Why Does Policy Change?

Lessons from British Transport Policy 1945-99

Dr Geoffrey Dudley, Geoffrey Dudley, Jeremy Richardson

• Publisher: Routledge
• Year: 2004
• eBook ISBN: 9781134701575
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

The tension between policy stability and change is a key political phenomenon, but its dynamics have been little understood. Why Does Policy Change? examines and explains the dynamics of major policy change by looking at case studies from British Transport policy since 1945. The significant contrasts between road and rail policies in this period lend themselves perfectly to the authors' theories of what brings about policy turnabout.

1
Government, interest groups and policy change

Organised interests, political parties, and policy experts do not simply ‘exert power’; they acquire power in part by trying to influence the political discourse of their day. To the degree they are able to do so, they may have a major impact on policy without necessarily acquiring the formal trappings of influence. The resultant flow of ideas is an important dimension of the process in which policy is made.

(Peter Hall, ‘Policy paradigms, social learning and the state’, p. 290)

Policy communities and policy networks

Concepts and models in political science probably reflect the politics of the period in which they were first formulated. This is both unsurprising and desirable, as their utility is to help understand, explain, and even predict events in the real world. An ambition to develop concepts more applicable to the realities of post-war British politics was the foundation of the British origins of what is now termed the network approach.

Humble British origins: analysing stability

Until the late 1970s, the study of interest groups in Britain, though very well established from the 1950s onwards, tended to play a relatively minor role in British political science (Richardson 1999). This all began to change in the late 1970s, however, with a shift in focus from the study of traditional institutions, towards a different world of power—essentially a post-parliamentary polity (Richardson and Jordan 1979). As Judge later commented, the ‘post-parliamentary’ thesis was, for over a decade, a characterisation which was largely unchallenged and found reflection in other important commentaries (Judge 1993:123). The core argument in the post-parliamentary thesis was that policy change generally took place only when the relevant ‘policy community’ agreed it was necessary and a consensus existed on the direction of change. As Judge noted, the policy community concept was descriptive rather than definitional (Judge 1993); similarly Dowding has suggested that it was a metaphor rather than a model (Dowding 1995). In fact, it had no great pretensions to being a new theory of British politics. Its purpose, as Judge observed, was to contrast established models of parliamentary and cabinet government with a new reality of governing (Judge 1993). The strength of the empirical tradition in British political science carried the simple, unrefined, concept forward. As Judge comments: ‘the concern for descriptive accuracy has led a whole generation of British scholars to follow’ (Judge 1993:121). This concern was still very much alive in the 1990s but was reflected in more systematic attempts to build theories around the network concept. (For British examples see Rhodes 1986, 1988, 1990, 1996; Marsh and Rhodes 1992; for other, more quantitative examples, see Knoke et al. 1996; Pappi and Henning 1998).

Whatever the merits of subsequent attempts to theorise on the basis of network analysis (for an excellent review see Thatcher 1998), current empirical descriptions of British policy-making seem to have changed rather little. Back in 1979, the new emphasis was on the shift of public policy-making away from conventional democratic institutions towards a somewhat hidden world where those deemed to be the legitimate ‘stake-holders’ met and decided both what the problems were and how they should be resolved:

In describing the tendency for boundaries between government and groups to become less distinct through a whole range of pragmatic developments, we see policies being made (and administered) between a myriad of interconnecting, interpenetrating organisations.

(Richardson and Jordan 1979:73–4)

Nearly twenty years later, the current emphasis on the notion of ‘governance’ appeared to stress the very same tendencies. ‘Governance’ is seen as signifying:

a change in the meaning of government, referring to a new process of governing; or a changed condition of ordered rule; or the new method by which society is governed. I employ a stipulative definition; it refers to self-organising, interorganisational networks.

(Rhodes 1997:35, emphasis in the original)

The Achilles heel of these closely related concepts of policy community, policy networks and new governance is the implication of stable policies, as well as stable relationships and a stable membership. Thus, it was argued that:

The logic of negotiation also suggests that policy-makers in both government and groups will share an interest in the avoidance of sudden policy change. Working together they will learn what kind of change is feasible and what would so embarrass other members of the ‘system’ as to be unproductive.

