/ S 7/7 RESEARCHPAPER SERIES q Z ENTERPRISEBEHAVIORANDECONOMICREFORMS: A COMPARATIVE STUDYIN CENTRALANDEASTERNEUROPE AND INDUSTRIALREFORM ANDPRODUCTIVIrYIN CHINESEENTERPRISES RESEARCH PROJECTS OF THE WORLD BANK EasternEurope and FSU NumberEE-RPS#8 Enterprise Adjustment in Transition Economies: Czechoslovakia, Hungary and Poland Saul Estrin LondonBusinessSchool Mark Schaffer Performance Centerfor Economic LondonSchoolof Economics Inderjit Singh World bank Division Transitionand Macro Adjustment Policy ResearchDepartment World Bank Washington,D.C. CONTENTS ACKNOWLEDGEMENT ...................... i I. INTRODUCTION ......... 1 II. OVERVIEWOF THE EMPIRICALEVIDENCE..3 A. INDUSTRIAL OUTPUT AND SALES .......................... ..... 3 B. EXPORTS .7 C. EMPLOYMENT ............................................... 8 D. PRICES .............. 11 E. SUMMARY.................................. 16 III. INDUSTRIAL PROFITABILITY AND UNIT LABOR COSTS . .17 IV. ENTERPRISE RESPONSE: MICRO EVIDENCE .. 25 A. INTRODUCTION .25 B. DATA .26 C. ENTERPRISE EXPORT SUPPLY RESPONSE . 27 D. EMPLOYMENT AND WAGES IN TRANSITION .29 E. CONCLUSIONS.32 REFERENCES ........................ 46 LIST OF TABLES TABLE 2.1 5 TABLE 2.2 7 TABLE 2.3 .............................................. 10 TABLE3.1 .21 TABLE 3.2 .............................................. 23 TABLE4.1 .............................................. 34 TABLE 4.2 .............................................. 38 TABLE4.3 .............................................. 40 TABLE4.4 .............................................. 43 LIST OF FIGURES FIGURE 1. 4 FIGURE2 .8 FIGURE3 .11 FIGURE4 .... 12 FIGURE5 ............................................... 12 FIGURE6 .. 14 FIGURE7 ..... 14 FIGURE 8 ............................................... 15 FIGURE9 .. 17 FIGURE 10 ..... 18 ACKNOWLEDGEMENT The research projects on "EnterpriseBehaviorand EconomicReforms: A ComparativeStudy in Central and Eastern Europe", and "Industrial Reforms and Productivity in Chinese Enterprises" are research initiatives of the Transition and Macro Adjustment Division (PRDTM) of the World Bank's Policy Research Department and managed by I.J. Singh, Lead Economist. These projects are being undertaken in collaboration with the following institutions: for the project in China, The Institute of Economics of the Chinese Academy of Social Sciences (IE of CASS), The Research Center for Rural Development of the State Council (RCRD), and The Economic Systems Reform Institute (ESRI), all in Beijing; and for the projects in Central and Eastern Europe The London Business School (LBS); Reforme et Ouvertures des Systemes Economiques (post) Socialistes (ROSES) at the University of Paris; Centro de Estudos Aplicados da Universidade Cat6lica Portuguesa (UCP) in Lisbon; The Czech Management Center (CMC) at Celakovice, Czech Republic; The Research Institute of Industrial Economics of the Janus Pannonius University, Pecs (RIIE) in Budapest, Hungary; and the Department of Economics at the University of L6di, in Poland. The research projects are supported with funds generously provided by: The World Bank Research Committee; The Japanese Grant Facility; The Portuguese Ministry of Industry and Energy; The Ministry of Research and Space; The Ministry of Industry and Foreign Trade, and General Office of Planning in France; and the United States Agency for International Development. The Research Paper Series disseminates preliminary findings of work in progress and promotes the exchange of ideas among researchers and others interested in the area. The papers contain the views, conclusions, and interpretations of the author(s) and should not be attributed to the World Bank, its Board of Directors, its management or any of its member countries, or the sponsoring institutions or their affiliated agencies. Due to the informality of this series and to make the publication available with the -least possible delay, the papers have not been fully edited, and the World Bank accepts no responsibility for errors. The authors welcome any comments and suggestions. Request for permission to quote their contents should be addressed directly to the author(s). For additional copies, please contact the Transition and Macro Adjustment Division, room N- 11065, World Bank, 1818 H Street, N.W., Washington, D.C. 20043, telephone (202) 473-1442, fax (202) 676-0083 or 676-0439. The series is also possible thanks to the contributions of Donna Schaller, Vesna Petrovic, Cecilia Guido-Spano and the leadership of Alan Gelb. 1. INTRODUCTION The purpose of this paper is to provide an analysis of the adjustments of industrial enterprises to economic reforms in transition economies. We do this firstly by examining the available empirical evidence from Hungary, Czechoslovakia and Poland, the countries most advanced in the transition process. We go on to analyze more formally enterprise adjustment on the basis of data at the industry and enterprise level. We are concerned with impact of transition on enterprise profitability, and therefore ability to survive and invest. We also investigate the extent to which firms have responded to changing labor and product market conditions in a manner consistent with profit maximization. The paper is distinguished by being the first application of econometric techniques to an enterprise-level panel from an economy in transition. We hope that this will be the precursor of a series of analyses of the behavior of micro-economic agents during transition using panel-level data. The paper has four main findings. The first is that despite the disparity in industrial structure, preconditions to reform, sequencing and so forth, the actual process of industrial adjustment has been remarkably similar in Poland, Czechoslovakia and Hungary. There have been sharp falls in output and employment cumulating over the reform period in each country of broadly similar orders of decline. Our second focus of attention is enterprise profitability in transition. Estrin and Hare (1992) have stressed the importance of movements in profits as an indicator of the causes of the output decline. The profound unreliability of the profits data in all three countries led us to approach this by studying unit labor costs; one can conceive of a wage-profit trade-off in which, as unit labor costs rise, economic profits must be falling, whatever is indicated by the accounting data. We find that transition has been associated with rising