Progress 09/01/06 to 08/31/10
Outputs
OUTPUTS: This project has generated a series of studies designed to improve our understanding of how to manage livestock disease problems, particularly when livestock are subjected to infection by a wildlife host reservoir. One set of studies analyzes the problem of bovine tuberculosis (bTB) in Michigan cattle and white-tailed deer. Our particular focus is how to target management efforts cost-effectively so as to simultaneously reduce infection risks and management costs. Options for managing risks within the livestock sector include test-and-slaughter of infected animals, investments in biosecurity measures (i.e., fencing) that prevents transmission of bTB from deer to cattle, and reduction of cattle herds to reduce the number of animals at risk of infection. Options for managing risks within the deer sector include reducing overall deer densities, differentially reducing densities of male and female deer (as infection rates differ among these groups), reducing supplemental feeding of deer, and investing in biosecurity (fences) to reduce transmission from cattle to deer. Each of these options manages risks very differently and will therefore have different epidemiological impacts and different costs to various groups in society. Second, we analyze the problem of brucellosis in bison, elk, and cattle in the Greater Yellowstone Area. Here we examined several existing proposals for how to manage the disease problem. These included closing the elk feed-grounds (or reducing their magnitude), elk population controls, vaccination of cattle and bison, and test-and-slaughter of bison. Though the management options here are different than those for the bTB problem, there are certain similarities in the types of controls being used. Overall, the results from both sets of studies improve our understanding about how different types of controls are more or less cost-effective in reducing risks to valued animal populations. Third, in performing the work indicated above, we developed a conceptual bioeconomic framework for animal disease management that differs from traditional epidemiology frameworks. This led us to examine the relation between the approaches so as to integrate the two theories. Fourth, we examine farmers' private incentives to invest in biosecurity and livestock health management. We also examine how public policies, such as indemnity payments for infected animals and herd testing requirements, can affect producers' incentives for on-farm disease management. In particular, we examine how indemnity payments affect producers' decisions to invest in preventative biosecurity measures and to report disease quickly after an infection occurs. We then examine how an alternative compensation plan can be developed to provide better incentives for biosecurity investments and reporting. Our work on indemnity payments has led to discussions with members of USDA-APHIS about how to alter indemnity compensation schemes to improve the incentives while also reducing disease risks and expected payouts. PARTICIPANTS: Richard D. Horan and Christopher A. Wolf, both in the Department of Agricultural, Food, and Resource Economics, Michigan State University, were co-PIs for this work. Fang Xie (PhD student in the Department of Agricultural, Food, and Resource Economics at Michigan State University) worked on the problem of brucellosis in bison in the Greater Yellowstone Area, and bTB in U.S. cattle. Her PhD Thesis is based on the results of this research. Richard Melstrom (PhD student in the Department of Agricultural, Food, and Resource Economics at Michigan State University) worked on the problem of species translocation versus disease control in the wild. TARGET AUDIENCES: The target audiences for our work this past year have primarily been academic economists and ecologists, and APHIS personnel and other policy-makers. Our work with APHIS is described under Outcomes/Impacts. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
Our bioeconomic work on managing wildlife and livestock populations indicates cost-effectiveness is improved when policies can better target the source of risk. Because infected wild animals are often not easily identified prior to the kill, many wildlife controls do not target wildlife disease risks very well. In contrast, it may be easier to target livestock controls to reduce disease risks. This is because individual livestock are often easier to test for disease. Also, because livestock are confined to a given location, biosecurity investments may be able to prevent contacts between wild and domestic animals. When controls can selectively target livestock risks but not wildlife risks, then cost-effectiveness is improved by doing comparatively little to manage wildlife populations (the disease persists in the wild) and instead investing in on-farm protections to reduce, but not eliminate, infection risks. Sometimes it is possible to manage wildlife risks more selectively. For instance, male deer are much more likely to become infected than female deer, due to behavior, and males are observable. Also, vaccinations for wildlife, when they are available (e.g., for brucellosis in bison but not for bTB in deer), can more effectively target wildlife disease risks. In these cases, it becomes cost-effective to shift the focus towards the wildlife population. Eradication may be optimal in some instances. Our bioeconomic work has shown that the disease ecology metrics traditionally used to guide disease management recommendations, such as the basic reproductive ratio of the pathogen, are not appropriate metrics once an outbreak occurs. Such metrics do not account for economic and epidemiological tradeoffs. Indeed, we find that managing only in response to disease ecology metrics is consistent with managers having a particular set of preferences over disease and economic outcomes - and when one actually evaluates those preferences, it seems unlikely that managers would actually have adopted them. We find disease risks and control costs can both be reduced if management is derived from a decision-theoretic framework that incorporates both costs and epidemiological relations. Our work on farmers' private incentives to invest in livestock health management shows there is value in promoting management practices having positive spillover effects that reduce risks for multiple disease problems. Our work on indemnity payments shows that the current structure of these payments does not create adequate incentives for farmers to take preventative actions or to quickly report infection. We developed an alternative approach that creates the proper incentives, by shifting some of the infection risks to producers. This work was part of the impetus for the Alternative Methods for Livestock Compensation Workshop in July 2010 at USDA-APHIS. During and after the workshop, we worked with APHIS' Appraisal-Indemnity- Compensation Specialist to consider how incentives from indemnities affect preventative and reporting behaviors by livestock producers, and how indemnity payments might be implemented to account for these incentives.
