u.s. dairy industry at a crossroad: biotechnology …u.s. dairy industry at a crossroad:...

U.S. Dairy Industry at a Crossroad:Biotechnology and Policy Choices

May 1991

OTA-F-470NTIS order #PB91-198085

Recommended Citation:

U.S. Congress, Office of Technology Assessment, U.S. Dairy Industry at a Crossroad:Biotechnology and Policy Choices--Special Report, OTA-F-470 (Washington, DC: U.S.Government Printing Office, May 1991).

For sale by the Superintendent of DocumentsU.S. Government Printing Office, Washington, DC 20402-9325

(order form can be found in the back of this report)

ForewordIn the 1990s the U.S. dairy industry will experience a technological revolution that will place the

industry at a crossroad. This industry will be the first to experience the biotechnology era in Americanagriculture. New animal health, reproduction, and food processing technologies are being developed.Advanced scientific techniques will be used to produce transgenic animals. These technologies can beused to increase milk production, improve the efficiency of food processing, develop new milk products,increase herd quality, and improve animal health.

Many of the new technologies may create some controversy. But in the early 1990s, the mostpervasive and controversial technology will be bovine somatotropin (bST) produced throughrecombinant DNA technology. Research has shown that the annual gain in milk output per cow from bSTwould take 10 to 20 years to achieve using current breeding methods. The technology is presently underreview by the Food and Drug Administration. Public concerns have been raised about recombinantlyderived bST that include the safety to humans of dairy products produced from bST-supplemented cows,the safety of the technology to the animal, and the economic consequences for many dairy farm operatorsin this country. Some States have placed a moratorium on the use of this technology, even if approvedby FDA, and some large retail food chains have refused to sell milk and dairy products from bST testherds even though FDA has approved their sale.

Congress requested the Office of Technology Assessment to examine the emerging technologiesthat will potentially be available to the dairy industry in the 1990s. This Report analyzes thesetechnologies with special attention to bST. The analysis includes an assessment of bST, a discussion ofother emerging technologies in this decade, and an economic and policy analysis of the impact that thesetechnologies, including bST, will have on the dairy industry.

The report concludes that, based on today’s research findings, bST poses no additional risk toconsumers and does not produce adverse health effects to cows. However, if approved by FDA, bST willaccelerate trends that already put additional economic stress on dairy farm operators in many areas of thecountry. Other new technologies that may become available during the decade may also have similarimpacts as bST and raise similar issues. The industry in the decade of the 1990s will be at a crossroadwith important decisions concerning new technologies and public policies.

This report was requested as part of a larger study examining emerging agricultural technologies andrelated issues for the 1990s. The study was requested by the Senate Committee on Agriculture, Nutrition,and Forestry, the House Committee on Government Operations, and the House Committee onAgriculture. The first report issued from this study was Agricultural Research and Technology TransferPolicies for the 1990s. Two remaining reports are in progress. Findings from this report are relevant tospecific legislation regarding dairy policy that was debated for the 1990 Farm Bill. The informationcontained in this report was made available to Congress for that debate.

OTA appreciates the support this effort received from the contributions of many individuals. Inparticular, we are grateful to workshop participants, contractors, reviewers, and informal advisers whoprovided invaluable assistance in analyzing the issues on this subject. OTA, however, remains solelyresponsible for the contents of this report.

~ f ~a’iL# ‘ >

JOHN H. GIBBONS~ D i r e c t o r

OTA Project Staff--U.S. Dairy Industry at a Crossroad

Roger C. Herdman, Assistant Director, OTAHealth and Life Sciences Division

Walter E. Parham, Program ManagerFood and Renewable Resources Program

Michael J. Phillips, Project Director

Marie Walsh, Analyst

Kathy Hudson, Analyst 1

Susan Wintsch, Editor

Administrative Staff

N. Ellis Lewis, Office Administrator

Nellie Hammond, Administrative Secretary

Carolyn M. Swann, P.C. Specialist

Consultants/Contractors

Dale E. Bauman, Cornell UniversityWilliam Hansel, Louisiana State University

Susan K. Harlander, University of MinnesotaRonald D. Knutson, Texas A&M University

J. Thomas McGuckin, New Mexico State UniversityBennie I. Osburn, University of California, Davis

Derrell S. Peel, Oklahoma State UniversityJames W. Richardson, Texas A&M University