(Jordan and Richardson 1982:93–4)

Jordan later further emphasised both the stability of policy communities and the existence of shared views. He sees policy communities as:

A special type of stable network, which has advantages in encouraging bargaining in policy resolution. In this language the policy network is a statement of shared interests in a policy problem: a policy community exists where there are effective shared ‘community’ views on the problem. Where there are no such shared views no community exists.

(Jordan 1990:327, original emphasis)

The explicit assumption of stability of relations and stability of actor participation (almost exclusiveness) is evident in most attempts to refine the original policy community and policy network concepts. Thus, as Judge points out, Rhodes repeatedly states (Rhodes 1985:15; 1988:78; 1990: 204; Rhodes and Marsh 1992a:182; 1992b:13) that:

policy communities are networks characterised by stability of relationships, continuity of a highly restrictive membership, vertical interdependence based upon shared delivery responsibilities and insulation from other networks and invariably from the general public (including Parliament).

(quoted by Judge 1993:122)

More recently, Rhodes has stressed that, among the six reasons that policy networks are important is the fact that ‘they decide which issues will be included and excluded from the policy arena’ (Rhodes 1997:34).

Our central thesis in this chapter is, however, that the picture of order and especially control which these characterisations of the policy process imply is at variance with the changing nature of policy-making at both the national and supranational levels in Europe. Policy-making within European states and at the European Union level is often much more fluid and unpredictable—and less controllable—than seems to be implied by enthusiasts of the network approach. Whilst there are undoubtedly policy communities and networks which exhibit both stability and exclusiveness and do control policy agendas, there appear to be counter-tendencies which lead to lack of control, policy instability, and unpredictable outcomes.

Incentives and costs of group participation

The emphasis on stable relationships and shared views fails to capture the reality of many cooperative arrangements between policy stakeholders. In many cases the ‘glue’ which enables the stakeholders to cooperate is often a recognition that, despite conflicting values and objectives, joint gains can result from cooperation either in the short or long run. Indeed, at the EU level (see below) the development of short-term ad hoc coalitions of extremely strange bedfellows—who certainly do not have common values—#8212;is now quite common. As Sebenius suggests, uncertainty is an important feature of policy-making for all participants. Cooperation, even with enemies, can reduce that uncertainty. He argues that alongside uncertainty in the policy process, there are opportunities for mutual learning and joint problem solving. By combining the politics of uncertainty and the politics of learning, Sebenius probably captures the core meaning of ‘policy community’ as a concept. Thus, he states that ‘beyond understanding technical uncertainties, finding joint gains also requires that each party learn about the other’s priorities in order to craft mutually beneficial trades’ (Sebenius 1992:329). Sebenius quotes Raiffa as follows:

In complicated negotiations where uncertainties loom large, there may be contracts that are far better for each negotiating party than the non-contract alternative, but it may take considerable skill and joint problem solving to discover these possibilities. Without the right atmosphere and without some reasonably trustful communication of values, such jointly acceptable contracts might never be discerned.

(Raiffa quoted by Sebenius, 1992:329, emphasis added)

There may be considerable and bitter disputation, yet the cooperative game continues to be played in order to secure mutual gains or to reduce individual losses. Lax and Sebenius have emphasised that the bargaining process indeed exhibits both conflict and consensus. Thus:

the competitive and co-operative elements are inextricably entwined. In practice they cannot be separated…. There is a central, inescapable tension between co-operative moves to create value jointly and competitive moves to gain individual advantage. This tension affects virtually all tactical and strategic choice.

(Lax and Sebenius 1986)

Participating in joint policy-making activity either on a temporary or more permanent basis has the potential to maximise benefits to the parties involved. Using concepts from negotiation analysis, Sebenius points out that outcomes can be influenced by favourably affecting the zone of possible agreement between the parties. The ‘zone of possible agreement’ means ‘a set of possible agreements that are better for each potential party than the non-co-operative alternatives to an agreement’ (Sebenius 1992:333).