Publications
- Fenichel, E.P., R.D. Horan, and G.J. Hickling (2010), "Bioeconomic Management of Invasive Vector-Borne Diseases", Biological Invasions, 12: 2877-2893.
- Horan, R.D., E.P. Fenichel, C.A. Wolf, and B.M. Gramig (2010), "Managing Infectious Animal Disease Systems", Annual Review of Resource Economics, 2:101-124.
- Fenichel, E.P., R.D. Horan, and G. Hickling (2010), "Management of Infectious Wildlife Diseases: Bridging Conventional and Bioeconomic Approaches", Ecological Applications, 20: 903-914.
- Gramig, B., C. Wolf and F. Lupi (2010), "Understanding Adoption of Livestock Health Management Practices: The Case of Bovine Leukosis Virus." Canadian Journal of Agricultural Economics 58:343-360.
|
Progress 09/01/08 to 08/31/09
Outputs
OUTPUTS: Several bioeconomic models have been developed or are under construction to explore the management of disease problems that affect wildlife and livestock. One published study focuses on the potential private vaccination responses to and the ecological impacts of policies proposed to confront the problem of brucellosis (Brucella abortus) being spread from elk to cattle in the Greater Yellowstone Area (GYA) in Wyoming. Specifically, we explore combinations of two proposals for managing brucellosis in the elk herd: closing feedgrounds and population control. These proposals affect disease risks to local cattle herds, which causes producers to alter their vaccination choices. We model these responses and analyze how the elk management proposals might affect the livestock sector. In a second study, we analyze the optimal design of livestock disease indemnity payments. The difficulty in designing these payments is that, if done improperly, they can discourage producer investments in preventive biosecurity actions and they can discourage early reporting of infection. A particular problem is that both biosecurity investments and herd disease status are largely unobservable without costly government investments in monitoring. If the government cannot observe when producers undertake risky actions, and if payments effectively insure producers against all risks, then farmers will have no incentives to invest in controlling risk. We investigate whether indemnity payments can be structured to alleviate these problems. If not, we might expect livestock disease problems to grow. Indeed, risks to livestock will increase as more diseases become transmitted via foreign trade and non-conventional sources like wildlife. And if farmers have no incentives to reduce the risks they face, then infections are likely to rise along with indemnity payments - something that has been observed in recent years as APHIS funded most of its indemnity payments using emergency funding. In a third study, we have begun to explore how pathogens may evolve in response to on-farm disease control measures such as test-and-slaughter of herds. We also explore the optimal approach to disease control when these evolutionary impacts are taken into consideration. This work is preliminary and on-going, and will likely involve a numerical application to the problem of avian influenza in poultry. A fourth study addresses whether disease management, particularly of valued (and potentially threatened) wildlife species, should involve translocation of healthy animals to disease-free reserves or whether control should focus on managing the disease within the original populations. A numerical analysis focuses on the Tasmanian Devil, because the data was readily available, but the issue of translocation has arisen for diseased wildlife in North America as well. For instance, the wood bison in North America is infected with brucellosis and bTB and poses risks to livestock. Another study examined strategies to eradicate or control TB in US cattle. The study considered the trade implications and simulated disease spread in dairy herds, beef cow-calf herds and feedlots. PARTICIPANTS: Richard D. Horan and Christopher