unit labor costs in Hungary, sharply falling and then sharply rising unit labor costs in Poland and sharply falling unit labor costs in Czechoslovakia thus far. A central phenomenon driving this process is found to be the wedge betwen consumer and producer prices. The "big bang" approach to reform therefore appears to raise profitability but only in the short term. Once reform begins to take place at a gradual pace, unit labor costs rise and profits fall. Consistency between falling real wages and rising unit labor costs comes from increases in the wedge. The last two findings of the paper come via the use of panel data to analyze enterprise behavior in transition. We first address the question of labor demand responses in the first year of the transition. Though we have appropriate data only for Poland, the findings about the similarity of industrial adjustment give some confidence in the generality of our results. We successfully estimate a "neoclassical" labor demand function for the first year of transition in Poland, in which employment changes are positively related to output adjustments, but inversely associated with changes in labor costs. This is strong a priori evidence for the emergence of cost-minimizing behavior at the level of the firm. 1 2 Enterprise Adjustment in Transitional Economies Finally we focus on enterprise supply responses in transition, by estimated a reduced form export supply function. This yields the striking result that changes in output at the enterprise level are n= related to changes in prices. This brings into question the relevance of macroeconomic policy formulated on the assumption that firms will respond rationally to price signals after liberalization. It also strengthens the suggestion of micro- level perversity in enterprise adjustment during transition, though we are unable to identify any "credit-crunch" effects in the export supply equation. An important deficiency of the paper is its near-exclusive focus on the state-owned enterprise sector, whether at the industrial or micro-panel level of aggregation. This is because most of the available data encompass little or none of the emerging private sector. Nevertheless, the state sector still comprises the vast majority of industrial output and employment in these countries, and is likely to do so for some time.' The direction and magnitude of the response of the state sector is therefore crucial in evaluating the likely transition path. Moreover, because there are critical links between the response of industrial firms and macroeconomic policy, there exists the real possibility that macroeconomic prescriptions will fail because of weak or perverse responses at the enterprise level. We now briefly sunmmarize the main features and differences in the recent reforms in these three countries since 1989, the period that provides the focus for our analysis.2 In all cases reforms began with price liberalization. At one extreme is Poland which began to liberalize food prices in mid-1989. Following these reforms, in the second half of 1989 about 50% of turnover took place at free market prices. By the end of the year, near hyper-inflationary conditions prevailed. In January 1990, Poland implemented the so-called "Big Bang" package of reforms with the simultaneous implementation of stabilization and further liberalization measures. Most remaining price controls were lifted (90% of all prices were liberalized); direct controls on foreign trade were reduced leading to a virtually free trade regime; stabilization measures included a heavy devaluation of the exchange rate, "domestic convertibility" of the zloty, introduction of high nominal interest rates and wage controls and very strict measures to control the fiscal I There is an already existing private sector in Hungary and Poland, and a very small nascent one in Czechoslovakia. A complete analysis of the changes in the industrial sector would include these "private" activities. By all accounts these firms do not yet account for a large share of industrial output, but the most important feature of the transition is that their share is bound to grow. This limitation should be borne in mind throughout the paper. 2 For a detailed review of the stabilization reforms in Poland see Andrew Berg and Olivier Blanchard, "Stabilization and Transition: Poland 1990-1991"; for those in Hungary see Kemal Dervis and Timothy Condon, "Hungary: An Emerging Gradualist Success Story?"; and for Czechoslovakia see Karel Dyba and Jan Svejnar, "Stabilization and Transition in Czechoslovakia", all papers presented at the NBER Conference on Transition in Eastern Europe, Cambridge, Mass., February 26-29, 1992. EnterpriseAdjustmentin Transitional Economies 3 deficits. Indeed so successful was this last measure that by the first half of 1990 there was a large budget surplus. At the other end, although Hungary began a series of partial reforms two decades earlier, the first wave of major price liberalization took place only in early 1989. Price liberalization has continued at a steady pace, and was nearly complete by the end of 1991. Import and export licensing was abolished in late 1991 and by the end of 1991 approximately 90% of turnover was taking place at free market prices. In Czechoslovakia the reforms began essentially in mid-1990 with the removal of the "negative turnover tax" which was a subsidy on consumer goods. Thereafter Czechoslovakia undertook its own "Big Bang" liberalization starting on January 1991, lagging Poland by exactly a year. This included a lifting of nearly all price controls, and was followed in-miediatelyby sharp price increases as in Poland. The accompanying stabilization measures designed to bring down inflation to the pre-big bang levels were also similar - tight credit controls, tough fiscal controls alongside wage controls and a freeing up of the trade regime. Again about 90% of turnover now takes place at free market prices, and the stabilization measures were effective - there was a budget surplus in the first half of 1991 and prices had stabilized by mid-year. Further, and as a consequence of events not related to these policy induced reforms, trade between ex-CMEA countries began to collapse and starting in January 1991 whatever such trade took place was conducted in hard currencies. This affected all three countries severely, according to their exposure to CMEA trade. II. OVERVIEWOF THE EMPIRICAL EVIDENCE In this section we provide a brief factual overview of the developments in the industrial sector in the three countries for the period 1989-1991. This is to provide a background to the more detailed empirical analysis in Sections III and IV. We attempt to provide a comparative framework. This is important because, as we noted in the introduction, these three countries have undertaken reforms at very different paces, and at different stages in the reform process. A. Industrial Output and Sales The changes in total real industrial output in the three countries are shown in Figure 1, and by sector in Table 2.1. We note that despite diversity of policies, starting points, and timing of external shocks, all three countries find themselves at roughly the same point at the start of 1992: industrial output is 60-70% of its average 1989 level and there are signs that the bottom has been reached. The decline has been general across sectors, though with some variation. 