Invasive Species and the Depth of Bioeconomic Integration

In the past, researchers have used an approach that assumes the economic system and the ecosystem affect each other in a one-sided way, which causes them to separate risk assessment from risk management. A change in the economic system is viewed as only changing the pressure on the ecosystem, or a change in the ecosystem is viewed as only changing the economic system. This approach does not address the idea of co-evolution - the two-way interactions and feedbacks between human and natural systems. Ecosystem changes alter human behavior and productivity in the economic system. People recognize the change in their productivity, and they adapt to this change, either by adapting the environment or by adapting to the environment. When people adapt, they alter the pressure they put on the ecosystem leading to further changes in the ecosystem. The co-evolutionary cycle continues.

Co-evolution can be addressed by integrating ecological and economic modeling into a single cohesive framework. The motivation behind integration is to get more precise estimates of invasive species damages on human and natural systems. Integration accounts for interdependencies, or feedback loops. Traditionally, economists have captured the notion of feedback loops using dynamic models. With a few exceptions, most standard bioeconomic models consider at most one or two feedback loops and operate at a relatively aggregate level. In many cases such models provide the needed insight into the underlying problem at hand. In other cases, however, more ecological or economic detail is needed to help avoid the unintended consequences of poorly advised policy. This challenge of balancing model tractability with more realism is not new in science, but it hits with full force when addressing the economics of invasive species management.

Herein we address two common questions that arise when doing integrated bioeconomic modeling for invasive species management: (1) what do we gain by integrating the web of life into economic analysis? and (2) if integration is worthwhile, how deep should we go?

What do we gain by integrating the web of life into economic analyses?

i. Yellowstone Lake

Using the population of cutthroat trout as a yardstick, we found that ignoring feedbacks biases risk estimates by overestimating cutthroat populations in the worst case and underestimating them in the best case. The difference arises from fishermen behavior. Without feedback, fishermen continue to fish as before. With feedback, fishermen adapt by fishing less and visiting other attractions more. Interestingly, the findings also revealed a troubling result from a species protection perspective in that only a small difference arises between the net benefits between the best- and worst-case scenarios, which suggests that gill netting for lake trout is inefficient. People preferred improvements in other park amenities (e.g., roads, wildlife viewing) relative to increased populations of cutthroat trout.

ii. Zebra mussels

The second dimension is the feedback between the resource manager and power plant. Removing the feedback causes the manager to act as if the power plant does not respond to changes. For example, following a successful invasion, the manager ignores the private control actions of the power plant. This has direct welfare consequences as resources may not be allocated efficiently, but the magnitude of the consequence depends on the actual response of the power plant. The results suggest that feedbacks can matter for this case—but not in every dimension and in varying degrees. Both biological and economic consequences of not addressing feedbacks are sensitive to the initial environment, behavioral perceptions about the state of the environment, and the completeness of the manager's beliefs.

iii. Leafy spurge

The results show that without humans, the native grasses are most likely to win. When humans enter and introduce cattle to the grazing ecosystem, the native grasses are placed at a competitive disadvantage, and leafy spurge generally becomes dominant depending on grazing intensity. The model illustrates the importance of accounting for grazing decisions when forecasting the further spread of leafy spurge.

If integration is worthwhile, how deep should it go?

Deciding just how deep to dig within and between the economic and ecological systems depends on the number of contact points between the systems and the indirect effects within the systems. For cases with one or two points of contact, a shallow, or abridged form of integration might suffice. But in cases with multiple contacts or important indirect effects, a deep integration is necessary. But in doing so it is necessary to make other simplifying assumptions. Such deeply integrated models may not be more realistic if the feedback loop or other representations do not conform accurately to reality. Addressing the challenge of adding more realism and being forced to solve a problem computationally rather than analytically requires one to work with a solid theoretical framework that guides the depth of integration. We illustrate the depth of integration challenge by using an example based on Finnoff and Tschirhart (2005) that examines the Alaskan economy and a marine ecosystem comprising Alaska's Aleutian Islands (AI) and the Eastern Bering Sea (EBS). Figure 1 shows the ecosystem and economic interactions and illustrates the thirteen key ecological descriptors and the feedback loops. The economy consists of Alaskan households and producing sectors linked to one another and the rest of the world through commodity and factor markets. All species in the food web are linked through predator-prey relationships and several species provide inputs to economic production. The prominent groundfish of the system, pollock, support a substantial fishery, and marine mammals including Steller sea lions (an endangered species), killer whales, blue whales, sperm whales, northern fur seals and sea otters. All of these species provide non-use inputs to the state's recreation sector. For a policy issue, we focus on the endangered Steller sea lion recovery via alternative pollock harvest quotas.

The first level of analysis is to understand the behavior of the actors in Figure 1. Economists study the behavior of individual consumers and producers. Consumer behavior has people within the household sector box making choices over combinations of goods and services. In Figure 1 this is a focus. Producer behavior is likewise captured by individual firms within the fish harvesting sector box choosing both their optimal mix of inputs and their optimal output level. Alternative quota levels are interpreted as changes in the prices faced by the households or producers. Similarly, ecologists study the behavior of individual animals; they would consider an individual pollock's optimal foraging behavior, and how foraging changes impact pollock populations as depicted within the pollock box. The alternative quotas would be interpreted as changes in the pollock populations.

The next level of analysis is to integrate all economic and ecological agents directly affected by pollock quotas through a bioeconomic harvesting model. In the economic system, individual consumer demands for pollock are aggregated to derive market demand, required for producers' decisions. Producer supplies are in part determined by the availability of pollock, which is derived from the aggregation of individual pollock behavior and population dynamics. Therefore, this level requires integration across the household, fish harvester, and pollock components. Linking these three components allows the derivation of market demand and pollock supply, which allows an assessment of how alternative quota policies impact the whole system.

But this level of integration is insufficient if we are interested in how the repercussions of the policies impact all of Alaska. In this case, deepening the analysis a further step within the economic system is necessary to include the other producing sectors of the Alaskan economy (recreation and composite goods in Figure 1), all other household demands, and trade flows into and out of the region. A complication arises, however, because the recreation industry depends on the marine mammals. Still further depth of integration is needed to increase depth within the ecosystem to account for the predator-prey relationship shown in Figure 1.

Finally, another level of integration is needed with nonmarket valuation. Nonmarket valuation involves assessing the total values (e.g., existence values) associated with scenarios of reduced human and environmental damages posed by some invasive species so we can better understand the net benefits of policy. The idea is that valuation work needs integration models to develop credible valuation scenarios. In turn, integration models need the parameters as defined by valuation work to capture the full range of benefits associated with the web of life. For instance, in the Yellowstone Lake case, Settle and Shogren (2006) integrated a valuation experiment within their bioeconomic model. They developed the Yellowstone Interactive Survey to ask people to value alternative scenarios designed to inform their integrated model. They determined the value for seeing/catching each species and used these estimates to parameterize the value to see/catch each species in measuring the visitor's welfare from Yellowstone National Park. The disquieting result that people preferred fixing the roads to protecting the native cutthroat trout emerged directly from this integration. Both valuation and bioeconomic modeling can likely be more relevant for policy if the scenarios people are asked to value are valid and if the scenarios created were informed by values stated by actual people. There are gains from joint production of values and feedback loops between economic and ecological systems.

Concluding Comments

For More Information