Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Agricultural Household Response to Avian Influenza Prevention and Control Policies Presented by Robert H. Beach, Christine Poulos, and Subhrendu K. Pattanayak RTI International Presented at Southern Agricultural Economics Association Annual Meetings Mobile, Alabama, February 4-6, 2007 3040 Cornwallis Road Phone 919-485-5579 ■ P.O. Box 12194 ■ Fax 919-541-6683 Research Triangle Park, NC 27709 e-mail [email protected] RTI International is a trade name of Research Triangle Institute Avian Influenza Avian Influenza (AI) viruses occur naturally among wild birds and generally cause only minor symptoms from which the birds rapidly recover Many subtypes of AI that vary in pathogenicity and infectivity Classified by the combination of two proteins on the surface of the virus z Hemagglutinins (H1 through H16) z Neuraminidases (N1 through N9) Numerous strains generally categorized into z Low pathogenicity avian influenza (LPAI) Generally causes only mild illness in birds z Highly pathogenic avian influenza (HPAI) Extremely infectious, causes severe illness, and can reach mortality rates above 50% and often reaching 90-100% in domesticated poultry, often within 48 hours 2 Strains of Avian Influenza LPAI is endemic in wild birds and outbreaks in poultry are not unusual and have occurred in countries throughout the world z 3 Certain birds, particularly migratory waterfowl, often carry the virus while showing no outward signs of illness, but shedding virus in saliva, nasal secretions, and feces However, some strains of LPAI (H5 and H7) can rapidly mutate and become highly pathogenic, especially when moving between bird species Domestication of poultry has created species subtypes that can be infected by an AI virus adapted to waterfowl and have it mutate rapidly into a strain of HPAI HPAI Infection Poultry can become infected through: z Primary introduction of HPAI z Secondary transmission 4 Contact with infected wild birds Live bird markets Contamination of people, equipment, or vehicles Infectivity (amount of virus produced by infected flock) Susceptibility (amount of virus needed to infect susceptible bird) Amount of virus transferred during contact Contact rate Number of flocks that make contact H5N1 5 Since 2003, outbreaks of A(H5N1) have led to the death of hundreds of millions of birds through disease mortality or culling (primarily chickens and ducks on farms), primarily in Southeast Asia z FAO estimates $10 billion in losses to SE Asian poultry farmers Adding to economic impacts are trade restrictions and negative demand shocks z Livestock disease (particularly zoonotic) potentially results in both short-run and long-run changes in consumption patterns In addition, the disease has infected humans in 11 countries z 270 confirmed human cases of H5N1 in humans since 2003 and 164 deaths (60.7% fatality rate) (WHO, January 2007) There is concern that H5N1 could mutate into a form that could spread between humans, which could lead to a global pandemic z Estimates that millions of people would be killed (WHO, 2004) and global economic losses in the hundreds of billions or even trillions of dollars (e.g., McKibbin and Sidorenko, 2006) Outbreaks of H5N1 in Poultry 6 Large Potential Impacts for Southern U.S. 7 HPAI can have major economic impacts on the poultry industry and regional agricultural economy U.S. is the world’s largest producer and second-largest exporter of poultry meat and one of the world’s largest egg producers z Farm sales of poultry meat and eggs over $28 billion in 2005 U.S. poultry production is concentrated in the south z Georgia – 1.32 billion broilers (15%) z Arkansas – 1.21 billion broilers (14%) z Alabama – 1.06 billion broilers (12%) z Mississippi – 853 million broilers (10%) z North Carolina – 735 million broilers (8%) z Texas – 628 million broilers (7%) z Kentucky – 298 million broilers (3%) Poultry and Eggs as Percentage of Total Agricultural Market Value 8 9 Source: Obtained from http://www.topix.net/search/ on December 22, 2006 using "avian or bird" and "flu or influenza" as search terms. 