Participation in cooperative policy-making institutions is, however, not exclusively about securing direct public policy gains for the participants or maximising the zone of possible agreement. It is also concerned with straightforward information gathering. As Austen-Smith and Wright (1992) argue ‘lobbying is a two-stage process in which the first stage involves the acquisition of specialised information’ (p. 231). Thus, it is perfectly rational to participate alongside opponents and enemies if this delivers useful information about their positions and possible policy outcomes—even if neither can be influenced. Knowing what is coming, however unattractive that might be, is much better than being surprised. At the European level, especially (see pp. 14–15), the need to acquire information seems a driving force behind much so-called ‘lobbying’ and the frenetic merrygo-round of committees, seminars, conferences and other meeting opportunities that now characterises politics in Brussels. All of this helps explain why actors will construct cooperative institutions of various types, in the face of uncertainty.

The very success of policy community politics might, however, be the cause of its erosion over time. As Heinz et al. observed in the US, interest groups became more active in order to reduce their uncertainty. However, interest group activity begets yet more interest group activity thus increasing, not decreasing, uncertainty as the number of stakeholders increases. Interest groups ‘by creating structures to control or adapt to uncertainty…have contributed to the development of a more complex and rapidly changing policy environment’ (Heinz et al. 1993:371). Thus, ‘although greater numbers of organisations of all types…actively monitor and participate in the making of national policy, and although the scope and intensity of their efforts has increased significantly, the returns on these efforts are not at all clear’ (Heinz et al. 1993:369). Uncertainty is also increased for all stakeholders by the tendency for closer linkages between policy problems across different policy sectors. The current Labour Government of Mr Blair is keen on what it calls ‘joined-up government’, but the reality of joined-up government is that even larger numbers of policy stakeholders are brought together round the bargaining table. This can result in a degree of ‘overcrowding’ of each, hitherto autonomous, policy sector as stakeholders from other policy communities demand and get entry. Policy communities and networks may become linked in a rather messy and unpredictable chain of actors, who do not know each other well and who do not speak the same ‘language’. Of special importance is that they may bring quite different ‘policy frames’ to the table, i.e. they have very different policy or cultural frames through which they view the real world (Schön and Rein 1994; Fligstein 1997). This leads them to make sense of reality in quite different and conflicting ways. Such large and diverse collections of stakeholders may be a ‘network’ only in the very loosest of senses. They inhabit the same policy area or domain but only minimal interaction occurs.

This shift—from a world of policy-making characterised by tightly knit policy communities and/or well-structured and stable networks, to a more loosely ‘organised’ and therefore less predictable collection of stakeholders—#8212;was noted in the US as early as the 1970s. In 1978 Heclo noted that policy problems often escape the confined and exclusive ‘worlds’ of professionals and are resolved in a much looser configuration (if indeed such a structured term can be used) of participants in the policy process. Heclo argued that the nature of power in Washington had begun to change. Exercising power was not as much fun as it used to be in the ‘clubby’ days of Washington politics (Heclo 1978:94). Politics was less ‘clubbable’ because more and more groups had entered the policy process. Thus ‘as proliferating groups have claimed a stake and clamoured for a place in the policy process, they have helped diffuse the focus of political and administrative leadership’ (Heclo 1978:94–5). In a now classic formulation, he argued that:

Looking for the few who are powerful, we tend to overlook the many whose webs of influence provoke and guide the exercise of power. These webs, or what I will call ‘issue networks’, are particularly relevant to the highly intricate and confusing welfare policies that have been undertaken in recent years…. An issue network is a shared-knowledge group having to do with some aspect (or, as defined by the network, some problem) of public policy…. But knowledge does not necessarily produce agreement. Issue networks may or may not, therefore, be mobilised into, second, a shared-notion group (creating a coalition) or, third, a shared-belief group (becoming a conventional interest organisation). Increasingly, it is through networks of people who regard each other as knowledgeable, or at least as needing to be answered, that public policy issues tend to be refined, evidence debated, and alternative options worked out—though rarely in any controlled, well-organised way.

