An Environmental Study of Bovine Somatotropin Use in the U.S. Impacts on Methane Emissions by Christine Kaestle Douglas Williams Michael Gibbs ICF Incorporated Prepared for Tom Wirth, Ruminant Livestock Methane Program Atmospheric Pollution Prevention Division U.S. Environmental Protection Agency August 21,1996 ------- An Environmental Study of Bovine Somatotropin Use in the U.S. Impacts on Methane Emissions Table of Contents Executive Summary 1 Preface 5 1. Introduction 5 2. Background 5 3. Results of rbST Use on Individual Herds 7 4. Consumer Concerns 8 4.1 Key consumer issues 8 4.2 Impact on dairy product demand. 10 5. Current and Future rbST Use 12 5.1 Current market penetration 12 5.2 Predicted adoption rates through 2010 15 6. Potential Consequences of Future rbST Use 17 6.1 Economic impact ofrbST use on the dairy industry. 17 6.2 Methane emissions model and results 18 6.3 Impact ofrbST on dairy industry methane emissions through 2010 21 Tables and Figures Figure 1: illustrative lactation cycle with rbST and without rbST treatment 7 Figure 2: mature cow populations through 2010 22 Figure 3: emissions factors through 2010 23 Figure 4: kg methane emitted per kg milk produced 24 Figure 5: methane emissions through 2010 25 Figure 6: illustration of supply curve shift and resulting increase in production 27 Figure 7: comparison of emissions reduction in 2000 with and without the supply curve shift 28 Table 1: rbST adoption and methane emissions summary for the years 2000 and 2010 3 Table 2: Hoard's survey - rbST use in 1995 13 Table 3: Hoard's survey - percent of herd treated with rbST 13 Table 4: Hoard's survey - rbST milk production response 13 Table 5: Hoard's Survey - months of rbST use 14 Table 6: Hoard's survey - plans for 1996 rbST use 14 Table 7: Regional Predicted Farmer Adoption Rates 15 Table 8: USDA and Executive Branch predicted rbST adoption rates 16 Table 9: Assumed mature cow rbST treatment under various adoption scenarios 16 Table 10: Baseline Methane Emissions - No rbST Adoption 19 Table 11: Baseline Comparison with EPA (1993) 20 Table 12: rbST adoption and methane production Low scenario 20 Table 13: rbST adoption and methane production High scenario 21 Table 14: rbST adoption and methane production Full scenario 22 Table 15:1995 dairy methane emissions 22 Table 16: EPA 1 993b report rbST impact compared to this report 25 Table 17: rbST adoption and methane emissions summary for the years 2000 and 2010 29 August 1996 ------- An Environmental Study of Bovine Somatotropin Use in the U.S. Impacts on Methane Emissions Executive Summary Action #39 of the Climate Change Action Plan targets methane emissions from dairy and beef cattle and lists production enhancing agents as an important tool for improving management and reducing emissions. One such production enhancing method is the treatment of dairy cows with bovine somatotropin (bST). This report summarizes current issues surrounding bST use and assesses current and future impacts of bST use on methane emissions. It includes estimates of current and potential future rates of bST adoption, a summary of consumer concerns about bST, and scenarios of the impact of bST use on methane emissions from the dairy industry. bST is a hormone produced naturally in cows which encourages milk production. This hormone can also be produced through the use of modem biotechnology in an essentially identical form called recombinant bST (rtoST). rbST can be injected into cows to increase milk production by about 1,800 pounds per year. To achieve good results, farmers must have good management practices and must adjust the cow's rations to match the higher milk production level. Consumers have many concerns over rbST use involving health and social issues. At this time, health risks associated with milk from rbST treated cows have not been substantiated with scientific research, although additional research in some areas may be warranted. Also, to date, these consumer concerns have not reduced the demand for dairy products. The impact of rbST adoption on the dairy industry will generally reinforce current trends. Milk production efficiency will be improved, thereby reducing the cost of producing milk. Depending on other demand and supply factors, this increase in efficiency may contribute to an eventual decline in farm milk prices while contributing to improved profitability of producers. rbST use will probably accelerate the move toward larger farms. Between 10 to 27% of mature U.S. cows are currently treated with rbST, and treatment levels vary regionally. Three scenarios of future adoption rates through 2010 were estimated: • Low scenario represents a low adoption rate, reaching 40% of mature U.S. cows by 2010. • High scenario represents a high adoption rate based on USDA (1995) predictions, reaching 75% of mature cows by 2010. • Full scenario represents a very high adoption rate, reaching 95% of mature cows by 2000, designed to demonstrate rbST's full potential impact. These adoption scenarios were compared to a baseline scenario without rbST to determine the impact of rbST on cow numbers and methane emissions. The baseline scenario includes the continuation of historical increases in milk production per cow. August 1996 1 ------- As milk production per cow rises as a result of rbST adoption, the size of the U.S. dairy herd will decline. This will cause methane emissions to decrease compared to a baseline of no-rbST-use. This decline is demonstrated in Figure 1 for each of the three adoption rate scenarios. However, because of predicted increases in the quantity of milk produced to meet future demand, even the Full adoption scenario shows a small emissions increase by the year 2010 compared to 1990 levels. Figure 1: Dairy Cattle Methane Emissions through 2010 • The baseline assumes no-rbST-use. Production per cow is assumed to increase gradually from other industry improvements. From 1994 through 2010, baseline methane emissions increase from 1.48 Tg to 1.54 Tg per year. • Low scenario is a low adoption rate scenario. In 2010, Low scenario methane emissions are 1.3% lower than baseline emissions for that year. • High scenario, a high adoption rate scenario, has emissions 2.6% less than baseline emissions in 2010. • Full scenario, a very high adoption rate, demonstrates rtST's potential. Emissions for 2010 are 3.2% lower than baseline rates. Even with full adoption, methane emissions are estimated to increase slightly by 2010 from 1990 levels. Current adoption indicators and predictions of future use indicate that adoption of rbST will most likely fall between the Low scenario and High scenario. This would represent a decrease of 0.01 to 0.03 Tg methane (0.06 to 0.18 MMTCE) from this study's baseline of no rbST use in the year 2000. Table 1 gives summary information for the years 2000 and 2010 for the baseline no-rbST-use and the three scenarios. The Climate Change Action Plan (CCAP) sets a goal of a 1.8 MMTCE reduction in methane emissions from all U.S. ruminant livestock compared to baseline calculations by the year 2000.1 The use 11n this report all MMTCE (Million Metric Tons Carbon Equivalent) conversions use a methane Global Warming Potential of 22. 2 August 1996 ------- of fbST between the Low scenario and High scenario adoption levels could reduce methane emissions by 3.3% to 10% of the CCAP goals for ruminants in the year 2000. Table 1: rbST adoption and methane emissions summary for the years 20! 10 and 2010 scenario year rbST adoption mature cow population methane Tg per year 1990 0% 10,008,000 1.48 baseline 2000 0% 9,186,000 1.48 2010 0% 8,800,000 1.54 1990 0% 10,008,000 1.48 low scenario 2000 20% 9,004,000 1.47 2010 40% 8,508,000 1.52 1990 0% 10,008,000 1.48 high scenario 2000 45% 8,786,000 1.45 2010 75% 8,269,000 1.50 1990 0% 10,008,000 1.48 full scenario 2000 95% 8„381,000 1.41 2010 95% 8,137,000 1.49 These scenarios were obtained using a series of assumptions that should be considered when evaluating the results. These assumptions are as follows: • There is significant uncertainty regarding the future adoption rate of rbST; the baseline and these scenarios are designed to cover a range of possible outcomes. The actual adoption rate of rbST will depend upon the final judgment of farmers and consumers. • Future milk production will be driven by demand and supply conditions, including feed costs, other farming alternatives, international trade conditions, and meat prices. To estimate the impact of rbST adoption on methane emissions, a single scenario of future milk production was adopted. To the extent that rbST use reduces consumer prices for milk products, demand (and hence production) may increase, thereby offsetting a portion of the estimated emissions reduction. Because the price elasticity of demand is relatively low for milk products in the U.S., only a small portion of the emissions reduction is likely to be offset by increased production. This issue is discussed further in section 6.3. • Based on USDA projections through 2005 (USDA, 1995), this analysis assumes that imports and exports remain generally stable. However, the dairy industry is working to increase exports. If successful, increased exports would likely increase total U.S. milk production and hence, increase methane emissions from the U.S. dairy industry over the scenarios presented in this report. • For the baseline, historical increases in milk production per cow continue at a rate of 320 pounds per cow per year. This number is based on historical trends in cow productivity (Blaney et a/., 1995). There is significant debate over the magnitude of future increases in cow productivity. August 1996 3 ------- • Response to rbST use is estimated by assuming that adopting farmers successfully obtain a 1,800 pound increase in milk production per cow per year for each cow treated with rbST. This may be a conservative assumption because if a percent increase had been assumed, then as the baseline production per cow rose each year, the amount by which rbST increased a cow's production would have risen as well. • Methane emissions from dairy cattle are influenced by animal and feed characteristics. This report assumed generally representative and constant animal and feed characteristics, such as cow weight, pregnancy rate and feed digestibility. The model incorporates increases in feed intake when milk production rises. The model does not include any additional methane emission decreases that might result from higher feed quality. The use of more digestible, higher quality feeds may be encouraged by bST use. August 1996 ------- Preface President Clinton's Climate Change Action Plan (CCAP) addresses the problem of increasing greenhouse gases such as methane. The CCAP is a comprehensive response designed to reduce greenhouse gas emissions while strengthening the U.S. economy. The CCAP identifies ruminant livestock as a target for methane emission reduction strategies. In 1990, U.S. dairy and beef cattle were responsible for about 32.4 MMTCE (5.4 Tg)2 per year of methane emissions (EPA. 1995). Action #39 of the CCAP outlines the reduction of projected livestock methane emissions levels in the year 2000 by 1.8 MMTCE (0.3 Tg). To accomplish this, the EPA is studying the impacts of dairy production and management practices on methane emissions and supporting the use of methane reducing practices through a variety of programs. This report was prepared to study the influence of rbST, a production enhancing agent, in accordance with Action #39. 