Homework answers / question archive / Question: Discuss the main explanations of the European Golden Age

Question: Discuss the main explanations of the European Golden Age

Sociology

Share With

Question: Discuss the main explanations of the European Golden Age. Which one do you find the most convincing?

This essay has an upper word limit of 1500 words, so I suggest the word count is 1450-1500 except references.

Chicago Format.

This course is topic oriented, so you should stand as an ignorant to write it to make others understand my argument and related evidences.

The citations should be 8-10.EH483 The Development and Integration of the World Economy in the 19C and 20C Lecture MT11 The Golden Age of Economic Growth Dr Natacha Postel-Vinay December 7, 2020 The world after the two world wars “Never, never again.” ? Two very destructive world wars. ? Interwar Great Depression. • • • • Web of reparations and debts. Stock market booms, busts and panics. Still debate about panic vs. fundamentals (see EH430). Relatively slow recovery in places (eg. US). Spectacular growth post-WW2 US UK France Germany Italy Spain 1950 GDP per capita 100 72 55 41 37 25 Growth in GDP per capita, % pa 19501973 2.5 2.4 4.1 5 5 5.8 Table: GDP per capita index relative to the US, and growth (Source: Maddison 2001). Golden Age of economic growth Table: European economic growth (Source: Crafts & Toniolo, 1996). Can “spring-back” (also called catch-up) from the wars explain most of this growth performance? Today 1. The “spring-back” or natural catch-up explanation. 2. The high investment rate explanation. Today 1. The “spring-back” or natural catch-up explanation. 2. The high investment rate explanation. Catch-up Figure: Correlation of 1950 GDP per man-hour with 1950-73 growth (Source: Eichengreen 1996). Relationship with war-time destruction? Figure: Real GDP growth rate, 1948-51 (Source: DeLong and Eichengreen 1991). A lot of catch-up to do Figure: German log of GDP per capita, with West Germany post-1945 (Source: DeLong and Eichengreen 1991). How to catch up Recall the Solow growth model. Labour productivity is: ? ? =? ? ? ? It thus depends on: ? • Capital intensity ( ? ) (also called investment). • Technology ? (also called Total Factor Productivity or TFP). → According to Abramovitz (1986), the key to catch-up is ?. Technology transfer matters. Conditions for catch-up From Lecture 8, recall Abramovitz’s condition for effective technology transfer: social capability. ? A relatively vague notion, with links to institutional economics (Douglas North). ? Right “institutional mix” • • • • Protection of property rights Checks on monopoly powers Frictionless trade Skilled labour force. ? Otherwise, need to find “substitutes” (Gerschenkron). Social capability in 1945 After the war, countries are destroyed BUT: ? Democratic legacy with protection of property rights. • Or at least strong will to democratise (West Germany). ? Highly skilled labour force. ? Marshall Plan (1947-51) comes with strings attached (market economy). → Technological transfer from the US was easy (also thanks to free trade). [No need to find “substitutes for pre-requisites”!] International environment The international economic and monetary environment was relatively stable: ? Strong will to avoid the mistakes of the past. • Ensure fixed E but keep domestic monetary control. • Only forego capital mobility (Bretton Woods). • But note: keeping E fixed still difficult because of large trade deficits. ? Institutions mitigate this problem: • EPU provides stabilization fund with exchange reserves. • Marshall Plan sends $14 billion to Europe. However… Figure: German log of GDP per capita, with West Germany post-1945, relative to 1880-1913 trend (Source: DeLong and Eichengreen 1991). France Figure: French log of GDP per capita, with West Germany post-1945, relative to 1880-1913 trend (Source: DeLong and Eichengreen 1991). Today 1. The “spring-back” or natural catch-up explanation. 2. The high investment rate explanation. A missing story? Figure: Correlation of investment shares and GDP growth across developed countries (Source: Eichengreen 1996). Context: interwar conflicts After World War 1, conflicts between labour and capital: ? Karl Polanyi’s The Great Transformation (1944). ? Eichengreen’s “war of attrition”. • • • • High public debt. Labour unwilling to be taxed, demanding higher wages. Capital also unwilling to be taxed. Partly responsible for high inflations/hyperinflations. → High wage demands could impede capital investments. What could have made labour become less demanding post-1945? A post-1945 social contract Eichengreen (1996) has posited the idea of a social contract (initially Maier’s idea from 1987). 1. Labour accepted to moderate wage demands for now… 2. … as long as capital was reinvested in the company. 3. Hope is that this would enable wages to rise in the future. How come this came about around1945? The Marshall Plan vs. the welfare state DeLong and Eichengreen (1991) emphasise the role of the Marhsall Plan: • $14 billion for foreign aid (not debt). • Makes it easier for workers to moderate their wage demands. • Strings attached: governments had to induce moderation via conditional subsidies to firms. Milward (2000) puts more emphasis on the welfare state itself: • Further development in the UK with the Beveridge Report (1942) and the creation of the National Health Service (NHS). Is wage moderation really the main story? Figure: GDP per capita growth relative to 1901-13 trend (Source: Eichengreen and Ritschl 2009). Conclusion Trying to explain exceptional European growth rates 1950-1973. ? Clearly some role for pure “spring-back” or catch-up. ? Facilitated by existing social capability. ? Role of international institutions for stable macro environment. ? But also need to explain “surpassing”, if any. ? Wage-moderation may be part of the story. ? May not be end of the story though (eg. France). 2 Institutions and economic growth: Europe after World War II BARRY EICHENGREEN 1 Introduction The quarter century that ended around 1973 was for Western Europe a Golden Age of economic growth. Real GDP rose nearly twice as rapidly as over any comparable period before or since.l Understanding the sources of this admirable performance would shed important light on the causes of the slowdown through which Europe has suffered subsequently. Part of the explanation is surely 'catch-up', as Abramovitz (1986) emphasized. The gaps that had opened up vis-d-vis both the United States and Europe's own prewar trend as a result of nearly two decades of depression and war offered exceptional scope for growth after 1945. Figures 2.1 and 2.2 summarize the standard evidence on the operation of these effects.2 But cross-section regressions relating growth rates to per-capita GDP differentials show that 'catch-up' and 'spring-back' explain only part of the postwar acceleration: purged of their effects, growth from 1950 to 1973 was still more than 50 per cent faster than it became subsequently.3 And even in so far as these factors provide the explanation, understanding what enabled post-World War II Western Europe so effectively to exploit the opportunity for closing the gap can have important implications for countries in Eastern Europe and the developing world currently seeking to join the 'convergence club'. Aside from catch-up, the proximate cause of postwar Europe's growth miracle was high investment. Net investment rates in Europe were nearly twice as high in the 1950s and 1960s as before or since.4 Scatter plots like those of Figure 2.3 suggest that increasing the gross investment share of GDP from 20 to 30 per cent increased the growth rate by as much as 2 percentage points. Countries with different investment rates were not identical in other respects, of course; multivariate regressions for this period, using Maddison's sample of sixteen advanced countries, which attempt to control for some of these differences, suggest that an extra 10 points on the investment rate translated into a more modest half a point on the growth rate.5 Still, together with catch-up this gets us a long way towards 'explaining' in an accounting sense postwar Europe's rapid growth. 38 Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth JUU - 250- 39 • Germany • Austria • Italy \ \ • France 200- Finland \ , O B e l g i u m ^ Netherlands Denmark* • \. Norway y * \^ 150" Sweden \^ Switzerland • \. • UK 100" \ • USA \ \ 50100 200 300 Initial 400 500 Source: Maddison (1982: 212). Figure 2.1 Initial (1950) GDP per man-hour and growth, 1950-73 J\J\J \^ • Germany N. • Austria • Italy 250Finland ^ y 2002 O 150" • France \ J^Netherlands Belgium • Denmark Norway N. •• Sweden Switzerland • > • UK 100- USA^V 50-25 \ i 25 Shortfall 50 75 Source: Maddison (1982: 212). Figure 2.2 Wartime and postwar (1938-50) change in GDP per man-hour and growth, 1950-73 Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 40 Barry Eichengreen o.io0.090.08- 0 0.07- 1 0.06| 0.05 0.040.03 0.02 0.15 0.20 0.25 Investment share 0.30 0.35 Regression /-statistic 3.00. Finland from 51, Japan 52. (a) 1950-73 0.045 0.0400.035 Q 0.030 0.025 0.020 0.015 0.010 0.15 0.20 0.25 Investment share 0.30 0.35 Regression /-statistic 3.31. (b) 1973-89 Figure 2.3 Investment share and growth rates of GDP: 16 countries, 1950-89 Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 41 Two things then remain to be understood: what made this high investment possible, and what made it so productive? This directs our attention to the other elements of the postwar growth recipe: wage moderation and export growth. Wage moderation stimulated both the supply of and the demand for investment - demand by making investment profitable, supply by making available the profits to finance it.6 The openness of European economies and the growth of their exports, whose volume expanded in the 1950s and 1960s by more than 8 per cent a year, allowed investment to be allocated to the sectors where its contribution to productivity growth was greatest. Nations could exploit their comparative advantage without being constrained by the composition of domestic demand. (Figure 2.4 displays the export-GDP growth nexus, indicating, suggestively, that it collapsed after 1973.) Having stripped another layer off the onion, what next must be explained is wage moderation and the growth of trade. Both were achievements by the standards of the interwar period, when exports stagnated and wage pressure was intense.7 Part of the explanation is surely that postwar policy-makers and market participants had learned from the disasters of the interwar years and determined not to repeat them. But the desire for a better outcome may not suffice; mechanisms are needed to achieve it. The argument of this chapter is that postwar growth benefited from the presence of institutions singularly well suited to reconstruction and growth. Those institutions solved commitment and coordination problems in whose presence neither wage moderation nor the expansion of international trade could have taken place. On the domestic side, social and economic institutions disseminated information and monitored the compliance of economic interest groups with the terms of their agreement to moderate wage claims and boost investment.8 They helped to lock in the bargain by creating bonds that would be lost in the event that someone reneged. They worked to coordinate the terms of the agreement across sectors of the economy. On the international side, institutions were created to coordinate national programmes of economic restructuring along export-oriented lines, and to lend credibility to European governments' commitment to openness. This encouraged countries to