4 Enterprise Adjustment in Transitional Economies ~~~ a , c ~~ ,,, ~ .~, ~ r c5 , E _5, G ,, 9 , p.., ,,,, .. . I : .' .. 9 ,. . .. . . trv !~~~~~~~~~r' X3 aCd n, _ ,r r X Figure 1 During the past two years real aggregate industrial sales have declined by about 40% in Poland. There was a huge and sudden drop in sales of approximately 30% in January 1990 associated with the "'big bang", after which output was roughly steady for the rest of the year. The further large, but gradual, contraction beginning in early 1991 was associated with the collapse of the CMEA trade regime and re-tightening of monetary policy. There are some signs of a pickup in industrial output in the first quarter of 1992. The decline in industrial output in Poland is huge even if we allow for private sector sector growth (which was substantial in 1991 - nearly 50% growth in recorded industrial output in the private sector proper, though a smallish portion of this is due to the inclusion of newly privatized industrial enterprises). In Hungary, there was a 4% drop in industrial output in 1989, a 7% drop in 1990, and then a very large fall of 22% in 1991. As in all things, output has declined gradually, with no sudden movements as in Poland. Production bottomed out in the third quarter of 1991 at about 65% of the average 1989 level, remarkably close to the Polish level. As in Poland, the state sector is dominant, and inclusion of the private sector would not change the aggregate picture. Enterprise Adjustment in Transitional Economies S In Czechoslovakia, the decline in industrial production has been more recent. There was a small 4% drop in industrial output in 1990, followed by a very large decline of 23% in 1991. The drop in output was gradual, as in Hungary. Industrial production had bottomed out in the third quarter of 1991 at 65-70% of the average 1989 level, as in Poland and Hungary. The private sector is of recent origin only and remains very small. TABLE 2.1: Changes in Output & Employment by Sub-Sectors HUNGARY Output: Level Employment: Level Growth (% p.a.) (1988 Growth (% p.a.) (1988 =100) =100 Year 1989 1990 1991 1991 1989 1990 1991 1991 Total -1.0 -9.2 -21.5 70.6 -1.7 -9.1 13.4 77.4 Industry by Branch: Mining -5.2 -11.8 -10.9 74.5 -8.0 17.3 15.9 64.0 Electricity 2.2 0.2 -8.0 94.2 -1.1 4.7 -5.2 98.2 Metals 4.4 -19.0 -32.7 56.9 -8.1 17.9 -18.4 61.6 Engineering 0.2 -16.2 -34.9 54.7 -1.4 -9.9 16.8 73.9 Building Materials -1.6 -5.0 -33.0 62.6 -1.8 -4.9 12.8 81.4 Chemicals -3.9 -5.4 -18.5 74.1 2.3 -4.1 -7.3 90.9 Light Industry -4.8 -11.7 -24.9 63.1 -0.3 10.3 13.5 77.4 Food Processing 1.0 -0.9 -9.7 90.4 0.2 -3.3 -7.1 90.0 Havi K6zlemdnyek, Sourc: Statisztikai various issues;and own calculations. 6 Enterprise Adjustment in Transitional Economies TABLE 2.1 (continued) POLAND Output: Level Employment: Level Growth (% p.a.) (1988 Growth (% p.a.) (1988 =100) =100 Year 1989 1990 1991 1991 1989 1990 1991 1991 Total -0.5 -24.2 -11.9 66.4 0.3 -4.9 -7.3 88.4 Industry by Branch: Fuel and Power -2.4 -22.1 -8.4 69.6 0.8 -8.2 -6.4 86.6 Metals -4.8 -19.7 -22.4 59.3 -3.8 -4.6 -8.1 84.3 Electro-engineering 0.7 -22.0 -22.4 61.0 -1.4 -4.8 11.0 83.5 Chemicals 2.6 -24.6 -13.5 66.9 2.1 -6.4 -6.4 89.4 Minerals 5.5 -21.5 -2.6 80.7 0.1 1.1 -3.1 98.1 Wood and Paper 6.9 -24.9 -1.4 79.2 4.0 -2.2 4.6 106.4 Light Industry 3.3 -33.8 -13.0 59.5 2.7 -5.4 -13.1 84.4 Food Processing -5.9 -23.7 -0.9 71.2 -0.4 -0.2 5.0 104.4 RocznmkStatvstvczny Przemvs u 19, 191; Biuletvn Statzstvcznv, various issues; Informacja o sytuaci spoleczno-gospodarczej kraju, January 1992; and own calculations. TABLE 2.1: Changes in Output & Employment by Sub-Sectors (continued) CSFR Output: Level Employment: Level Growth (% p.a.) (1988 Growth (% p.a.) (1988 =100) =100 Year 1989 1990 1991 1991 1989 1990 1991 1991 Total 0.7 -3.5 -24.7 73.2 -0.6 -3.2 12.1 84.6 Industry by Branch: Fuel and Power -0.3 -4.1 -5.7 90.2 -0.4 -3.8 -7.5 88.7 Metals 0.5 -1.8 -26.0 73.0 -0.9 -3.0 10.1 86.4 Electro-engineering 0.8 -3.4 -32.4 65.8 -0.8 -5.5 13.5 81.1 Chemicals 0.7 -8.7 -22.7 71.1 0.0 -2.2 -9.4 88.6 Minerals 2.2 -3.7 -30.3 68.6 0.0 -4.5 12.6 83.5 Wood and Paper 0.9 -0.8 -22.4 77.7 -1.2 -3.1 11.1 85.2 Light Industry 1.1 -1.1 -36.0 64.1 -1.0 -2.9 14.6 82.1 Food Processing 2.5 -2.0 -16.4 84.0 0.5 -1.4 -10.3 88.9 DurCe: i Ukazatele v Prtimyslu v Roce 1990; 'Stri?kiLa Statistlka Ro enka, 1990, 1991; "Vybran a Efektivnost teskoslovaenskeho Pr.myslu v Roce 1991"; and own calculations. EnterpriseAdjustmentin Transitional Economies 7 B. Exports The export performance of these countries has been well covered by Rodrik (1992a, 1992b).3 The pattern has been similar in all three countries: collapse of CMEA exports (and imports), and rapid growth in exports to (and imports from) hard currency countries. It is useful to note here that the decline in industrial output cannot be attributed simply to the collapse of the CMEA exports as is sometimes argued because this decline was partially offset in all three countries by an increase in hard currency exports. This can be seen from Table 2.2, which gives some basic data on changes in industrial production and exports in all three countries. In Poland and the CSFR the 1991 decline in exports was much less than the fall in total industrial sales, and in Hungary the fall in exports was not much greater than the fall in total sales. The adjustment of hard currency exports in response to the massive changes in the foreign trade regime associated with transition is one of the few bright spots in the early years of transition. In Section IV we investigate formally whether changes in export supply at the enterprise level represent a rational economic response by firms to the changing economic environment, or the consequence of extreme supply response and chance. TABLE 2.2: Industrial Sales and Exports % growth per annum 1989 1990 1991 Poland: Industrial sales -0.5 -24.2 -11.9 Total exports 0.2 13.7 -1.4 Hungary: Industrial sales -3.9 -9.5 -17.9 of which, exports 0.7 -16.5 -21.9 CSFR: Industrial sales 0.2 -0.8 -30.3 of which, exports -0.7 -6.8 -4.8 Source: National statistical otfices. 