The paper develops and estimates a theoretical model of wage determination and union-nonunion wage differentials. In order to overcome the institutional ctiticisms of the formal bargaining literature, the paper generalizes the Nash-Zeuthen-Harsanyi model by linking the solution to the institutional concepts of bargaining power and fear or cost of disagreement and by making the outcome depend not only on endogenous but also on exogenous factors. An operational specification of bargaining power and fear of disagreement allows the model to be estimated with data covering twelve companies and trade unions during the period from mid-1950's to the late 1970's. While giving limited support to the NashZeuthen-Harsanyi solution, the empirical analysis indicates that the bargaining outcome usually deviates from the Nash-Zeuthen-Harsanyi point and, in accordance with the institutionalist claim, that it varies significantly with exogenous factors. Contrary to the traditional labor economics view, the results do not support the general conclusion that the bargaining solution lies on the marginal revenue product curve of labor. Instead, the relevant coefficients suggest that for many firms and unions the outcome might be better characterized by the efficient contract (vertical contract curve).

T HE present study estimates the effects of trade unions, Hitler's regime, and codetermination on relative wages in Germany. I Each of these establishments is considered to have had important implications for Germany's labor market. Moreover, a thorough understanding of their effects can be valuable for comparative purposes. Following a brief discussion of these phenomena and their hypothesized effects on wages, the analytical model is developed in section II. The empirical results are reported in section III and section IV contains the summary and conclusions. The extensive literature on the relative wage effects of trade unions has focused almost exclusively on the union impact in North America2 and, more recently, in Great Britain.3 Since the German institutions of industrial and labor relations differ considerably from their AngloAmerican counterparts4 and often have been studied as a successful (possibly superior) system which might be adopted by others, it is worthwhile exploring the magnitude of the union wage effect in the German institutional setting. In particular it has been argued that, compared to their American and British counterparts, the continental unions tend to place more emphasis on political and social, rather than economic achievements.5 If this hypothesis is correct, then, ceteris paribus, the estimated relative wage effect of German unions ought to be smaller than that found in the United States and Great Britain. With the advent of Hitler's regime in 1933 there was a dramatic transformation of the existing political and economic institutions. Trade unions, which in 1932 represented over 40% of the industrial labor force, were promptly abolished6 and replaced by a government-operated Front.7 Strikes and lockouts were forbidden and wages were determined centrally.8 While the effects of these changes on the cultural, political and social life have been well documented, few economic studies attempted to analyze the effect of Hitler's regime on wages and incomes. Moreover, those that did, used a qualitative or only a loose quantitative approach, partially as a result of the lack of a systematic wage series prior to 1934.9 W. Krelle (1962, p. 17) has shown, for example, that labor income Received for publication July 30, 1979. Revision accepted for publication May 1, 1980. Cornell University. I would like to thank Orley Ashenfelter, Albert Rees and two anonymous referees for helpful comments. I have also benefited from discussions with J. S. Butler, William Greene, Louis Phlips and Katherine Terrell. Any remaining errors are, of course, my own. This research was in part supported by a grant to the Princeton University Economics Department from the Sloan Foundation. I The term codetermination refers to the German participatory system of management, as it was originally established in the Federal Republic by the 1951 Codetermination Act and the 1952 Works Constitution Act. 2 The most important work and collection of references on the subject is still Lewis (1963). Among the later studies in the private sector are Ashenfelter (1972), Ashenfelter and Johnson (1972), Bloch and Kuskin (1978), Boskin (1972), de Menil (1971), Rosen (1969) and Schmidt and Strauss (1976). Public sector studies are summarized by Lewin (1977). X See especially Pencavel (1974), Mulvey (1976) and Metcalf (1977). 4 Needless to say there are important differences between the U.S. and British systems of industrial and labor relations. They exhibit considerable homogeneity, however, when compared to the systems in continental Europe. Historically, the relatively most salient features of the German system have been (1) trade union affiliation with political parties and/or religioiis organizations, (2) industrial unionism. (3) governmental interference in industrial and labor relations, and (4) paternalistic management. World War II led to the unification of the formerly splintered unions in the Deutsches Gewerkschafts Bund (DGB), close collaboration with the Social Democratic Party (SDP) and the establishment of codetermination. For institutional references see Almanasreh (1977), Fiirstenberg (1969, 1977), Schregle (1978) and Vollmer (1976, 1979). 5 See Kassalow (1969, 1980) and Windmuller (1969). 6 Employers associations followed their lead in 1934. 7 The Labor Front included all employees and employers. It represented the Nazi government in the factories and possessed broad powers over its member subjects. 8 In fact, the entire economy underwent considerable centralization as Hitler proclaimed his first Four Year Plan in 1933 and a second one in 1936. 9 Among the most thorough and informative of these conventional studies is that of Bry (1960). For a discussion of the data see Gerss (1977).

