Prioritizing Invasive Species Threats Under Uncertainty

Agricultural and Resource Economics Review, Apr 2006 by Moffitt, L Joe, Osteen, Craig D

The traditional model can be further simplified by assuming constant marginal social welfare and minimizing expected social cost (EC), measured as the sum of expenditures for preparedness actions and expected damage, which is a standard approach for pest control decision making:

If probabilities of events are not available, cannot be estimated, or are not appropriate for representing uncertainty, the decision considers only one θ^sub k^ at a time. In this framework, c(x^sub i^, Q^sub i^) is interpreted as the cost of an action or program. Many actions or programs have aspects of both mitigation and adaptation. For example, eradication or containment can reduce damages at one location and prevent or slow the spread of damage to another. Effective prevention programs have a mitigating effect, but they prevent damages even if the probabilities of damages are unknown. For example, if a practice, such as methyl bromide fumigation or irradiation, is known to control a pest, treating potentially infested imports would prevent damages even if the probability of pest entry is unknown.

Unfortunately, for many invasive species threats, information needed to apply the minimax criterion, as expressed in (3), may not be available. Program managers may face situations where there are few or no known preventative or control actions available or where there is little or no information about the effect of actions on pest damage. In these cases, selecting prevention and control actions for species becomes problematic. As a result, initial preparedness decisions will often focus on collecting information about pest threats and on the availability, cost, and effectiveness of prevention or control actions, in order to later facilitate the selection of actions.

The relative cost criterion, which is implicit in some economics of terrorism literature and based on the difference between damage estimates and action costs, could be combined with the minimax criterion to cope with this uncertainty. Some issues faced in the context of terrorism are remarkably similar to issues in crop protection (e.g., Cauley and Im 1988, Enders and Sandier 1993, Enders and Sandier 1996, Lapan and Sandier 1993, Lee 1988, Schwartz 1998, Slone 2000). A critical question is the prioritization of defense of potential targets, as well as prioritization of resources between preemptive and defensive measures (Endress 2002). For example, a comparison of damages resulting from a successful terrorist attack with the costs to terrorists of mounting an attack has led some economists to conclude that the marginal product of destructive activity greatly exceeds the marginal product of defense. In other words, potential damages to a target are thought to be large, and defensive measures are not expected to prevent all successful attacks. Therefore, some economists have concluded that resources should be directed at terrorist resources in a preemptive rather than a defensive effort (Madrick 2002). In such analyses, the likelihood that preemptive or defensive actions will prevent destructive activity is difficult to quantify, though the cost of the actions can be more easily determined.