1. Introduction When CCAP was developed in 1993, rbST was new and its potential penetration and impacts were unknown. Because rbST will potentially have significant impacts on methane emissions, EPA is reviewing the results of its use to date and assessing its likely impacts on future methane emissions. The purpose of this report is to examine the issues surrounding rbST adoption in the United States and determine its current and future impact on methane emissions from the dairy industry. The report is organized as follows: • Section 2, Bovine Somatotropin Background, gives background information on rbST, including cow response, the technology behind recombinant bST production, and how rbST is administered to the cow. • Section 3, Results of rbST Use on Individual Herds, outlines the changes that are expected when a farmer adopts rbST, including the role of management and diet and the resulting milk increase. • Section 4, Consumer Concerns, discusses the possible effects of rbST on human or cow health and examines the possible impact of these concerns on dairy product demand and rbST use. • Section 5, Current and Future rbST Use, presents data on current market penetration of rbST and predicted rbST adoption rate scenarios through 2010. • Section 6, Potential Consequences of Future rbST Use, describes the impact of predicted rbST adoption on the dairy industry economy and on methane emissions. The adoption rate scenarios developed in Section 5 are used with a methane emissions model to determine changes in methane emissions due to rbST use. 2. Background Bovine Somatotropin (bST), also known as Bovine Growth Hormone (BGH), is a protein naturally produced in the pituitary gland of cows. This growth hormone is similar to the growth hormones produced in all animals and is important in the regulation of growth, development and nutrient use. In 2 The conversion of Tg of methane to MMTCE (Million Metric Tons Carbon Equivalent) was performed using a GWP of 22. Tg = 1,000,000,000 kg. August 1996 5 ------- the 1930s, it was discovered that when additional bST was injected into lactating cows their milk production increased. We now know that the mechanism by which bST causes increased milk production is quite complex. When bST is injected into lactating cows, it coordinates a redirection of nutrients to the mammary glands and improves milk production efficiency (Tyrrell, 1982). Many coordinated changes in the metabolism of various organs and tissues and in the allocation of all nutrient classes are involved in creating this production increase (Bauman, 1992). Despite these benefits, bST was not commercially exploited between the 1930s and the 1980s because supply was limited. Extracting the bST from the pituitary glands of slaughtered cows was the only known method of supplying bST, and only small amounts could be produced. Therefore, it was not economically feasible to produce bST on a large scale. More recently, biotechnology has developed the tools for producing recombinant bST on a commercial scale. In the 1980s, new biotechnology provided the opportunity to produce large amounts of recombinant bST (rbST) economically. rbST is produced by inserting the genetic material coding for the protein into a proper bacterial host. This bacteria can be grown in large numbers to produce large quantities of rbST. Recombinant bST is structurally and functionally the same as pituitary-derived bST, although 1 to 8 extra amino acids may become attached to the end of the molecule. The current commercially available rbST product contains one additional amino acid (Hard, personal communication, 7/9/96). These extra amino adds cause no change in bSTs biological activity (OTA, 1991). The new rbST production method made large scale bST use commercially feasible. Four companies originally sponsored rbST products for approval by the FDA. Monsanto's rbST product, called sometribove or Posilac, is the only rbST product currently on the market. Posilac was released on the market in February 1994. The product release followed a 90-day moratorium after FDA approval to study the socio-economic impact of rbST use. FDA approval was based on hundreds of formal scientific studies and hearings on safety related issues (Executive Branch, 1994, p4). rbST is administered during a portion of a cow's lactation cycle. Well-managed dairy cows give birth every 12 to 14 months. After giving birth, cows lactate for about 305 days. The cow's milk production level gradually increases, and peaks about 60 to 90 days after calving. Then milk production begins a steady decline. rbST produces the largest increase in milk production when it is administered following the peak in the lactation period for the remainder of the lactation period, i.e., for about 215 to 245-days. It is recommended to start treatment when cows reach their peak milk production in early lactation because at this time in their lactation period, cows are usually in a positive energy balance and have more milk secretory cells in the mammary gland than late lactation cows (Smith, 1995). When rbST is injected, milk yield increases gradually during the first 6 days of treatment, and increases in voluntary feed intake lag slightly behind. A 14-month calving interval is recommended for cows with rbST treatments because rbST extends the period in which cows produce milk efficiently (Iowa State University, 1993). Producers can administer rbST to their cows without a veterinarian. The Monsanto Posilac product is a prolonged release formula injected every 14 days. It is supplied in disposable single dose syringes and costs $0.42 per cow per day of treatment, or $0.38 with Monsanto's discount for producers who treat 6 August 1996 ------- more than 60% of their herd for 6 months or more. Data from farmers indicate that the recommended full treatment period of 215 to 245 days is often not carried out in the field. Depending on management practices, a dairy farmer may treat cows for 100 to 200 days per lactation. Thus, Posilac use costs range from about $40 to $100 per lactation, depending on the length of treatment. Biotechnology advancements may make the current rbST product obsolete. Improved delivery systems and more specific agents are being researched. In the far off future, transgenic cattle that are genetically engineered to produce higher levels of bST may be developed. Although transgenic cattle can now be produced, the techniques are too inefficient and expensive to use for commercial production at this time (OTA, 1991). 3. Results of rbST Use on Individual Herds Use of rbST can boost milk output per cow by an average of 1,800 pounds, or 816.5 kg per year (about 12%), for successful adopters. Response varies between individual cows. (Executive Branch 1994, p22). Upon rbST treatment, there is an immediate increase in milk yield, and then there is a reduction in the normal production decline over the lactation period (improved persistency) (Figure 1). The improved persistency is rbST's most important impact on total productivity (Iowa State University, 1993). The total milk response of a herd also depends on the duration of treatment per lactation and the percent of the herd treated. milk lactation period Cow management, including health programs, nutrition programs, environmental conditions, and milking practices, is the principal factor affecting the magnitude of increased milk production resulting from rbST use. Feed quality and availability and cow comfort are the most important aspects of cow management that influence rbST response (Hard, personal communication, 7/9/96). Inadequate feed or lack of nutrient balance can result in no response to rbST use on milk production. In this way, a lack of response to rbST use can be an indicator of management deficiencies. Figure 1: illustrative lactation cycle with rbST and without rbST treatment Peak: begin treatment production per day with bST treatment no bST treatment August 1996 7 ------- Use of rbST does not alter feed digestibilities or the energy required for maintenance or milk synthesis. Thus, existing nutrient requirement tables based on milk production levels remain applicable (Executive Branch, 1994, p49). Cows injected with rbST need more nutrients to support the higher production level: treated cows eat 10 to 20% more grain and forage (Iowa State University, 1993). Successful adopters emphasize the importance of maximizing dry matter intake early so the cows lose less weight in early lactation. Ideally, the cow should not lose over one body condition score3 (on a five point scale) during the entire lactation, and feed should be adjusted accordingly (Smith, 1995). Along with higher feed intakes, production efficiency is increased because a higher proportion of energy intake goes to milk production rather than maintenance. Each cow requires more energy, but the energy used per unit of milk produced decreases because the maintenance energy use is diluted by the higher milk production. Although positive results with rbST use have been observed over a wide range of ration types including nothing but pasture, future widespread rbST use could lead to an increase in the energy density in diets. Cows with higher milk production require more concentrates (high energy foods such as grains) (Bauman, 1992). Currently, however, farmers do not appear to be changing the composition of rations, just increasing the volume of feed (Steven Berry 3/4/96 and David Dickson 3/6/96, personal communication). 