A. Wolf, both Professors in the Department of Agricultural, Food, and Resource Economics, Michigan State University, were co-PIs for this work. Fang Xie (M.S. student in the Department of Agricultural, Food, and Resource Economics at Michigan State University) worked on the problem of brucellosis in elk and deer in the Greater Yellowstone Area. Her MS Thesis was based on the results of this research. Richard Melstrom (PhD student in the Department of Agricultural, Food, and Resource Economics at Michigan State University) worked on the problem of species translocation versus disease control in the wild. Nicole Olynk (PhD student in the Department of Agricultural, Food, and Resource Economics at Michigan State University) worked on simulating the spread of TB in the US cattle industry. TARGET AUDIENCES: The target audiences for our work this past year have primarily been academic economists and ecologists, and APHIS personnel and other policy-makers. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
For our work on the Brucellosis problem in the GYA, our results confirm prior disease ecology work that population controls, even when feeding is eliminated, are unlikely to eradicate brucellosis in elk. We also find that some farmers would likely vaccinate some on their own, though at rates much less than those arising under vaccination mandates. Closing down public feedgrounds, which farmers support doing, may actually increase disease risks by reducing the number of resistant elk and by encouraging elk to move out of the feedgrounds in search of food. Farmers respond to these increased disease risks with increased investments in vaccinations, with the costs and benefits of vaccination nearly offsetting so that farmers are not significantly worse off. Producers are likely to gain when elk population controls are implemented, though the gain would probably be small. In a separate study, we analyze the optimal design of livestock disease indemnity payments. We find that indemnity payments can be structured to encourage producers to undertake appropriate biosecurity investments and to report disease when it occurs, even if these actions cannot be monitored by the government. The key is to shift some of the risk to producers, which current indemnity programs fail to do. Specifically, we find that producers should be awarded a bonus if herd prevalence levels at the time of reporting are small, as this suggests producers undertook appropriate actions to protect their herds. But if herd prevalence is too high, then producers should be penalized by receiving a smaller indemnity. In our third study, our preliminary work on the evolution of pathogens in response to livestock disease control measures shows that test-and-slaughter programs could cause pathogens to become more virulent. In our fourth study, on the concept of translocation of healthy animals from infected populations, our preliminary work shows that one approach or the other (translocation or disease control) should be adopted, but not both. Our numerical work suggests that translocation may be an inferior strategy under many circumstances, due to the costs involved relative to those of a disease control program. The TB eradication program study concluded that they key aspects were the detection and trace-back rates in the current program. Low trace-back rates in the beef cow-calf sector were particularly problematic in achieving eradication. Given the high marginal cost of achieving eradication, the current program had the highest benefit/cost ratio.
Publications
- Gramig, B.M., R.D. Horan, and C.A. Wolf (2009), "Livestock Disease Indemnity Design when Moral Hazard is Followed by Adverse Selection", American Journal of Agricultural Economics, 91: 627-641.
- Xie, F. and R.D. Horan (2009), "Disease and Behavioral Dynamics for Brucellosis Control in Elk and Cattle in the Greater Yellowstone Area", Journal of Agricultural and Resource Economics, 34: 11-33.
