The recent literature on endogenous growth has developed various theories in which institutions affect long-run growth prospects, but there have been few empirical studies, at least for the European economies. A CEPR joint workshop with the Centre for Economic Performance (CEP) at the London School of Economics, on `Institutions and Economic Growth', held at the CEP on 20/21 November, considered the effects of education, bargaining systems, industrial policies and payments unions on the long-run, post-war growth performance of several European economies. This workshop formed part of CEPR's research programme on `Comparative Experience of Economic Growth in Postwar Europe', supported by the Commission of the European Communities under its SPES programme. The workshop was organized by Nicholas Crafts, Professor of Economic History at the University of Warwick and Research Fellow in CEPR's Human Resources programme, and Charles Bean, Professor of Economics at the LSE, CEP Deputy Director, and Research Fellow in CEPR's International Macroeconomics programme.

Esmond Birnie and David Hitchens (Queen's University, Belfast) and Karin Wagner (Wissenschaftszentrum Berlin für Sozialforschung) opened the workshop with a two-part paper on `Institutional Impacts on Productivity: A Comparison of Ireland, East and West Germany'. They first attributed the poor long-run performance of both parts of Ireland to their `British' systems of education and training, the two economies' lack of integration, prolonged large subsidies to inefficient firms and industries, industrial policy and `the troubles'. They then assessed the German vocational training system and suggested that this may well be transferable to the UK, although its reinstatement in Eastern Germany has proved difficult.

Stephen Broadberry (University of Warwick and CEPR) doubted whether investment in specialized skills is appropriate, since Ireland is most productive in standardized agricultural goods. Hitchens replied that the high productivity achieved in its foreign-owned industrial sector indicated the need to promote the productivity of indigenous firms.

In the second paper, Barry Eichengreen (University of California, Berkeley, and CEPR) asked `Was the European Payments Union a Mistake? Would a Payments Union for the Former Soviet Union be a Mistake as Well?'. He showed that EPU was clearly superior to trade bilateralism, but current account convertibility was also largely feasible in the 1950s. He found little evidence that the EPU had distortionary effects on trade, relative prices or capital flows; it was chosen to tip the terms of trade in Europe's favour, protecting its living standards and the post-war social consensus. These `special' reasons for the EPU's success suggest, however, that a payments union would be unsuitable for Eastern Europe and the former USSR today.

Alan Milward (LSE) stressed the methodological problem of choosing appropriate price indices, the overstatement of US financial leverage, the importance of black markets in ensuring de facto convertibility, and the link between commercial and industrial policies allowed by the EPU.

James Symons (University College London and CEP) presented his joint paper with Manfred Keil and Andrew Newell, `Growth, Convergence and Education', which developed a model in which capital is interpreted as the extent of adapting to (exogenous) new world knowledge, adjustment of such `capital' is costly, and cultural attitudes towards technological change differ across countries. He derived a growth-productivity-gap equation and estimated national adjustment costs and output gaps relative to the US for 17 OECD countries. Adjustment costs were determined by union density and education expenditure and were lowest in the US followed by Germany and Canada but surprisingly high in Japan.

Giorgio Brunello (Università di Venezia) stressed the importance of learning-by-doing effects, while various participants questioned whether the estimated high Japanese adjustment costs and output gap might reflect a conservative attitude to change.
Lars Ekdahl (Arbetslivcentrum, Stockholm) and Alf Johansson (University of Uppsala) presented their paper on `Labour Market Institutions and Productivity in Post-war Swedish Industry'. They maintained that the dissolution of the `Swedish model' the breakdown of institutional structures and industrial relations and hence of the social consensus on which it had been based contributed to Sweden's relatively slow productivity growth since the mid-1970s.

Howard Gospel (University of Kent at Canterbury) questioned the Swedish model's compatibility with productivity growth and attributed its apparent success to the high inter- and post-war productivity of Swedish manufacturing. Several participants suggested further research on the effects of openness to international trade, wages, and effects of training systems on sectoral productivity.