restructure their economies and to exploit more fully their comparative advantages, enhancing the productivity and profitability of investment. Institutions were not equally well adapted to the needs of growth in all European countries. Some, notably the UK and Ireland, failed to develop the relevant domestic institutions. Others, such as France and Italy, managed to do so only with delay. Some countries, like France in the 1950s, failed to align domestic arrangements with the evolving international framework and to exploit the opportunities offered by the external sector. These different institutional responses go a fair way towards accounting for variations across countries and over time in European growth performance. To complete the story it is necessary to account for these different responses. The argument of this chapter is that the relevant socioeconomic institutions necessarily displayed considerable inertia. Their function, in part, being to serve as coordinating mechanisms, their very nature created coordination problems for altering them. Institutions function as standards, giving rise to network externalities that tend to lock in their operation. The exceptional circumstances of war and reconstruction provided singular opportunities for coordinating wholesale adjustments in institutional Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 42 Barry Eichengreen 0.10-j 0.090.08- § 0.07 -I 3 0.062 | 0.05 0.040.030.02 0 0.05 0.10 0.15 Growth rate of exports 0.25 0.20 Regression ^-statistic 5.35. (a) 1950-73 0.0450.040 0.035 - i 0.030 0.025 0.020 0.015 0.010 0.035 0.040 0.045 0.050 0.055 0.060 Growth rate of exports 0.065 0.070 0.075 Regression f-statistic 1.51. (b) 1973-89 Figure 2.4 Growth rates of GDP and exports: 16 countries, 1950-89 Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 43 arrangements. Even under these extraordinary conditions, however, radical changes in coordinating institutions were necessarily difficult to organize. Inevitably, the important institutional changes of the postwar period were only marginal adaptations. They were feasible only where considerable progress had already been made in developing the institutional structures required for growth after World War II. Because of the evolving structure of the institutions analysed in this chapter, it is important to be precise about the period under consideration. I concentrate here on the institutional arrangements of the recovery period (1945-9) and thefirstpostwar decade of sustained growth (1950-9), and on the critical departures from pre-1945 institutional arrangements that made possible the inauguration of Europe's postwar golden age. Subsequently, one sees in many countries the institutional structures of the postwar years weaken and break down. I examine institutional responses to these difficulties in the 1960s, which typically involved heightening the centralization of wage bargaining and expanding the role of government in concertizing sectoral negotiations. I describe the problem to which these initiatives were the response - how the end of 'catch-up' growth made cooperative agreements more difficult to sustain - and explain why the response ultimately proved inadequate. My goal is to show that the perspective developed here is capable of explaining both why postwar institutions were initially so conducive to growth and why they ultimately broke down. 2 The model 2.1 Domestic institutions Lancaster (1973), Grout (1984) and van der Ploeg (1987) model a dynamic game between capital and labour with a common general structure.9 Welfare is maximized when capitalists and workers both agree to defer current compensation in return for future gains. Workers moderate their wage claims in order to make profits available to enterprise and to make profitable their investment in capacity modernization and expansion. Capitalists restrain dividend payout in order to reinvest. Investment stimulates growth, raising the future incomes of both capitalists and workers. In the cooperative equilibrium in which both workers and capitalists exercise restraint, the costs of forgoing current consumption are dominated by the benefits of the future increase in incomes accruing to both. This cooperative equilibrium may be impossible to sustain, however, for the sequencing of events renders it time inconsistent. Consider the problem for labour created by uncertainty concerning subsequent investment. If investment requires liquidity, and liquidity requires profits, then workers must restrain wages now in order to make profits available to capitalists for investment later. Once the wage restraint has occurred, however, capitalists are best off if they renege on their agreement to invest, paying out profits as dividends instead. Since investment is no higher than if workers had failed to moderate their wage demands, labour has no incentive to exercise restraint. In this non-cooperative equilibrium, workers pursue wage increases, management pays out profits as dividends, and investment and growth are depressed. Even if workers can be assured of capital's willingness to invest, this is not a Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 44 Barry Eichengreen complete solution to the problem. Unions may be able to recontract after investment has taken place. Once the investment has occurred, workers can renege on their agreement to restrain wages, seeking to capture the surplus instead. Since profits are not higher than if management had failed to invest, management has no incentive to plough profits into investment. Again, in the non-cooperative equilibrium, workers pursue wage increases, management pays out profits, and investment and growth lag. A contract that binds capitalists to invest profits and workers to exercise wage restraint overcomes the problem of dynamic inconsistency, rendering both groups better off. The social and economic institutions developed in Europe after World War II can be thought of as mechanisms to enforce this agreement.10 Institutions were developed to monitor the compliance of capitalists with their deferred contribution to the bargain and to disseminate evidence of non-cooperation; by reducing the likelihood that shirking went undetected, these mechanisms reduced the returns to doing so. Institutions were used to create bonds that would be lost in the event of reneging, effectively increasing the stakes and providing a further deterrent to shirking. By credibly committing capital to invest the profits made available by wage moderation, they provided labour with the incentive to be moderate. Institutions also deterred workers from appropriating the profits of firms that undertook investment Long-term contracts, social pacts between labour, management and government, and statutory wage and price controls are three mechanisms that could be used to precommit unions to wage moderation and thereby to induce management to invest. Unemployment, health and retirement programmes - the institutions of the welfare state, in other words - served as bonds that would be jeopardized if labour reneged. Centralization and concertation of sectoral wage negotiations further encouraged moderation on the part of unions. In so far as onefirm'searnings could pass through the capital market and finance another firm's investment, the benefits of wage moderation by one group of workers then accrued to other workers. Since the level of wages affected economy-wide determinants of investment like the interest rate, there was a need to coordinate wage demands across sectors to render a bargain to moderate wage claims attractive to each party to negotiations.l x Hence the need for institutions to centralize or concertize sectoral bargaining. On the employer side, any one firm contemplating investment had reason to worry that its decision to invest would encourage its workers to raise their wage demands in order to appropriate the profits thereby generated. But if wages were determined in economy-wide rather than at enterprise-level negotiations, an individual firm's investment decision no longer affected the wages it had to pay. In these circumstances, centralized wage negotiations led to a higher level of investment and, in so far as labour productivity was raised, to higher wages in equilibrium.12 2.2 International institutions For deferring consumption to be worthwhile, investment has to be productive. For investment to stimulate growth, in other words, there has \o be a market for the goods produced by domestic industries whose capacity is augmented and whose efficiency is enhanced. Here the postwar expansion of trade was key. International Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 45 trade, and intra-European trade in particular, allowed countries to specialize in the production of goods in which they had a comparative advantage without regard to limits on the demand for those products existing at home. It allowed them to rely on cheap foreign supplies of raw materials and intermediate inputs that were costly to produce domestically. But the expansion of trade created further coordination and commitment problems. Restructuring along export-oriented lines was costly. Sinking the costs of reallocating resources along lines of comparative advantage could turn out to be an expensive mistake if one's trading partners reneged on their commitment to openness. Encouraging the expansion of steel production on the assumption that coal and iron ore could be imported from abroad could be a costly error if foreign supplies were not forthcoming. Augmenting the capacity of such industries would not pay if other countries ultimately refused to draw down their import tariffs on final goods. Before encouraging the rationalization of domestic production along lines of comparative advantage, governments had to be convinced that their partners' commitment to openness was permanent. Here again institutions solved these commitment and coordination problems. The European Coal and Steel Community (ECSC) created monitoring and surveillance technologies that guaranteed the French steel industry access to German coal, and the German industry access to French iron ore. A Joint High Authority monitored the compliance of participating countries to the terms of the agreement. The European Payments Union (EPU) coordinated the simultaneous move of European countries to currency convertibility for intra-European current account transactions and precommitted the participants to a sequence of trade liberalization measures. An EPU Managing Board was created to monitor the policies of member countries and to discourage them from reneging on their commitments. The participants contributed hard currency and credit to the EPU's central fund; access to these resources was contingent on their adherence to the terms of the EPU agreement, which served as an effective bond. Compared to unilateral convertibility, then, the payments union was a more credible commitment mechanism.13 3 The structure of domestic institutions 3.1 The nature of the bargain That wage restraint and high investment were the dual cornerstones of the postwar settlement is evident in statements of the time. The Dutch case is prototypical. The Netherlands perceived itself as suffering from overpopulation (partly because of the return of expatriates from Indonesia), the solution for which was wage moderation and investment to enhance international competitiveness and raise living standards (Shonfield, 1965: 212). Between 1947 and 1954, consumer prices and the gross weekly earnings of adult male workers rose in lock step - by 32 and 33 per cent, respectively - this despite the fact that productivity increased by nearly 50 per cent between 