IFrom September through December 1990

iv

ContentsPage

Chapter I: Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Chapter 2: Overview of the Dairy Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Chapter 3: An Emerging Technology: bovine Somatotropin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Chapter 4: Emerging Technologies in the Dairy Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Chapter 5: Economic and Policy Impacts of Emerging Technologies on the U.S. Dairy Industry . . . . . . . 73

AppendixA: A National and Regional Analysis of the Adoption of bovine Somatotropin . . . . . . . . . . . . . 89

Appendix B: Detailed National and Regional Impacts of bovine Somatotropinand Other Emerging Technologies Under Alternative Dairy Policies . . . . . . . . . . . . . . . . . . . . 93

Appendix C: Detailed Farm Level Impacts of bovine Somatotropin and OtherEmerging Technologies Under Alternative Policy and Demand Scenarios . . . . . . . . . . . . . . . 102

Appendix D: Workshop Participants and Reviewers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Chapter 1

Summary

ContentsPage

AN EMERGING TECHNOLOGY: BOVINE SOMATOTROPIN (bST) . . . . . . . . . . . . . . 3Production Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Food Safety Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Effect on Milk and Meat Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Effects on Bovine Reproductive Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Effect on Bovine Health and Stress... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Commercial Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Product Labeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

OTHER EMERGING TECHNOLOGIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6ECONOMIC AND POLICY IMPACTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Dairy Industry Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Technology Adoption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7National Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Farm Level Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Science Policy and Emerging Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

FigureFigure Pagel-1. Projected Impact of a 10-Percent Permanent Demand Reduction on U.S.

Government Milk Purchases Under Alternative Dairy Policies, 1990-98 . . . . . . . . . . 10

TablesTable Pagel-1. Level of Milk Production, With and Without bST, Under Alternative Policy

Scenarios, 1990-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9l-2. Level of CCC Purchases, With and Without bST, Under Alternative Policy

Scenarios, 1990-98, Milk Equivalent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9l-3. Impacts of bST Adoption on the Economic Viability of Moderate-Size

Representative Farms, Assuming No Change in Demand for Milk Due to bST,Trigger Price Policy, by Region, 1989-98. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

l-4. Impacts of bST Adoption on the Economic Viability of Large RepresentativeFarms, Assuming No Change in Demand for Milk Due to bST,Trigger Price Policy, by Region, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . ,.... . . . 11

l-5. Impacts of bSTAdoption on the Economic Viability of Representative Large(125-Cow) Upper Midwest Farms Under Alternative Dairy Policies, AssumingNo Change in Demand for Milk, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

l-6. Impacts of bST Adoption on the Economic Viability of Moderate-SizeRepresentative Farms, Assuming Small Decrease in Demand for MilkDue to bST, Trigger Price Policy, by Region, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Chapter 1

Summary

The dairy industry will lead U.S. agriculture intothe biotechnology era of the 1990s, and also will feelthe first profound impacts of emerging technologies.Recombinant DNA techniques, cell culture andantibody methods are but a few of the new bio-technology techniques that will produce technol-ogies that will sustain or accelerate the historical2-percent annual increase in milk output per cow.

Whereas farmers once had no choice but topasture bulls with cows and let nature run its course,artificial insemination has provided a means ofcontrolled breeding since about 1950. In the nearfuture, farmers will potentially exercise even morecontrol over herd reproduction and genetics and overthe health and milk-producing potential of theiranimals, For example, embryos produced by in vitrofertilization (of ova from selected females withsexed sperm) and placed at predetermined times intothe uteri of estrous-cycled animals can result inhigher conception rates than are now obtained byartificial insemination. This will accelerate geneticgains. Monoclinal antibodies used as diagnosticagents will greatly reduce the cost, time, and laborrequired to maintain animal health. Bovine soma-totropin produced with recombinant DNA technolo-gies has the potential to greatly enhance milkproduction per cow.

The emerging biotechnologies will require con-siderable management expertise on the part offarmers. Information technologies will be powerfulaids to farm operators. Expert systems, for example,make onfarm consulting accessible via a microcom-puter and can aid farmers with decisions regardingmanagement and new technology adoption.