In their paper, `Productivity and Bargaining in the Post-war Engineering Industry', Alan Booth and Joseph Melling (University of Exeter) investigated the impact of trade unions on productivity in UK engineering during 1945-55. They found that this specific institutional framework weakened the support given by industry-wide organizations to productivity-improving measures by influencing bargaining practices and shop-floor attitudes. This prevented the emergence of a German-style `productivity coalition' in the UK.

Geoffrey Owen (CEP) emphasized the high degree of collusion and complacency that characterized the industry at the time. Howard Gospel wondered whether decentralized, plant-level bargaining might have improved its economic performance. Booth replied that this would have required strong individual unions and truly independent firms.

Giovanni Urga (Queen Mary and Westfield College, London) concluded the first day with his econometric study, `Trade Unionism and Manufacturing in Italy in Recent Decades', based on a panel of 45 private, state and foreign-owned firms for 1958-85. Irrespective of ownership, he found no evidence of direct union effects in the estimated employment and investment functions, but the positive labour-cost coefficient in the latter suggests that there is a substitution effect between capital and labour and that strong trade unions extract some of the increased rents from capital investment.

In the first paper of the second day, `The British Electronics Industry from 1960 to the 1990s', Geoffrey Owen challenged the view that poor performance in the electronics sector has been uniquely `British', finding a similar pattern in other European countries. He linked the UK's relative strength in telecommunications to the absence of international competition and argued that `electronics envy' has focused too much attention on large firms while the good performance of smaller companies in the UK has been undervalued.

In discussion, Barry Eichengreen cautioned that comparisons of industrial policy with Japan must take account of the effects of greater competition in the UK.

Pedro Fraile Balbín (Universidad Carlos III, Madrid) presented his paper, `Regulation and Capture in Peripheral Europe: the Spanish Steel Industry, 1941-1981. An Institutional Analysis', This assessed the government's establishment of a state-owned, integrated steel mill in order to break the steel industry's highly cartelized structure and reduce prices following the Civil War. Engineers and managers that remained loyal to the private cartel gradually gained control of the regulatory agencies, however, and the state mill became a Stackelberg follower to the private cartel. This foiled the government's plans and led to a rise in prices after 1940.

In discussion, Stephen Broadberry noted that production had become highly differentiated so that steel prices were difficult to compare. Alan Milward added that most European integrated steel plants had been failures anyway.

Giorgio Brunello then presented `Markups in the Labour and Product Markets and the Relative Performance of Industry and Services: Italy, 1931-1990', which developed an integrated two- sector model with monopolistic competition in product markets. His results indicated that employment growth was explained by mark-ups in product markets and more importantly in labour markets. The lack of data on self-employment posed problems, however, while the modelling of endogenous productivity growth as a function of mark-ups was left for future research.

Ronald Schettkat (Wissenschaftszentrum Berlin für Sozialforschung) presented `Productivity Trends in Germany and the US: Some Speculations on the Influence of Institutions', which showed that Germany's high productivity growth relative to the US was produced by only a small part of its population, while the US distributed work (but not wages) more evenly as a result of its low productivity growth. Germany's high productivity growth did not lead to a zero-sum game, with gains paid into transfer payments, but to real gains that were large enough for winners to compensate losers.

In his paper, `Reward Structures and the Allocation of Talent', Daron Acemoglu (CEP) developed Baumol's model relating the allocation of talent between unproductive activities and more productive entrepreneurship to the rewards offered by making the latter endogenous in a steady-state analysis. In a dynamic model, past allocations of talent also affect the reward structure, so historical dependence explains the simultaneous determination of rewards and talent allocation.

Gianni Toniolo (Università degli Studi di Venezia and CEPR) questioned the distinction between rent-seeking and simply unproductive activities. Geoffrey Owen noted that `individualism' in the 1980s had made entrepreneurship more respectable.

In the final paper, `Rent-Seeking and Rent-Resistance in Postwar Britain: Evidence from the Cotton Mills and the Aircraft Factories', John Singleton (University of Manchester) considered how British industry had pursued rents after the war. From a case-study of these two industries in the 1950s and 1960s, he found that governments had not been easily influenced by business interests in marginal constituencies. Relations with overseas trading partners and a reluctance to take stakes in declining industries favoured a mostly non-interventionist industrial strategy; the few cases of government intervention were largely unrelated to lobbying by the industries concerned.