which sometimes militated against innovation, was reduced’. So a first issue is that it might be concluded that realism and conventional views are different in rhetoric but that their empirical claims are similar.2 Such inferences need to be avoided.
Some points stressed by Lawson are valid and important. Worker resistance, to the extent that it occurred, was both rational and defensive. Workers’ powers grew up at the level of the shop floor, and they were based in the specifics of the production process. This potentially gave workers leverage over the details of the process. But this leverage was not articulated with the powers of other workers, either within the politics of production or in relation to national politics (Fox, 1985). Faced, moreover, with managements that often insisted on their own right to manage, workers responded rationally by playing the same game, of seeking advantage in shop floor bargaining when conditions were favourable.
Yet this argument says that workers should not be blamed for doing what they did, while still accepting, indeed stressing, that they did in fact do it. A substantial amount of IR research has criticized both Kilpatrick and Lawson and the conventional view (Nichols, 1986; Nolan and Marginson, 1990; Nolan and O’Donnell, 2003). Some key points are the following.
Even at the height of their power, trade unions organized no more than half the UK working population. Even more to the point, the kind of trade unionism associated with poor productivity (namely, a strongly organized group of workers using shop floor power to resist management) was in a minority. Its influence on productivity elsewhere is not at all clear.
Such unionism flourished during the 1960s and 1970s, yet poor productivity preceded and followed this period. Most notably, a phase of union weakness since the 1980s has not seen productivity leap to the levels of other countries.
Unions in many sectors where innovation was weak, such as steel and the railways, co-operated in changed work organization, and more generally there was welcome of, and not hostility towards, technological change (Hyman and Elger, 1981; Daniel, 1987). More recent research, exploring the links between new human resource management practices and trade unions, finds a positive, not a negative, association (Cully et al., 1999).
The pattern of relative productivity does not correlate with union strength. The sectors accounting for most of the productivity gap with the US are wholesale and retail trade and financial intermediation (ESRC, 2004: 8), neither of which is a bastion of unionism.
Periods of marked trade union defeat did not lead to productivity growth, a clear example being the coal mining industry during the 1920s and 1930s (Fine and Harris, 1985).
So how do we develop a more complex picture than the conventional story of trade union ‘constraints’ on management? In respect of unions themselves, it is the case that British unions had a more ‘restrictivist’ character than their counterparts in many other countries (Fox, 1985). This term refers to a focus on the specifics of work organization at shop floor level and a tendency to defend traditional, informal, work practices together with the customary rules regulating those practices.
2 There is an interesting parallel here in IR. Hugh Clegg (1975) argued that Marxist critiques of conventional approaches were long on rhetoric but offered no new empirical analysis. There are in fact differences of both questions posed and answers given, as argued elsewhere in relation to workplace industrial relations (Edwards and Scullion, 1982: 277-84).