12 /2 1/ 6 06 00 20 /2 06 6 06 00 20 6 6 6 6 06 00 00 00 00 20 /2 3/ /7 /2 12 11 6/ 6 06 20 /2 /2 /2 /2 6 06 00 20 2/ /9 /2 /1 28 14 31 17 11 10 10 9/ 9/ 8/ 8/ 6 6 6 6 6 06 00 20 /2 3/ 20 8/ 7/ /2 6/ 22 7/ 6/ 00 00 00 00 00 6 06 00 20 /2 /2 /2 /2 /2 /2 8/ 25 11 27 13 30 16 6/ 5/ 5/ 4/ 4/ 3/ 3/ 6 06 00 20 /2 2/ 16 3/ 2/ 20 6 06 05 00 20 20 /2 2/ 19 2/ 1/ 2/ 5/ /2 1/ 12 Global Media Index for Avian Influenza 600 500 400 300 200 100 0 10 19 / 2/ 20 06 2/ 20 2/ 0 16 6 /2 0 3/ 06 2/ 20 3/ 0 16 6 /2 3/ 006 30 /2 4/ 006 13 /2 4/ 006 27 /2 5/ 006 11 /2 5/ 006 25 /2 0 6/ 06 8/ 2 6/ 006 22 /2 0 7/ 06 6/ 20 7/ 0 20 6 /2 0 8/ 06 3/ 20 8/ 0 17 6 /2 8/ 006 31 /2 9/ 006 14 /2 9/ 006 28 / 10 200 /1 6 2/ 10 200 /2 6 6/ 2 11 006 /9 / 11 200 /2 6 3/ 2 12 006 /7 / 12 200 /2 6 1/ 20 06 1/ 05 00 6 /2 0 5/ 2 /2 2 1/ 12 Media Index for Avian Influenza: United States 300 250 200 150 100 50 0 Source: Obtained from http://www.topix.net/search/ on December 22, 2006 using "avian or bird" and "flu or influenza" and country name as search terms. Outbreaks in 2007 Europe/Near East z Azerbaijan – suspected human case z Britain – first case of HPAI in captive birds this week (2,600 birds died, >160,000 to be culled); has increased fears of spread across Europe z Hungary – H5N1 in goose farms (90 birds died, 3,265 culled) Africa/Middle East z Egypt – 1 human death z Nigeria – 1st human case; concerns that official quarantine is not being enforced Asia z z z z z z z 11 Hong Kong – 2 poultry outbreaks suspected Indonesia – 6 human cases (5 deaths), poultry outbreaks, mass culling in Jakarta Japan – 3 poultry outbreaks (7,032 birds died, 69,468 culled) Russia – 1st outbreak of the season this past week South Korea – 5 outbreaks (9,757 birds died, 319,746 culled) Thailand – 3 provinces with poultry outbreaks (336 died, 3,734 culled) Viet Nam – 52 poultry outbreaks (10,314 birds died, 15,485 culled) AI Outbreaks: Turkey Source: http://declanbutler.info/blog/?cat=6 12 13 Animal Outbreaks Human Cases Fresh Chicken Media Index/10 Sep 25 2006 Sep 04 2006 Aug 14 2006 Jul 24 2006 Jul 03 2006 Jun 12 2006 May 22 2006 May 01 2006 Apr 10 2006 Mar 20 2006 Feb 27 2006 Feb 06 2006 Jan 16 2006 Dec 26 2005 Dec 05 2005 Nov 14 2005 Oct 24 2005 Oct 03 2005 AI Events/Media Index 70 60 50 40 30 20 10 0 1,000,000 900,000 800,000 700,000 600,000 500,000 400,000 300,000 200,000 100,000 0 Fresh Chicken Sales (kg Turkey - Fresh Chicken AI Outbreaks: Italy Source: http://declanbutler.info/blog/?cat=6 14 15 European Media Source: Lexis-Nexis Academic database Italy Specific Sep 10 2006 Aug 13 2006 Jul 16 2006 Jun 18 2006 May 21 2006 Apr 23 2006 Mar 26 2006 Feb 26 2006 Jan 29 2006 Jan 01 2006 Dec 04 2005 Nov 06 2005 Oct 09 2005 Sep 11 2005 Aug 14 2005 Jul 17 2005 Jun 19 2005 May 22 2005 Apr 24 2005 Mar 27 2005 Feb 27 2005 Jan 30 2005 Jan 02 2005 Dec 05 2004 Nov 07 2004 Oct 10 2004 Sept 12 2004 Aug 15 2004 European Media Index 3,000 600 2,500 500 2,000 400 1,500 300 1,000 200 500 100 0 0 Italy Media Index Avian Influenza Media Indices Used in Analysis of Italian Meat Consumption Elasticities of Poultry Demand with Respect to Media Reports on Avian Influenza 16 Calculated elasticity with respect to media reports at sample means for the two media indices used Long-run elasticities (40-week period) z Fresh (-0.088 avrow; -0.061 avit) z Frozen/processed (-0.024 avrow; -0.118 avit) Estimates indicate that news outside of Italy had more impact on consumption of fresh poultry than Italy-specific news z No domestic outbreaks until Feb 2006, almost 18 months after the start of our sample z Italian consumers may have already partially adjusted to AI news from other regions The opposite is found for frozen/processed poultry, where Italy-specific news had a much larger impact on consumption 17 European and Italy-specific media European media only Sep 24 2006 Aug 13 2006 Jul 02 2006 May 21 2006 Apr 09 2006 Feb 26 2006 Jan 15 2006 Dec 04 2005 Oct 23 2005 Sep 11 2005 Jul 31 2005 Jun 19 2005 May 08 2005 Mar 27 2005 Feb 13 2005 Jan 02 2005 Nov 21 2004 Oct 10 2004 Weekly sales as a proportion of simulated sales without media reports o Simulated Effects of AI Concerns on Italian Fresh Poultry Sales, Proportionate Change 1.00 0.90 0.80 0.70 0.60 0.50 18 European and Italy-specific media European media only Sep 24 2006 Aug 13 2006 Jul 02 2006 May 21 2006 Apr 09 2006 Feb 26 2006 Jan 15 2006 Dec 04 2005 Oct 23 2005 Sep 11 2005 Jul 31 2005 Jun 19 2005 May 08 2005 Mar 27 2005 Feb 13 2005 Jan 02 2005 Nov 21 2004 Oct 10 2004 Weekly sales as a proportion of simulated sales without media reports on Simulated Effects of AI Concerns on Frozen/Processed Poultry Sales, Proportionate 1.00 0.90 0.80 0.70 0.60 0.50 European Media Index Poultry Production and Avian Influenza Table 1. Classification of Poultry Production Systems System Biosecurity Bird and product marketing Sector 1 Sector 2 Sector 3 Sector 4 Industrial integrated Commercial Commercial Village or backyard High Moderate to high Low to minimal Minimal Commercial Usually commercial Birds usually sold in live bird markets Birds and products consumed locally Source: FAO (2004). 