Friday, February 26, 2010

New Paternalism on the Slippery Slopes, Part 9: Framing in Public Policy

And after another long interruption, I’m finally going to finish my series of excerpts from Mario Rizzo’s and my article, “Little Brother Is Watching You: New Paternalism on the Slippery Slopes.” There are three more posts, including this one.

As discussed in an earlier post, the new paternalists use the notion of framing -- that is, the idea that people’s choices respond to seemingly irrelevant differences in how the choice situation is presented -- to justify a variety of policy interventions. But what happens when we apply the notion of framing to the choices of the policymakers themselves? There is a natural human tendency to frame decisions narrowly “because immediate and concrete effects are more psychologically accessible than remote and abstract ones” (p. 726), and this tendency has worrisome implications for public policy. Specifically, paternalist policy-makers will tend to ignore the indirect and longer-term and implications of their policy choices (p. 726-727):
Narrow framing leads decisionmakers to consider choice-options simply as they arise, framed by present circumstances, the crisis of the moment, and perhaps the activities of rent-seekers. Their actions will often be ad hoc solutions to particular problems, and the narrow framing produces a tendency not to see important interrelationships. In Kahneman’s words again, “[t]he decision of whether or not to accept a gamble is normally considered as a response to a single opportunity, not as an occasion to apply a general policy.” For example, the interaction of biases may be ignored. This means the problem is not simply one of discounting long-term effects, but also of discounting effects that occur through longer and more complex chains of causality.
Narrow framing will enhance every variety of slope we have discussed so far, because all slopes occur in part from a failure to take a global perspective on policy. Altered incentives slopes, for instance, occur because policymakers tend to focus on one issue at a time—in this case, a single cognitive or behavioral bias, or a single means of correcting a bias. Simplification and distortion slopes occur because policymakers enact policies to address a specific problem, while failing to see how the new policy could empower experts and rent-seekers to advance less desirable policies in the future. To the extent that narrow framing inhibits policymakers’ awareness of such possibilities, it exacerbates the slippery-slope risk.
Furthermore, we argue that the new paternalist framework itself frames policy choices in a manner that encourages ever greater intervention (p. 727-728):
As presented in the behavioral literature, framing does not result from the deliberate choices of the decisionmaker; instead, it is an aspect of decision-making that is passively accepted. It is the result of unconscious processes whereby the conscious mind sees options or events with particular features accentuated; framing alters “the relative salience of different aspects of the problem.” Here we suggest that the particular way in which the new paternalists (most notably Camerer and coauthors and Sunstein and Thaler) have framed the issue of paternalism gives rise to an inherently expansionist dynamic. If irrational or boundedly rational policymakers accept the new paternalists’ approach, they will have accepted a paternalism-generating framework. Thus future policymakers, or the same policymakers in future situations, will tend to see more opportunities for paternalistic intervention than they otherwise would.

The decisions of targets are not intrinsically different from those of the policymakers. Framing is thus important in the policy context as well. The public-policy framework produced by the new paternalists directs policymakers’ attention to intrapersonal preference conflicts, that is, conflicts between operative preferences (choosing the sugary dessert) and deeper or more important preferences (maintaining good health). The framework then labels as paternalism any plan that alters the decision problem with the intent of improving welfare.

Therefore, if there is to be any solution to the target’s problem, paternalism is inevitable. Thus, the decision problem is framed not as “whether or not paternalism is desirable,” but as “what form of paternalism shall we have?” Sunstein and Thaler, for example, urge us to “abandon the less interesting question of whether to be paternalistic or not, and turn to the more constructive question of how to choose among the possible choice-influencing options.”
To summarize (p. 729):
Therefore, the Sunstein and Thaler approach is expansive not only in the sense that adoption of specific policies today will make the adoption of further, even more interventionist, policies more likely in the future, but also because their basic framework of analysis frames the overall issue as one in which some form of paternalism is “inevitable.” Sunstein and Thaler adopt a paternalism-generating public-policy framework. If policymakers accept this framework, they will be led by the framing to produce more and more paternalistic policies.
(As usual, full citations are available in the full paper. Cross-posted at ThinkMarkets.)

No comments: