growth in European Former Centrally Planned Economies. Final Research Project, Central European University Foundation, Prague, 1992. Bulantsev V.Yu., Ivanov S.L. Macrostructure of Employment and GNP Per Capita: General Tradeoff and East European Transition Case. Joint Research Project, Central European University, Department of Economics, Prague, 1991. Burda M . Labour Markets in Eastern Europe. Economic Policy, April 1993.  261 Burda M . Wait Unemployment in Europe. Economic Policy, No.7,1988,p.393425. Burda M.,Wyplosz C. Gross Labour Market Flows in Europe: Some Stylized Facts. INSEAD Discussion Paper 90/51/EP, 1990. Burmeister E . , Dobell A. Mathematical Theories of Macmillan Series in Economics, NY, 1970.  Economic Growth.  Calmfors L . , Nymoen R. Real Wage Adjustment and Employment Policies in the Nordic Countries. Economic Policy, 1991. Calvo G., Coricelli F. Stabilizing a Previously Centrally Planned Economy: Poland 1990. Economic Policy, 1992. Campeanu P. Transition in Eastern Europe. Social Research,1990, Vol.57, No.3, p.587. Charap J. and Dyba K. Transition to a Market Economy: the Case of Czechoslovakia. European Economic Review, 1991, V.36, No.2-3, pp.581-590. Chenery H . , Syrquin M. Patterns of Development 1950-1970. Oxford University Press for the World Bank, L . , 1975. Chilosi A. Economic Transition and the Unemployment Systems, Vol. 17, No. 1, 1993, pp.63-78.  Issue. Economic  Church G.J. Shock of Reform. Time, February 17, 1992, V. 139, No.7, p.38. Clark C. The Conditions of Economic Progress. L . , 1957. Clark C. Development Economics: The Early Years. In: Meier G.M. and Seers D. (eds.) Pioneers in Development, Oxford, 1984, pp.59-77. Collins S.M.,Rodrik D. Eastern Europe and the Soviet Union in the World Economy. Washington, DC, 1991. Commersant-DAILY, October 22, 1993. Commersant-DAILY, October 13, 1994. Comisso E. Where Have We Been and Where Are We Going? A paper presented to the Midwest Political Science Assiciation in Chicago on 14 May 1989.  262 Cukor E.,K6vari G. Wage Trends in Hungary. International Labour Review, 1991, Vol.130, No.2, pp. 177-190. Daniels P.W. Service Industries: Growth and Location. Cambridge University Press, Cambridge, 1982. Dewatripoint M. and Roland G. The Virtues of Gradualism and Legitimacy in the Transition to a Market Economy. Economic Journal, 1992, No.401, p.291. Domar E. D. Essaysin the Theory of Economic Growth, Oxford University Press, NY, 1957. Dornbusch R. and Fischer S. Macroeconomics. 1991. Dornbusch R. Policies to Move from Stabilization to Growth. In Proceedingsof the World Bank Annual Conference on Development Economics, 1990, The World Bank, Washington, D.C., 1991, pp. 19-48. Dziennik Ustaw, No.4, January 27, 1990. Ekonomika Rossii v 1993 godu. (Economy of Russia in 1993). Akademiya Narodnogo Khozyaistva pri Pravitel'stve Rossii. (Academy of National Economy of the Russian Government). Vysshaya Shkola Mezhdunarodnogo Biznesa. (The Highest School of International Business), Moscow, August 1993. Eichengreen B. and Uzan M . The Marshall Plan: Economic Effects and Implications for Eastern Europe and the former USSR. Economic Policy, No. 14, April 1992, pp. 13-75. Emerson M . , ed. Europe'sStagflation. Clarendon Press, Oxford, 1984. Eriksen T . H . Ethnicity versus Nationalism. Journal of Peace Research, 1991, V.28,No.3, p.263. Faria C.R. The Origins of Economic Inequality between Nations. L . , Unwin Hyman, 1991. Fei J., Ranis G. Development of the Labour Surplus Economy: Theory and Policy. Irwin, Homewood, IL, 1964. Feiner S., Roberts B. Hidden by the Invisible Hand: Neoclassical Economic Theory and the Textbook Treatment of Race and Gender. Gender and Society, 1990, v.4, No.2, p. 159.  263 Finansovye Izvestiya (Financial News), July 13, 1994. Finansovye Izvestiya (Financial News), March 30, 1995. Finansovye Izvestiya (Financial News), May 04, 1995. Fisher A.G.B. A Note of Tertiary Production. Economic Journal. December 1952. Fisher A.G.B. The Clash of Progressand Security. Macmillan, London, 1935. Fischer G., Standing G., eds. Structural Change in Central and Eastern Europe: Labour Market and Social Policy Implications. OECD, Paris, 1993. Flanagan R.J. Labour Market Behaviour and European Economic Growth, in R.Z.Lawrence and C.L.Schultze (eds), Barriers to European Growth: a Transatlantic View, The Brookings Institution, Washington, 1987. Flanagan R.J. Wages and Wage Policies in Market Economies: Lessons for Central and Eastern Europe. OECD Economic Studies, No. 18, Spring, 1992. Freeman C , Clark J., and Soete L . L . G . Unemployment and Technical Innovation: a Study of Long Waves and Economic Development. Frances Pinter, London, 1982. Freeman R., Medoff J. What Do Unions Do? NY., Basic Books, 1984. Friedman M . Inflation and Unemployment. Journal of Political Economy, 85, 1977,pp.451-472. Friedman M . The Role of Monetary Policy. American Economic Review, 1968, Vol.58, pp. 1-17. Galenson W. Economic Development and the Sectoral Expansion Employment. International Labour Review, No.6, 1963.  of  Genov N. The Transition to Democracy in Eastern Europe: Trends and Paradoxes of Social Rationalization. International Social Science Journal, 1991, V.43,No.2, p.331. Goldfarb J.C. Post-Totalitarian Politics: Ideology Ends Again. Social Research, 1990, Vol.57, No.3, p.533.  264 Gomulka S. Stabilizacja i Wzrost: Polska 1989-2000.(Stabilization and Growth: Poland 1989-2000) in Politykafinansowa- nierownowaga - stabilizacja (II). (Financial Policy-Disequlibrium-Stabilization (II)), Ministerstwo Finansow i Instytut Finansow, Wilga, May 30-31,1990, pp.303-321. Gordon L. Social Policy on the Renumeration of Labour. Soviet Sociology,1988, No.2, pp.59-88. Gora M . Shock Therapy for the Polish Labour Market. International Labour Review, 1991, V. 130, No.2, pp. 145-164. Gora M . , Kotowska I., Panek T., Podgorski J. Labour Marek Trends and Policies: Poland. In Structural Change in Central and Eastern Europe:Labour Market and Social Policy Implications. OECD, Paris, 1993. Granick D. Job Rights in the Soviet Union: Their Consequences. Cambridge University Press, 1987. Gyurak K. Az Atszervezettek Gazdalkodasa. Figyelo, 1984, August 09, p.3. Hare P.,ReveszT. Hungary. Economic Policy, April 1992, pp.227-264. Harrod R. F. An Essay in Dynamic Theory. Economic Journal, vol. 49, 1939, pp.14-33. Harrod R. F. Towards a Dynamic Economics, Macmillan, L . , 1948. Hars A. Hopeless Prospects - Unemployment in Hungary. Acta Oeconomica (Budapest), 1989, Vol.41, pp. 193-236. Hars A., Kovari G., Nagy G. Hungary Faces Unemployment. Labour Review, 1991, Vol. 130, No.2, pp. 165-176.  International  Hars A., Nagy G. Changes in Unemployment and the Composition of the Unemployed in the Light of Statistics, Budapest, Labour Institute, 1991. Helgeson A. Geographical Mobility: Its Implications for Employment. In D.Lane (ed.): Labour and Employment in the USSR, Hassocks, Wheatsheaf, 1986. Herrick B., Kindleberger C. Economic Development. McGraw-Hill, 1983. Howell C. The Dilemmas of Post-Fordism: Socialists, Flexibility and Labour Market Deregulation in France. Politics and Society, 1992, v.20, No. 1, p.71.  