3 His general findings are that trade export performance has been impressive in all three countries and that import booms are under way; that there is no evidence that exporters who lost CMEA markets were able to find Western marketsfor the same products;that the Soviettrade shock was very seriousaccountingfor lossesamountingto 7-8 % of GDP in Hungaryand Czechoslovakia and 3 112% of GDP in Poland; that the export performance is attributableto the exchangerate policies althougha declinein domesticdemandalso playeda part and that so far there is little evidencethat the very open trade regimeshavebeen able to imposeany price desciplineon domesticindustriesor forcedthem to restructure.See Dani Rodrik, "ForeignTrade In EasternEurope's Transition", paper presented at the NBER Conference (op. cit). 8 Enterprise Adjustment in Transitional Economies C. Emplovment The relevant data on quarterly changes in industrial employment are shown in Figure 2. Industrial employment has fallen gradually and substantially in all three countries. Most labor has been shed in Hungary, and the least in Poland. These quarterly data are somewhat problematic, however, since individuals leaving the state sector for the private sector often go to small firms that don't have to report monthly or quarterly. E u5 3 89 CC3,Xl '99 30 3 9 9 CYQ IC)03 19 9 G c a: d a a r y _C Fp Figure 2 Employment has fallen less than real output in all three countries, so labor productivity has also declined. Several explantions have been offered. One is that insider power plays a role as firms try to keep on workers unless survival is at stake. Absence of effective owners (the state) means that firms in transition have become, to varying degrees, "worker-controlled firms". Poland is the best example of this effect, where the machinery of Workers' Councils and the way they operate formalized worker control over state firms. One can argue that a similar situation exists also in Hungary and Czechoslovakia, although to a lesser degree. We return to the question of the relationship between changes in output and employment in transition by estimating neoclassical labor demand functions in Section IV. Enterprise Adjustment in Transitional Economies 9 The evidence on labor turnover in both Hungary and Poland suggests that labor shedding is largely via attrition (see Table 2.3 for detailed Polish labor turnover figures by sub- sector). Unemployment data show that starting from zero or very low levels, recorded unemployment stood at 7% in Czechoslovakia, 9% in Hungary and 12% in Poland by the end of 1991. 10 Enterprise Adjustment in Transitional Economies TABLE 2.3: Labor Turnover in Socialized Industry in Poland ~1 Hirings Departures IndUStrY Number As a % of Number As a % of (thousands) employment (thousands) employment 1989 1990 1989 11990 1989 j_1990 1989 1990 Total 698.7 485.0 15.8 10.7 878.7 937.8 19.3 22.0 Energy and 99.5 76.6 14.3 11.5 127.7 115.6 18.0 17.5 power 70.3 51.0 13.8 10.9 95.7 90.7 18.5 19.6 Coal 8.2 6.2 14.1 10.0 9.1 6.4 15.3 10.9 Fuel 21.1 19.4 16.5 14.6 22.8 18.4 17.1 12.3 Power 26.4 20.8 12.1 9.0 35.0 30.9 15.5 14.2 Metals 18.0 14.7 11.6 9.1 25.5 22.4 16.0 14.7 Iron and steel 8.5 6.1 13.2 8.9 9.5 8.5 14.2 13.1 Non-ferrous 221.3 155.8 15.5 10.6 290.1 320.8 19.9 23.5 Electro-engineeri 42.4 26.6 16.5 10.1 54.2 60.3 20.7 24.0 ng 68.2 49.2 14.7 10.4 95.9 102.4 20.5 24.0 Metal products 10.7 7.1 13.7 8.1 15.4 16.6 19.5 22.2 Engineering 53.0 46.4 14.5 12.8 74.0 76.8 20.0 22.4 Precision insts. 47.1 26.5 17.9 9.1 50.6 64.7 18.1 24.0 Transport 44.5 24.7 15.0 7.6 53.6 48.9 17.6 16.7 equip. 43.8 32.6 19.4 13.5 54.8 56.3 23.6 25.0 Electronics etc. 29.0 22.7 20.4 15.0 36.9 39.4 25.3 27.3 Chemicals 9.6 6.5 17.4 11.1 11.5 10.8 20.4 20.5 Minerals 5.3 3.4 17.8 10.1 6.3 6.1 20.5 20.9 Building 41.2 27.9 18.6 12.1 50.9 56.9 22.6 26.7 materials 33.0 20.5 19.1 11.2 41.1 45.2 23.5 27.0 Glass products 8.2 7.4 16.7 15.4 9.8 11.8 19.5 25.8 Ceramics 108.4 68.3 14.7 8.5 140.0 184.4 17.8 26.3 Wood and paper 54.4 28.3 14.5 6.5 69.9 88.4 18.3 24.8 Wood 31.0 24.2 15.0 11.6 41.9 58.6 16.8 29.9 Paper 23.0 15.8 14.8 9.5 28.1 37.4 18.6 25.4 Light industry 82.0 63.1 19.2 13.9 84.1 86.8 25.1 19.2 Textiles Clothes Food -l:RoczMikSt-atystycn TT urzFn o,T 1991_____ ________ ___ EnterpriseAdjustmentin Transitional Economies 11 D. Prices Figures 3-5 show the changes in consumer and producer prices in the three countries. The most dramatic changes have been in Poland, where both consumer and producer prices increased fifteen-fold between the first quarter of 1989 and the first quarter of 1990. The stabilization program slowed these price increases dramatically, but following the "big bang" in 1990, prices have continued to increase and at the end of 1991 stood at a level almost twice that at the start of 1990. After the "big bang" consumer prices rose faster than producer prices, creating a "wedge". 'ABE Cr1- 0. Firms whose profit orientation is slight will display no significant behavioral response to higher prices. Finally, if Polish firms have become effectively self-managed in the transition period, then we might find evidence of the traditional perverse supply response (see Ward (1957), Domar (1966), Meade (1972)) with O% < 0. EnterpriseAdjustmentin Transitional Economies 29 Equation (1) is estimated using enterprise level data for the first year of transition, 1990. against pre-reform data for 1989. The results are reported in Table 4.2. The equation provides a reasonable explanation of the change in export volumes, with an adjusted R2 of 0.48 and an F statistic of 7.41. However, almost all the explanation derives from the industry dummies controlling for firm specific cost and demand elasticity changes. The estimated coefficient on prices is positive, but very low (less than unity) and statistically insignificant. This implies that the dramatic changes in relative