Search

As autonomous and intelligent systems continue to grow in importance, key stakeholders from government, civil society, and the private sector have released draft governance frameworks for these technologies. This piece makes an argument for how one might construct a unified framework for thinking about accountability in the regulation of autonomous and intelligent systems. It proposes a simple, cost- based framework for deciding whether to add (or subtract) the n ’th regulatory body to the existing regulatory mix. The development of such a framework is important to ensure the optimal development of mixed regulatory systems. In a space that is evolving as quickly as automated systems accountability, a broad set of stakeholders make a broad set of demands as to which costs or harms should be addressed by regulation. As a result, there can be an impulse to continue to add on layers of regulation, without giving appropriate weight to the fresh costs that those layers themselves can create. This framework addresses that worry.

Critics frequently argue that nudges are more covert, less transparent, and more difficult to monitor than traditional regulatory tools. Edward Glaeser, for example, argues that “[p]ublic monitoring of soft paternalism is much more difficult than public monitoring of hard paternalism.” As one of the leading proponents of soft paternalism, Cass Sunstein, acknowledges, while “[m]andates and commands are highly visible,” soft paternalism, “and some nudges in particular[,] may be invisible.” In response to this challenge, proponents of nudging argue that invisibility for any given individual in a particular choice environment is compatible with “careful public scrutiny” of the nudge. This paper offers first of its kind experimental evidence that tests whether nudges are, in fact, compatible with careful public scrutiny. Using two sets of experiments, the paper argues that, even when made visible, nudges attract less scrutiny than their “hard law” counterparts.

How do the main theoretical approaches in the social sciences — rational choice theory, the three institutionalisms, and a prominent variant of interpretivism, constructivism — explain regulatory governance instruments and mechanisms? In order to answer this question, we need an account of regulatory governance and an account of instruments and mechanisms. Rather than conceptualize regulatory governance in broad patterns, we will attempt to provide a framework for conceptualizing instances of regulatory governance at a more primitive level. This more primitive level is that of instruments and mechanisms. We will conceptualize instances of regulatory governance as mechanistic structures composed of component instruments designed by individuals with particular resources in order to influence other individuals. Each of the theoretical approaches provide an alternative causal explanation of the instruments and mechanisms that constitute regulatory governance.

Network analysis provides a powerful toolkit to operationalize, measure, and model core theoretical concepts in research on the policy process. Prominent among these concepts is that of an advocacy coalition, or a group of actors who coordinate with one another to pursue common policy goals. How to identify advocacy coalitions is an important empirical question, and crucial to the testing of several theoretical frameworks including the Advocacy Coalition Framework (ACF). However, the best way to detect advocacy coalitions is still an open question in the literature. Based on most conceptual definitions, we argue that advocacy coalitions are defined not only by clustered attributes (such as shared incentives, beliefs, or policy preferences), but also by clustered link structures since common attributes raise the probability of coordinated action in the network. We propose how one specific clustering method, k-means clustering, may be used to identify clusters of collaborative networks that are relatively homogenous in terms of policy beliefs, and illustrate the importance of clustering simultaneously on collaborative ties and belief systems. This method is used to identify advocacy coalitions within the climate change policy subsystem in Switzerland (N = 70). Results suggest that the delineation of coalition membership is very sensitive to whether researchers choose to cluster on attributes, link structures, or the union of the two.