Second, institutions do not merely constrain autonomous individuals; they actively shape the ways in which people think and act.
Third, knowledge is a key resource of organizations, but it is embedded in customs and routines rather than being simply the property of any one group.
Such insights are strongly in tune with empirical IR, and illustrate what is meant by contextualized comparison. Let us for simplicity consider the three points from Hodgson. First, much of IR traditionally addressed the creation of rules governing the employment relationship, notably a series of classic studies of custom and practice rules (Brown, 1973). Such rules can be seen as institutions, and research demonstrated the varying degree of fixity of different kinds of rule (Edwards, 1988). For example, some are firmly grounded in workers’ expectations, and they are defended if challenged; an example from the past would be demarcation rules governing which trades could do which jobs. Other rules are merely weak understandings that can readily be overturned.
Second, workplace institutions shape how people behave. Classic studies of occupational crime showed that the relevant practices were features of occupations, not individuals (Mars, 1982). A worker who entered an occupation with a strong set of norms would be shaped by those norms and come to accept them as natural. Lupton (1963) opens his account of workplace politics by describing his first day as an engineering apprentice; as he was labouring at a task at the end of the day, he noticed a group of his mates gathering around him, who silently conveyed the message as to when he should stop work. Individual preferences are embedded in institutional norms and assumptions.
Third, tacit knowledge has been a theme of workplace studies (Kusterer, 1978; Halle, 1984). Such knowledge means, first, that efforts at rationalization are bound to be incomplete at best, and, second, that knowledge can be contested. Workers can retain knowledge and bargain about its use. Even in modern and highly rationalized settings such as call centres, workers use their judgement as to how to perform tasks. In doing so, they sometimes make work proceed more smoothly, and indeed rescue management from its own rules, by for example responding flexibly to a customer rather than sticking to their instructions. But they can also use their tacit knowledge to create their own social space independent of management (Korczynski, 2002).
Such empirical results can, then, be readily expressed in institutionalist terms. The link the other way, from institutional analysis to CR, does not need spelling out, save to say that the emphases on history and context are clearly compatible with CR.
Illustrations of explanatory development
IR in particular and institutionalism in general are, as noted above, criticized for the absence of theoretical progress. They may be able to explain some of the complexities of workplace rules or trade union behaviour, but is this any more than descriptive detail and ad hoc explanation? Developments in two areas illustrate ways in which further progress has been made (for other illustrations, see Edwards, 1995).
Teamwork has been a major innovation in workplace relations in the past 20 years, though it has important historical precedents. Debates on it neatly illustrate different empirical and theoretical approaches in IR. Teams have often been promoted as means to increase efficiency while also meeting workers’ interests in job autonomy. Some IR researchers demonstrated that this was far from being the inevitable result,