Many biotechnologies will be controversial, mostnotably bovine somatotropin (bST). Although bSTcan boost milk yield per cow significantly---doing in1 year what it would take 10 to 20 years to achievewith current reproductive technologies, concernshave been raised about its safety for humans andanimals and about the economic consequences of itsuse for the industry. In response to these concerns,two States placed a moratorium on the use of bST ifapproved by the Food and Drug Administration

(FDA); up to four States are seriously consideringlaws that would require milk and dairy productsproduced from bST-supplemented cows to be solabeled. Major retail food chains have curtailed salesof milk and dairy products from bST test herds eventhough FDA has approved their sale.

In addition, issues concerning science policy havebeen raised in conjunction with biotechnology—including bST. These issues include the social needsbeing met by these new technologies, the appropri-ateness of public sector investment in their develop-ment, and lack of information about benefits andrisks of a new technology prior to commercializa-tion. This report analyzes the major questionsconcerning the use and safety of bST, examinesother technologies that will affect the dairy industryin this decade, and evaluates the economic andpolicy implications of these issues.

AN EMERGING TECHNOLOGY:BOVINE SOMATOTROPIN (bST)Some 50 years ago, research showed increased

growth rates in rats injected with a crude pituitaryextract. Later it was discovered that the extract,which contains a protein hormone called somato-tropin, also affects lactation, and research withlactating cows ensued. Prior to the 1980s, progresswas slow in bST research because: 1) the availabilityof bST was restricted to that which could beextracted from pituitary glands of slaughtered ani-mals, limiting studies to a few cows and shorttimeframes; and 2) the mechanism of action for bSTwas thought to be acutely stimulated use of body fatreserves: scientists believed it would only work infat cows with a low milk yield. No studies used highmilk-producing cows because it was assumed thatacute mobilization of body fat reserves would causeketosis l and other adverse health effects.

In the late 1970s, new research showed that thephysiological basis for more efficient milk produc-tion in genetically superior cows was better use ofabsorbed nutrients. Scientists recognized the needfor new concepts regarding nutrient regulation in

1A me~bolic disorder which occurs when production of ketones exceeds the ability of the body to use tiem. OCCUrS in dairy cows when he ne~

for glucose exceeds the production of glucose.

–3–

4 ● U. S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

animals. Recent work has demonstrated that somato-tropin exerts key control over nutrient use. Whenadministered exogenously, bST markedly improvesproductive efficiency in lactating cows. In the lastdecade, as the important role of somatotropin hasbeen established, bST produced by recombinantDNA technology has replaced pituitary-derived bSTin research with cows. Since that time, the quantityand scope of the research with bST has increasedexponentially.

Production Response

The impact of bST on milk production will varyaccording to quality of management on individualfarms, but a reasonable expectation is that successfuladopters would experience, on average, a 12-percentboost in production. However, the increase in outputper cow tends to be absolute (in number of pounds)rather than proportional to normal production. Thus,approximately the same increase in pounds of milkproduced might be expected (in comparably man-aged herds) from all cows producing 12,000 to20,000 pounds of milk per year. Supplementationwith bST not only results in an immediate increasein milk yield, it also reduces the normal decline inmilk yield during the lactation period.

Because bST is rapidly cleared from the blood-stream and is not stored in the body, exogenous bSTis needed every day to sustain the increase in milkyield. This requires daily injections or use of aprolonged release formulation of bST. Severalprolonged release formulations have been developedand are administered by subcutaneous injection atintervals ranging from 2 or 4 weeks.

Obtaining a milk response to bST does not requirespecial diets or unusual feed ingredients. Substantialmilk responses have been observed on diets rangingfrom pasture to the more typical forage/concentratediets used in the United States. However, voluntaryintake of feed increases in bST-supplemented dairycows. This increase in voluntary intake occurs aftera few weeks of bST supplementation and persiststhroughout the interval of bST use. It has beenconsistently observed across a wide range of diets.

Poor management results in a near zero responsefrom bST supplement. Facets that contribute to thequality of management (and milk response to bST)include the herd health program, milking practices,nutrition program, and environmental conditions.

Food Safety Considerations

Somatotropin is produced by the anterior pituitarygland and is transported by the blood to various bodyorgans where it has certain biological effects. Ifsomatotropin is given orally it is broken down to itsconstituent amino acids in the digestive process justlike any other dietary protein. Thus, somatotropinmust be injected to be biologically active.