1950 and 1954 alone, very considerably raising capital's share of national income (Flanagan et a/., 1983: 107).14 In Germany, trade unions observed 'significant wage restraint' throughout the 1950s in conjunction with the adoption Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 46 Barry Eichengreen of investment-friendly policies designed to stimulate growth (Maier, 1984). Norwegian unions followed policies of wage moderation into the 1950s, in return for government policies attaching an overriding priority to capital formation. Even in the UK, hardly a paragon of labour/management harmony, contemporary observers were impressed by 'the wisdom, the moderation and sense of responsibility of the great bulk of the trade union movement' (to quote Sir Walter Monckton, Churchill's Minister of Labour, in 1954).15 This wage moderation was part of a bargain in which labour agreed to restraint now in return for more investment, faster growth and higher incomes later. In the Netherlands, the unions conceded that the fruits of all productivity increases in the first half of the 1950s should be used tofinanceinvestment.16 They allowed wages to lag behind productivity in the 1950s 'so that industry could earn profits which would pay for expansion and modernization of the productive apparatus' (from a union publication, cited in Windmuller, 1969:350-1). A clearer statement of a wage moderation-for-investment trade-off is hard to imagine. The movement extended beyond this one case, however. In Belgium, wage restraint was part of a Social Pact (Projet d'Accord de Solidarity Sociale) initialled in 1944, in which unions and employers agreed to seek means of enhancing the competitiveness and productivity of enterprise.17 Workers agreed to moderate their wage demands and to refrain from striking except as a last resort, in return for a management commitment to boost productivity. In the mid-1950s unions and employers concluded a series of'Productivity Agreements' in which workers agreed to link their wage demands to productivity increases in return for a management pledge to stimulate the latter.18 In the 1960s, as part of the programmation sociale, labour agreed to renewed wage moderation in return for agreement by industry to a programme of capacity expansion and the provision by government of investment subsidies and guarantees. In Norway, unions agreed to restrain wages in return for promises from government and management to limit distributed profits to acceptable levels (Esping-Andersen, 1985). The Labour government mobilized the populace around the 'Work for Everybody' programme, under which unions agreed to restrain wages in return for a pledge that profits would be ploughed back into capital formation.19 Price, profit and dividend controls, maintained through 1953-4, provided a guarantee that capital would keep to its part of the bargain. The government used profit taxes and other income sources to underwrite the highest investment rates in Europe. Labour embraced the bargain because it was promised higher future incomes and full employment. Industry acquiesced because it was promised cheap credit and wage restraint (Esping-Andersen, 1985: 216-19). In Austria, interest groups came together in a 'social partnership' whose goal, from the end of World War II to the late 1960s, was 'the creation of a favourable climate for growth by preventing an outbreak of inflation from uncoordinated and mutually inconsistent income claims by the various economic interest groups (Flanagan et aL, 1983: 63). Wage growth was moderate, fuelling the economy's relatively high investment rates. In Germany in the 1960s, as part of a programme of concerted action, 'the unions were urged to restrain their wage demands, not to puncture the entrepreneurs' improved profit expectations. Unions were promised a later settlement, which would include these delayed wage increases and restore Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 47 "social symmetry"' (Bergmann and Muller-Jentsch, 1975: 259). In other countries, attempts to cut such a bargain proved a failure. In the UK, Ireland, France and Italy, wage pressure was intense, squeezing capital's share of the national income and limiting the availability of domestic investment finance. Through the 1950s, investment ratios and economic growth rates were disappointing by the standards of northern European countries, where cooperative solutions were reached, and even by the standards of countries like France and Italy in the 1960s. 3.2 Commitment mechanisms The immediate problem for postwar Europe was how to mobilize the funds to finance capital formation. International lending remained depressed; while it was possible to tap direct foreign investment flows from the United States, access to foreign capital was limited by the legacy of interwar default and the persistence of wartime controls. For growth it was therefore necessary that wages be moderated so that profits could be generated domestically to finance investment. Hence, workers had first to restrain wages in order to make profits available to capitalists for investment later. But once the wage restraint had occurred, capitalists could renege on their commitment to invest the profits, paying them out as dividends instead, leaving workers no reason to exercise restraint. And capitalists had no incentive to keep their part of the bargain in the absence of reassurance that workers would continue to exercise restraint in the future, appropriating the entire surplus. Workers and management thus confronted a prisoner's dilemma which could be resolved only through the introduction of institutional restraints on opportunistic behaviour. These institutions restrained such behaviour by disseminating information, monitoring compliance, and offering bonds. 3.2.1 Monitoring compliance and disseminating information One set of institutions monitored compliance and disseminated evidence of non-cooperation; by reducing the likelihood that shirking would go undetected, this reduced the returns to doing so. Workers were allowed to monitor management decisions relating to investment. Unions and employers' associations were encouraged to exchange information on wage and investment decisions through governmentsanctioned peak associations. The representation of labour unions on advisory and administrative committees of industry and government was made compulsory. German workplace co-determination, giving labour input into firms' investment strategies, provides a clear illustration of the operation of this mechanism.20 The 1951 Co-Determination Law mandated the appointment to the executive boards of all corporations of significant size a representative of labour, with the same rights as other members. By the late 1950s some 100firmshad labour representatives on their boards (Dahrendorf, 1959: 264-6). Streeck (1984: 147) argues that granting labour institutionalized influence on the management of the enterprise was a way of reassuring them that wage restraint would actually translate into investment. Works councils (required in all workshops with five or more employees) played a similar information-disseminating role in small German firms not covered by the Co-Determination Law. Further examples abound. In Austria, German-style wage moderation and high Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 48 Barry Eichengreen investment were secured by regular consultation between representatives of labour, management and government (Katzenstein, 1984). In Holland, representatives of government, firms and workers sat together on PBOs (Publiek Rechtelijke Bedrijfsorganisatie) to develop common employment and investment policies for Dutch firms.21 In Norway, the Social Democrats introduced a system of Bransjerad (planning councils) and Produktionsutvalg (production committees) to 'promote democratic participation in matters of investment, planning and industrial organization'.22 In the Netherlands, the government used leverage acquired through the administration of price controls to inspect the books of individual companies, mainly regarding prices and costs, but on other matters as well. In Belgium, works councils were installed in every company with more than 100 employees under the provisions of the 1948 law on the 'Organization of Industry', and union and industry representatives sat together for the first time on the boards of public and semi-public industrial and financial institutions. Even France, which otherwise failed to put the institutional prerequisites for growth in place, made some progress in this one respect. Labour-management plant committees (comites d'enterprise) were required by law for all enterprises employing fifty or more workers, starting in 1945. Two members were entitled to attend the meeting of the board of directors of limited liability companies. Though the comites have been dismissed as ineffectual in shaping managers' production and investment decisions, to do so is to misunderstand their role. Never intended to give workers veto power over investment decisions, they simply provided a conduit for information regarding investment behaviour. An Act of 1946 required management to inform the comite and receive its opinion before finalizing investment decisions (ILO, 1950: 206). Limited liability companies had to inform the committee of the profits made by the undertaking, permit it to audit its accounts, and allow it to examine the record of discussions of the board of directors. Information relevant to the investment bargain was thereby disseminated.23 3.2.2 Bonding A second set of institutions locked in the bargain by creating 'bonds' that would be lost in the event of reneging.24 A variety of devices were utilized to achieve these ends. In Austria, for example,firmswere guaranteed raw materials and semi-finished products at submarket prices from public enterprises in return for their willingness to adhere to the terms of the bargain. Private business was given representatives on the boards of the state holding company to lend credibility to this arrangement (Kurzer, 1993: 35-6). Government subsidies for firms similarly functioned as bonds that would be forfeited in the event that capitalists failed to plough profits back into investment. In Germany, for example, subsidies for the steel and the aircraft industry were seen as part and parcel of the social market economy that provided a social safety net for workers.25 The tax and transfer system could also be used. The Austrian government threatened to levy tax penalties onfirmsthat failed to keep their part of the bargain. That labour possessed a powerful voice in postwar coalition governments right-wing parties having been discredited during the final years of the First Republic and the Nazi occupation - rendered this threat credible. Similarly, in the Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 49 late 1940s the Swedish government introduced regulations limiting the payments of dividends by public companies and taxing away nearly half of supernormal profits. An investment reserve scheme, established before the war but greatly expanded in the 1950s, 'invited' corporations to place as much as 40 per cent of their annual pre-tax profits into a closed public account, to be released with government approval. Business received substantial tax savings on such deposits. A clearer example of bonding is hard to imagine. The institutional initiatives enumerated above were all designed to bond capital, since in an environment of scarce liquidity thefirstproblem was to get management credibly to commit to invest, so that labour had the incentive to restrain wages and make requisite profits available. But these bonding