Can Bureaucracies Change Policy?

In the 1990s, policymakers at Yellowstone and Banff National Parks enacted two of the most controversial programs in the history of protected lands. At Yellowstone, the U.S. National Park Service (NPS) and the U.S. Fish and Wildlife Service (FWS) personnel reintroduced wolves into the Yellowstone ecosystem. This program restored a crucial element to the park ecosystem that had been eliminated decades before and not returned since extermination. At Banff, federal authorities imposed strict limits to growth of the town of Banff. This action reversed a policy dating to the park's establishment in the late nineteenth century of allowing and encouraging growth and development of the town within Banff. How did these policy changes occur?

The question is important not just for these two preeminent national parks but also for the policy process in general. Typically, public agencies and their political superiors are reluctant, if not unwilling, to pursue reversals of long-standing policy traditions, particularly when such reversals involve enacting controversial policies. Yet, as these two cases show, under certain conditions bureaucracies can change policy even when that means reversing decades of tradition. This article uses literature from the fields of public policy and public administration to posit those conditions and then examines them in the two case studies.

Theoretical Expectations

The Yellowstone and Banff cases are obviously important in many ways. At the broadest level of generalization, they reflect upon important theoretical arguments regarding policy change.

The Unlikelihood of Controversial Changes

Perhaps the most dominant perspective on policy change is that change is most likely to occur only incrementally.¹ A large and expanding literature has recently questioned that perspective with compelling theoretical arguments for dramatic change and some impressive empirical evidence.² Even given that growing literature, however, much conventional wisdom still casts doubt on the likelihood of dramatic changes.

In particular, a great deal of literature suggests that most policymakers want little to do with changing traditional goals, if for no other reason than such changes will likely be controversial. The literature on the reluctance of Congress to deal with contentious issues has deep roots. Indeed, many scholars argue that Congress uses the ability to delegate policymaking to public agencies precisely to avoid making decisions that may anger some group or constituent that has become accustomed to traditional behavior.³ These delegations are not received with open arms. Indeed, as one scholar argues, many bureaucrats "shun situations in which they might have to make troublesome decisions."⁴ Many other widely cited studies of public agencies suggest that bureaucracies tend to be conservative or at least risk-averse, and bureaucrats try to avoid antagonizing traditional constituents through contentious decision making.⁵ So, under what conditions can bureaucracies change policy?

Conditions Conducive to Controversial Policy Changes

Given that most policymakers prefer to avoid controversial changes to long-standing policies, the literature describing successful change efforts in these areas is somewhat limited. However, some classic works in the public policy literature do discuss conditions conducive to effective implementation in ways that can be extended to propose two broad hypotheses regarding contentious policy change to traditional goals. I posit two conditions as necessary, neither alone as sufficient.

Since public agencies are unlikely to pursue controversial policy change in traditional goals on their own, such actions depend upon meaningful intervention by formal authorities external to the implementing agency. The basis for such an argument dates at least to Schattschneider's concept of expanding the sphere.⁶ Conflicts always have the potential to expand. If they do expand, then they are more likely to engage higher levels of authority. Schattschneider states this in his discussion of federal-state relations: "One way to restrict the scope of conflict is to localize it, while one way to expand it is to nationalize it."⁷ The dynamics of any situation change substantially as those previously outside the conflict become involved. This is particularly relevant when those who become involved have authority. Matland's synthesis of a great deal of implementation literature describes the outcome of the most contentious policies as determined by the actions of those with power.^8...