265 Howitt P. Transaction Costs in the Theory of Unemployment. American Economic Review 75, 1985. Hughes G . , Hare P. Industrial Restructuring in Eastern Europe: Policies and Prospects. European Economic Review, 1992, Vol.36, pp.670-676. Ivanov S.L., Ushkalov I.G. Makrostruktura Zanyatosti i Uroven' Ekonomicheskogo Razvitiya v SSSR i Stranakh Vostochnoi Evropy. (Macrostructure of Employmemt and Level of Economic Development in the USSR and Eastern European Countries). Izvestiya Akademii Nauk SSSR, Seriya Ekonomicheskaya. Proceedings of the USSR Academy of Sciences, Series "Economics", No.4, 1991, pp.74-87.In Russian. Ivanov S.L.,Ushkalov I.G. Struktura Zanyatosti v SSSRi Stranakh Vostochnoi Evropy: Neobkhodimost' Korennykh Reform. (Employment Structure in the USSR and Countries of Eastern Europe: Radical Reform Needed). Mezhdunarodnyi Zhurnal Problemy Teorii i Praktiki Upravleniya (International Journal on Problems of Theory and Practice of Management). No.5-6, 1991a. In Russian. ILO (International Labour Office). Economic Transformation and Employment in Hungary. Report of study mission, Geneva, 1991. IWTAB - Electronic Library, Prospect and Research Corporation, 1987-1988, special number 126-000-001,vers ion 1.26. Jones D. C. The Bulgarian Labour Market in Transition. International Labour Review, 1991, Vol.130, No.2, pp.211-226. Kabaj M . Elementy Pelnego i Racjonalnego Zatrudnienia w Gospodarce Socjalistycznej. (Elements of Full and Rational Employment in Socialist Economy). Warszawa, 1972. Kaldor N. A Model of Economic Growth. Economic Journal, December 1957, pp.591-624. Kanbur A. and Mcintosh J. Dual Economy Models: Retrospect and Prospect. University of Essex Discussion Paper No. 288,1986. Kapl M.,Sojka M.,Tepper T. Unemployment and Market-Oriented Reform in Czechoslovakia. International Labour Review, 1991, Vol. 130, No.2, pp. 199-210. Karl T.L. and Schmitter P.C. Modes of Transition in Latin America, Southern and Eastern Europe. International Social Science Journal, 1991, V.43, No.2,  266 p.269. Kemme D. Productivity Growth in Polish Industry. Journal of Comparative Economics, 1987,11, No. 1-20. Kerr W. The Decline of Pleasure,S\mon and Schuster, NY, 1972. King I. Sectoral Shift Models of Unemployment: Measurement Ahead of Theory. Bulletin of Economic ResearchA5:2>, 1993. Kiyono K. Post-war Industrial Policy of Japan: A Brief Economic Perspective. Paper presented to the conference on Industrial Policy, Ministry of Industry and Trade (MITI)and Japanese External Trade Organization (JETRO), Vienna, May 25-26,1992. Klodt H. Government Support for Reconstructing the East German Economy. Kiel. Working Paper, 450, 1990. Kolodko G.W. Economic Growth Cycles in the Centrally Planned Economies: A Hypothesis of "Long Cycle". BEBR, Faculty Working Paper, No. 1228, University of Illinois at Urbana-Champaign, February 1986. Kolodko G.W. From Output Collapse to Sustainable Growth in Transition Economies. The Fiscallmplications. IMF, Washington, D . C . , December, 1992. Kolodko G.W. From Recession to Growth in Post-Communist Economies: Expectations versus Reality. Communist and Post-Communist Studies, Vol.26, No.2, June 1993, pp. 123-143. Kolodko G.W. Reform, Stabilization Policies and Economic Adjustment in Poland. WIDER, Working Papers, World Institute for Development Economic Research of the United Nations University, Helsinki, 1989. Kurihara K. The Keynesian Theory of Economic Development. L . , 1959. Kuznets S. Economic Growth of Nations: Total Output and Production Structure, London, 1971. Kwiatkowski E. Polish Unemployment Demand Shocks. Mimeo, July 1992.  in Transition: the Role of Supply and  Lane D. Soviet Labour and Ethic of Communism. Full Employment and Labour Processin the USSR. Colorado, 1987.  