4. Consumer Concerns As a new biotechnology product for use in food production, rbST was put under intense scrutiny before receiving FDA approval. David Kessler, MD, the FDA Commissioner, stated that rbST is tone of the most extensively studied animal drug products to be reviewed by the agency .... The public can be confident that milk and meat from bST treated cows is safe to consume."(FDA, 1994). Despite these tests, many objections were made to the marketing and use of rbST. To the extent that these concerns persist, the future use of rbST and future demand for dairy products may be affected. 4.1 Key consumer issues Consumer concerns focus mainly on the possibility of humans ingesting increased amounts of certain substances from milk produced by rbST injected cows. One substance is rbST itself, which some people fear will be present in increased amounts and will be active in the human body. The second substance is IGF-1, which is a hormone that rbST treatment increases in cows. Some people fear that IGF-1 will have biological activity in the human body. The third substance is antibiotics that might be used at higher levels on rbST treated cows. Increased use of antibiotics raises many questions currently debated in the health field. Some consumers are concerned over animal welfare issues involved in rtST use. There is also some concern over the socio-economic effects of rbST. This subsection examines each concern in turn. rbST in milk: bST is not found in significantly greater amounts in milk from rbST treated cows, although all milk has trace amounts. In fact, at this time there is no practical way to test milk to determine if the cow that produced it was injected with rbST. bST consumed by humans is immediately broken down by 3 Body condition scoring is a method of cow evaluation based on weight which helps producers to make management decisions. 8 August 1996 ------- the digestive system into its component amino acids and peptides, which are biologically inactive. In the 1950s, bST was injected directly into the human blood stream of children with growth defects in an attempt to stimulate growth; however there was no response. We now know that bST differs from human somatotropin by about 30% of its amino acid sequence, so it cannot function in the human body (Iowa State University 1993). Ingestion and even accidental injection of bST into humans will result in no response. Insulin-like Growth Factor 1: Insulin-like Growth Factor 1 (IGF-1) is a naturally occurring hormone that may mediate some of bST's affect on cows. IGF-1 is found in increased levels in the blood and milk of rbST injected cows, and unlike somatotropin, bovine IGF-1 is identical to human IGF-1. The FDA requested additional research from the rbST sponsors on IGF-1. This research concluded that: • the natural IGF-1 content of cow's milk varies widely; • although the milk from rbST treated cows has IGF-1 levels above those found in non-treated cows, the IGF-1 levels appear to fall within the broad range found in milk from non-treated cows; • data indicate that milk from rbST treated cows has IGF-1 levels within the physiological range found in human breast milk; • IGF-1 is present naturally in human saliva and is apparently inactive orally and in the upper gastrointestinal tract; • the amount of IGF-1 consumed through milk from rbST-treated cows would be less than that which enters the stomach from saliva and other sources in the body; • even if IGF-1 was not destroyed through digestion, the FDA believes it would enter the human bloodstream in physiologically insignificant amounts. Based on its research, the FDA concluded that there are no human safety problems related to the increased IGF-1 levels in milk from rbST-treated cows (Juskevich and Guyer, 1990 and Executive Branch, 1994). However, some concern over IGF levels has recently surfaced. Dr. Samuel Epstein, a well-known critic of rbST use, published a literature review in the January 1996 issue of the International Journal of Health Services stating that there is evidence that IGF-1 from rbST-treated cow milk has cancer promoting effects on colon and breast cells. Dr. Epstein criticized previous studies of IGF-1's oral toxicity, and presented evidence linking IGF-1 to various cancers. Dr. Epstein also accused the FDA of active complicity in the release of a poorly tested product (Epstein, 1996). This article has been widely criticized by many health organizations. Dr. Stephan Sundlof, the director of the FDA Center for Veterinary Medicine, stated that Dr. Epstein's conclusions were tased on assumptions we consider to be blatantly false," and an American Medical Association official called the paper "bunk." (Steyer, 1996). Antibiotics: There is concern that rbST will increase the incidence of mastitis in dairy cows. Mastitis is a common infection of the udder which is treated with antibiotics. Some consumer activists fear that this will increase the chances of antibiotic residues to contaminate milk. The FDA concluded that use of Monsanto's sometribove product (rbST) is linked to a slightly higher occurrence of mastitis, about 0.1 case more per cow per year (Glausiusz, 1995). Other studies deny the connection between rbST use and mastitis, attributing any increase in mastitis occurrence to the increased milk production, not rbST directly (OTA, 1991). Either way, an increase in mastitis could increase the use of antibiotics for some cows. August 1996 9 ------- Milk is subject to stringent purity requirements to prevent the kind of contamination some consumer activists fear. Mastitis is usually fought with a class of drugs called beta-lactams, which includes penicillin. Each tanker load of milk is tested for beta-lactams before it is sold. Other drugs that are sometimes used to fight mastitis include tetracyclines, sulfonamides, animoglycocides and macrolids. The dairy industry, individual states and the FDA each test for these drugs periodically. It is illegal for a producer to put milk tainted with antibiotics into the bulk tank to be sold, and if drugs are detected in the milk, the producer faces significant economic and regulatory damages. For these reasons, use of antibiotics in the dairy industry has been minimized when possible. (Payne, personal communication 5/24/96). In addition to drug residue regulations, producers must also keep the somatic cell count (SCC) in the milk under federal and local standards. SCC rises dramatically during mastitis, providing another reason not to mix the milk of cows with mastitis with the milk to be sold. With both the drug residue regulations and the SCC standards, producers have incentives to catch mastitis early, isolate the cow and its milk, and manage the recovery carefully. Animal welfare concerns: Many animal rights activists object to many current farm practices and view rbST use as further manipulating the life of another creature. Animal welfare issues involve concern over the stress inflicted on the animal. When cows are under stress, they tend to produce less milk and use more energy for maintenance. Neither of these two indicators are apparent with rbST use even under poor management conditions. However, there is possible stress related to increases in mastitis and other health problems, and sometimes the cows develop permanent swellings up to 10 cm at the injection sites (Monsanto Posilac label, 1993). In Germany, veterinarians formally oppose the use of rbST as a violation of their code of ethics because of the these and other medical problems treated cows sometimes experience (Montague, 1996). Socio-economic issues: Some publicity has been given to the socio-economic impacts of rbST which might affect consumer opinion of rbST. Socio-economic concerns have focused on the impact of rbST on traditional small farms, different regions of the U.S., and total dairy income. Each of these concerns is examined in section 6.1 Economic Impact of rbST Adoption. Although there is some public concern over potential changes in the structure of the dairy industry, it should be noted that the U.S. federal government has never banned a new technology because of its socio-economic consequences (Executive Branch, 1994). 4.2 Impact on dairy product demand General consumer resistance to the use of rbST on dairy cows is strong. In many surveys before FDA approval of rbST, respondents expressed general concerns about biotechnology and rbST, support for labeling milk from rbST injected cows, and a significant number indicated that they would decrease their milk consumption when rbST was released on the market. Other surveys indicate that many consumers continue to believe the milk supply is safe despite rbST use and that education can have an impact on consumer opinions about safety issues (Executive Branch, 1994, p.43). Based on these surveys and examples of past biotechnology, the Executive Branch (1994) study concluded that they expect no significant reduction in demand for dairy products. Wisconsin Senator Russ Feingold (1994) issued a press release criticizing the study's conclusions. Senator Feingold stated that consumers do not want 10 August 1996 ------- dairy products produced with rbST and that it was likely that consumers would translate this concern into a decrease in demand. If those surveyed were to act on these intentions consumption of fluid milk would drop 4 to 20% (Executive Branch, 1994, p.43). Although the potential for rbST to decrease demand for dairy products does exist, it has not materialized. The Food and Agricultural Policy Research Institute (FAPRI) (1995) and Doane's Agricultural Report (1995) found that commercial rbST use has not created a decrease in dairy sales. This result seems to validate studies such as Dr. Thomas Hoban's rbST acceptance survey of consumers (1994), which indicated that rtST would Tiave no real impact on projected future consumption of milk" The survey also showed relatively little general interest in rbST's economic impacts on farmers. However, there was considerable interest expressed in additional independent scientific information. As more factual information about rbST was presented, acceptance increased. Hoban found the American Medical Association is considered a reliable source, but activist groups have very low credibility (Hoban, 1994). It seems that as more factual information about rbST