The Central and East European (CEE) countries are in their sixth year of a dramatic transition from a centrally planned to a market-based system. In the first phase of the transition, most of these economies have achieved macroeconomic stabilization but also experienced a major decline in officially measured output and a slower but significant decline in employment. The attention has thus shifted to the ability of governments to check the rapid rise in unemployment, induce efficient behavior of firms, and improve the functioning of the infant markets. In particular, while a fundamental feature of the centrally planned economies was full employment, with state enterprises hoarding unproductive labor, a distinguishing feature of the transition has been the emergence of a double-digit unemployment rate, together with varying degrees of restructuring, privatization, and birth of firms. An understanding of these phenomena is essential for grasping the process of transition and formulating appropriate policies. In this paper, I provide a step in this direction by discussing some recent econometric evidence for CEE on (i) enterprise behavior (in the areas of restructuring and privatization, as well as employment and wage setting) and (ii) the flow of individuals from unemployment into employment.

This article presents a static and dynamic intertemporal analysis of employment and income distribution in unionized and labor-managed firms. Motivated by theoretical considerations and institutional features of Western trade unions and labor-managed firms, we examine firms in which worker-members share the risk of layoffs by (acting as if) compensating laid-off members. In the static framework we show that, although the firms do not behave perversely, their behavior depends crucially on the initial membership. Since the issue of optimal initial membership has been virtually ignored in the literature, we analyze it next within a dynamic framework.

We use two large samples of firms to assess the effects of business environment constraints, competition, export orientation, and ownership on firm performance. We deal with omitted variables, errors in variables, and endogeneity, and find that few business constraints affect performance. Replicating the analysis with Doing Business and Heritage Foundation indicators of the business environment yields similar results. In fact, country fixed effects, reflecting time-invariant differences in the business environment as well as other factors such as health care and education, matter more for firm performance than differences in the business environment across firms within countries.

Strategic restructuring of firms through investment is key to a transition from plan to market. Using data on industrial firms in the Czech Republic during 1992-1998, we find that foreign-owned companies invest the most and cooperatives the least, that private firms do not invest more than state-owned ones, and that cooperatives and small firms are credit rationed. Given the large volume of nonperforming bank loans to firms and the high rate of investment of large state-owned and private firms, our findings also suggest that these firms operate under a soft budget constraint. Estimates of a dynamic model, together with the support for the neoclassical model, suggest that firms started to behave consistently with profit maximization.

Czechoslovakia provides a unique example of a country that became underdeveloped as a result of an externally imposed system. Before World War II, Czechoslovakia was a democracy, with GNP per capita similar to that of Belgium and Austria. Its industries were on the technological edge and its products were known worldwide for their superb workmanship. By 1990, Czechoslovak GNP per capita is estimated by the World Bank at $3,300, thus being comparable to that of Venezuela, Gabon, and Yugoslavia, but only slightly above one-fifth of that of Austria and Belgium.1 Most Czechoslovak products are now of mediocre quality and selling at a discount, if at all, in the West. This remarkable transition occurred over approximately 40 years. During the postWorld War II reconstruction, the country was still run as a market economy, although major parts of industry, banking, and insurance were already nationalized. After the 1948 Communist takeover, Soviet-type economic planning was imposed, the remaining private enterprises were nationalized,2 and priority was given to heavy industry. Czechoslovak foreign trade was reoriented from world markets toward Soviet bloc countries. The Czechoslovak government adhered to the Soviet-type planning system faithfully throughout the 1950's. The economic slowdown in the early 1960's led to a series of reform attempts, that culminated during the Prague Spring of 1968 with a short-lived and partial program of price liberalization, separation of economic policy from political decision making, enterprise autonomy, and workers' participation in enterprise management. However, central planning was reimposed after the 1968 invasion and remained virtually intact until the late 1980's.

Using firm-level data, we estimate the effects of the major wave of 1991 breakups of Czechoslovak state-owned enterprises on the subsequent performance of the ‘master enterprises’ and spun-off divisions. We estimate the performance effects of spinoffs by comparing the performance of enterprises that remained intact throughout the 1990–1992 period to the performance of the master enterprises that experienced spinoffs and the newly spun-off subsidiaries. Our estimates suggest that the breakups had a significant immediate effect on the productive efficiency and on the profitability of industrial firms in 1991, and that the effect became much less significant in 1992. The effect is a negative function of the size of the spinoff, being positive for small to slightly above average-sized spinoffs and negative for very large ones. We cannot reject the hypothesis that the estimated effect was identical for the spun-off subsidiaries and the master enterprises that experienced the spinoffs. Our 1991 estimates suggest that the large firms created under the centrally planned system suffered from inefficiencies that were alleviated by the breakups. The 1992 estimates are consistent with increased competition and the appropriation of profits by managers.