Thus, we might expect productivity performance in the post-war British economy to be adversely affected by the interaction of the institutional legacies in respect of corporate governance and industrial relations together with full employment and weak competition. This would seem to be a recipe for a malfunctioning LME. And indeed the empirical evidence lends considerable support to this diagnosis.
Econometric studies have shown the following. First, greater competition raised productivity growth where firms did not have a dominant external shareholder (Nickell et al., 1997). In this (typical) case, a fall in supernormal profits from 15 to 5 per cent of value added raised TFP growth by 1 per percentage point per year. Moreover, increases in interest payments relative to cash flow also promoted significantly faster productivity growth.13 Second, greater competition was good for innovation. Geroski (1990) found that, once differences in technological opportunity across industries were taken into account, in the 1970s the positive effects of market power working through expected profits were heavily outweighed by negative effects on managerial effort. Third, Bean and Crafts (1996) examined the consequences of multiple unions for TFP growth in an endogenous-innovation setting where, in the absence of binding contracts, it is predicted that innovative effort will be reduced by the expectation that the workforce will appropriate some of the gains. They found that the presence of multiple unions lowered TFP growth by 0.75 to 1 percentage point per year in the 1954–1979 period.
Case studies strongly implicate bad management and restrictive labour practices resulting from bargaining with unions in poor productivity outcomes. Pratten and Atkinson (1976) reviewed 25 studies of which 23 reported inefficient use of labour, in 21 cases from failings of management and in 14 instances from restrictive labour practices. Inefficient use of labour and industrial relations problems accounted for a significant productivity gap in multinational companies between British plants and those in Germany or the United States in 1972 (Pratten, 1976) which would not have surprised the business respondents to an Oxford survey in the late 1940s who thought such problems were prevalent (Andrews and Brunner, 1950). A strong theme in several studies which highlight low effort bargains is that they were sustained by the weakness of competition; for example, this emerges clearly in the study by Zweig (1951) and also in the seminal work in industrial sociology on restrictive labour practices (Lupton, 1963). A notable aspect of these bargains is that potential productivity gains from new technology were impaired (Prais, 1981). In a well-known and egregious case, motor vehicles, management completely lost control of effort norms in the switch to measured day work with disastrous consequences for productivity (Lewchuk, 1987).
In sum, the evidence is that multi-unionism and the separation of ownership and control adversely affected United Kingdom productivity performance during the Golden Age. Corporate governance and industrial relations were clearly recognizable as the grandchildren of their Victorian predecessors but having mutated into more problematic forms and with a greater downside in the environment of weak competition that prevailed in these early post-war decades. In terms of Figure 1.1, this unfortunate combination of institutions and policy significantly reduced λ.
5.5 Conclusions
The years from 1950 to 1973 saw the fastest ever growth of real GDP per person in the United Kingdom. Even so, it was a period of acute relative economic decline during which rivals like France and West Germany grew much faster and eventually overtook the United Kingdom in terms of income levels. Seen in terms of the scope for catch-up growth, the United Kingdom underperformed to the extent of a shortfall of probably around 0.8 per cent per year with the implication that the level of GDP per person was at least 20 per cent below what would have been reasonable to expect at the end of the Golden Age.
By the 1960s, slow growth was recognized as a problem by British governments but the design of supply-side policy left a lot to be desired throughout the period. Marginal income tax rates were too high, the productivity record of the nationalized industries was undermined by principal-agent problems, investment subsidies were badly targeted, selective industrial policy was politicized and deservedly got a bad name, industrial relations continued to be unreformed, competition policy was too weak and protectionism was reversed much too slowly. In terms of Figure 1.1, this policy configuration implied a downward shift of the Schumpeter line. There were a number of reasons for government failure but an important constraint on policy design resulted from politicians’ fear of unemployment and the perceived need to persuade trade unions to exercise wage restraint.
A key feature of this policy configuration is that it implied that competition was impaired across much of the economy. The interaction of weak competition with institutional legacies in the corporate governance of large companies and the system of industrial relations was not favourable for productivity performance. It allowed bad management to persist in an era of weak shareholders and provided supernormal profits which were shared with multiple unions in higher wages and lower effort. This implied that the productivity potential of new technology was less than fully realized. Nor was a cooperative equilibrium (with high investment by firms and wage restraint by workers) feasible. In the policy environment engendered by the inescapable experience of the 1930s and the shifting centre ground of politics, the outcome was a malfunctioning LME. Rather than the early twentieth century, the Golden Age after the Second World War was the point at which the downside of the early start was felt most severely.
Appendix 1
Cameron and Wallace (2002) provide a game-theoretic version of Eichengreen’s argument that a co-operative equilibrium which entailed high investment in return for wage moderation underwrote high investment in Golden-Age Europe.