becomes known, consumer concerns about its use diminishes. Monsanto research indicates that media coverage of rbST has dropped dramatically since January 1994, demonstrating that public interest in the issue may be declining (Barton, personal communication, 7/8/96) If significant consumer rejection of milk from rbST treated cows were to materialize, then the issue of labeling milk from rbST treated cows could have a significant impact on demand. The FDA.has determined that it has no legal basis to require labeling of milk from rbST treated cows. Because a practical test to distinguish between milk from rbST injected and non-injected cows has not yet been developed, labeling can be difficult to enforce. However, at this time the FDA permits the labeling of milk that does not come from rbST injected cows unless the label is misleading. The Dairy Coalition claims that an increasing number of brands are marketing milk labeled as from cows not treated with rbST (Weiss, personal communication, 7/22/96). Future labeling of milk from cows not treated with rbST may need to involve farm inspections or the development of practical testing to maintain consumer confidence. The existence of dairy products from non-rbST treated cows on the market may provide an option for consumers who would otherwise consider decreasing consumption (Executive Branch, 1994). In 1994, the Pure Food Campaign listed over 75 companies that stated they would not accept or sell milk from cows treated with rbST. Some farmers are asked by milk marketing organizations to sign letters stating that they do not use rbST. In areas where this type of pressure exists, many farmers do not want to discuss rbST, and rbST use can fall below the national average (Varner 3/6/96, personal communication, and Montague 1996). However, pressure on farmers not to use rbST has decreased gradually (Eastwood 3/7/96, personal communication). August 1996 11 ------- 5. Current and Future rbST Use 5.1 Current market penetration To understand rbST's impact on methane emissions we must estimate how many cows are treated with rt>ST. Monsanto has not released sales figures for 1995. However, several sources give clues to current rbST use in the U.S. Dairy Today magazine surveyed 400 producers in June and July of 1995 about rbST use on their farms: 20% had tried rbST and 12% continued to use rbST; 5% of producers who had not tried rbST planned to use it in the future; and another 7.5% were undecided. 70% of the respondents reported that they did not intend to use rbST in the future. However, Wisconsin was heavily sampled for this survey and is an area of especially strong opposition to rbST, so national adoption rates may be higher. Those that have used rt>ST report an average production level increase of 5.6 pounds per cow per day is required to recover the costs of using rbST. The average reported response to rbST was 10 pounds per cow per day. This indicates that rbST is an economically profitable technology for many farms. The survey also indicates that farm size is an important factor in rbST adoption. 11% of herds of 40 to 99 cows used rbST, but 33% of herds of over 100 cows used it. The percent of cows treated with rbST may be larger than the percent of producers using it, because adopting producers have a greater number of cows per farm (Dickrell, 1995). Hoard's Dairyman, a long-established dairy trade magazine based in Wisconsin, recently conducted an unpublished survey of their readers concerning rbST use on their farms. The survey asked farmers: "have you used bST during 1995?" Of 747 respondents nation wide, 160 answered yes, and were asked the following additional questions: • Give the "percent of herd you are treating." • "Was the increase in milk production less, about the same, or more than expected?" • "How many months did you use bST in 1995?" • "What are your plans for using bST in 1996?" The answers to these questions are presented in the tables below, divided by the region where the respondent lived: East, Central and West. About 34% of the 747 responding producers were from East, 49% from Central and 17% from the West. The size of each farm is not known, so these percentages do not necessarily reflect the number of cows represented from each area. For example, respondents from the West might represent larger farms and thus more cows than 17% of the total cows. Table 2 shows the answers to the question "have you used bST during 1995" by region. Of the respondents, it appears that rbST use is distributed evenly. The regions do not differ statistically in their answers. Of all respondents, 21.4% used rbST in 1995. 12 August 1996 ------- Table 2: Hoard's survey - rbST use in 1995 rbST use East Central West All Respondents YES 24.7% 19.6% 20.3% 21.4% NO 75.3% 80.4% 79.7% 78.6% Source: Hoard's Dairyman unpublished survey (1996). Of those farmers who used rbST, Table 3 shows what percent of their herd they were treating. This question may be unclear, because it can be interpreted to mean what percent is being treated at any one time. Because cows are only treated for 100 to 245 days out of every 12 to 14 months, a 100% adoption rate could mean treating only a third of the herd at any one time. Alternatively, producers who reported treatment of low percentages of their herd may also practice selective treatment of only certain cows. Table 3: Hoard's survey - percent of herd treated with rbST Percent treated East Central West All Respondents 1-24 32.7% 26.1% 40.0% 31.4% 25-49 26.9% 30.4% 15.0% 26.3% 50 or more 40.4% 43.5% 45.0% 42.4% Source: Hoard's Da ryman unpublished survey (1996). Table 4 illustrates whether the increase in milk production from rbST use was less, the same or more than farmers expected. Only 14% of all respondents answered that the production benefits were less than they expected. Most farmers achieved the response they were expecting. Of all respondents and in each region, producers achieved greater results than they expected more often than they achieved less than expected. It seems rbST has succeeded in producing the expected milk yield gains in the field. Table 4: Hoard's survey - rbST milk production response rbST response East Central West All Respondents Less than expected 14.3% 11.1% 20.8% 14.0% Same as expected 63.5% 73.0% 50.0% 65.3% More than expected 22.2% 15.9% 29.2% 20.7% Source: Hoard's Dairyman unpublished survey (1996). Table 5 shows the number of months during which farmers used rbST in 1995. Almost half of the responding farmers used rbST for 10-12 months of the year. Respondents who used rbST for less than 7 to 8 months may have just started or ended rbST treatments this year, or could be treating cows for shorter periods of time than the recommended treatment period. Table 6 presents the future plans of farmers who use rbST for 1996. 85.8% of respondents plan to continue using rbST on some level. August 1996 13 ------- Table 5: Hoard's Survey - months of rbST use Months East Central West All Respondents 0-3 18.8% 15.6% 25.9% 18.7% 4-6 18.8% 23.4% 29.6% 22.6% 7-9 10.9% 14.1% 7.4% 11.6% 10-12 51.6% 46.9% 37.0% 47.1% Source: Hoard's Dairyman unpublished survey (1996). Table 6: Hoard's survey - plans for 1996 rbST use rbST use plans East Central West All Respondents Use more rbST 11.3% 15.0% 15.4% 13.5% Use less rbST 11.3% 15.0% 0.0% 10.8% Use the same amount rbST 64.6% 58.3% 61.5% 61.5% Stop using rbST 12.9% 11.6% 23.1% 14.2% Data has been normalized to only include rbST users. Source: Hoard's Dairyman unpublished survey (1996). From this survey, it seems that 21% of the farmers used rbST on some portion of their herd in 1995, and most plan to continue to do so. Last year a similar survey by Hoard's Dairyman found 16.6% of respondents said they used rbST, indicating that rbST use is expanding. The survey information does not give the actual percentage of cows receiving full rbST treatments. However, these numbers seem to indicate that rbST use was lower than the USDA predicted adoption rate of 27% of all cows for FY95-96. The low adoption rates implied by the survey could have several explanations. The survey sampled only Hoard's Dairyman readers who were willing to respond to the survey, and may represent a subgroup that is less likely to adopt, such as smaller, more traditional farms. Adopting farms tend to be larger farms and represent a greater number of cows. Also, several factors could be contributing to lower adoption rates nation wide. Grain prices have been high, making it difficult for many farmers to cover input costs. In addition to the cost of the rbST itself, farmers must also consider the increase in feed required by higher producing cows. Farmers may lower rbST use and function with only the bare basics when feed costs are high. Adoption rates will probably increase when feed prices decline again. Some experts blame poor weather for low adoption rates (Montague, 1996 and Larson, personal communication 3/6/96). Heavy rain in some areas and extreme heat in others caused stress to cows that made farmers reluctant to use rbST. Also, early bad press for rbST may have had an initial impact on adoption that may fade as the public leams more about rbST. A survey in Wisconsin found that 80% of farmers who did not use rbST cited concerns about consumer reaction and possible herd health problems as reasons they do not adopt rbST. Also, 75% cited concern over the effect of higher milk production on milk prices (ATFFI, 1995). These concerns may have less influence in the future. Consumer reaction has not been manifested in reduced dairy sales so far, and improved management and veterinary care can control health problems. As individual farmers make the economic decision to use rbST and increase their milk production, reluctant farmers may need to adopt rbST to remain competitive. 