export prices in 19906 played no significant role in enterprise adjustment during the transition. As such, it is fairly striking evidence against the view that firms in the state sector were operating according to the principles of profit maximization during the first year of reform. It is also surprising that no significant explanation is offered by the export orientation and diversion variables, which might have been expected to segment the data set according to managerial motivation and experience with markets. Our findings therefore provide no evidence of a positive supply response to profitable new opportunities, or indeed negative responses to loss-making areas, by the Polish state-owned enterprise sector in the first year of transition. D. Employment and Wages in Transition We normally consider adjustment in factor markets to be slower than in product markets. However, even though the Polish state sector does not appear to have adjusted output in line with prices, transition may still have led them in the direction of cost minimization, in the sense of adjusting employment in line with output, and inversely with cost. Unlike positive supply responses, such behavior does not rely on profit maximization but could be consistent with alternative objectives for finns in transition (prior to privatization), e.g. size maximization (see Estrin and Hare (1992)). We make no attempt to break new ground in the formulation of employment and wage equations. We therefore draw on the standard literature (see e.g; Layard, Nickell and Jackman (1990)) to specify a labor demand and wage system of the general form, Ld= Ld (X W, Z) (2) w= w(II, Z, R,I) (3) where Ld denotes labor demand, w is wages, Z is a vector of variables controlling for the firm's exposure to foreign trade, measured separately for hard and soft currency areas, H 6 The standard deviation of hard currency export price growth (CPP3E2) is 0.17. (The coefficient of variation is 0.09, but only because the mean is 1.89, corresponding to a six-fold increase in the price level on average.) 30 Enterprise Adjustment in Transitional Economies is company profits and equation (3) also includes regional (R) and industrial (I) dummy variables. As before, Z is included to capture the possible "learning by doing" and motivation effects of foreign trade. In the wage equation, enterprise profits are included to capture "insider" effects while external market pressures, as well as labor compositional and skill factors, are controlled by the industry and regional fixed effects. Equations (2) and (3) are estimated in loglinear first difference form on enterprise level data, 1989-90. Hence, lnAL, = Bo + BElrAX. + BlnAw. + B3AZ4+ )2i and 8 6 (5 lAw. = yo + y,A(profit/sales)1989 + jyR. + E k=lY3kik + 43i ) w here & 21 and 43i are independently distributed error terms. Our basic hypothesis is that firms will adjust employment according to the changes in output (B 1 > 0). However, it should be noted that this is not the only possible outcome. If Polish state-owned firms in the early phase of transition were not motivated to minimise costs, there is no reason a priori to expect any positive relationship between output and employment; for example, if budget constraints are soft and managers aim primarily at maintaining employment, there is no reason to expect a relationship between the two variables. An insignificant B 1 would therefore imply widespread perverse behaviour at the enterprise level, and would have considerable ramifications for macroeconomic policy and the speed of the privatization programme. On the other hand a negative significant B2 would strongly imply the emergence of some cost-minimizing considerations in enterprise decision-making. Wages can probably be regarded as exogenous for the firm in a planned economy, and perhaps even in a market socialist economy such as Poland in the 1980s. However, in a market system one typically assumes wages to be in part also determined by enterprise- specific factors influencing the company's ability to pay (see Krueger and Summers (1988) Nickell and Wadhwani (991)). In the absence of any information about unionization, bargaining procedures and pay setting structures in Poland post-reform, we followed Carruth and Oswald (1990) in assuming that insider forces are best proxied by enterprise profitability. The emergence of insider forces in wage setting would imply 'y, > 0. Outsider effects are picked up by Y2 and y3. Equations (4) and (5) are estimated simultaneously using two stage least squares. Both output and profits are endogenous to this system, and are instrumented. Output is EnterpriseAdjustmentin Transitional Economies 31 instrumented in the labor demand equation because output and employment are jointly determined. 7 Instruments for the profit equation focus on changes to both the revenue and the cost sides. Revenue effects are identified through export prices, changes in the capital stock to proxy for output effects, as well as changes in the ratio of consumer to producer prices for a good to reflect the changes in subsidy/tax regime during transition (when the change in the price wedge is greater than unity, the authorities are removing subsidies from firms so we would expect profits to fall). The firm's domestic market share in 1989 (domestic sales as a percentage of all domestic sales at the 3-digit level) also enters, to capture market power effects. Cost side factors are proxied by the share of material and energy costs to sales in 1989, to pick up the impact of relative factor price changes. Finally, the potential impact of the credit squeeze on profits is proxied by the turnover credit to inventories and dues ratio at the end of 1989. This last variable is treated as endogenous and is instrumented on the end-1988 "own funds" to inventories ratio.' The results are reported in Tables 4.3 and 4.4. The wage equation provides a good explanation of the changes in wages during the first year of the transition, the bulk of which comes from "outsider" forces proxied by regional and industry dummies. The industry dummies can be regarded as indicating the impact of emerging industry-level bargaining on wages, as well as reflecting industry-specific job-type, skill and labor supply characteristics. Regional labor market effects are already significant in the first year of transition, which is encouraging for the emergence of local labor markets, but their impact on the level of wages is modest. There is some evidence for the impact of insider forces on wages, but the relationship is perverse; firms with higher changes in profits raised wages less than their apparently more successful counterparts. This might be seen as consistent with a credit crunch interpretation of the perversities in the profit 7 Instruments include (the change in) prices, capital stock, export shares and the ratio of enterprise "own funds" to inventories at the end of 1988 (the greater are "own funds", the less turnover credit the enterprise needs), to pick up the potential effects of a credit squeeze on production (see Calvo and Coricelli (1992)). 