19 Scale and production technology vary enormously between and within countries Sectors 3 and 4 generally considered to be more susceptible to infection due to low levels of biosecurity, although losses may be larger in the event of an infection in Sector 1 and concentration of birds may increase probability of virus circulation and mutation Poultry Systems in Southeast Asia Country Cambodia Sector 1 Sector 2 Sector 3 Sector 4 Believed not to exist ~0.4 million poultry 0.4 million chickens 0.8 million ducks 99.9% of farms and 90% of poultry (12 million chickens and 2.7 million ducks) Indonesia 9.7 million poultry, export oriented but large proportion for national consumption 58.2 million poultry for the national market 32.4 million poultry 174 million poultry Lao PDR Believed not to exist Relatively insignificant 10% of poultry 90% of poultry Thailand 70% of national production, important export market 20% of national production 36% of farms and 7% of production 61% of farms and 3% of production Viet Nam Relatively insignificant 20-25% of production, but very few producers 10-15% of production, but very few producers 65% of production with possibly 70% of the country’s population involved Source: Rushton et al. (2005). 20 Duck Production Systems in Thailand 21 A) Closed system with high biosecurity, an evaporative cooling system, and strict entrance control. B) Open system but with netting to prevent entrance of birds. Biosecurity was not strictly enforced. This system is no longer approved for the raising of poultry. C) Free-grazing ducks. Biosecurity is never practiced in this system. D) Backyard Muscovy ducks; no biosecurity is practiced in this system Free-grazing duck husbandry in Asia 22 Found to be important risk factor in spreading HPAI in duckproducing regions of Asia Ducks infected with H5N1 may show few clinical signs while shedding enough of the virus to form a potential reservoir for continued infections Frequently rotated between rice paddy fields after rice harvest to feed on leftover rice grains, insects, and snails Come into contact with wild birds and then brought together with other flocks of free-grazing ducks and other poultry as they are rotated H5N1 in Thai Production Systems 23 Songserm et al (2006) conducted tests for H5N1 during the 2004 outbreak in Thailand that resulted in 60 million poultry being killed Found no H5N1 in the closed industrial system (about 1% of every flock was sampled before slaughter, over 10,000 birds) H5N1 was found in all three open systems: z Open house system: 4/17 flocks (23.5%) tested positive z Free-grazing duck system: 28/61 flocks (45.9%) tested positive 10 flocks studied in more detail; no evidence of H5N1 while they were in brooding houses but all 10 tested positive within 12-63 days of moving to rice fields z Backyard poultry: 56% of backyard chicken flocks and 27% of backyard duck flocks tested positive Responses to Outbreak in Viet Nam 24 Delquigny et al. (2004) examined three outbreaks in Viet Nam and found highly varied responses In the first outbreak, the company notified the Vietnamese veterinary services of high death rates and the birds in the region were destroyed, although the level of destruction of birds and eggs varied across villages In a second outbreak that took place in a village with no official veterinary services, data collected on the fate of sick and apparently healthy birds showed that almost 29% of sick birds and 56% of healthy birds were consumed, 70% of sick birds and 8% of healthy birds were destroyed, and 1% of sick birds and 36% of healthy birds were sold The third outbreak took place on two commercial farms. In a 5,000 bird broiler farm, chickens began dying on day 1 with 50% mortality by the end of day 2. On day 3, the remaining birds were killed and buried using quicklime. In a 1,200 bird layer farm, 50% of hens died by the second day of the outbreak, but