14 August 1996 ------- Because rbST sales were not as high as some initial predictions anticipated, some analysts believe Monsanto might stop manufacturing rbST (Galloway, 1996). Although sales last year were estimated at over $100 million, the product is not breaking even yet. However, Monsanto claims steady increases in sales, number of cows receiving rbST and percentage of cows within herds (Messey, 1996). Monsanto believes that various changes in the dairy industry will improve future rbST sales. Over the next calendar year, Monsanto expects growth in sales of 25 to 40% (Montague, 1996). In 1995, the USDA and FAPRI made estimates of future rbST use. The USDA estimated 27% adoption for FY 95-96. FAPRI estimated 25% for 1995 and 35% in 1996. The OTA (1991) predicted that during rbST's first year of commercial availability, which was 1994, no more than 17% of farmers would use it. In summary, Dairy Today magazine's survey found 12% of their respondents continued to use rbST on some portion of their cows as of June 1995. Hoard's Dairyman found that 21% of their respondents used rbST on a portion of their cows in 1995. USDA and FAPRI predicted that 27% and 25% of cows respectively would be treated in 1995. Based on these surveys and estimates and on discussions with dairy specialists from many regions, rbST adoption rates probably fall within a low of 10% and high of 27% of all cows for 1995, and may gain speed as conditions and the industry itself change. 5.2 Predicted adoption rates through 2010 To determine rbST's long term impact on methane emissions, future adoption rates must be estimated. This subsection presents three adoption rate scenarios based on information from current trends discussed in subsection 5.1 and from various industry predictions. The OTA (1991) estimated rbST adoption regionally. Table 7 displays the OTA predictions for the first ten years of commercial rbST availability. After ten years, adoption ranges from 67% of herds in the Pacific to 31% of herds in the Corn Belt. Table 7: Regional Predicted Farmer Adoption Rates region first year fifth year tenth year available available available Appalachian 15% 32% 46% Corn Belt 13% 25% 31% Upper Midwest 15% 35% 46% Northeast 15% 31% 43% Pacific 17% 46% 67% Southeast 15% 29% 39% Southern Plain 13% 34% 42% Source: Ol 'A (1991) FAPRI (1995) predicted that 25% of cows would receive rbST treatment in 1995 and that adoption would increase to 35% in 1996, to 42.5% in 1997 and an additional 5% each year until 2000 when adoption increases 2.5% each year, reaching 65% adoption by 2004. The USDA (1995) predicted adoption rates are displayed in Table 8. August 1996 15 ------- Table 8: USDA and Executive Branch predicted rbST adoption rates market year adoption rate (cows) 1994-1995 17% 1995-1996 27% 1996-1997 30% 1997-1998 32% 1998-1999 34% 1999-2000 38% 2000-2001 43% 2001-2002 49% 2002-2003 55% 2003-2004 58% 2004-2005 60% 2005-2006 61% Source: USDA (1995) and Executive Branch (1994). Using the information available, a baseline and three adoption scenarios were created for this report. The baseline assumes no-rbST-use but does include a predicted 320 pound increase in milk per cow per year due to other industry advances, and is used for comparison with the adoption scenarios. The Low scenario is a low adoption rate scenario, starting with 10% of U.S. cows treated with rbST in 1995 and reaching 40% adoption by 2010. The High scenario is based on the USDA (1995) predicted rates and is considered the high adoption rate scenario. The Full scenario assumes that adoption reaches 95% by the year 2000. The Full scenario is used to analyze rbST's full potential. The yearly adoption rates for the three scenarios are presented in Table 9. These scenarios are used extensively in section 6. Table 9: Assumed mature cow rbST treatment under various adoption scenarios Year Low adoption High adoption Full adoption 1994 5% 17% 20% 1995 10% 27% 30% 1996 12% 30% 45% 1997 14% 34% 60% 1998 16% 36% 75% 1999 18% 39% 85% 2000 20% 45% 95% 2001 22% 51% 95% 2002 24% 58% 95% 2003 26% 62% 95% 2004 28% 65% 95% 2005 30% 67% 95% 2006 32% 69% 95% 2007 34% 70% 95% 2008 36% 72% 95% 2009 38% 74% 95% 2010 40% 75% 95% 16 August 1996 ------- 6. Potential Consequences of Future rbST Use This section describes the impact of predicted adoption rates on the dairy industry economy and on methane emissions. Subsection 6.1 examines the economic impact of rbST on several aspects of the dairy industry. Subsection 6.2 introduces the model used in this report to calculate the methane emission impacts of rbST use. First the baseline results from the model are presented, including cow population and methane emission levels. Then the results for each rbST adoption scenario are presented. Subsection 6.3 compares the baseline emissions to the three rbST adoption scenarios. These results are also compared to other studies of rbST's impact on methane emissions. 6.1 Economic impact of rbST use on the dairy industry The use of rbST may influence several aspects of the dairy industry through market forces. Farm size, regional advantages and total dairy income are discussed in turn below. Size of farm: There is concern that rbST use will damage the chances for survival of small farms in the United States. Many people have called rbST a "size neutral" technology in that it does not involve an economy of scale: there is no capital investment, the product can be purchased in small amounts, and it is relatively inexpensive. Although good management and financial records to judge returns are necessary for obtaining the best results, these farm characteristics do not necessarily depend on size. However, the use of rbST on a national level will not affect each farm equally because of differences in who adopts it, what response they achieve from it, and how the ultimate economic consequences will affect them. The Agricultural Technology and Family Farm Institute (ATFFI) (1995) surveys of Wisconsin farmers characterized farmers who adopt rbST. They are generally younger farmers with more years of education. They have larger herds and are more likely to use intensive management practices that are associated with the best returns from rbST use. Larger, more progressive farms are adopting rbST at a higher rate than many smaller farms (Executive Branch 1994 and Dickson 3/6/96, personal communication). In the long run, this could give adopting farmers a competitive advantage that might pressure non-adapters to increase herd size, adopt intensive management practices, and use rbST (ATFFI, 1995). Because rbST reduces production costs, its widespread adoption could put downward pressure on farm milk prices. It is commonly believed that smaller farms are less competitive or will not have the resources to withstand a potential drop in prices and that many will be pushed out of business. Based on the premise that smaller farms are less cost competitive, rbST could accelerate the current trend toward fewer, larger farms. Regional impacts: National use of rbST may affect regional dairy producers in different ways. Regions that adopt more slowly will be put at a competitive disadvantage to successful adopters. Technology adoption in the dairy industry has historically been most rapid in the Pacific region and much slower in traditional production regions such as the Com Belt and Upper Midwest (OTA, 1991). However, the downward pressure on milk prices that is likely to result from rbST use may be especially hard on those dairy farms with a higher proportion of farm income generated solely from milk production. This could result in some decrease in cow and farm numbers in the Pacific, Southeast and Southern Plains. Despite these possible regional impacts, rbST use is not expected to have a large effect on where milk will be produced in the U.S. (Failed et al., 1987). August 1996 17 ------- Dairy income: In a study by the Executive Branch (1994, p.30), which assumed a continuation of the current price support system, the estimated rbST use would result in a total drop in net farm income of $1.3 billion by the year 2000, with the annual reduction growing to $546 million. This represents a 1% decline in income for dairy farmers over the next several years, which could have a significant impact on rural communities (Feingold, 1994). This drop was calculated before the enactment of the 1996 Farm Bill, which outlines the gradual removal of dairy price supports. The price of milk may fall even further without the support of government purchases. Although widespread adoption of rbST is expected to put downward pressure on farm milk prices, the market price will be affected by the overall demand and supply conditions. Also, it should be emphasized that rbST use puts downward pressure on prices because it reduces production costs. As a consequence, rbST has the potential to improve the profitability of those dairy farms that remain in the industry while simultaneously reducing consumer prices. 6.2 Methane emissions model and results As a productivity enhancing agent, rbST has the potential to contribute toward reaching the methane emission reduction goals of the Climate Change Action Plan. To determine the effect of rbST adoption on methane emissions, the dynamics of milk production in the U.S. must be considered. To accomplish this, a model of total milk production, cow population, productivity and methane emissions was created. The following procedure was used to determine methane emissions from the U.S. dairy herd for the baseline and three adoption scenarios: Step 1. Estimate milk production and consumption: USDA data are used to estimate total milk production from 1990 through 2005 (USDA, 1995 and 1996). Population forecasts and USDA (1995) forecasted utilization levels are used to predict yearly milk production from 2005 through 2010. Imports and exports are assumed to remain at a constant level. Step 2. Estimate per cow milk production: Even without the use of rbST, per cow production is expected to rise because of other technical and management improvements. This improvement rate is assumed to add 320 pounds of milk to the production rate per cow each year in the baseline scenario. This estimate is based on historical tends in cow productivity (Blaney et al., 1995). In the adoption scenarios, rbST is assumed to improve a cow's annual milk production by 1,800 pounds. Thus, non- treated cows are given baseline values for milk production and treated cows are given an additional 1,800 pounds per year. The proportion of rbST-treated cows is determined by the adoption scenarios outlined in subsection 5.2. Step 3. Estimate animal population: Dairy cow populations are obtained by dividing the total milk production by milk production per cow (calculated in step 2). Estimates of replacement populations are based on the mature cow populations. Replacement populations are the animals being raised to replace mature milk producing cows that are culled from the herd. The replacement population is directly tied to the mature cow population numbers. The dairy herd numbers are estimated for three groups: mature cows, 0-12 month old replacements, and 12-24 month old replacements. 