devices were likely to suffice only if supplemented by mechanisms also locking in future as well as current wage restraint; otherwise management, fearful that an aggressive labour movement would appropriate future profits, would be tempted to disregard its forgone bonds and renege on its commitment to invest. A variety of postwar institutional initiatives can be understood as devices to bond labour. In Belgium, the first postwar government pushed through a scheme for social security in return for labour's adherence to the 1944 Social Pact (Dancet, 1988: 98). The welfare state was greatly extended between 1964 and 1969 (to entail expanded provision of health insurance, child and unemployment benefits, and centralized pensions), coincident with labour and capital's agreement to a new and more ambitious cooperative agreement (the so-called programmation sociale). In Norway, the government in the mid-1950s offered legislation improving work conditions, including vacations and the length of the working week, in implicit compensation for wage restraint (Schwerin, 1980: 84). In Sweden, the government's role in supporting the Rehn model of wage concessions for firms that succeeded in raising productivity was to protect workers against redundancies by providing active manpower policies entailing retraining, education, job placement services, and public job creation.26 The Austrian government agreed to extend tax and social insurance concessions to labour in return for its agreement to moderate wages.27 Governments surmounted problems of intergenerational equity by establishing pension schemes. Wage moderation now which translated into higher incomes later might benefit future generations at the expense of present ones, a fact which might cause those currently working to hesitate to defer their gratification. Governments therefore indexed pensions not just to changes in the cost of living, but to increases in the living standards of the currently employed. In Germany, state pension schemes were revised in the 1950s to 'ensure that persons retiring on a pension of a given money value should have the right to an adjustment in line with the increase in the general prosperity of the country, which was confidently expected to result from the rising productivity of those at work' (Shonfield, 1965: 92). 3.2.3 Coordinating mechanisms A third set of institutions coordinated bargains acrossfirmsand sectors, internalizing the externalities from wage moderation in one sector for investment and hence future wage growth in others.28 Either bargaining was centralized in the hands of a trade union federation and a national employers' association, or the government took an active role in the concertation of bargains reached by individual unions and Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 50 Barry Eichengreen employers. Under these 'corporatist arrangements', the major labour and management associations enjoyed quasi-public status, and negotiations in different industries were coordinated under state auspices (see Bruno and Sachs, 1985; Newell and Symons, 1987). Bonding was important here as well: wage compression or 'solidarity' enticed sectoral unions to accept the loss of autonomy implied by centralized bargaining, while promises to provide private companies with inputs produced by public enterprises at submarket prices elicited the cooperation of employers. Thus, the Rehn model in Sweden made wage compression an explicit part of the bargain. In Holland the compression of wage differentials was a goal of the Board of Mediators from the immediate aftermath of World War II until the end of the 1950s; a skilled/unskilled differential of 10 per cent for weekly wages and 20 per cent for hourly earnings was maintained virtually unchanged to 1959. In Belgium, the progranimation sociale of the 1960s entailed a narrowing of regional and intrasectoral wage differentials and of disparities between men and women (Windmuller, 1969: 340; Dancet, 1988: 103-4). It is important to highlight two respects in which this framework differs from the recent literature on labour markets and macroeconomic outcomes. First, the present framework departs from the 'hump-shaped' model (Calmfors and Driffill, 1988), in which it is argued that both highly centralized and highly decentralized labour markets produce efficient outcomes. The argument here is that, while both highly centralized and highly decentralized structures may have facilitated adjustment to changing supply and demand conditions, competitive decentralized wage setting was ill-suited for internalizing the externalities for investment that spilled across sectors and over time. To the extent that such externalities prevailed, coordinated wage bargaining was required for efficiency. In the presence of externalities, competitive, decentralized markets did not suffice. Second, the term 'corporatism' is used differently here than in the recent literatures in both macroeconomics (e.g. Bruno and Sachs, 1985; Crouch, 1985) and comparative politics (e.g. Schmitter, 1981).29 Here it refers to the concertation of negotiations between labour and management by government. Those who follow the above-mentioned authors use it instead to refer to countries where labour is centralized and where there exists low shop floor autonomy. Here I refer to these factors as 'centralization'.30 This term as well is used differently than in the recent literature, where in practice it refers to the centralization of labour representation.3 * Here it refers instead to centralization on both the labour and employers' sides. 4 The evolution of domestic institutions Authors like Olson (1982) emphasize the cleansing role of World War II in dissolving the quilt of interest groups and institutional restraints that had grown up in previous years and stifled growth. Their removal, in this view, freed the market to operate efficiently. The perspective I develop here is different. I emphasize continuity between pre- and postwar institutional arrangements. Postwar arrangements, rather than being created from scratch to fill a temporary institutional vacuum, were lineal descendants of earlier institutional structures. As I argue in more detail below, coordinating institutions are resistant of change. While the Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 51 disruptions of war and reconstruction afford more scope for change than do normal peacetime periods, this does not alter the fact that the scope of such change is necessarily incremental, implying continuity with the past.32 4.1 An overview of the historical dynamics As Panitch (1979) and Maier (1984) remind us, efforts to institutionalize labourmanagement cooperation under state aegis can be traced to the late-nineteenth century. Protectionist responses to the crisis of 1873 in many European countries constituted a part of a political bargain between nascent interest organizations, the alliance of iron and rye in countries like Germany and France affording the most prominent example. With the rise of unionization, labour was introduced into discussions between producers' organizations. World War I was a watershed: it created excess demands for labour and enhanced the bargaining power of unions, conscious of the special vulnerability of governments and societies to work stoppages. Unions and management were brought together by government in an effort to conclude economy-wide wage negotiations that might head off labour unrest. Public officials used the authority of ministries of munitions to encourage this process without resorting to legislation (see Maier, 1984: 43). Labour's leverage diminished with the termination of hostilities and the re-emergence of unemployment. In some countries, new initiatives- the Stinnes-Legien Agreement in Germany, the NIRA in the USA, Popular Front labour policies in France, the Saltsjobaden agreement in Sweden - ensured continuity. Such frameworks were concessions to the labour movement by governments seeking to head off more radical alternatives. Frequently they were negotiated with the political arm of the labour movement, which had acquired a parliamentary presence. Maier suggests that these arrangements are more accurately characterized as 'supervised pluralism' (separate, decentralized negotiations conducted under broad procedural guidelines set down by government) than as economy-wide concertation.33 Centralized negotiations and encompassing organizations - 'state corporatism' to use Schmitter's (1979) phrase - were established only in Mussolini's Italy and Hitler's Germany.34 To minimize work stoppages and mobilize resources for the pursuit of war, between 1940 and 1945 many countries centralized bargaining between labour and capital under the heavy hand of government. Where postwar reconstruction was particularly difficult, these arrangements were allowed to persist into the postwar years. By the 1950s, however, reaction had set in. Governments withdrew, allowing labour and capital, often still federated into peak associations, to concentrate on bilateral bargaining. Prominent exceptions notwithstanding, intervention was limited to granting recognition and a representational monopoly to the leading interest organizations, using moral suasion to facilitate their cooperation, and posting bonds that would be lost in the event that any party reneged on the terms of its long-term commitment. In the 1960s, as bilateral cooperation between capital and labour became increasingly problematic, for reasons discussed below, governments intervened directly with statutory sanctions and controls.35 Maier (1984) argues that corporatist structures were especially precocious in small, open economies whose vulnerability to external disturbances encouraged the development of cooperative response mechanisms. Where wartime damage had Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 52 Barry Eichengreen been extensive, as in Norway, the case for government involvement in the economy was particularly compelling. Germany and Austria, as subjects of occupation, felt vulnerable for different reasons, but their response was similar. In such countries, sectoral unions were more willing to delegate authority to peak associations, while labour and management were willing to tolerate government guidance of negotiations (Esping-Andersen, 1985: 217). It might be added that the large-numbers problem that bedevils attempts to arrange cooperative solutions was less debilitating in smaller than larger European countries.36 4.2 Overall trajectories Most countries experienced increases in centralization over the two postwar decades, as employers' associations were rebuilt and union membership expanded. The growing centralization of wage negotiations can be seen as a mechanism for preventing individual groups of workers from responding to additional investment by their particular employer, by escalating their wage demands in order to appropriate the surplus generated by that investment. That this danger had mounted with time reflected the decline of unemployment to extremely low levels by the mid-1960s and fading memories of high unemployment in the 1930s, which had worked to subdue labour militancy; hence the institutional response. In all countries but Portugal and the Netherlands, arrangements moved in more corporatist directions over time. In the cases of Norway and Austria this movement was slight, since the highly developed state of corporatism in the 1950s left relatively little scope for further articulation. In Portugal, the authoritarian Salazar regime applied a strong corporatist hand throughout the period, while the Dutch government, so heavily involved in bargaining at the beginning of the period, reduced its role over time. The growing involvement of government in the economy-wide concertation of sectoral negotiations can r5e seen as a way of supplementing efforts at centralization, with the goal of preventing individual groups of workers from appropriating any surplus generated by decisions to invest. A third direction taken by institutional developments in the 1960s was the expansion of the welfare state. In many countries, health, unemployment, retirement and training programmes were elaborated. Governments more aggressively pursued their commitment to policies of full employment. This can be seen as an attempt to provide new bonds for labour to contain the resurgence of labour militancy. 