the farmer sold the rest to a trader. The farm also had 600 chicks that all died on the third day and were disposed of in a public rubbish area. Thus, in addition to the importance of incentives for reporting disease and ensuring that infected birds are destroyed, proper incentives are also necessary for encouraging adequate procedures for disposal of infected carcasses Prevention and Control Measures 25 Provision of information and technical assistance Disease surveillance Improved biosecurity practices Culling and disposal Vaccination Information and Technical Assistance 26 Poultry producers may not have sufficient information on HPAI and appropriate disease prevention and control strategies to make efficient decisions on adoption of control measures Backyard growers may not be aware of the potential severity of HPAI and could have more difficulty identifying an outbreak because villagers in many developing countries may accept significant poultry losses as “normal” (Rushton et al., 2004) Development and dissemination of this information lowers the private cost of control measures for producers by reducing the time and human capital required to identify and adopt appropriate actions, thereby increasing expected adoption Disease Surveillance Another important measure for controlling the spread of disease is surveillance of poultry operations to monitor birds for disease and immediately report any suspected cases of AI Proper incentives are vital for inducing producers to report suspected outbreaks z 27 In some Asian countries, low or no compensation for birds lost is a significant impediment to disease reporting As part of surveillance, it is important to understand the epidemiology, ecology, and economics of the disease and assess spatial and temporal patterns to improve the effectiveness of control programs Improved Biosecurity Practices 28 Strict biosecurity practices that prevent exposure to any animals or other items potentially contaminated with AI are vital for preventing and controlling the spread of disease z Closed poultry housing z Ensure wild birds cannot access poultry feed and water supplies z Isolate new birds or avoid their introduction into existing flocks z Limit access to poultry houses and thoroughly clean all clothing, shoes, equipment and vehicles before and after coming in contact with birds z Restrictions on transportation and marketing, especially at live bird markets Culling and Disposal 29 Culling infected birds as well as birds that may have come in contact with the sick birds has been a typical response to an HPAI outbreak and is often credited with limiting the spread of the disease, particularly in countries such as Hong Kong and Thailand, which moved quickly to quarantine and destroy potentially affected flocks Reduces flock infectivity by removing the flock altogether In conjunction with surveillance, early detection and culling reduces amount of virus produced by infected birds Compensation level is vital for producer cooperation Vaccination Vaccination reduces the probability of infection and the amount of virus produced by an infected flock Currently being used in several countries, including Indonesia, Viet Nam, and China, which has instituted a major initiative to inoculate poultry in areas considered susceptible to AI infection z 30 China inoculated 2.68 billion birds, primarily chickens, ducks, and geese between February 2004 through January 2005 Potentially costly to administer Concerns that vaccination may suppress the symptoms of the virus for vaccinated birds that get infected, allowing it to continue spreading without notice – e.g., Thailand and South Korea are not vaccinating Farm-level behavior 31 Increase in risk of poultry production will provide incentives to shift away from poultry production towards non-poultry farm production and non-farm income, other things being equal Because of potential income losses (and the fact that human cases have generally been poultry workers in direct contact with live infected birds), there are clearly incentives for poultry farmers to take precautions against HPAI Nonetheless, there are externalities that will not be fully taken into account Adoption of Control Measures 32 Producers adopt control measures to the point that the private marginal benefits are equal to the private marginal costs z Tend to under-invest relative to