267 Laszlo E. Changing Realities of Contemporary Leadership: New Opportunities for Eastern Europe. Futures, V.24, No.2, 1992. Layard R. Wage Bargaining and Incomes Policy: Possible Lessons for Eastern Europe.Centre for Economic Performance Discussion Paper #2, London School of Economics, London, 1990. Lewis A. Economic Development with Unlimited Supplies of Labour. Manchester School of Economics and Social Studies, Vol.22, pp. 139-91,1954. Lilien D. Sectoral Shifts and Cyclical Unemployment. Journal of Political Economy, Vol.90,1982, pp.777-93. Lin G. Ch. Development Paradigms in Transition: a Review of the Development Thought. Unpublished article, UBC Department of Geography, 1994. Lindbeck A. Macroeconomic Theory and the Labour Market. European Economic Review 36, 1992, pp.210-235. Linder S.B. The Harried Leisure Class, Columbia University Press, NY, 1970. Lipton M . Why Poor People Stay Poor: a Study of Urban Bias in World Development. L . , 1977. Logvinov L.D. VseobshchayaZanyatost'i Razdelenie Truda pri Sotsializme. (Full Employment and Division of Labour under Socialism), Moscow, 1972. In Russian. Lucas R. and Prescott E. Equilibrium Search and Unemployment. Journal of Economic Theory, 1974, Vol.77, pp. 188-209. Mandelbaum K. The Industrialization of Backward Areas, Oxford, 1961. Meade J. E. A Neo-Classical Theory of Economic Growth. 2nd ed. Unwin University Books, L . , 1962. Micklewright J. The Reform of Unemployment Compensation: Choices for East and West. European Economic Review, April 1991, V.35, No.2-3, p.419. Nee V. A Theory of Market Transition: From Redistribution to Markets in State Socialism. American SociologicalReview, 1989, V.54, No.5, p.663. Nesporova A. Labour Market Trends and Policies: The Chech and Slovak Federal Republic. In Structural Change in Central and Eastern Europe.Labour  268 Market and Social Policy Implications. OECD, Paris, 1993. Narodnoye Khoziaystvo SSSR (Economy of the USSR), Moscow, Goskomstat, various years. Novoye Russkoye Slovo, (New Russian Word), 1993, September 19. Nishimizu M . , Page J.M. Total Factor Productivity Growth, Technological Progress and Technical Efficiency Change: Dimensions of Productivity Change in Yugoslavia in \965-\91% Economic Journal, 1982, Vol.92, No.368, pp.920-936. Nuti D.M. Lessons From the Stabilization Programmes of Central and Eastern European Countries, 1989-1991.Economic Papers.EEC, Brussels, 1992. O'Donnell G., Schmitter P.C., Whitehead L . , eds. Transitions from Authoritarian Rule. Baltimore: The Johns Hopkins University Press, 1986. OECD Employment Outlook, OECD, Paris, 1988. OECD Employment Outlook, OECD, Paris, 1992. OECD. Occupational and Educational Structures of the Labour Force and Levels of Economic Development. Paris, 1970. Offe C. Capitalism by Democratic Design? Democratic Theory Facing the Triple Transition in East Central Europe. Social Research, 1991, V.58, No.4, p.865. Offe C. Das Dilemma der Gleichzeitigkeit, Demokratisierung und Marktwirtschaft in Osteuropa. (The Dilemma of Simultaneousness, Demacratization and Market in Eastern Europe). Merkur 45, 1991a, pp.279-291. Oxenstierna S. From Labour Shortages to Unemployment: the Soviet Labour Market in the 1980s, Almquist & Wicksell, Stockholm, 1990. Oxenstierna S. Labour Market Policies in the Baltic Republics. International Labour Review, 1991, Vol.130, No.2, pp.255-274. Pasinetti L. Structural Change and Economic Growth. Cambridge University Press, 1981. Paunov M . Labour Market Transformation in Bulgaria. Communist Economies and Economic Transformation, Vol.5, No.2, 1993, pp.213-228.  