4.3 Recapitulation The preceding sections have traced the evolution of institutions developed in Western Europe after World War II to coordinate, monitor, bond and enforce the compliance of economic interest groups with the terms of their bargain to moderate wage claims and boost investment. For reasons rooted in national histories and politics, European countries differed in the articulation and subsequent trajectory of such institutions. France, Italy, Ireland and the UK possessed neither the centralization nor the corporatization required to coordinate, monitor and enforce such bargains, though they moved in more corporatist directions over time. Danish and Belgian arrangements started out more centralized, though there too the initial, extent of Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 53 corporatism was modest; they moved more quickly in the corporatist direction subsequently. The Netherlands, another country characterized by an intermediate degree of centralization, was initially highly corporatist and became less so over time. The role of government in supporting corporatist arrangements increased significantly in the more centralized German and Swedish economies, and modestly in the already highly centralized and corporatized Norwegian and Austrian ones. 5 The structure of international institutions 5.7 The nature of the bargain For deferring consumption to be worthwhile, investment had to be productive. To put it another way, for investment to stimulate growth, there had to be a market for the goods produced by domestic industries whose capacity was augmented and whose efficiency was enhanced. Otherwise the terms of the domestic bargain would be rendered unattractive; if investment failed significantly to boost productivity and future incomes, workers and managers would be unwilling to defer current consumption for future gains. International trade, and intra-European trade in particular, allowed countries to specialize in the production of goods in which they had a comparative advantage without regard to limits on the demand for those products existing at home. By enhancing efficiency, this improved the intertemporal terms of trade between current and future incomes, encouraging wage restraint and investment. But efforts to expand trade had to surmount a further set of coordination and commitment problems. Restructuring a national economy along export-oriented lines was costly. Sinking the costs of reallocating resources along lines of comparative advantage could be an expensive mistake if one's trading partners reneged on their commitment to openness. Having incurred the costs of shifting resources into the production of exportable goods, countries mightfindtheir access to export markets blockaded, rendering their investments uneconomical. Before reorienting policy in this direction, governments therefore had to be convinced that their partners' commitment to openness was credible. This problem of collective action, though relevant to all European countries, was particularly pressing in the German case. Other European countries were particularly sceptical of its commitment to openness, given memories of the Schachtian policies of the 1930s and the Second World War (Berger and Ritschl, 1995). Germany had been Europe's dominant supplier of capital goods and the single largest demander of raw materials and consumer goods produced by other European countries. If it could not be relied on to supply the capital goods needed for the expansion and modernization of industry in other European countries, and to purchase the consumer goods and other merchandise which would be produced with that capacity, sacrificing efficiency for self-sufficiency would be the sensible strategy. Institutions which rendered credible Germany's commitment to intra-European trade could thus go a long way towards reconstituting the traditional pattern of comparative advantage and towards curing the dollar shortage (the balance of payments deficits of European countries vis-d-vis the USA, due mainly to their excess demand for capital goods). Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 54 Barry Eichengreen For Germany, recovery and growth required the consent of the occupying powers to the removal of controls on industrial production. Postwar policy-makers required in return that Germany be integrated into the European economy and that barriers to exit be erected to prevent that commitment from being reversed. German policy-makers required guarantees of their country's access to imported raw materials, industrial intermediates and foodstuffs in light of the prominence the Nazis had lent to Germany's dependence on foreign supplies. Institutional solutions to these commitment and coordination problems were provided by the European Payments Union and the European Coal and Steel Community. The ECSC provided monitoring and enforcement technologies guaranteeing national steel industries access to one another's coal and iron ore. The EPU entailed the establishment of institutions to monitor member country trade policies and the posting of bonds increasing the cost for any European country of reneging on its commitment. As an internationally coordinated initiative, it solved the 'after you Alphonse' problem discouraging unilateral steps towards convertibility. 5.2 Locking in intra-European trade The effort to rebuild Europe's trade proved a remarkable success. The volume of Western European exports rose by nearly 9 per cent annum in the 1950s and 1960s. The exports of all Western European countries but the UK grew at rates in excess of 5 per cent, with those of West Germany and Italy expanding at a double-digit pace. The ability of European countries to exploit their comparative advantage heightened the efficiency of investment, contributing to the postwar growth miracle. In this development the European Payments Union played an important role. Participating countries were required to agree to a schedule of intra-European trade liberalization. A Code of Liberalization formalized their commitment. By February 1951, less than a year after the EPU went into effect, all existing trade measures were to be applied equally to imports from all member countries. Participants were required to reduce trade barriers by one-half initially, and then by 60 and 75 per cent. The share of quota-free intra-European trade was to rise to 90 per cent by the beginning of 1955. Countries failing to comply with this schedule, or employing policies to manipulate the terms or volume of trade in undesirable ways, could expect to be denied access to EPU credits. Operating the EPU required creating a set of institutions (the Organization for European Economic Cooperation (OEEC), which worked in tandem with the Bank for International Settlements) to monitor compliance and impose sanctions. Drawings on the system were embedded in a mechanism minimizing the likelihood that a country could use EPU credits to exploit its partners by remaining in persistent deficit. No conditions were attached to a country's drawings on its quota of 15 per cent of its intra-EPU trade. But additional credits could be obtained only if a country agreed to conditions set down by the EPU's Managing Board. Discussions were often initiated well before a country's quota was exhausted, and it was made clear that the provision of exceptional assistance was contingent on the country's early adoption of policies of adjustment. Officials of governments receiving exceptional credits were required to appear at meetings of the Board for questioning and to submit memoranda regarding their progress for its review. Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 55 Moreover, US Marshall Plan administrators supported the EPU by providing $350 million of working capital to finance its operation. (In fact, the Economic Cooperation Administration and the State Department had to overcome the opposition of the Treasury and the US Executive Director to the IMF, who feared that the EPU would allow Europe to discriminate against US exports.) The fact that Europe and the EPU depended on Marshall aid reduced the likelihood that a debtor would renege on its agreement with the Managing Board and fail to take corrective action to eliminate its deficit. The provisions of Marshall aid and Europe's financial dependence on the USA allowed the United States to convince a reluctant UK to participate in the system rather than proceeding toward convertibility unilaterally (see Dickhaus, 1993).37 Indeed, for those concerned to construct a commitment technology, the EPU was preferable to unilateral current account convertibility, the other basis on which postwar Europe's trade might have been rebuilt. Convertibility was not technically infeasible, but, as a unilateral policy, it was too easy to reverse (Eichengreen, 1993a). It lacked the multilateral surveillance and conditionality that rendered the EPU an effective institutional barrier to exit. 5.5 Industrial self-sufficiency and national security The ECSC further enhanced the credibility of Germany's commitment to openness by ensuring the French steel industry access to the Ruhr coal that was indispensable to its survival, and by providing German steel producers with guaranteed access to the iron ore of Lorraine. Coal and steel were viewed, rightly or wrongly, as essential to national security and to the rehabilitation of Europe's industrial base. As Pollard (1981: 86) put it, 'it had been precisely these industries which had become a focus of international hostility and national armaments and war-mongering'. By 1950 the inevitability that Allied control of German heavy industry would soon be terminated could not be denied. The question was whether Germany would use its industrial capacity benignly and allow other European nations free access to its products, or whether the rest of Europe would have to build up its self-sufficiency. In response to these dangers, the Schuman Plan of May 1950 proposed to create a common market in coal and steel. Its goal was to abolish protection in trade in coal, iron and steel among the six member states. The discussions culminating in the ECSC banned price discrimination between domestic and foreign customers. A Joint High Authority was created to monitor compliance with the terms of the agreement. Its decisions, within circumscribed limits, were binding on the national governments and private citizens of the partner countries. To render the High Authority accountable to its constituency and thereby lend it legitimacy, a Consultative Committee of industrialists, trade unionists, consumers and dealers was created, and a body of delegates from the parliaments of the member countries (the Common Assembly) was established with the power to vote 'no confidence' in the High Authority and thereby force its members to resign. Finally, a Community Court was established to adjudicate disputes between member governments, individual enterprises and the High Authority. Thus, by