(Heclo 1978:102–3, emphasis added)

We dwell on Heclo’s analysis simply because it focuses our attention so clearly on the core motivation for interest group participation—the need to acquire and share knowledge about both the real world and the political opportunities for changing it. Following the assumption that interest organisations are rational actors (Christiansen and Rommetvedt 1999:210), we have argued that participation in cooperative structures—be they policy communities, issue networks, or even looser configurations—is perfectly rational, even though its consequences may not necessarily reduce uncertainty. However, interest groups also have to deal with other actors in the policy process—especially governments. As Christiansen and Rommetvedt argue, ‘powerful actors may choose to change institutions if they have the power to do so and if they calculate change to be in their favor’ (Christiansen and Rommetvedt 1999:210). Governments can seize power over established institutions such as policy communities and networks. The British case, post-1979, seems to be such an example.

Challenging policy communities and networks

Heclo’s emphasis on lack of control and organisation, and greater unpredictability in policy-making, was based on empirical observation. He had noticed that Washington politics was different by the 1970s. However, the typical ‘European polity’ (Heisler and Kvavik 1974) was, at that time, still relatively unaffected by the erosion of policy communities which he described. Old style politics based on policy communities and neocorporatist arrangements appeared to continue for most of the Western European states. Yet this established policy style has shown signs of change in Western Europe in the 1980s and 1990s. We have seen two decades of sometimes radical policy reform, leaving no Western European nation untouched and few policy sectors undisturbed. After this wave of reform, at least two outcomes can be observed. First, actors in many policy sectors now operate under radically changed rules of behaviour. Labour market policies are an obvious specific sectoral example, although many other policy areas have been greatly affected by the deregulatory and privatisation fashion which has swept Europe. Second, the distribution of power within and across policy sectors has seen major changes. The mix of winners and losers has changed amongst traditional stakeholders as a result of public policy change. Also, new stakeholders have sometimes been introduced—either by importation from cognate (but hitherto excluded) policy areas or the emergence of completely new sets of actors (environmentalists and women’s organisations being obvious examples) and institutions. Many deeply entrenched interests, who had been granted the ‘franchise for public policy’ seem to have been challenged by more assertive governments. For once, Britain is the first mover, rather than the laggard, in Europe. As an extreme example, it is an important test case of the capacity of network approaches to explain the process of policy change.

Britain: the shift from ‘governing under pressure’ to ‘policy communities under pressure’

In the British case it seems reasonable to argue that the consensual policy style was a clear enough feature of post-war politics (Beer 1956, 1965). This is not to say that groups were never challenged by governments (the clash between the Heath Conservative Government and the coal miners in 1972/3 being an obvious example). However, confrontation and imposition was not the preferred policy style. As Wyn Grant has noted, the policy community ‘model’ caught on because it seemed to be a good fit with reality (Grant 1989). It took Margaret Thatcher’s succession of Conservative governments to give policy community politics a real jolt as the new policy style unfolded. It is quite true, as Jordan and Maloney argue, that consultation with interest groups continued under the Thatcherite ‘revolution’. However to argue, as they do, that ‘consultation is the predominant policy-making routine in British politics—a pattern scarcely touched by supposed Thatcherite antipathy to groups’ (Jordan and Maloney 1997), is misleading. It places insufficient emphasis on the extent to which the Conservatives systematically changed the underlying bases of the consultations which they continued to conduct with the ‘affected interests’. The consultations were often only after the extensive re-writing, by government, of old public policy ‘franchises’. Consultation yes, but after what and on whose terms?

A hidden assumption of much of the policy community approach (less so of the network literature...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Preface
• 1: Government, interest groups and policy change
• 2: The dynamics of the rail ‘hollow core’: The prelude to Beeching
• 3: Occupying the ‘hollow core’: The Beeching era at British Rail 1960–5
• 4: Roadbuilding as a policy idea and a weak advocacy coalition 1900–54
• 5: Ministers and motorways: The making of a hegemonic policy community 1954–9
• 6: Undermining a hegemonic policy community: Trunk roads and the first wave of environmentalism 1973–9
• 7: ‘Roads for prosperity’ versus ‘sustainable mobility’: New challenges and confrontations 1989–93
• 8: The destruction of Roads for Prosperity: From hegemony to ignominy 1993–9
• 9: The ‘hollow core’ revisited: Rail privatisation 1987–92
• 10: Why does policy change?: The ‘Four Is’ and the significance of time
• Bibliography