The idea is to model a situation where it is profitable for a firm to invest if the union selects a low wage claim but not if the union makes an aggressive pay claim. In addition, if the union knows for sure that the firm would invest rather than pay profits out as dividends a lower pay claim would be optimal. The wage moderation/high investment equilibrium is Pareto-dominant. This requires the two conditions
(Firm) δFπ(w0, L0) > I > δFπ(wH, L(wH))
(Union) (1 + δU)w0L0 > wHL0 + δUwHL(wH)
where I is the investment, δF and δU are, respectively, the firm and union discount factors where a higher value implies greater patience, π is the profit function, and w0, L0 and wH, LH are the wage bills associated with the profit-maximizing employment level chosen by the firm at low and high wages, respectively. The two conditions imply that discounted profits and the discounted sum of total wages are higher with investment and wage moderation.
If, as is likely, the payoffs are subject to some uncertainty, then the equilibrium that is chosen will be based on risk-dominance. This implies that the wage restraint/ high investment equilibrium is chosen if and only if
[δFπ(w0, L0) – I][(1 + δU)w0, L0 – wH, LH – δUwH, L(wH)] > [I – δFπ(wH, L(wH))](wH – w0)L0
In the case where f(L) = ALα this condition can be re-stated as
(δF/δU)[(1 + δU)w0 – wH]L0 > |(α/(1 – α))
where L0 = (αA/w0)1/(1 – α).
This formulation implies that defection from the wage restraint/high investment equilibrium is more likely if δF, δU, or A decrease or wH increases. Thus, the good outcome is less likely when firms and/or workers become more impatient, in the case of adverse technology shocks or when unions’ bargaining power goes up.
The Eichengreen hypothesis that centralized wage bargaining is conducive to the high investment/wage moderation equilibrium might be interpreted as a prediction that δU and δF will be high. However, other aspects of the Golden Age economic environment are clearly relevant and these might include restricted capitalmobility as positive for δU the fixed exchange rate regime as a constraint on wH, and nice technology surprises as positive for A.
Key implications of this formulation are that the good equilibrium is quite fragile and that formulating a decisive test
of the Eichengreen hypothesis is not straightforward. However, this set-up does help to explain why the end of the restricted capital-mobility, fixed-exchange rate Bretton Woods system in the early 1970s may have undermined the co-operative equilibrium.
1 Trade liberalization had only a transitory effect on growth and did not raise the long-run trend growth rate, cf. Badinger (2005).
2 Business R & D was 1.5 per cent of GDP in 1967 compared with 1.1 per cent in both France and West Germany. Verspagen (1996) found the impact of R & D on output growth was not statistically significant in the United Kingdom and the estimated elasticity was tiny, in sharp contrast with the other countries.
3 Converted into modern Programme for International Student Assessment (PISA) equivalents, the United Kingdom was in the low 490s while France and West Germany were around 500 (Woessmann, 2016). The regression estimates in Hanushek and Woessmann (2012) indicate that the United Kingdom would suffer a growth penalty of about 0.1 per cent per year.
4 This is large enough to be a real cause for concern but it is also fair to say that ‘decline’ is an ideological construct which has been associated with the politicization of economic policy (Tomlinson, 1996).
5 However, this was not the only route to rapid catch-up growth. In other countries, for example, Italy, growth was promoted by industrialization based on elastic supplies of labour and undervalued currencies which underpinned investment and TFP growth and also restrained wage increases at least until the late 1960s (Crafts and Magnani, 2013).
6 In the endogenous innovation framework the ‘hold-up’ arises when after a successful innovation workers use their bargaining power to extract a share of the profits. This reduces the incentive to innovate and thus the rate of growth. The more unions are involved in the bargaining, the more profits are reduced. The problem can be eliminated if a binding contract prevents renegotiation or there is no union or if a cooperative equilibrium is achieved with a single union. For a formal model and empirical evidence, see Bean and Crafts (1996).
7 For a textbook treatment see Wallis and Dollery (1999).
8 It is in fact quite predictable that declining industries will take the lion’s share of support from selective industrial policies (Baldwin and Robert-Nicoud, 2007) but this was not perhaps obvious before these policies were tried and failed.
9 Concorde and the Advanced Gas-Cooled Reactor were egregious policy errors (Henderson, 1977).
10 The existence of significant market power in the UK but not in West Germany at this time is confirmed by the similarity of the primal and dual measures of TFP in the latter but not in the former; see Crafts and Mills (2005) for further elaboration.
11 A classic example is the abandonment in 1952 by the new Conservative government of the ROBOT scheme to float the pound which was seen by its supporters as integral to strengthening the role of market forces in the economy (Bulpitt and Burnham, 1999).
12 Nyman and Silberston (1978) actually reported an estimate of 52.6 per cent but they used different criteria. If the data in their paper are analysed on a similar basis to Florence, this falls to 38.5 per cent.