joining the Community, member states committed themselves to renouncing certain unilateral initiatives. As Gillingham (1995) puts it, the ECSC Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 56 Barry Eichengreen 'was based on a new idea, supranationality. Membership required transference of sovereign powers to a new European authority.' It is hard to imagine a more effective barrier to exit.38 5.4 Extra-European institutions The EPU and the ECSC were just two of the institutional arrangements committing countries to openness and international trade. They were tailored to Europe's special economic and security needs, and they spoke to the historically specific fears aroused by the continent's experience in the period after World War I, when the commitment to openness had proved ephemeral. At the same time, they were buttressed by the global framework in which they were embedded, especially the General Agreement on Tariffs and Trade (GATT) and the Bretton Woods system. In contrast to the ad hoc tariff truce conferences of the 1920s, GATT was quickly recognized as an ongoing process. Even before completing one round of negotiations, signatories agreed to another. The repeated-game nature of their interaction mitigated against the temptation to engage in non-cooperative behaviour. The many GATT rounds, aimed at slashing tariffs, proved effective in dealing with the ever-present protectionist pressures' (Bhagwati, 1988: 41). Another important institutional innovation was to transfer tariff-setting authority from the US Congress to the Executive Branch, which was less susceptible to narrow constituency pressures.39 As epitomized by the Smoot-Hawley Tariff, US commercial policy had been dominated by Congressmen who proposed measures that afforded protection to industries in their districts without taking into account the negative externalities imposed on other districts. The 1934 Reciprocal Trade Agreements Act delegated authority to an executive with a national constituency more likely to take into account the external effects of district-specific trade policy measures. The executive was authorized to negotiate and implement pacts with other nations in which each agreed to cut tariffs on items of interest to the other by up to 50 per cent without further recourse to Congress. New centres of expertise like the State Department were thereby created which diluted the protectionist agenda with broader international economic and political considerations. Given US dominance of the postwar trading order and its leverage over multilateral trade negotiations, this lent impetus to the GATT process. Perhaps most importantly from the present point of view, GATT avoided conflicts between regional and global trade liberalization, conferring institutional blessings on discrimination in trade when pursued with the goal of establishing a regional customs union. Countries were allowed to offer one another intraregional trade preferences so long as in the course of doing so they did not raise barriers against imports from the rest of the world. This provision worked to the special advantage of Europe, where for historical reasons the scope for regional liberalization initiatives was greatest. The role in the early postwar years of the International Monetary Fund and the Bretton Woods system of pegged exchange rates should not be exaggerated. Efforts to have members establish par values and declare their currencies convertible produced little of more than symbolic value. Initially, foreign exchange rationing Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 57 remained widespread. France experimented with multiple exchange rates. Devaluations were undertaken in 1949 and on other occasions without the prior consultation written into the IMF Articles of Agreement. Starting in 1950, Europe's international monetary relations were shaped by the EPU, not the Bretton Woods institutions. The strength of the Bretton Woods Agreement was its ability to prove compatible with specially tailored arrangements like the EPU. After 1958 and the termination of the EPU, the Bretton Woods system came into its own. For nearly a decade and a half, the international monetary system operated without a major crisis (Bordo, 1993). The UK's 1967 devaluation and revaluations by Germany and the Netherlands perturbed but did not disrupt the system. Exchange rate stability encouraged the expansion of trade. Large-scale private foreign lending resumed. Inflation, though accelerating in the 1960s, remained moderate by the standards of prior and subsequent decades. The import of the Bretton Woods system,firstas a structure into which the EPU happily fitted and then as the framework for Europe's international monetary affairs, was that it provided a nominal anchor that stabilized price expectations. So long as the USA remained committed to pegging the dollar to gold at $35 an ounce, and European countries remained committed to pegging their currencies to the dollar at declared par values, clear limits were placed on inflationary tendencies. Persistent high inflation in the USA would drive the American balance of payments into deficit and exhaust the country's gold reserves. Persistent high inflation in Europe would drive a European country's balance of payments into deficit and threaten the viability of its par value. Initially, ample US gold reserves and the dollar's role as a reserve currency relaxed the first of these constraints, while the maintenance of capital controls placed limits on the operation of the second. But while such factors relaxed the constraint on policy, they did not remove it. Hence, governments' investments in the Bretton Woods system of pegged exchange rates lent credibility to their stated commitments to the maintenance of price stability. The stability of price expectations under Bretton Woods in turn enhanced the effectiveness of macroeconomic stabilization policy, averting serious recessions and preventing cyclical declines in investment expenditure.40 So long as the pegged exchange rates of the Bretton Woods system remained credible, workers agreeing to a sequence of wage bargains did not have to worry that their nominal value would be inflated away. Any acceleration of inflation could be regarded as temporary, rendering compensating wage increases less urgent. Thus, when governments used Keynesian demand stimulus to counter a recession, the pressure of demand was less likely to translate into wage inflation and more likely to encourage production. Demand management policy proved effective. The consequent absence of serious recessions sustained investment at high levels.41 Indeed, the main difference in investment behaviour between the 1950s and 1960s and the interwar period was not that postwar investment rates were higher during expansions, but that governments succeeded in preventing investment from collapsing in recessions. This encouraged firms to contemplate sequences of related investment projects which would yield high returns if not interrupted by recessions. And the high returns on investment improved the terms of the trade-off between current and future consumption for workers and capitalists contemplating policies of moderation. Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 58 5.5 Barry Eichengreen Implications Institutions like the EPU and the ECSC, buttressed by extra-European arrangements like the Bretton Woods system and GATT, solved commitment and coordination problems that had to be surmounted for intra-European transactions to be rebuilt and rapid growth to be sustained. Still, not all countries took equal advantage of the opportunities afforded by this new institutional basis for intra-European trade. The best counterexample is France, whose governments pursued policies that cloistered industry from international economic pressures until the 1960s. The OEECs Code of Liberalization anticipated removing three-quarters of quotas on private imports among the member countries and their dependencies by February 1951. While France temporarily reached this level in April of that year, the subsequent deterioration of its balance of payments led the government to reimpose quotas on all intra-European imports in February 1952. The OEECs 'minimum goal' of 75 per cent was only reached in February 1955 (Table 2.1), and even then a special 10-15 per cent 'provisional compensatory tax' was applied to most of the freed commodities to insulate French producers from import competition (Baum, 1958: 102-3).42 In manufacturing, imports accounted for only 8 per cent of domestic consumption as late as 1959. For half of all manufacturing industries thefigurewas less than 5 per cent. In thirteen of thirty-two industries, imports were essentially nil (Adams, 1989: 155). These low levels of import penetration reflected the relatively restrictive non-tariff barriers to trade maintained by France even during the EPU years. Only following the establishment of the European Common Market did import exposure rise, from 8 per cent of domestic consumption of manufactures in 1959 to fully 25 per cent in 1980, with most of the change occurring in the 1960s. The impact on domestic market conditions was profound: all French industries experiencing an increase in import penetration due to the establishment of the European Economic Community (EEC) saw their product prices fall in thefirstfive Common Market years. All industries with an increasing propensity to export saw their product prices rise (Adams, 1989: 172). A clearer example of restructuring along the lines of comparative advantage is hard to imagine. With this restructuring came a dramatic increase in the rate of French economic growth, plausibly reflecting higher returns on investment due to a greater tendency to invest along lines of comparative advantage, which in turn enhanced the incentive and hence the tendency to invest. 6 The decline of the postwar settlement Eventually the Golden Age drew to a close. Output and productivity growth decelerated. Wages exploded, investment rates slumped, and the growth of trade no longer outstripped the growth of output to the same extent as in the 1950s and 1960s. The domestic and international settlements that had provided the basis for the postwar growth miracle ceased to function. A satisfactory explanation for the rapid growth of the Golden Age must also account for this subsequent deceleration. Afirstset of arguments is based on the idea that the postwar growth process contained the seeds of its own destruction. On the Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Table 2.1. Intra-European trade: percentage liberalized at selected dates0 Country 30 June 30 June 30 June 30 June 30 June 30 June 30 June 30 June 30 June 30 June 30 June 30 June 1954 1952 1957 1950 1953 1955 1956 1959 1951 1958 1960 1961 Austria 53 56 55 53 58 47 — — 64 54 39 53 — 53 81 42 57 — 75 66 50 75 — — 41 75 76 51 83 — 75 85 63 90 — 75 75 68 — 77 — 41 75 100 75 100 — 75 88 63 46 36 87 92 76 — 90 (90) — 75 100 75 93 — 91 92 — 58 76 87 93 76 51 90 (90) 29 77 100 76 93 — 91 92 — 80 82 88 93 76 75 90 (95) 29 77 99 75 93 — 91 92 — 84 90 91 — 86 82 92 (95) 29 90 99 78 94 — 93 92 — 94 90 96 — 86 — 93 (95) 29 90 99 81 94 — 93 91 — 94 90 96 — 86 — 94 (95) 29 90 98 81 94 — 93 91 — 94 90 96 — 96 91 92 (76) 29 90 98 81 94 — 93 91 — 95 90 97 — 95 (97) 92 (76) 85 90 98 81 94 61 93 91 — 97 90 97 — 95 (99) 93 (72) 85 90 98 85 85 70 98 91 (52) 97 56 65 66 71 81 84 89 83 83 91 91 (94) BLEU Netherlands Denmark France Germany (FR) Greece Iceland Ireland Italy Norway Portugal Spain Sweden Switzerland Turkey United Kingdom M e m b e r countries combined a Percentage of private trade freed from quantitative restrictions. Base year is 1948, except for: Germany (FR) (1949), Spain (1950), Austria, Greece (1952), Turkey (1959), Iceland after 1960 (1958), and Benelux after 1957 (1953). In the case of Benelux, the percentages shown for 1959 and 1957 reflect the change in the base year rather than any effective change in liberalization. Source: Eleventh Annual Economic Review, Europe and the World Economy, Paris: OEEC, 1960, p. 185. 