|
Progress 09/01/07 to 08/31/08
Outputs
OUTPUTS: Several bioeconomic models have been developed or are under construction to explore the management of disease problems that affect wildlife and livestock. One published study focuses on the joint management of wildlife and livestock when both populations are already infected, and includes a numerical example applied to the case of bovine tuberculosis (bTB) in Michigan white-tailed deer and cattle. In a second study, we investigate the potential private vaccination responses to and the ecological impacts of policies proposed to confront the problem of brucellosis (Brucella abortus) being spread from elk to cattle in Wyoming. Specifically, we explore combinations of two proposals for managing brucellosis in the elk herd: closing feedgrounds and population control. Farmers respond to changes in risk through private decisions regarding herd vaccination. This study became a student's M.S. Plan B paper, which was successfully defended this year. The paper has also been presented at professional meetings and was submitted to a journal. In a third study, we investigate the design of livestock indemnity payments and how these payments influence farmers' decisions to invest in biosecurity prior to a disease outbreak, and to report infections as soon as they have occurred. Since the government cannot observe biosecurity efforts, these cannot be regulated directly and hence we have a case of moral hazard. Infection status is also not observable prior to government testing, and so this is a case of adverse selection. As more diseases become transmitted via foreign trade and non-conventional sources like wildlife, the risks to livestock will increase. However, if indemnity payments effectively insure farmers against these increased risks, without taking farmer responses into account, then farmers have no incentives to undertake preventive actions. The result is that infections are likely to rise along with indemnity payments - something that has been observed in recent years as APHIS funded most of its indemnity payments using emergency funding. This study yielded a chapter of a student's PhD Dissertation, which was successfully defended this past year, and has since been accepted for publication in the American Journal of Agricultural Economics. Finally, PI-s Wolf and Horan have also performed an analysis of the U.S. TB Eradication Program and communicated the findings to USDA-APHIS. PARTICIPANTS: Richard D. Horan and Christopher A. Wolf, both in the Department of Agricultural, Food, and Resource Economics, Michigan State University, were co-PIs for this work. Also involved were Eli P. Fenichel (recent graduate of MSU; currently at Arizona State University), Bengamin M. Gramig (recent graduate of MSU; currently at Purdue University), K. Mathews Jr. (USDA-ERS), and Fang Xie (M.S. student in the Department of Agricultural Economics at Michigan State University). This research project has provided training for the current and former students listed in the areas of bioeconomic modeling, dynamic optimization and simulation, and econometrics. Benjamin Gramig's dissertation and Fang Xie's MS Thesis were based on the results of this research. TARGET AUDIENCES: The target audiences for our work this past year have primarily been academic economists and ecologists, and APHIS personnel. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
By incorporating human responses to disease risks, and the joint-determination of economic and ecological systems, we have illustrated both why it is important to recognize human responses and how this information can be used to improve disease management. Most of our work focuses on how to allocate disease prevention and control efforts, across species and over time, in an economically efficient manner for disease systems involving wildlife and livestock. The conventional wisdom has been that control efforts should be directed primarily at wildlife disease reservoir hosts to protect livestock. We find this may not be the case under certain circumstances. The problem is that infected wild animals cannot be identified prior to the kill. If they could, then controlling the disease would be simply a matter of hunting down all infected animals. But with disease status being unobservable, harvesting becomes non-selective with respect to disease status. As a result, harvesting results in offtake of both healthy and infected deer, which increases the cost of disease control if the host species is of economic value, as deer are in Michigan. In contrast, livestock systems are highly managed, with the possibility of testing to identify infected animals and of biosecurity and livestock management practices that can reduce cross-species contacts. Due to these differences in manageability, we find it more cost-effective to invest more heavily in livestock-sector controls. For our work on the Brucellosis problem in the GYA, our results confirm prior disease ecology work that population controls, even when feeding is eliminated, are unlikely to eradicate brucellosis in elk. We also find that some farmers would likely vaccinate some on their own, though at rates much less than those arising under vaccination mandates. Finally, we find that closing down public feedgrounds, which farmers support doing, may actually increase disease risks by reducing the number of resistant elk and by encouraging elk to move out of the feedgrounds in search of food. Farmers respond to these increased disease risks with increased investments in vaccinations, with the costs and benefits of vaccination nearly offsetting so that farmers are not significantly worse off. Our work on the design of indemnity payments illustrates how farmer responses can be taken into account when constructing indemnity payments, so that these payments can be designed to encourage farmer investments in biosecurity measures. The resulting indemnity payments would satisfy the constitutional requirement of paying fair market value for diseased animals, while also reducing infection risks and therefore the likelihood that any payments will actually need to be made. Finally, our work analyzing the TB eradication program illustrates the potential economic and epidemiological impacts of various proposed strategies for dealing with this disease problem.
Publications
- Horan, R.D., C.A. Wolf, E.P. Fenichel, and K.H. Mathews, Jr. (2008), "Joint Management of Wildlife and Livestock Disease", Environmental and Resource Economics, 41:47-70.
- "Economic Analysis of US TB Eradication Program Options," Christopher Wolf, Joleen Hadrich, Richard Horan, Phil Paarlberg, and John Kaneene. A Report to the U.S. Department of Agriculture, Animal and Plant Health Inspection Service, 2008.
- Xie, F., Disease and Behavioral Dynamics for Brucellosis in Elk and Cattle in the Greater Yellowstone Area. MS Plan B paper, Michigan State University, 2008.