18 August 1996 ------- Step 4. Estimate emissions: Methane emission factors (kg per head per year) for mature cows were developed using a simplified model of energy intake required to support maintenance, milk production, growth and pregnancy (IPCC, 1994). The model was calibrated to match the results of a detailed emissions modeling conducted for EPA (1993). Using this model, the implications of increased milk production per cow are reflected in increased feed intake requirements and hence increased emissions per head. Emissions factors for the replacement population were held constant based on the estimates in EPA (1993). Total emissions were estimated by multiplying the emissions factors by the relevant populations for each year, and summing. Baseline emissions: The model described above was used to estimate a baseline scenario, including the cow population, milk production per cow, and methane emissions for 1990 through 2010. Table 10 shows the baseline estimates. As shown in the table, the population of mature cows is projected to decline from about 10 million in 1990 to 8.8 million in 2010. This decline in cows numbers occurs despite an estimated 25 percent increase in total milk production: milk production per cow increases by over 40 percent over the same period. As a result of the increase in milk production per cow, methane emissions per cow per year also increase, from 115 kg/yr to 142 kg/yr in 2010. Total methane emissions are relatively flat from 1990 through 2000, and are estimated to increase by about 0.06 Tg by 2010. Table 10: Baseline Methane Emissions - No rbST Adoption year mature cow population average milk kg/cow/day emissions factor kg CIVcow/yr Mature cow CH4 emissions Tg/yr replacement population replacement CH4 emissions Tg/yr total ChU emissions Tg/yr 1990 10,008,000 18.3 115 1.15 8,406,000 0.33 1.48 1991 9,883,000 18.6 116 1.14 8,301,000 0.33 1.47 1992 9,714,000 19.3 118 1.15 8,160.000 0.32 1.47 1993 9,679,000 19.3 119 1.15 8,130,000 0.32 1.47 1994 9,677,000 19.7 120 1.16 8,129,000 0.32 1.48 1995 9.608.000 20.1 121 1.16 8,071,000 0.32 1.48 1996 9,480,000 20.5 123 1.16 7,963.000 0.31 1.47 1997 9,330,000 20.9 124 1.16 7,837,000 0.31 1.46 1998 9,272,000 21.3 125 1.16 7,789,000 0.31 1.47 1999 9,240,000 21.7 127 1.17 7.761,000 0.30 1.48 2000 9,186,000 22.1 128 1.18 7,716,000 0.30 1.48 2001 9,123,000 22.5 130 1.18 7,663,000 0.30 1.48 2002 9,084,000 22.9 131 1.19 7,631,000 0.30 1.49 2003 9,052,000 23.3 132 1.20 7,604,000 0.30 1.50 2004 9,015,000 23.7 134 1.21 7,573,000 0.30 1.50 2005 8,985,000 24.1 135 1.22 7,548,000 0.30 1.51 2006 8,939,000 24.5 137 1.22 7,509,000 0.29 1.52 2007 8,902,000 24.9 138 1.23 7,478,000 0.29 1.52 2008 8,867,000 25.3 140 1.24 7,448,000 0.29 1.53 2009 8,833,000 25.7 141 1.24 7,419,000 0.29 1.54 2010 8,800,000 26.1 142 1.25 7,392,000 0.29 1.54 This baseline differs in several key aspects from the baseline presented in EPA (1993), Anthropogenic Methane Emissions in the United States: Estimates for 1990. Report to Congress. Most importantly, the baseline scenario presented in Table 10 includes the continuation of past trends in milk production per cow. The baseline in EPA (1993) was developed assuming the milk production practices remain unchanged from conditions in 1990. Also, the scenarios of total milk production over time were August 1996 19 ------- developed differently. In both studies, the estimates of total milk production are used to estimate the size of the cow herd overtime. Consequently, the differences in the milk production estimates contribute to the differences in the emissions estimates. Table 11 summarizes the differences in the baseline estimates. Table 11; Baseline Comparison with EPA (1993) year milk production per cow (kg/cow/day) total US milk production (million metric tons) total ch4 emissions (Tg methane per year) EPA 11993) This studv EPA (1993) This studv EPA (1993) This studv8 1990 18.0 18.3 66.5 67.0 1.5 1.5 2000 18.0 22.1 74.4-75.7 74.2 1.7 1.5 2010 18.0 26.1 78.6-88.0 83.8 1.8-2.0 1.5 Values are rounded to match EPA 1993 precision levels. In this study's baseline, methane emissions rise from 1.48 to 1.54 from 1990 to 2010. Low scenario emissions: Low scenario is a low rbST adoption rate scenario. Adoption only reaches 40% by 2010. The model described above was used to calculate dairy herd characteristics through 2010. The resulting milk production per mature cow, mature cow and replacement populations, emission factors and total methane emissions are presented in Table 12. Table 12: rbST adoption and methane production Low scenario year rbST adoption mature cow population average milk kg/cow/day emissions factor kg ChVcow/yr Mature cow CH4 emissions Tg/yr replacement population replacement CH4 emissions Tg/yr total CH4 emissions Tg/yr 1990 0% 10,008,000 18.3 115 1.15 8,406,000 0.33 1.48 1991 0% 9,883,000 18.6 116 1.14 8,301,000 0.33 1.47 1992 0% 9,714,000 19.3 118 1.15 8,160,000 0.32 1.47 1993 0% 9,679,000 19.3 119 1.15 8,130,000 0.32 1.47 1994 5% 9,623,000 19.8 120 1.16 8,083,000 0.32 1.47 1995 10% 9,503,000 20.4 122 1.16 7,982,000 0.31 1.47 1996 12% 9,358,000 20.8 124 1.16 7,861,000 0.31 1.46 1997 14% 9,192,000 21.2 125 1.15 7,721,000 0.30 1.45 1998 16% 9,119,000 21.7 127 1.15 7,660,000 0.30 1.45 1999 18% 9,072,000 22.1 128 1.16 7,620,000 0.30 1.46 2000 20% 9,004,000 22.6 130 1.17 7,563,000 0.30 1.47 2001 22% 8,928,000 23.0 131 1.17 7,500,000 0.29 1.47 2002 24% 8,876,000 23.5 133 1.18 7,456,000 0.29 1.47 2003 26% 8,831,000 23.9 135 1.19 7,418,000 0.29 1.48 2004 28% 8,783,000 24.3 136 1.20 7,378,000 0.29 1.48 2005 30% 8,742,000 24.8 138 1.20 7,343,000 0.29 1.49 2006 32% 8,686,000 25.2 139 1.21 7,296,000 0.29 1.50 2007 34% 8,638,000 25.7 141 1.22 7,256,000 0.28 1.50 2008 36% 8,593,000 26.1 142 1.22 7,218,000 0.28 1.51 2009 38% 8,550,000 26.5 144 1.23 7,182,000 0.28 1.51 2010 40% 8,508,000 27.0 146 1.24 7,147,000 0.28 1.52 High scenario emissions: High scenario is a high rbST adoption rate scenario based on the USDA predicted adoption rates for rbST. The model described above was used to calculate the milk production, mature cow and replacement populations, emission factors, and total methane emissions, which are presented in Table 13. 20 August 1996 ------- Full scenario emissions: Full scenario is a very high adoption rate scenario designed to evaluate rbSTs full potential impact on methane. In Full scenario, adoption reaches 95% by the year 2000. The resulting milk production, mature cow and replacement populations, emission factors, and total methane emissions calculated from the model are presented in Table 14. Table 13: rbST adoption and methane production High scenario year rbST adoption mature cow population average milk kg/cow/day emissions factor kg CH4/cow/yr Mature cow CH4 emissions Tg/yr replacement population replacement CH4 emissions Tg/yr total CH4 emissions Tg/yr 1990 0% 10,008,000 18.3 115 1.15 8,406,000 0.33 1.48 1991 0% 9,883,000 18.6 116 1.14 8,301,000 0.33 1.47 1992 0% 9,714,000 19.3 118 1.15 8,160,000 0.32 1.47 1993 0% 9,679,000 19.3 119 1.15 8,130,000 0.32 1.47 1994 17% 9,494,000 20.1 121 1.15 7,975,000 0.31 1.48 1995 27% 9,328,000 20.7 123 1.15 7,836,000 0.31 1.46 1996 30% 9,180,000 21.2 125 1.15 7,711,000 0.30 1.45 1997 34% 9,003,000 21.7 127 1.14 7,562,000 0.30 1.44 1998 36% 8,935,000 22.1 128 1.15 7,505,000 0.29 1.44 1999 39% 8,883,000 22.6 130 1.15 7,462,000 0.29 1.45 2000 45% 8,786,000 23.1 132 1.16 7,380,000 0.29 1.45 2001 51% 8,683,000 23.7 134 1.16 7,294,000 0.29 1.45 2002 58% 8,597,000 24.2 136 1.17 7,222,000 0.28 1.45 2003 62% 8,544,000 24.7 137 1.17 7,177,000 0.28 1.46 2004 65% 8,494,000 25.2 139 1.18 7,135,000 0.28 1.46 2005 67% 8,459,000 25.6 141 1.19 7,105,000 0.28 1.47 2006 69% 8,410,000 26.1 142 1.20 7,064,000 0.28 1.47 2007 70% 8,376,000 26.5 144 1.20 7.035,000 0.28 1.48 2008 72% 8,336,000 26.9 145 1.21 7,002,000 0.27 1.48 2009 74% 8,298,000 27.4 147 1.22 6,971,000 0.27 1.49 2010 75% 8,269,000 27.8 148 1.23 6,946,000 0.27 1.50 6.3 Impact of rbST on dairy industry methane emissions through 2010 The scenarios and the baseline (no-rbST-use) developed in subsection 6.2 provide insights into how methane production by dairy herds in the U.S. responds to rbST use. It is likely that current rbST use is between 10 and 27% of cows, or in the range of adoption scenarios 1 and 2. While production per cow is variable from year to year, depending on other technologies, weather, feed prices and other changes, for this analysis it is assumed that, in the absence of rbST, industry development would have followed the baseline. Comparing baseline herd characteristics and methane emissions to Low and High scenarios for 1995 demonstrate that rbST may already be influencing the dairy industry. USDA (1996) estimates a mature cow population and milk per cow production level for 1995 between those calculated for this report's Low and High scenarios. The USDA values are closer to those predicted for the Low scenario. Assuming that rbST is responsible for any Increased production over the baseline levels predicted in this report for 1995, then the improvement corresponds with that expected with a Low scenario adoption or higher. Such an adoption level would indicate that rbST use has reduced methane emissions from dairy cattle by 0.01 to 0.02 Tg this year (see Table 15). August 1996 21 ------- Table 14: rbST adoption and methane production Full scenario year rbST adoption mature cow population average milk kg/cow/day emissions factor kg CIVcow/yr Mature cow CH4 emissions Tg/yr replacement population replacement CH4 emissions Tg/yr total CH4 emissions Tg/yr 1990 0% 10,008,000 18.3 115 1.15 8,406,000 0.33 1.48 1991 0% 9,883,000 18.6 116 1.14 8,301,000 0.33 1.47 1992 0% 9,714,000 19.3 118 1.15 8,160,000 0.32 1.47 1993 0% 9,679,000 19.3 119 1.15 8,130,000 0.32 1.47 1994 20% 9,463,000 20.2 121 1.15 7,949,000 0.31 1.46 1995 30% 9,298,000 20.8 124 1.15 7,811,000 0.31 1.45 1996 45% 9,037,000 21.5 126 1.14 7,591,000 0.30 1.44 1997 60% 8,767,000 22.3 129 1.13 7,365,000 0.29 1.42 1998 75% 8,596,000 23.0 131 1.13 7,221,000 0.28 1.41 1999 85% 8,496,000 23.6 134 1.13 7,137,000 0.28 1.41 2000 95% 8,381,000 24.2 136 1.14 7,040,000 0.28 1.41 2001 95% 8,336,000 24.6 137 1.14 7,003,000 0.27 1.42 2002 95% 8,313,000 25.0 139 1.15 6,983,000 0.27 1.43 2003 95% 8,296,000 25.4 140 1.16 6,968,000 0.27 1.43 2004 95% 8,274,000 25.8 141 1.17 6,950,000 0.27 1.44 2005 95% 8,257,000 26.2 143 1.18 6,936,000 0.27 1.45 2006 95% 8,226,000 26.6 144 1.19 6,910,000 0.27 1.46 2007 95% 8,202,000 27.0 146 1.20 6,890,000 0.27 1.47 2008 95% 8,180,000 27.4 147 1.20 6,871,000 0.27 1.47 2009 95% 8,158,000 27.8 149 1.21 6,853,000 0.27 1.48 2010 95% 8,137,000 28.2 150 1.22 6,835,000 0.27 1.49 Table 15: 1995 dairy methane emissions rbST adoption mature cow population average milk kg/cow/day emissions factor kg CH4/cow/yr total cm emissions Tg/yr baseline 0% 9,608,000 20.1 121 1.48 low scenario 10% 9,503,000 20.4 122 1.47 high scenario 27% 9,328,000 20.7 123 1.46 The scenarios modeled in subsection 6.2 illustrate future trends in the dairy herd and methane emissions as they are influenced by rbST adoption rates. With rbST adoption, milk production per cow increases from our baseline. As a result, fewer cows are required to produce the milk needed for projected consumption. This is illustrated in Figure 2. Although cow numbers are decreasing, the amount of feed each cow requires increases with rbST use and other baseline production increases to support higher production levels. This increase in energy intake causes the emissions factor (kg methane emitted per cow per year) to increase, as illustrated in Figure 3. This is because a portion of the energy consumed by a cow is converted to methane in the rumen. 