60 Barry Eichengreen domestic side, wage moderation was supported not just by centralized bargaining and corporatist arrangements, but by relatively elastic supplies of underemployed labour and memories of high unemployment in the 1920s and 1930s.43 The passage of time allowed memories of unemployment a generation before to fade, taking with them their restraining influence. At first, ample supplies of labour flooding into Germany from its East and into Holland from Indonesia encouraged docility on the part of their unions. Underemployed labour in the agricultural sector had similar effects in countries like Italy and France. But a decade and a half of rapid growth transformed Germany and the other countries of Western Europe from high- to low-unemployment economies, encouraging the unions to flex their muscles. A problem with this explanation is that it underestimates the market power of unions in many European countries. As early as the 1950s, supplies of skilled workers were inelastic to industries. Accounts are unanimous in ascribing market power to unions, aside from a few countries, such as Italy and France prior to 1960. Indeed, had elastic labour supplies stripped workers of market power, there would have been no need for institutions to lock in wage moderation once the growth process got under way, since wages would have been determined by competitive labour market conditions; there would have been no danger of wage rises designed to appropriate future profits to discourage investment. Many readily observed domestic institutions would have been without a clear rationale. On the international side, the success of GATT in stripping away tariff barriers may have encouraged governments to cultivate less transparent and tractable non-tariff barriers to trade with which the GATT process was less able to cope. The growing number of GATT signatories created a large-numbers problem hindering negotiations. The scope for the United States to encourage other countries to liberalize by allowing them to discriminate against it diminished hand in hand with declining US industrial predominance and Europe's catching up. While this explanation has appeal when applied to the 1980s, it is anachronistic to attribute mounting tensions on the international side in the late 1960s to the breakdown of the GATT process. The most important voluntary export restraints came later. US economic leverage, while in decline, was still considerable. The large-numbers problem cited by commentators on the Uruguay Round was not yet pressing. Although there may still be a sense in which the very success of the postwar growth miracle contributed ultimately to its downfall, the framework developed above, with its emphasis on institutions, points to five additional factors which together helped to ring down the curtain on the Golden Age. It directs attention to one of those five factors as possessing special importance. 6.1 Capture An obvious place to start is Olsonian capture (Olson, 1982). Olson's argument is that the passage of time enables special interest groups to gain leverage over the operation of institutions and to turn them to their selfish ends. The new-installed union leaders of the postwar period, who initially identified with the workforce as a whole, may have come to favour senior workers at the expense of their junior counterparts as they themselves gained seniority, creating the insider-outsider Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 61 problems that undermined the smooth operation of the labour market (Lindbeck and Snower, 1988). Governments initially able to use public programmes to bond unions and firms may have fallen prey to capture by the very agents whose behaviour they sought to influence, as the latter acquired knowledge of the workings of government bureaucracy. While this is an appealing story, the general formulation does not explain the timing of events - why the Golden Age of rapid growth drew to a close in the late 1960s or early 1970s - or their abruptness. Perhaps public institutions are especially susceptible to capture; if so, then the increasingly prominent role played by government in the operation of domestic institutions starting in the 1960s, as it sought to paper over cracks in the postwar settlement through the use of wage and price controls and direct intervention in labour-management negotiations, could have opened up avenues for capture where these had not existed before, thereby helping to explain the timing of events. Unfortunately, like other variants of the Olson thesis, this one does not easily lend itself to formal statistical tests. 6.2 The oil shocks A second explanation for the end of the Golden Age emphasizes the oil shocks of the 1970s. A large literature addresses the question of whether these have the capacity to explain the productivity slowdown, generally reaching agnostic to negative conclusions.44 An institutional framework suggests a different perspective. By reducing even modestly the returns on investment (by raising the relative price of energy, or even just by making that price more variable and therefore requiring firms to diversify away the associated risks by investing in a wider variety of different kinds of equipment), the oil shock worsened the intertemporal terms of trade for workers offering wage concessions now in return for higher investment, productivity and incomes later. Thus, even a modest impact of the oil shock on the productivity of new investment could have tipped the balance for workers deciding whether to offer wage moderation now in return for higher incomes later, and thus had major implications for the level of investment and rate of growth. The problem with this explanation is in the timing. Thefirstoil shock came along in 1973, but the wage explosions that mark the breakdown of the postwar settlement were already evident at the end of the 1960s. The oil shocks might help to explain why the postwar settlement, once shattered, proved so hard to put back together, but they are less obviously a factor in explaining its downfall. 6.3 The breakdown of Bret ton Woods A third potential explanation is the collapse of the Bretton Woods system of pegged but adjustable exchange rates. With the breakdown of Bretton Woods, not just inflation but the business cycle grew increasingly volatile.45 The elimination of par values removed a nominal anchor that had stabilized price expectations. With the commitment to par values removed, agents had no reason to regard an acceleration of inflation as temporary. When governments stimulated demand in the effort to offset a recession, this provoked compensating wage increases; aggregate demand policies therefore elicited inflation rather than stabilizing output. Governments Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 62 Barry Eichengreen having lost the capacity to counteract incipient recessions, the business cycle and investment grew more volatile.46 The institutional perspective developed above points to channels through which this increased volatility could have depressed the rate of economic growth. Workers were willing to trade wage moderation now for higher incomes later, so long as they had confidence that sacrifices now would deliver investment, growth and higher incomes later. Recessions were an obvious threat to this confidence. Wage moderation might not succeed in stimulating investment if in the interim the level of activity, profits and the incentive to invest collapsed into a recession. As noted above, investment rates were higher in the 1950s and 1960s than they had been in the 1920s and 1930s not because those rates were significantly higher during expansion phases of the business cycle, but because serious recessions in which investment rates collapsed were successfully averted after World War II.47 So long as serious recessions were absent, the trade-off between wage moderation now and investment later remained attractive to European labour. Once the Bretton Woods nominal anchor was removed and Keynesian stabilization policy lost its power, allowing recessions to re-emerge, labour lost the incentive to keep its part of the bargain. The timing of events is suggestive: although the breakdown of the Bretton Woods system only took place in 1971-3, those events were widely anticipated, and the system was already showing signs of instability in the late 1960s.48 The limitations of this explanation are two. There is reverse causality: labour militancy developing for other reasons could have increased the responsiveness of wages to prices and the pressure of demand, implying that the change in labour market behaviour was a cause rather than a consequence of the breakdown of Bretton Woods. In addition, the view emphasizing the effectiveness of countercyclical stabilization under Bretton Woods may exaggerate the prevalence of such policies in Europe prior to the 1970s. In the 1950s Keynesian demand management policies were little used outside the UK. In Germany, to take one example, they were quite inconsistent with the prevailing ideology of ordoliberalism. In the 1960s they may have gained popularity in countries like France, but over wide portions of Europe they were still scarcely used. 6.4 Capital mobility A fourth explanation for the end of the Golden Age is rising capital mobility. This is the factor whose particular importance is highlighted by the institutional framework of previous sections.49 Section 2 suggested that the postwar growth process had to surmount two problems of dynamic inconsistency. Thefirstwas that capitalists had to be precommitted to invest their profits in the future, so that workers would restrain their wage demands now and the liquidity would be made available to finance investment. The second was that workers had to be precommitted to moderate their wage demands later, so that capitalists would invest now in the expectation of reaping future profits. Institutions undertaking monitoring, informationdisseminating and bonding functions were developed with both of these problems in mind. It can be argued that the structure of the relevant institutions was predicated on limited international capital mobility. In a situation of limited capital mobility, it Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 63 was possible, for example, to threaten the imposition of taxes on distributed profits at rates that would not be feasible when capital could evade them by moving abroad. As capital mobility began to increase in the 1960s, the bonding mechanisms of the 1950s therefore lost much of their effectiveness. Where wage moderation might once have guaranteed investment, international capital mobility now gave management the option of investing abroad rather than at home. Domestic workers no longer assured that wage moderation would translate subsequently into higher productivity and labour incomes had less incentive to be moderate. The problem with this explanation, advanced by Thelan (1991) and others, is that there exist equally compelling arguments working in the other direction. Capital's possession of an exit option, by providing investors with a credible threat, should have worked to encourage the acquiescence of labour. The knowledge that labour militancy might cause capital to move abroad was precisely the kind of information that should have induced labour to exercise restraint in order to prevent capital from exercising its exit option. Indeed, if capital was mobile inward as well as outward, wage restraint could induce foreign capital to flow in, augmenting the plant and equipment with which labour could work and eventually rendering it better off. Thus, it is still possible to use the capital mobility argument to explain why dividend payout rates might have increased and investment rates slumped in the 1970s. Capital was able to indulge its preference for consuming a higher share of profits without provoking wage demands sufficient to prevent the accrual of those profits, labour's militancy having been restrained by capital's exit option. What this hypothesis cannot explain, given that capital mobility should have exercised a restraining influence on labour militancy, is the simultaneous explosion of wage demands. 