8 We were able to explain some 21% of the variance, via changes in both output prices and input costs. On the revenue side, the estimated coefficient on the wedge conforms to expectations, with profits inversely correlated to subsidy cuts. The market power and export price variables are insignificant, however. The change in capital stock has a perverse sign and is also insignificant. More surprising is the impact of the material and energy intensity variables; these are both positively associated with the change in profits, rather than inversely as theory would lead us to expect. The finding of a positive and significant coefficient on material intensity is robust to changes in formulation of the equation, and holds even if (inflation-biased) historical cost profits are used as the dependent variable. There are two possible explanations here. If firms continued to mark up their prices to costs, initial input prices rise in transition and the change in profits could be positively correlated through change in the output price. Alternatively there is some evidence for a Calvo-Coricelli credit squeeze in the profit equation; the credit exposure variable is positively associated with the change in profits at the 1% significance level. It might have been that firms with higher energy and material intensity were more exposed to the credit squeeze, and were therefore forced to increase profits (by whatever means) more than firms with lower levels of credit exposure. 32 Enterprise Adjustment in Transitional Economies equation. Firms with high credit exposure, which is related positively to the change in profits (see footnote 9), react by holding down wages more their less constrained counterparts, in order to boost liquidity. The labor demand equation suggests that firms in transition adjust employment according to the same factors as their Western counterparts. Thus there is a positive association between the change in employment and in output, though at 0.21 the estimated short-run elasticity is rather low; with several years of data, one would expect the estimated long- run elasticity to be rather higher. There is also convincing evidence of cost-minimizing behaviour, coming from the negative estimated coefficient on wage changes. Once again the coefficient of 0.16 is rather less than one might observe for Western firms, but is strongly significant. There is however no evidence that employment adjustment is in any way affected by exposure to foreign trade (hard (NCCE) or soft (NCCE) currency) or by the share of export sales going through foreign trade companies (FTEE). This finding is consistent with our earlier output supply results. E. Conclusions Our econometric work represents a first cut at the enterprise level. Our approach has been to use simple-minded neo-classical theory as the intellectual foundation to modelling enterprise adjustment. One might argue that the problem with this is that it abstracts from one of the fundamental issues in transition - managerial motivation. In fact this is not correct; if firms do not respond in the manner predicted, this says something about enterprise motivation. Perversity of response could be consistent with labor management. By focusing seperately on the product and labor market, our analysis was also structured to distinguish between two interpretations of enterprise response; cost minimization and profit maximization. None of the firms in our sample had been privatized, so our results also cast some light on the question of-whether the freeing of prices and commercialization of the state enterprise sector leads to adequate adjustment by firms to the new market conditions, or whether privatization is necessary from the outset. We find very little evidence of purely perverse behaviour. Polish state-owned firms may be characterized by weak management and powerful workers, but that does not appear to have led to Ward-type income maximizing behaviour. Output responses in the short run are not perverse. Outsider forces appear to influence wages. Employment does adjust in line with output, though the elasticity is low. The scale of previous exposure to foreign trade is never a significant determinant of company behaviour. However, we find more evidence for cost minimization than for profit maximization. The significant determinants of labor demand are the same as those pertaining in Western economies. The inverse relationship between wages and employment, in the context of endogenous wage determination, is particularly striking. However, when one turns from factor demand to product supply, the picture is less encouraging. The simple correlation between export supply and price is in fact negative; once firm specific demand and cost factors have been controlled for, the two variables are not significantly associated. Polish EnterpriseAdjustmentin Transitional Economies 33 state-owned firms have failed to increase supply to activities whose relative price and therefore profitability have increased; exports have not been significantly reduced in activities whose relative profitability fell. To explain these apparently inconsistent findings, we need to consider the management of the typical state-owned firm throughout Eastern Europe. The managerial function was concentrated on production engineers. Our findings suggest that management took the reforms seriously, in the sense of altering their behavior as a consequence. However, these changes were more marked in functions in which existing managers were already skilled, such as reorganizing production and adjusting inputs in line with output. Managers have been much less successful in responding in output markets, where the necessary responses depend more on skills in marketing and finance. One may conclude from our findings that Polish state-owned firms in the first year of the transition were not responding in a satisfactory manner to the problems and opportunities opened up by the reform program. It is an open question as to whether a major deficiency of the program was therefore the failure to prioritize privatization of the state sector from the outset, or whether the inadequate human capital found in these firms would lead to unsatisfactory performance in the short run whatever their form of ownership. 