- Gramig, B., Essays on the Economics of Livestock Disease Management: On-Farm Biosecurity Adoption, Asymmetric Information in Policy Design, and Decentralized Bioeconomic Dynamics. Ph.D. Dissertation, 2008.
|
Progress 09/01/06 to 08/31/07
Outputs
OUTPUTS: Several bioeconomic models have been developed or are under construction to explore the management of disease problems that affect wildlife and livestock. On study focuses on the joint management of wildlife and livestock when both populations are already infected, and includes a numerical example applied to the case of bovine tuberculosis (bTB) in Michigan white-tailed deer and cattle. This paper has been submitted for publication in Environmental and Resource Economics, and we have been invited to submit a revised version. A second study, which is only at the theoretical stage, expands the problem to examine management prior to infection when there is a positive risk of infection. This paper was presented as a principal paper at the 2007 annual meetings of the American Agricultural Economics Association in Portland, OR, in July (co-PIs HOran and Wolf organized the Principal Paper Session on Invasive Species). It has also been submitted to the American Journal of
Agricultural Economics. A third study, also based on the problem of bTB in Michigan, focuses on gender-based management of the deer population as a way of improving risk-management of the disease. This paper has been accepted for publication in the American Journal of Agricultural Economics (but has not yet been published and so is not listed in the published outputs below). A fourth study, which will be part of a PhD dissertation, econometrically estimates the spillover effects of various livestock biosecurity practices. Specifically, the analysis investigates whether producers take into account how recommended actions to protect against infection for one disease have an added benefit of protecting against additional diseases. This paper was presented as a principal paper at the 2007 annual meetings of the American Agricultural Economics Association in Portland, OR, in July (co-PIs HOran and Wolf organized the Principal Paper Session on Invasive Species). It has also been submitted
to the American Journal of Agricultural Economics.
PARTICIPANTS: Richard D. Horan and Christopher A. Wolf, both in the Department of Agricultural Economics, Michigan State University, were co-PIs for this work. Also involved were Eli P. Fenichel (PhD candidate in the Department of Fisheries and Wildlife at Michigan State University), Bengamin M. Gramig (PhD candidate in the Department of Agricultural Economics at Michigan State University), and Fang Xie (M.S. student in the Department of Agricultural Economics at Michigan State University). This research project is providing training for these students in the areas of bioeconomic modeling, dynamic optimization and simulation, and econometrics. Benjamin Gramig's dissertation and Fang Xie's MS Thesis will be based on the results of this research.
TARGET AUDIENCES: The target audiences for our work this past year have primarily been academic economists and ecologists, though we have consulted some with ecologists in the Michigan Department of Natural Resources.
Impacts
Most of our work focuses on how to allocate disease prevention and control efforts, across species and over time, in an economically efficient manner for disease systems involving wildlife and livestock. The conventional wisdom has been that control efforts should be directed primarily at wildlife disease reservoir hosts to protect livestock. We find this may not be the case under certain circumstances. The problem is that infected wild animals cannot be identified prior to the kill. If they could, then controlling the disease would be simply a matter of hunting down all infected animals. But with disease status being unobservable, harvesting becomes non-selective with respect to disease status. As a result, harvesting results in offtake of both healthy and infected deer, which increases the cost of disease control if the host species is of economic value, as deer are in Michigan. We find that investing in biosecurity measures that can prevent infectious contacts
between wildlife and livestock (e.g., deer fencing) may be a more economical approach. These investments directly target the source of damages to the livestock sector (i.e., infectious contacts), and they reduce the reservoir status of the wild herd. With these investments in place, we find it is actually optimal to reduce control efforts in the deer sector since livestock damages have been reduced. Alternatively, we also find that disease control costs in the deer sector can be reduced through gender-based management. While we cannot observe infected animals, we can observe gender. And behavioral differences among the genders result in differential disease risks for each gender. So by differentially targeting control efforts towards males and females, we can be more likely to reduce the force of infection among deer. We also consider pre-disease controls versus post-disease controls. We find that controls to prevent an outbreak are important, though adopting such stringent controls
so as to reduce the probability of infection to zero is not optimal. Finally, we find that producers do not take the spillover effects of biosecurity actions into account, possibly signifying an underinvestment in these activities. It is likely that they have not been made aware of these spillover effects in the past and therefore do not seek to take advantage of them. More research in this area is ongoing.
Publications
- No publications reported this period
|
|