socially optimal level z Decision to invest in avoiding primary introduction depends on the farmer’s ability to impact likelihood of introduction into region For regions where HPAI is not yet present, producers may determine it is optimal not to invest in avoiding introduction into the area if they have little impact on probability of regional introduction Public Policy 33 Providing information to improve private decision-making National surveillance program z Analyze the epidemiology, ecology, and economics of the disease and assess spatial and temporal patterns to improve the effectiveness of control programs. May be welfare-improving for government to compel or provide incentives for at least some producers (often largest) to adopt biosecurity (Hennessey, 2006) Once these producers have adopted, the probability that other farmers would be the source of introduction increases, which may make it optimal for them to adopt as well Similarly, vaccination of a certain fraction of the susceptible population by government may help avoid major epidemic In the event of an outbreak, government involvement in proper culling and disposal likely to be important Compensation Producer compensation for losses is a major issue Costly for governments to provide, but important tool for reducing severity of outbreaks Too low 34 z Less surveillance z Do not report suspected infections Too high z Little incentive for costly control measures to avoid HPAI z Possibly even incentive to get birds infected! Current Compensation Levels 35 Varies widely z Thailand: 70-100% of market price z Viet Nam: 20-30% before being raised to 50-60% in November 2005 z China: 80% of market price z Cambodia: 0% z Egypt and Nigeria provide compensation, but exclude smallholders z Canada: 100% market value after adjustments for age and salvage value z US: 100% of costs associated with eradication (for H5/H7 LPAI, those involved in the National Poultry Improvement Plan are also eligible for 100%, 25% for those that are not as of September 2006) The Economist reported in late January 2007 that z Indonesia is offering equivalent of $1.38 per bird but the central government has expressed concern that local authorities have been slow to pay z China’s central government has promised to spend $1 billion over the next two years on HPAI and other animal epidemics, but there have been concerns that local officials do not report all outbreaks, are slow to investigate others, and have been slow to cull infected birds due to the cost of poultry farmer compensation Conclusions 36 Poultry producers play a vital role in containing HPAI Spatial dimension of agricultural production is important for infectious animal diseases There are averting and mitigating options available to farm households, but numerous barriers to adoption especially in smallholder systems z Information and resource constraints z Social and cultural barriers z Disease control efforts substitute for those of other producers in the region z Difficulty of monitoring and enforcement Conclusions (2) 37 Public policy design is very important for providing socially optimal incentives for infectious disease control z Provision of information z Compensation level for culled poultry z Compulsory vaccination or biosecurity vs. market incentives Comparisons between alternative control strategies need to reflect the incentive structures under each strategy Appropriately designed policies that account for producer response can help overcome the coordination failure that otherwise arises under market equilibrium conditions and improve household welfare z Government resource limitations z Difficulty in monitoring and enforcement z Implications for poultry production systems Future Directions 38 Empirical research to quantify implications of farm behavior for the spread of avian influenza, the distribution of impacts, the extent and cost of private disease prevention behavior, and the effectiveness of public disease control programs z Requires detailed farm household-level data from before and after the recent H5N1 outbreaks as well as information on outbreaks and local ecological and market conditions z Examination of producer responsiveness to incentives in reporting disease and cooperation with depopulation Incorporation of potential human health effects Integration with economy-wide modeling of impacts using CGE model