34 Enterprise Adjustment in Transitional Economies TABLE 4.1: Description of Sample of Polish Firms (Means and standard deviations) (1) ~(2) (33 (4) (5) 6 Ind. No. Sales % change Empl. % change Monthly k change of in 1989 in sales in 1989 in empl. wage in real f_rms in 1989 product wage 1 11 352747 -19.6 6734 -2.9 280242 -38.6 371292 18.2 7948 3.5 52685 10.8 2 42 103465 -23.4 5298 -6.6 219579 -27.3 94674 32.1 5116 7.6 25307 15.2 3 20 144028 -18.3 3953 -2.0 233947 -40.4 82290 17.0 2351 4.0 28443 10.0 4 10 74133 -15.7 2725 -6.0 213666 -28.4 46897 22.6 1010 4.5 34117 12.1 5 14 70756 -34.1 3583 -11.3 212120 -16.3 27499 14.4 1444 7.6 20072 5.7 6 14 86738 -28.4 2586 -1.8 227813 -35.6 56029 13.2 2436 6.9 31403 14.0 Total 111 126600 -22.1 4408 -5.4 227745 -31.0 154617 23.0 4374 6.9 35317 14.4 Enterprise Adjustment in Transitional Economies 35 TABLE 4.1: Description of Sample of Polish Firms (continued) (Means and standard deviations) (7) (8) (9) (10) Ind. No. Hist. cost Hist. cost Cash flow Cash flow of profit/sales profit/sales /sales in /sales in firms in % in i % in 1989 W in 1990 in 1989 in 1990 1 11 34.4 23.1 19.6 8.7 6.1 6.0 8.0 9.3 2 42 44.4 29.7 27.6 11.0 15.2 16.2 14.5 20.9 .3 20 42.8 26.5 29.5 14.7 10.6 8.3 10.2 8.5 4 10 50.7 25.1 37.3 13.3 10.3 12.0 12.8 14.9 5 14 36.3 11.7 19.0 -6.6 11.9 10.1 16.1 14.2 6 14 34.7 14.6 -2.2 -8.4 20.1 8.2 30.9 21.1 Total 111 40.2 24.6 23.1 9.0 13.2 11.6 17.0 15.9 36 Enterprise Adjustment in Transitional Economies TABLE 4.1: Description of Sample of Polish Firms (continued) (Means and standard deviations) (11) (12) (13) (14) Ind. No. Material Material Energy Energy of intensity intensity intensity intensity firms in 1989 in 1990 in 1989 in 1990 1 11 64.4 68.5 3.0 4.2 11.5 9.8 3.7 4.8 2 42 45.5 49.6 1.8 3.0 11.9 12.2 0.9 1.7 3 20 62.8 62.2 4.8 6.3 14.8 9.6 3.4 4.0 4 10 52.1 61.2 2.8 4.9 9.0 9.7 2.3 3.2 5 14 43.7 55.9 1.2 2.8 10.8 8.4 0.9 1.9 6 14 89.6 69.6 0.8 1.4 14.7 8.7 0.7 1.1 Total 111 58.4 60.9 2.7 4.0 18.1 12.8 2.8 3.7 Enterprise in Transitional Adjustment Economies 37 TABLE4.1: Description of Sample of Polish Firms (continued) (Means and standard deviations) (15) (16) (17) (18) (19) Ind. No. Export/ Export/ Hard currency Hard currency Growth of of sales sales export/sales export/sales exports firms ratio in ratio in ratio in ratio in in 6 % in 1989 % in 1990 % in 1989 % in 1990 1 11 29.9 42.4 17.6 34.2 49.0 16.3 14.5 10.2 14.0 95.8 2 42 36.0 44.0 18.2 31.2 -10.4 19.9 22.2 11.7 22.6 60.6 3 20 26.2 44.2 24.1 43.2 114.9 11.2 16.0 12.1 17.1 115.7 4 10 21.0 31.5 20.2 31.0 15.5 12.4 14.0 10.7 12.8 35.6 5 14 17.6 34.6 13.7 24.2 7.9 7.8 12.8 7.4 10.5 44.3 6 14 46.2 51.2 44.5 49.8 -14.1 22.5 23.4 19.8 22.7 83.2 Total 111 31.1 43.1 21.3 36.2 28.2 17.8 18.0 14.1 18.8 94.4 38 Enterprse Adjustment in Transitional Economies TABLE4.2: Expor Supply Response Source Ss df MS Number of obs = 70 -------- + -------------- ----- F( 10, 59) = 7.41 Model 15.4176167 10 1.54176167 Prob > F = 0.0000 Residual 12.2796391 59 .208129476 R-square = 0.5566 --------- +------------------------------ Adj R-square = 0.4815 Total 27.6972557 69 .401409503 Root MSE = .45621 Variable j Coefficient Std. Error t Prob > |t| Mean -- - - - - - - - +---- - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ -__ _ CE_SC | .2076983 -- - - - - - - - +--- - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - _ _-__ - - - - _ - -- - _-_-_-__ _ _-_ CPP3E2 .4043154 .4910949 0.823 0.414 1.892548 HCDIV .0601937 .1204594 0.500 0.619 .0218898 CCE89 -. 8902907 .5809091 -1.533 0.131 .2121751 NCCE89 -1.0025 .6609087 -1.517 0.135 .0529984 FTEE89 -. 6891265 .5881 -1.172 0.246 .1214107 IND21 1.122443 .3004029 3.736 0.000 .1571429 IND22 -.1309057 .2806045 -0.467 0.643 .2285714 IND23 .4436616 .2826714 1.570 0.122 .1714286 IND24 -. 1192335 .2720151 -0.438 0.663 .1142857 IND25 -. 0686805 .2560588 -0.268 0.789 .1857143 IND26 (dropped) _cons -. 429247 1.070271 -0.401 0.690 1 --------- +-_--_________________--________--________--_____________-___ Variables: CE_SC (Exponential) growth in real exports to the hard currency area, 1989-90. 'PP3E2 (Exponential) growth in nominal hard currency export price index (3-digit level). Calculated from data in Handel-Zagraniczny, styczein-grudziefn 1990, GUS, May 1991. HCDIV Change in real level of ruble exports 1989-90, divided by level of hard currency exports in 1989. CCE89 Hard currency exports in 1989 divided by total sales in 1989. NCCE89 Ruble area exports in 1989 divided by total sales in 1989. FTEE89 All exports via foreign trade companies in 1989 divided by total sales in 1989. Enterprise Economies Adjustmentin Transitional 39 TABLE 4.3: Wage Equation (2SLS) Source SS df MS Number of obs = ill ------ +------------------------------ F( 14, 96) = 4.41 Model 1.64322291 14 .117373065 Prob > F = 0.0000 Residual | 3.04307623 96 .031698711 R-square = 0.3506 ------- +------------------------------ Adj R-square = 0.2559 Total | 4.68629915 110 .04260272 Root MSE = .17804 Variable j Coefficient Std. Error t Prob > |t| Mean --------- +-------------------------------------------------------- CWAGE | - .3819789 -__ _--__ +-__ __ ___ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _-__ __ _ _ _ __ _ _ __ _ DPROFS -. 446048 .2144495 -2.080 0.040 -. 1144015 CCE89 -. 0962175 .1548707 -0.621 0.536 .2065707 IND21 - .1256458 .0831717 -1.511 0.134 .0990991 IND22 - .0366515 .0672085 -0.545 0.587 .3783784 IND23 -. 1667198 .0740182 -2.252 0.027 .1801802 IND24 (dropped) IND25 .1762533 .0772192 2.283 0.025 .1261261 IND26 -. 0441217 .0923003 -0.478 0.634 .1261261 REGION1 -. 205496 .1245073 -1.650 0.102 .0630631 REGION2 .0364142 .113832 0.320 0.750 .1441441 REGION3 .001409 .1117921 0.013 0.990 .2792793 REGION4 -. 0854123 .1403334 -0.609 0.544 .036036 REGION5 -. 0464772 .1132438 -0.410 0.682 .2162162 REGION6 (dropped) REGION7 -. 0304341 .1139692 -0.267 0.790 .1531532 REGION8 .1605168 .1221113 1.315 0.192 .0810811 -cons -. 