6.5 The end of catch-up In the introduction to this paper, catch-up - the scope for exceptionally rapid expansion as European economies recovered prewar levels of output and closed the productivity gap vis-a-vis the United States - was offered as an alternative to institutional explanations for postwar Europe's growth. A different view is that the scope that existed for rapid growth during the catch-up period reinforced the institutions erected to solve commitment and coordination problems, and that once the catch-up phase passed, those institutional arrangements lost much of their effectiveness. Recall that at the centre of the institutional argument is a time-inconsistency problem. Workers and capitalists are both best off if they can agree to defer current consumption in return for future gains, which take the form of additional investment that results in higher productivity and income for all concerned, but neither is willing to agree to defer without an assurance that the other will do the same. Assume now a decline in the rate of return on deferring current consumption because, for example, the return on investment and the underlying rate of productivity growth both fall. The incentive for the parties to the agreement to resist the temptation to renege on its terms is correspondingly reduced. Institutions which Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 64 Barry Eichengreen were adequate to contain this temptation previously may no longer suffice, since the intertemporal terms of trade have changed. The end of the catch-up phase of postwar Europe's growth could have played precisely this role, by reducing the return on investment and weakening the incentive for capital and labour to adhere to their agreement to defer current consumption in return for now more meagre future gains. From this perspective, the scope for catch-up and the institutional arrangements developed to solve time-inconsistency problems following the war should be viewed as interacting positively with one another, rather than as competing explanations. The question then becomes why, if institutions were developed after World War II to solve the commitment and coordination problems that confronted the European economy at that time, it did not prove possible to reinforce them to meet these new circumstances. In fact, there was some institutional adaptation in the 1960s. Welfare state programmes to bond labour expanded greatly in the 1960s. Government acquired an increasingly prominent role in the operation of labour market institutions in many countries over the course of this decade. Growing use of statutory wage controls, concertation of sectoral negotiations, and direct intervention in wage bargaining are all characteristics of government policy in the 1960s that can be seen as a response to the need to deal with the second dynamic-inconsistency problem. But the increasingly prominent role of government was superimposed on existing institutions rather than superseding them. The underlying structures - craft versus trade unions, cohesive or fragmented trade associations, the degree of centralization of wage bargaining - changed only slowly. There are good reasons to expect socioeconomic institutions like those described in this paper to change only gradually. Institutional arrangements coordinate the actions of a large number of individual parties. It may not be in the interest of any one of those parties to modify existing arrangements unless the others can be made to go along, which poses formidable coordination problems in a decentralized setting. Even in a reasonably centralized market like Germany, where there are sixteen large sectoral unions and a matching number of sectoral employers' associations, it may be hard to get the social partners in any one sector independently to alter the timing and structure of their bargaining arrangements, if the cost is forgoing the coordination benefits provided by an established tradition of intersectoral pattern setting. As David puts it, 'Institutions generally turn out to be considerably less "plastic" than is technology.' (See David, 1993; Zysman, 1993.) In this light, the exceptional opportunity that is afforded by an extraordinary event like World War II is clearly evident. A wartime disruption which suspends the operation of normal peacetime institutions reduces the opportunity costs of coordinating a shift to new arrangements. That the most dramatic changes in domestic arrangements took place in countries like Norway, the Netherlands and Belgium, many of whose labour, management and government leaders were exiled, and who took the interlude as an opportunity to plan a wide-ranging package of institutional changes, is much more than a coincidence from this point of view. Yet what is remarkable given the extent of wartime dislocation is how durable early institutional arrangements remained. As section 3 shows, there was very considerable continuity between Europe's postwar institutions and the institutional developments Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 Institutions and economic growth 65 of the interwar years and even the pre-1913 period. The exceptional circumstances of World War II and its aftermath may have provided an opportunity to adapt Europe's institutions to the special needs of postwar economic growth, but this could be done only where the departure from prewar arrangements was not too great.50 It is not surprising that, when in the 1960s circumstances again changed but without an event like World War II to interrupt the coordination functions of existing institutions, innovation was slow to come. 7 Conclusion European economic growth in the quarter of a century that ended in 1973 outstripped growth in any period of comparable length before or since. The elements of Europe's growth miracle - wage moderation, high investment and rapid export growth - were delivered by a tailor-made set of domestic and international arrangements - on the domestic side, the social market economy; on the external side, international agreements and supranational institutions - that solved problems of commitment and cooperation that would have otherwise hindered the resumption of growth. Why then did growth slow after 1971? These exists a wealth of explanations, including the capture of socioeconomic institutions by special interest groups, the breakdown of the Bretton Woods system, the oil shocks of the 1970s and the rise in international capital mobility. While it is not possible to rule out these explanations, each of them has serious limitations. I therefore single out the role played by another factor: the interaction of postwar 'catch-up' with the sustainability of cooperative arrangements. The faster was growth, the greater was the willingness for workers and capitalists to defer current compensation in return for future gains. The period of rapid catch-up following World War II was therefore ideal for sustaining cooperative behaviour. As the scope for catch-up was spent, the incentive to renege on cooperative agreements was heightened, and agreements to cooperate broke down. Wage pressure intensified, investment slumped and the rate of growth was further depressed. Thus, whereas the beginning of the introduction to this chapter presented catch-up and institutional arrangements as two alternative explanations for postwar growth, my conclusion is that it would be more accurate to see them as complementary factors that reinforced one another. NOTES Prepared for thefinalconference of the Commission of the European Communities' Programme on 'Comparative Experience of Economic Growth in Postwar Europe', All Souls College, Oxford, December 1993. This chapter is part of a project on postwar recovery and growth in Western Europe, which receivesfinancialsupport from the National Science Foundation and the Center for German and European Studies of the University of California. It is an elaboration and extension of a short paper presented to the European Economic Association's Annual Meetings in August 1993, and published in the European Economic Review (June 1994). I thank Brian A'Hearn, Lisa Ortiz and Pablo Vasquez for research assistance, and Lars Downloaded from https://www.cambridge.org/core. London School of Economics Lib, on 03 Mar 2021 at 14:07:05, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511758683.003 66 Barry Eichengreen Calmfors, Paul David, Assar Lindbeck, Torsten Persson, Bob Powell, Gilles St Paul, Solomus Solomou, Rick van der Ploeg and Charles Wyplosz for helpful comments. 1 The unweighted average of the annualized growth rate of GDP per hour worked for eight European countries was 4.4. per cent in 1950-73 but only 2.4 per cent in 1922-37 and 2.3 percent in 1979-88. Calculated from Crafts (1992: table 1) and Boltho (1982: table 1.1). The contrast with recent years would be even more dramatic if the comparison were with 1973-88. 2 Note that the slopes of the regression lines in both of these figures differ significantly from zero at the 99 per cent level of confidence. Interestingly, when both effects are included in a single growth equation, the initial (1950) level of GDP per man-hour remains statistically significant, but not so the wartime and immediate postwar shortfall. 3 Crafts (1992) presents calculations of the growth bonus due to catch-up vis-a-vis the USA and spring-back to prewar levels for the same eight European countries, finding that purged of catch-up and spring-back, growth rates decelerated from 3.1 per cent in 1950-73 to 1.9 per cent in 1979-88. 4 The estimates of Maddison (1976) show the investment rate in Western Europe rising from 9.6 per cent in 1920-38 to 16.8 per cent in 1950-70. 5 See, for example, Crafts (1992: table 2). 6 As Shonfield (1965: 6) put it, The success of the modern capitalist society in reversing the pressures making for high consumption at the expense of investment is one of its outstanding achievements.' In the discussion that follows, I suggest that postwar policy-makers were especially concerned with the second, or supply-side, influence of wage restraint on investment. Moderate and slowly growing real wages can been seen as theflipside of the coin of the low and slowly rising real exchange rate that some authors (e.g. Boltho, 1982, 1993) emphasize as an element of Europe's postwar growth miracle. In simple open-economy macroeconomic models, in which domestic and foreign economies are completely specialized in the production of different goods that are imperfect substitutes for one another, there is a one-to-one correspondence between the real exchange rate and the relative real wage. 7 For an overview of the interwar situation in various European countries, see Maier (1975). A detailed and controversial an...