What ‘Works’ in Environmental Policy-Design? Lessons from Experiments in the Australian and Dutch Building Sectors

Abstract

There has been a long-standing academic interest in experimentation with policy-designs. Researchers such as Charles Sabel and Jonathan Zeitlin claim that such experimentation may help to understand how significant legal, social or economic barriers can be overcome. This appears to be particularly true in the field of environmental policy. This article explores the nature and outcome of this experimentation by evaluating 21 policy-designs that have been experimented with by governments, businesses and citizens in the building sectors in Australia and the Netherlands. Building on the existing literature, it addresses expectations related to policy learning, collaboration, policy-outcomes and experimental biases. The article finds limited support for many of the claims made for experimentation in the current and past literatures. This may be a result of a mismatch between real-world experimentation with policy-designs and how academics conceptualize this process.

Key Words:

• environmental policy
• policy-design
• policy experimentation

Acknowledgements

I am grateful to all participants in this research for their voluntary participation. I wish to stress that my assessment here is a critique to the oft normative arguing in the literature about policy-design experimentation and not to the individual experiments discussed. I wish to thank my colleagues at RegNet and the editors and anonymous reviewers of this journal for helpful comments to earlier drafts of this article.

Funding

The research presented in this paper is funded through a VENI early career researchers grant by the Netherlands Organization for Scientific Research [grant number 451-11-05].

Notes

1. I thank Mike Howlett for this insight.

2. The literature also discusses indirect outcomes of policy-design experiments, for instance, spill-over effects from participants to non-participants in the policy-design experiment, the setting of a focal point or the normalization of certain (intended) behaviour (cf., Darnall & Sides, 2008; Lyon & Maxwell, 2007). Such indirect outcomes are beyond the scope of this article.

3. The relatively small size of the Netherlands is promising in attracting participants from various backgrounds and from all over the country to a mini-symposium; due to its size, Australia unfortunately is not.

4. In line with the custom of qualitative social science research interviewees provided their insights confidentially. As such the identities of the interviewees (and the cases studied) cannot be provided. To give the reader insight into the variance of the interviews voice is given to interviewees by referring to each individual with a number (e.g. ‘int50’).

5. To give an example, at the time of study, the most successful case analysed (case #1) celebrated its 10-year anniversary. By then it had issued close to 400 certificates to buildings with high levels of environmental performance in Australia. This numbers is bleak when contrasted with the vast size of the Australian building sector. For instance, only in the Australian state of Victoria yearly about 45,000 residential buildings are built.