3613478 .1163198 -3.107 0.002 1 --------- +------------------------------------------------------------__ Variables: DPROFS Change in real net cash flow (nominal net cash flow in 1990 deflated by the PPI - net cash flow in 1989) normalized by sales in 1989. CCE89 Share of convertible currency exports in total sales in 1989. IND21-6: 2-digit industry REGION1-8: Regional dummy dummy variables: variables. 1. Metallurgy 1. Western Poland 2. Electro-machinery 2. West-central Poland 3. Chemicals 3. South-west Poland 4. Wood and paper . 4. Gdafisk and 5. Light industry surroundings 6. Food processing 5. Warsaw surroundings 6. Warsaw 7. South-east Poland 8. North-east Poland 40 Enterprise Adjustment in Transitional Economies TABLE4.4: Labor Demand Equation (2SLS) Source | SS df MS Number of obs = 111 ------- +------------------------------ F( 6, 104) = 4.67 Model .087832696 6 .014638783 Prob > F = 0.0003 Residual .543131417 104 .005222417 R-square = 0.1392 --------- +------------------------------ Adj R-square = 0.0895 Total .630964113 110 .005736037 Root MSE .07227 Variable Coefficient Std. Error t Prob > |t| Mean -- - - - - - - --+--- - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ -- - _- CL | -. 0533228 -______+-_____-___ _ ______ ___ __ ______-_________-_ _ -_ CWAGE -.1599871 .0464665 -3.443 0.001 -.3819789 CGVO .2081915 .0548455 3.796 0.000 -.2486044 WEDGE -.0510289 .0828775 -0.616 0.539 .015087 CCE89 -. 0256977 .0582745 -0.441 0.660 .2065707 NCCE89 -.045979 .0520996 -0.883 0.380 .0907687 FTEE89 .070191 .0531819 1.320 0.190 .1001673 -cons -.0594563 .0271902 -2.187 0.031 1 ------- +--_--------------------------------------------------------__.__ Variables: CL % change in employment. CWAGE % change in real product wage (nominal wage deflated by appropriate 3-digit industrial sales price index). CGVO % change in real gross value of output (nominal output deflated by the 3-digit industrial sales price index). v%f'lDGE % change in the ratio of sales at purchaser prices to sales at producer prices. A positive number indicates an increase in sales taxes/decrease in subsidies. CCE89 Convertible currency exports as a % of sales in 1989. NCCE89 Non-convertible currency exports as a % of sales in 1989. FTEE89 Exports via foreign trade enterprises as a % of sales in 1989. EnterpriseAdjustmentin Transitional Economies 41 TABLE 4.5: Other variables (exogenous/instruments) CK % change in real gross fixed capital, start-year. CKSQ CK*CK (instrument for CGVO). CP3I % change in the appropriate 3-digit import price index. CTS_I % growth in real sales of appropriate 3-digit industry, excluding the sales of the enterprise, i.e., growth in real sales by the rest of the 3-digit industry (instrument for CGVO). DOMMS89 Domestic market share in 1989. (Domestic sales / est. total domestic sales, by 3-digit industrial category.) ENERINT Energy intensity in 1989. Energy costs / gross output. INV89 Turnover credit as a % of total inventories and dues on 31 December 1989. MATINT Material intensity in 1989. Inflation-bias-corrected material costs (excluding energy and depreciation) / gross output. NOTINV88 "Own funds" of the enterprise as a % of total inventories on 31 December 1988 (instrument for INV89). RELXP Relative export price variable. Share of exports in total in 1989 * change in relative price of exports (at the 3-digit level). 42 Enterprise Adjustment in Transitional Economies REFERENCES Orley Ashenfelter and Richard Layard, eds. (1986), Handbook of Labor Economics, Amsterdam: North Holland. Andrew Berg and Olivier Blanchard (1992), 'Stabilization and Transition: Poland 1990- 1991', NBER Conference Paper, Cambridge, MA, February 26-29. Olivier Blanchard and Richard Layard (1991), 'Post-stabilization Inflation in Poland', MIT, mimeo. Olivier Blanchard et al. (1991), Reform in Eastern Europe, Cambridge: MA. Guillermo Calvo and Fabrizio Coricelli (1992a), 'Stagflationary Effects of Stabilization Programs in Reforming Socialist Economies: Enterprise Side and Household Side Factors', The World Bank Economic Review. Guillermo Calvo and Fabrizio Coricelli (1992b), 'Stabilizing a Previously-Centrally- Planned Economy: Poland 1990', Economic Policy, No. 14. Alan Carruth and Andrew Oswald (1989), Pay Determination and Industrial Prosperity, Oxford: Clarendon Press. Evsey Domar (1966), 'The Soviet Collective Farm as a Producer Cooperative', American Economic Review, Vol. 56, No. 4, pp. 734-757. Saul Estrin and Paul Hare (1992), 'Firms in Transition: Modelling Enterprise Adjustment', London School of Economics, Centre for Economic Performance Discussion Paper No. 89. Richard Freeman (1992), 'What Direction for Labor Market Institutions in Central and Eastern Europe?', NBER Conference Paper, Cambridge, MA, February 26-29. Stanislaw Gomulka (1991), 'The Causes of Recession Following Stabilization', Comparaitive Economic Studies, Vol. 33, No. 2, Summer. Mark G6ra (1991), 'Stages of Transition: The Labour Market', London School of Economics, Centre for Economic Performance Working Paper No. 141. Marek G6ra and Harmut Lehmann (1992), 'Flow and Stock Analysis of Polish Unemployment: January 1990-June 1991', Labour, Vol. 6, No. 1. A.B. Krueger and L.H. Summers (1988), 'Efficiency Wages and Inter-Industry Wage Structure', Econometrica,Vol. 56, No. 2, pp. 259-93. Enterprise Adjustment in Transitional Economies 43 Richard Layard, Stephen Nickell and Richard Jackman (1991), Unemployment: Macroeconomic Performance and the Labour Market, Oxford University Press. James Meade (1972), 'The Theory of Labour Managed Firms and Profit Sharing', Economic Journal, pp. 402-28, March. Stephen Nickell and Sushil Wadhwani (1991), 'Financial Factors, Efficiency Wages and Employment: Investigations Using UK Micro-Data', Review of Economic Studies. Vol. 58, No. 5, October. Dani Rodrik (1992a), 'Making Sense of the Soviet Trade Shock in Eastern Europe: A Framework and Some Estimates', Hoover Institution, mimeo. Dani Rodrik (1992b), 'Foreign Trade in Eastern Europe's Transition: Early Results", NBER Conference Paper, Cambridge, MA, February 26-29. Mark E. Schaffer (1992a), 'The Polish State-Owned Enterprise Sector and the Recession in 1990', Comparative Economic Studies, Vol. 34, No. 4, Spring, pp. 58-85. Mark E. Schaffer (1992b), 'The Enterprise Sector and the Emergence of the Polish Fiscal Crisis 1990-91', London School of Economics. Centre for Economic Performance Working Paper No. 191, August. B. Ward (1958), 'The Firm in Illyria: Market Syndicalism', American Economic Review, Vol. 48, No. 4, pp. 566-589.