DFG Research project: Information acquisition under fundamental uncertainty

DFG Research project: Information acquisition under fundamental uncertainty


2013 – 2015


Prof. Timo Goeschl, Ph.D.

Dipl. Math. Daniel Heyen


How much uncertainty are we willing to accept before making decisions like the approval of a new pesticide or the deployment of climate engineering technologies? And how does this answer depend on regulatory decision rules like cost-benefit analysis or the precautionary principle?



The increasing scale and innovative nature of human activities expose society to an increasing variety of complex risks. Two examples are the approval of new pesticides and the intentional change of the climate system, e.g. by means of stratospheric particles to counteract global warming. Two separate, yet linked characteristics render the task to find the right regulatory course of action very demanding and controversial. The first characteristic is the presence of fundamental uncertainties: Because of their novelty, there is frequently little scientific basis for an accurate assessment of new pesticides; likewise, the climate system involves many feedback mechanisms which preclude a clear probability estimation of outcomes (Weitzman 2009). The presence of fundamental uncertainties implies that the full ramifications of a regulatory decision will typically be poorly understood.


The second feature that characterizes the regulatory challenge is that it entails a choice about the desired state of knowledge available to the regulator (Sunstein 2002). A regulator may, for instance, specify a certain level of pre-tests that attest the harmlessness of the product before a company can request a decision. These measures improve the level of knowledge, but the research required typically involves significant costs. The costliness of information constitutes a substantial trade-off to the regulator: Although the regulator may improve her state of knowledge and by that enhance the regulatory decision, the information costs will typically cause only partial resolution of uncertainty. In other words, the regulation of complex environmental problems usually involves the choice of how much uncertainty a regulator is willing to accept.


As democratic societies devolve responsibility to regulate such risks to dedicated bodies, they need to provide these bodies with guidance on how society wants them to deal with the joint presence of fundamental uncertainty and costly provision of additional information. Important questions which naturally arise include: What are the appropriate decision rules under fundamental uncertainty? How much should society spend to improve its state of knowledge? What are the implications of certain decision rules under uncertainty on the research behavior?


Objectives and methods

The proposed research aims to bring recent advances in the field of decision-making under uncertainty to bear on these questions and thus fill a gap in the current environmental economics literature. We set up a simple decision-theoretic framework that combines uncertainty in the form of ambiguity (multiple priors) and active learning. The latter is realized by giving the decision-maker control over the precision of a signal that conveys noisy information over a payoff relevant parameter. With this simple framework at hand, we can tackle the questions above in a traceable and rigid manner. In particular, we will approach the following questions in detail


  • How does the demand for information depend on decision rules like CBA and the precautionary decision rule maxmin?
  • How do these results depend on the payoff structure of the decision problem, in particular for settings characterized by a 'risk-risk' trade-off?
  • What are the implications of our findings for the institutional design of regulatory decision and information acquisition?




Editor: Office
Latest Revision: 2013-10-10
zum Seitenanfang/up