13 This suggests that principal-agent problems were an issue when management was in the comfort zone. It also implies that industrial subsidies which eased financial pressures were potentially detrimental to productivity performance consistent with the prediction in Aghion et al. (1997) for the agency-problem case.
6
From the Golden Age to the Financial Crisis
After the early 1970s, growth slowed down markedly right across Europe. The end of the Golden Age had a number of unavoidable aspects including the exhaustion of transitory components of fast growth such as post-war reconstruction, reduced opportunities to redeploy labour out of agriculture, narrowing of the technology gap and diminishing returns to investment. Moreover, the United States itself experienced a productivity growth slowdown. All-in-all, the scope for catch-up growth was considerably reduced, although by no means eliminated, and in terms of Figure 1.1, both the Schumpeter and the Solow lines experienced adverse shifts. Contrary to what might have been expected in the 1970s, growth in the United Kingdom suffered less of a decrease in the following 30 years and relative economic decline ceased at least for a while.
Two key surprises shaped this reversal of fortunes, namely, the Thatcher experiment and the ICT revolution. These entailed, respectively, the ‘end of the post-war consensus’ and a new general purpose technology. Since New Labour largely accepted the reforms of the 1980s, there were not only significant but sustained changes in supply-side policy as well as opportunities for business to transform production, plus some synergies between these developments. In the context of the ‘third industrial revolution’, there was a distinct possibility that relative economic performance might once again change. Might ICT play more to the strengths of an LME than a CME?
Buoyed by the contribution of ICT, growth of real GDP per person was apparently quite robust from the mid-1990s to the financial crisis. Some even claimed that boom and bust had been abolished, but then it all ended in tears. Since 2007, productivity growth has been in the doldrums and the economy has endured its worst post-war recession. Now, there is a quite widespread belief that the ‘new normal’ is an era of low growth or even ‘secular stagnation’.
This discussion prompts several questions to be addressed in this chapter. Did United Kingdom growth performance actually improve after the Golden Age? How far was Thatcherism an effective antidote to relative economic decline? Do the 2007–2008 crisis and its aftermath reveal that growth prior to this was ‘too good to be true’?
6.1 Growth Performance: the End of Relative Economic Decline
After the 1970s, real GDP per person in the United Kingdom was no longer falling relative to its peer group. The income levels reported in Table 6.1 for 2007 show more favourable ratios for the United Kingdom relative to the other three countries than had been the case in 1973.1 As can be seen in Table 6.2, after the Golden Age each of these countries grew more slowly for the next 20 years or so. United Kingdom growth slowed down less than was the case in France or West Germany so relative performance improved. In the context of the ICT revolution, from the mid-1990s, growth of real GDP per person revived in the United Kingdom and USA but not in France or Germany so this was the point at which income gaps between the United Kingdom and continental Europe were reduced significantly.
Table 6.1 Real GDP per person
France Germany UK USA
1973 12824 13152 12025 16689
1995 (1) 18318 17127 17599 24712
1995 (2) 32636 36216 30232 41861
2007 40143 43558 40697 54093
Notes: 1973 and 1995 (1) are in 1990$GK. 1995 (2) and 2007 are in 2015$EKS. Estimate is for West Germany in 1973.
Sources: Maddison (2010) and The Conference Board (2016).
Table 6.2 Rates of growth of real GDP/person and real GDP/hour worked (% per year)
Y/P Y/HW
1950–1973
France 4.02 5.29
Germany 5.00 5.91
UK 2.42 2.81
USA 2.45 2.57
1973–1995
France 1.65 2.67
Germany 1.76 2.86
UK 1.76 2.40
USA 1.81 1.27
1995–2007
France 1.70 1.77
Germany 1.54 1.70
UK 2.41 2.09
USA 2.18 2.30
2007–2016
France 0.06 0.66
Germany 0.84 0.68
UK 0.19 0.09
USA 0.46 0.85
Note: Germany is West Germany prior to 1995.
Source: The Conference Board (2017).
Slower growth was accompanied by smaller shares of non-residential investment in GDP, as is reported in Table 6.3. The difference between the United Kingdom and its European peer group was much smaller at the turn of the century than at the end of the Golden Age (cf. Table 5.2). These figures, however, only cover tangible capital whereas, by the early t
wenty-first century investment in intangibles in the United Kingdom was nearly as big and was a larger share of GDP than in France or Germany so that the total United Kingdom investment rate exceeded these countries.2 By contrast, United Kingdom investment in R & D as a share of GDP had fallen compared with 1970 and was lower in 2000 than in France, Germany and the United States and less than half the proportion in the leading country, Sweden. Since R & D investments are estimated to have a very high social rate of return this represents a disappointing United Kingdom weakness (Frontier Economics, 2014).
Table 6.3 Investment in broad capital, c. 2000
France Germany UK USA
Non-residential investment (%GDP) 11.7 12.5 11.6 12.0
Forging Ahead, Falling Behind and Fighting Back Page 11