22 August 1996 ------- Figure 2: mature cow populations through 2010 Figure 3: emissions factors through 2010 Because the part of a cow's energy intake allocated to maintenance is unchanged by rbST, the maintenance energy requirement makes up a smaller proportion of total energy intake. This is called maintenance dilution. Even though each cow requires more energy, the total energy needed to produce a unit of milk goes down because the maintenance energy requirement is spread out over more units of milk. It is a proportion of the total energy intake that is converted to methane, so methane emissions are also decreasing per unit of milk, even though emissions per cow are increasing. In this way, rbST increases milk production per cow faster than methane production per cow. This means methane August 1996 23 ------- emissions per unit of milk produced will decline. The resulting decreases in methane emissions per unit of milk is illustrated for the baseline and three rbST adoption scenarios in Figure 4. Figure 4: kg methane emitted per kg milk produced 0.023 0.017 0.016 - baseline -Low scenario -High scenario -Full scenario 86.4% 81.8% 77.3% 72.3% 0.015 1990 1995 2000 2005 2010 Because emissions per unit of milk production decline with rbST use, methane emissions will decline compared to baseline emissions. Cow numbers will drop faster than methane per cow will rise. The resulting methane production levels for the three scenarios compared to the baseline are shown in Figure 5. • Low scenario is the low adoption rate scenario. In 2010, Low scenario has a methane emissions rate of 1.52 Tg methane per year, 1.3% lower than baseline emission levels for that year. • High scenario, the high adoption rate scenario, has an emissions rate of 1.50 in 2010. This is 2.6% less than baseline emission rates in 2010. • Full scenario, the very high adoption rate, demonstrates rbST's potential. Emission rates for 2010 are 1.49, 3.2% lower than baseline rates. The Climate Change Action Plan sets a goal of a 1.8 MMTCE reduction in methane emissions from all U.S. ruminant livestock compared to baseline calculations by the year 2000. The use of rbST between the Low scenario and High scenario adoption levels could reduce year 2000 methane emissions by 3% to 10% of the Climate Change Action Plan goals for ruminants. However, it should be noted that CCAP goals are based on a current practices baseline, not the baseline used in this report. This report's estimates can be compared to other published studies. EPA (1993b) examines the effects of rbST use on methane emissions. EPA (1993b) provides two rbST adoption scenarios: in the low adoption scenario, adoption reaches 45% in 2000 and 75% in 2010. In their high adoption scenario, adoption reaches 70% in 2000 and 95% in 2010. The resulting reductions in methane emissions per unit 24 August 1996 ------- product are given below for High scenario and Full scenario from our model and the EPA scenarios in Table 16. The reductions estimates of the two models are close. Figure 5: methane emissions through 2010 1.60 1.50 1.40 0 e 1 | 1.30 K 1.20 1.10 1.00 I 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1990 1995 2000 2005 2010 Table 16: EPA 993b report rbSI impact compared to this report year This study EPA 1993b This study EPA 1993b high scenario low scenario Full scenario high scenario pounds of milk/cow/yr 1800 1350 1800 1710 increase due to rbST adoption 2000 45% 45% 95% 70% 2010 75% 75% 95% 95% reduction in 2000 0.03 0.03-0.05 0.07 0.04-0.07 CKU emissions due to rbST (Tg/yr) 2010 0.04 0.05-0.06 0.05 0.06-0.08 Two other studies have examined the potential impact of rbST on dairy methane emissions. These studies assumed that the adoption rate was 100% and national milk production remained at a constant level to simplify the models. Hartnell (1991) estimated that, with 100% adoption at 1989 total milk production levels of 66 billion kg, there would be a 0.16 Tg reduction of methane produced by dairy cows in the U.S. Johnson et a/. (1992) also used 100% adoption at 65.4 billion kg total milk production assumptions, and estimated a 0.13 Tg decrease in the methane produced by the dairy cows and their replacements in the U.S. A close look at each model compared to the model used in this report is warranted. baseline Low scenario High scenario A—Full scenario August 1996 25 ------- Hartnell's estimate of a 0.16 Tg decrease in methane caused by rbST adoption includes estimates for methane from manure degradation in addition to that emitted directly from the dairy herd. Our model only considers emissions from the herd itself, not the manure. In his model, Hartnell assumed that all manure was fermented anaerobically, producing large amounts of methane. In reality, many farmers spread manure on fields where it can degrade aerobically without methane production. If the manure component of Hartnell's model is removed, predicted methane emissions reductions are 0.05 Tg. We can compare his model to our model by running Hartnell's 1989 scenario conditions through our model. Hartnell assumed 66 billion kg total milk production for 1989 and a production rate of 18.1 kg per cow per day. Our model predicts that with 100% adoption under these conditions, assuming total milk production is constant, there would be a 0.09 Tg reduction in methane emissions. This can be broken down into roughly 0.05 Tg reduction from mature cows and 0.04 Tg reduction from the replacements. Hartnell's model does not include estimates for the replacement population. When the estimates are adjusted for Hartnell's inclusion of manure and this report's inclusion of replacement population, they are very close. Both estimate a 0.05 Tg reduction from mature cow emissions if rbST had been adopted 100% in 1989. Johnson's model uses a similar starting point, with a slightly lower production level and slightly higher cow population for 1989. As in this report, Johnson does not calculate the impact of methane from manure and does calculate changes from the replacement population. However, Johnson estimates a 0.13 Tg methane emissions reduction, compared to 0.09 Tg estimated in this report for the same conditions. The difference is mainly the result of different mature cow emission factors. In our model, 100% adoption of rbST in 1989 increases the emission factor (kg methane emitted per cow per year) by 7%. This is because the cow must eat more to sustain the higher milk production levels. In Hartnell's model this emissions factor increased by 6.6%. But in Johnson's model this emissions factor only increases by 4%. Thus, in the Johnson model, there is not as much increase in emissions per cow to counter act the decline in cow population and so methane emissions fall more quickly. Differences in emissions factors arise because of different assumptions about feeds, energy requirements, and methane conversion factors. Most dairy methane emissions studies involving bST, including the one used in this report, assume that the total quantity of milk produced will not change as a result of using bST. This is possible because of the relatively inelastic demand for dairy products (it requires large shifts in price to produce small shifts in quantity demanded). Although milk demand is relatively inelastic, if bST causes a large enough price reduction, the increase in quantity produced could negate some of the methane emission reducing potential of bST. To estimate the extent to which increases in production could negate the emissions reductions from bST use, the potential impact of the shift in the supply curve for milk was assessed. This assessment was performed using the following assumptions: • bST causes the supply curve for milk to shift downward, so that the cost of supply is reduced at any given quantity of milk produced (see Figure 6). • The extent of the shift in the supply curve can be calculated based on the reduction in the cost of producing milk. To estimate the maximum potential supply shift (which would tend to over-estimate the extent to which increases in production will negate the emissions reductions), it was assumed that only increased feed costs are incurred as the result of the increase in milk production per cow. 26 August 1996 ------- The cost of bST itself, as well as potential marginal costs associated with equipment, labor and buildings are ignored. Based on this assumption, the shift in the supply curve was estimated as the change in the average cost per unit of milk produced resulting from the change in the average milk production per cow. • The supply curve is assumed to have a price elasticity of 0.5 (Miller, personal communication, 7/3/96). • The demand curve is assumed to have a price elasticity of -0.15 (making it relatively inelastic) (Miller, personal communication, 7/3/96). Using these assumptions, the shift in the supply curve causes the quantity of milk produced to increase slightly, as illustrated in Figure 6. The extent of the shift in the supply curve depends on the extent of bST adoption. However, overall the increase in production due to reduced production costs is found to