269 Penkin A. Bezrabotitsa v Rynochnoi Ekonomike. (Unemployment in Market Economy). Communist, No.5, 1990, p.88. In Russian. Petkov K. and Thirkell J.E.M. Labour Relations in Eastern Europe: Organizational Design and Dynamics. Routledge, London and NY, 1991. Pindyck R.S.and Rubinfeld A.S..Econometric Models and Economic Forecasts. Singapore, 1991. Pissarides C. Equilibrium Unemployment Theory. Basil Blackwell, Oxford, 1990. Pissarides C. Labour Hoarding in Industrial Countries: Concept and Measurement. In Employment and Unemployment in Economies in Transition: Conceptual and Measurement Issues.OECD, Paris, 1993. Pissarides C. The Role of Relative Wages and Excess Demand in the Sectoral Flow of Labour, Review of Economic Studies, Vol.45,1978, pp.453-67. Portes R. Structural Reform in Central and Eastern Europe. European Economic Review, April 1992, V.36, No.2-3, p.661. Prasnikar J., Svejnar J., Klinedinst M . Structural Adjustment Policies and Productive Efficiency of Socialist Enterprises. European Economic Review, 1992, 36, pp.179-199. Prebisch R. The Economic Development of Latin America and its Principal Problem, New York, 1950. Pudney S. Household Survey Data, Econometrics and Economic Transition. Economics of Planning 26: 55-80,1993. Rao V.K.R.V. Investment, Income and the Multiplier in an Underdeveloped Economy. IER, February 1952. Remes S. East European Futures Scenarios. Futures, V.24, No.2, p. 138. Rial J. Transitions in Latin America on the Threshold of the 1990s. International Social Science Journal, 1991, No. 128, p.286. Roeder P.G. Soviet Federalism and Ethnic Mobilization. World Politics, 1991, V.43,No.2, p. 196. Rosenstein-Rodan P.N. Natura Facit Saltum: Analysis of the Disequilibrium Growth Process. In Meier G . M . , Seers D. (eds.) Pioneers in Development,  270 Oxford, 1984. Ruccio D.F. Fordism on a World Scale: International Dimensions of Regulation. Review of Radical Political Economics, 1989, v.21, No.4, p.33. Sachs J.D. Crossing the Valley of Tears in East European Reform. Challenge, September-October 1991, V.34, No.5, p.26. Sacks S.R. Self-Management and Efficiency: Large Corporations in Yugoslavia. George Allen and Unwin, L . , 1983. Samorodov A. The Labour Market, Social Policy and Industrial Relations in the USSR under Transition. In Structural Change in Central and Eastern Europe: Labour Market and Social Policy Implications. OECD, Paris, 1993. Sandi A . M . Restoring Civil Societies in Central and Eastern Europe. Futures, V.24,No.2, p.110. Schmieding H. Die Ostdeutsche Wirtschaftskrise: Ursachen und Losungsstrategien. Institut filr Weltwirtschaft, Arbeitspapier, 461, Kiel, 1991. In German. Schopflin G. East Central Europe from Reform to Transition (book reviews). International Affairs, 1992, V.68, No. 1, p. 179. Schumacher P. Small Is Beautiful, NY, 1973. Seers D. The Birth, Life and Death of Development Economics. Development and Change, October, 1979. Selowsky M . Comment on "Policies to Move from Stabilization to Growth". In Proceedingsof the World Bank Annual Conference on Development Economics, 1990, The World Bank, Washington, D.C., 1991, pp.49-52. Shapiro C , Stiglitz J. Equilibrium Unemployment as a Worker Discipline Device. American Economic Review, 1984. Shcherbakov V.I. Renumeration of Labour in the USSR: Problems and Prospects. International Labour Review, 1991, Vol.130, No.2, pp.227-236. SiebertH. German Unification: the Economics of Transition". Economic Policy, April 1991. Sirakov S., Rainova M . , Radev T. Mezhdunarodni Sravneniya na Brutniya  271 Vtreshen Produkt: Metodologiya i Analiz. (International Comaprisons of the Gross Domestic Product: Methodology and Analysis). Pulishing House of the Bulgarian Academy of Sciences, Sofia, 1989. In Bulgarian. Solow R. Growth Theory: an Exposition. Oxford University Press, NY, 1970. Standing G., ed. In Search of Flexibility: the New Soviet Labour Market. ILO, Geneva, 1991. Standing G. Labour Market Policies for Economic Restructuring: Rapporteur's Report. In Structural Change in Central and Eastern Europe.Labour Market and Social Policy Implications. OECD, Paris, 1993. Standing G. Wages and Work Motivation in the Soviet Labour Market. Why a "BIP",not a "TIP" Is Required. International Labour Review, 1991, Vol.130, No.2, pp.237-254. Standing G., Tokman V.,eds. Towards Social Adjustment: Labour Market Issues in Structural Adjustment. ILO, Geneva, 1991. Stewart F. Technology and Underdevelopment. L . , 1977. Summers L.The Next Decade in Eastern and Central Europe. In Ch.Clague and G.C.Rausser, eds., The Emergence of Market Economies in Eastern Europe. Cambridge, MA and Oxford, UK, 1992, pp.25-34. Szelenyi I. SocialistEnterpreneurs.Embourgeoisementin Rural Hungary. Madison: University of Wisconsin Press, 1988. Sziraczki G. Employment Policy and Labour Market in Transition: from Labour Shortage to Unemployment. Soviet Studies, 1990, V.42,No.4, p.701-722. Terry S.M. Thinking about Post-Communist Transitions: How Different Are They? Slavic Review 52, No.2, 1993, p.333-337. Thirlwall A. P. Growth and Development with Special Reference to Developing Economies, L . , 1982. Tilly C. Big Structures, Large ProcessesJiuge Comparisons. NY: Russel Sage Foundation, 1984. The Transformation of Economies in Central and Eastern Europe:Issues,Progress and ProspectsThe World Bank, Washington, D.C., April 1991.  272 The Transition to a Market Economy, vol.1: The Broad Issues;vo\.2: Special Issues,OECD, Paris, 1991. Todaro M . Economics for a Developing World. An Introduction to Principles, Problems and Policies for Development. L . , 1977. Tomes I. Social Reform: a Cornerstone in Czechoslovakia's New Economic Structure. International Labour Review, 1991, Vol.130, No.2, pp. 191-198. Tyson L. The Yugoslav Economy System and its Performance in the 1970s. Research Series No.44, Institute of International Studies, University of California-Berkeley, 1980. Van Brabant J. Remaking Eastern Europe - On the Political Economy of Transition. Boston, Dordrecht, London, 1990. Vanek J. The General Theory of Labour-Managed Market Economies. Cornell University Press, Ithaca, NY., 1970. Vodopivec M . The Labour Market and the Transition of Socialist Economies. Working Paper WPS 561, The World Bank, December 1990. Wellisz S. The Lessons of Economic Reform: the Polish Case. Journal of International Affairs, Summer 1991,45, No.l, pp. 165-179. Wesolowski W. Transition from Authoritarianism Research, 1990, V.57, No.2, p.435.  to Democracy. Social  Whitesell R. The Influence of Central Plannings on the Economic Slowdown in the Soviet Union and Eastern Europe. Economica, 1985,52, pp.235-244. Winiecki J. Costs of Transition That Are Not Costs: On Non-Welfare-Reducing Output Fall". Rivista di Politico Economica. No.VI , June 1991, pp.85-94. Winiecki J. The Inevitability of a Fall in Output in the Early Stages of Transition to the Market: Theoretical Underpinnings. Soviet Studies, 1991a, V.43,No.4, p.669. WolfH. Miracle Prescriptions - Post-War Reconstruction and Transition in the 1990s. In The Economic Consequencesof the East, Centre for Economic Policy Research, London, 1992. Zycie Gospodarcze,No.36, September 06, 1992, p. 1.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0087006/manifest

Comment

Related Items