negate only about 10 to 20% of the emissions reduction estimated in this report, which assumes no change in production. Figure 7 summarizes how including this supply curve shift influences the emissions reductions estimated in the year 2000 for the three adoption scenarios analyzed in this report. As shown in the table, the emissions reduction estimates are about 10 to 20% smaller when the supply curve shift is considered. The methane emission estimates presented in this report only encompass the effects of rbST-induced changes in productivity, diet and emission factors. Other considerations include manure methane production, the use of dairy animals for beef, and the greenhouse gas emissions from manufacture of Posilac (Monsanto's rbST product). Johnson et at. (1992) determined that, considering the decrease in animal population and increase in feed intake, total manure production would decrease by 10% if rbST was adopted 100% at 1989 production August 1996 27 ------- Figure 7: comparison of emissions reduction in 2000 with and without the supply curve shift 0.08 Low High bST Adoption Scenario Full levels. Microbial decomposition of manure results in methane production under anaerobic conditions, such as lagoons or holding ponds used on some farms. Thus a decrease in manure production could decrease manure methane emissions. However, rbST may encourage the trend toward larger farms, which are more likely to hold manure in methane-producing anaerobic conditions. Methane recovery projects such as the EPA program AgSTAR provide promising opportunities to use manure methane emissions from many medium to large farms for electricity production. This can be profitable and also prevents the methane from being released into the environment. As programs such as AgSTAR grow, it appears that rbST use will help reduce methane from manure by reducing total manure production, despite the trend toward larger farms. The impact of a smaller dairy herd on the beef industry should also be considered. Currently about 25% of the U.S. beef supply comes from salvage dairy animals and male calves. In the Johnson et al. (1992) 100% adoption and production scenario, this supply source would decrease by 11%. To compensate for this loss the U.S. beef herd would need to increase by 3%. This is an extreme case scenario, but it should be noted that the substantial decrease in the dairy herd may be accompanied by some increase in the beef herd. In addition to manure and beef industry methane production, the manufacture of Posilac should also be examined for methane emissions. Unlike the microbial population in the rumen, the organism which actually produces recombinant rbST, E. Coli, is not capable of methanogenesis, so no significant amount of methane is produced in the production of Posilac. The treatment of waste from Posilac production involves a system that is essentially aerobic, converting waste to carbon dioxide and solid fertilizer. However, it is possible that some of this waste is converted to methane. In the highly unlikely scenario where all of the carbonaceous waste is converted to methane, this would produce 1.400 x 103 metric tons (0.0014 Tg) of methane (Irwin, 1992). 28 August 1996 ------- In conclusion, rbST has the potential to contribute to the Climate Change Action Plan goals to reduce methane emissions under Action #39. The Climate Change Action Plan sets a goal of a 1.8 MMTCE reduction in methane emissions from all U.S. ruminant livestock compared to baseline calculations by the year 2000. Future rtST adoption rates and the resulting impact on dairy methane emissions will most likely fall between the Low scenario and High scenario used in this report's model. A summary of some dairy herd characteristics by 2000 and 2010 is presented in Table 17 comparing these scenarios to the baseline. Adoption rates between Low scenario and High scenario would represent a decrease of 0.01 to 0.03 Tg methane (0.06 to 0.18 MMTCE) from this study's baseline of no rbST use by 2000. This would reduce methane emissions by 3.3% to 10% of the Climate Change Action Plan 1.8 MMTCE emissions reduction goal for ruminants in the year 2000. Table 17: rbST adoption and methane emissions summary for the years 2000 and 2010 scenario year rbST adoption milk production (kg/cow/day) mature cow population methane Tg per year 1990 0% 18.3 10,008,000 1.48 baseline 2000 0% 22.1 9,186,000 1.48 2010 0% 26.1 8,800,000 1.54 1990 0% 18.3 10,008,000 1.48 low scenario 2000 20% 22.6 9,004,000 1.47 2010 40% 27.0 8,508,000 1.52 1990 0% 18.3 10,008,000 1.48 high scenario 2000 45% 23.1 8,786,000 1.45 2010 75% 27.8 8,269,000 1.50 August 1996 29 ------- SOURCES ATFFI (1995), "Use of bST in America's Dairyland" Agricultural Technology and Family Farm Institute (ATFFI) Fact Sheet No. 3, March 1995, University of Wisconsin, Madison. Barton, G., Monsanto Biotechnology Communications, personal communication, 7/8/96. Bauman, D., E. (1992), "Bovine Somatotropin: Review of an Emerging Animal Technology," Journal of Dairy Science 75:3432-3451. Berry, S., Professor at University of California, Davis, Department of Animal Science, personal communication 3/4/96. Blaney, D., J. Miller, R. Stillman (1995), "Dairy Background for 1995 Farm Legislation," USDA Agricultural Economic Report Number 705. Dickrell, J. (1995), "BST Use Near its Peak?" Dairy Today, October, 1995. Dickson, D., Professor at University of Wisconsin, Madison, Dairy Sciences, personal communication 3/6/96. Doane's Agricultural Report (1995), "BST Update," Doane's Agricultural Report, 11/17/95, Vol. 58 No. 46-5. Eastwood, B., Cooperative State Research Education and Extension Services, USDA, Washington DC, personal communication 3/7/96. EPA (1993), Anthropogenic Methane Emissions in the United States: Estimates for 1990, Report to Congress, EPA 430-R-93-003. EPA (1993b), Opportunities to Reduce Anthropogenic Methane Emissions in the United States, Report to Congress, EPA 430-R-93-012. EPA (1995), Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990 - 1994, EPA-230-R-96-006. Epstein, S. (1996), "Unlabeled Milk from Cows Treated with Biosynthetic Growth Hormones: a Case of Regulatory Abdication," International Journal of Health Services, Vol. 26, no.1, pp.173-185. Executive Branch, U.S. Government (1994), Use of Bovine Somatotropin (BST) in the United States: Its Potential Effects, Executive Branch, U.S. Government, January, 1994. August 1996 31 ------- Fallert, R., T. McGuckin, C. Betts, and G. Bruner (1987)," bST and the Dairy Industry: A National, Regional and Farm-Level Analysis," Agricultural Economic Report Number 579, Economic Research Service, U.S. Department of Agriculture, Washington, DC. FAPRI (1995), FAPR11995 U.S. Agricultural Outlook, Food and Agricultural Policy Research Institute, Iowa State University and University of Missouri. FDA (1994), FDA Consumer January-February 1994. Feingold, R. (January 5,1994), "Feingold: Administration's Draft Report Confirms the Devastating Impact of BGH on the Federal Deficit and Dairy Industry," Press release from U.S. Senator Russ Feingold. Galloway, J. (1996), "Monsanto Losing Money with rBGH," Wisconsin State Journal, Money section, 4/14/96, Madison Wisconsin. Glausiusz, F. (1995), "Flap Over Milk" Discover, January 1995, p.96. Hard, D., Director of Technical Services, Protiva, Monsanto, personal communication, 7/9/96. Hartnell, G. (1991), "Potential Effects of Posilac Approval in the United States on Emissions of Carbon Dioxide and Methane from Dairy Cattle and Their Manure," Environmental Assessment-Sometribove- Zinc Complex, The Agricultural Group of Monsanto Company. Hoard's Dairyman (1996), unpublished survey on readership bST use. Hoban, T. (1994), "Consumer Awareness and Acceptance of Bovine Somatotropin (bST)," A survey for the Grocery Manufacturers of America. IPCC (1994), "Greenhouse Gas Inventory Reference Manual," IPCC Draft Guidelines of National Greenhouse Gas Inventories, IPCC/OECD Joint Programme, 1994. Iowa State University - University Outreach Program (December 1993), "Bovine Somatotropin (bST)," Biotechnology Information Series (Bio-3), North Central Regional Extension Publication. Irwin, G., H. (1992), "Emissions of the Greenhouse Gasses Carbon Dioxide and Methane Resulting from Manufacture and Transport of Posilac," Environmental Assessment-Sometribove-Zinc Complex, The Agricultural Group of Monsanto Company. Johnson, Donald E., Gerald M. Ward, and Joan Torrent (1992), "The Environmental Impact of Bovine Somatotropin Use in Dairy Cattle," Journal of Environmental Quality, vol.21, no.2, April-June, 1992. (7.1 Animal Economics and Management Practices). Juskevich, J., and G. Guyer (1990), "Bovine Growth Hormone: Human Food Safety Evaluation," Science. August 1990, 249:875-884. 32 August 1996 ------- Larson, S., Hoard's Dairyman Senior Associate Editor, personal communication, 3/6/96. Messey, J. (1996), "Report: rBST not breaking even," The Country Today, p. D1, 2/21/96. Miller, J., USDA dairy economist, personal communication, 7/3/96. Monsanto Agricultural Company (1993), Posilac label, St. Louis, MO. Montague, P. (1996), "Is BGH in Trouble?" Rachel's Environment and Health Weekly #483 February 29, 1996, Annapolis MD. Office of Technology Assessment (OTA), U.S. Congress (1991), U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices-Special Report, OTA-F-470. Washington DC: GPO, May 1991. Payne, M., University of California, Davis, School of Veterinary Medicine, personal communication, 5/24/96. Pure Food Campaign (1994), 1130 17th St. N.W., Suite 300, Washington D.C. 20036, electronic mail communication over Dairy-L listserver. Smith, J., F., (1995), "Managing your bST Program," The Western Dairyman, December 1995. Steyer, R., (1996), "Groups Defend BST, Assail Study," St. Louis Post- Dispatch, Five Star Lift, January 23,1996, p.6C. Tyrrell, H.F., A.C.G. Brown, P.J. Reynolds and G.L. Haaland (1982), "Effect of Growth Hormone on Utilization of Energy by Lactating Holstein Cows," Energy Metabolism: Proceedings of the Ninth Symposium on Energy Metabolism, Butterwoiths Publishers, London, p. 46-49. U.S. Department of Agriculture (USDA) (1995), Baseline Projections February 1995. U.S. Department of Agriculture (USDA) (1996), 1996 Dairy Outlook, 3/21/96. Vamer, M., University of Maryland, Dairy Extension Services, personal communication 3/6/96. Weiss, D., Director of Dairy Coalition, personal communication, 7/22/96. August 1996 33 ------- |