Somatotropin is species-limited, and the biologi-cal effects of somatotropin from one species onothers varies. In order to have any biological effect,a protein hormone first must bind to a specificcell-surface receptor. Studies have shown conclu-sively that due to its unique three-dimensionalshape, bST does not elicit any of its normalbiological actions in humans even if injected.

Recombinantly derived bST products may differslightly from the bST produced by the pituitarygland because in the manufacturing process a fewextra amino acids can become attached at the end ofthe bST molecule. The number of extra amino acidsvaries from one to eight depending on the particularmanufacturing process. Some manufacturing proc-esses produce no additional amino acids. Theadditional amino acids that may be produced do notchange the three-dimensional shape of the activepart of the molecule and, hence, do not alter thebiological activity of bST in dairy cows or the lackof activity of bST in humans.

Some biological actions of somatotropin in cowsmay be mediated by insulin-like growth factor 1(IGF-1). This protein hormone, a member of thesomatomedian family, normally occurs in tracelevels in milk and also in human saliva. Administra-tion of bST to dairy cows augments IGF-1 in milk,but the levels are still within the range typicallyobserved in early lactation of untreated cows.Similar to results with bST, studies with laboratoryanimal models have demonstrated that IGF-1 has nobiological activity if administered orally. The impor-tance of increased amounts of IGF-I in milk frombST-treated animals is uncertain. However, theamount of IGF-I ingested in 1 liter of milk approxi-mates the amount of IGF-I in saliva swallowed dailyby adults.

Chapter I--Summary ● 5

Effect on Milk and Meat Composition

The overall composition of milk (fat, protein, andlactose content) and meat is not substantially alteredby bST supplementation. There can be minorchanges, primarily in fat content of milk during thefirst few weeks of bST supplementation as the cow’smetabolism and voluntary feed intake adjust. How-ever, these changes are temporary and within thestandard variation that occurs naturally during alactation cycle. The meat derived from treated cowshas a lower fat content but is otherwise identical.

In manufacturing characteristics, milk from bST-supplemented cows does not differ from the milk ofuntreated cows. Characteristics that have beenevaluated include freezing point, pH, alcohol stabil-ity, thermal properties, susceptibility to oxidation,and sensory characteristics, including flavor. Simi-larly, no differences were observed in cheesemakingproperties, including starter culture growth, coagula-tion, and acidification or in the yield, composition,or sensory properties of various cheeses.

Effects on Bovine Reproductive Performance

Of special interest are bST effects on reproductivevariables such as conception rate (services perconception), pregnancy rate (proportion of cowsbecoming pregnant), and days open (days fromparturition to conception). As expected, cows ad-ministered bST show decreased pregnancy rates andincreased days open; these changes are associatedwith increases in milk yield and occur regardless ofwhether or not the high milk yields are achievedusing bST. The management of the reproductioncycle may need to be adjusted to account for thesephysiological changes. Conception rate is unchangedby bST supplementation.

Effect on Bovine Health and Stress

Catastrophic effects such as the incidence ofketosis (underproduction of glucose), fatty liver,crippling lameness, milk fever (feverish disorderfollowing parturition), mastitis (inflammation of theudder), sickness, suffering, and death have beenpostulated to occur with bST. However, no sucheffects have been observed with bST-supplementa-tion of dairy cows in any scientifically validpublished studies, nor have subtler health effectsbeen in evidence. From the hundreds of investiga-tions with bST, no study reported the lower milkyield and decreased productive efficiency likely to

be associated with increased sickness and suffering.Relevant studies include short- and long-term re-search and both chronic and acute toxicity studies. Inacute toxicity studies, dairy cows were given 30,000mg of bST over a 2-week period, an amount of bSTapproximately equaling what would be administeredin four lactation cycles.

Reduced resistance to infections has not beenfound to occur in bST supplemented dairy cowsalthough such an effect has also been postulated.Indeed, basic biological studies have demonstratedthat rather than reducing resistance to infection,somatotropin plays a key role in several aspects ofmaintaining immune competence.

Animal stress is more difficult to evaluate thandisease, but several indices exist that demonstrate nostress effects due to bST supplementation. Dairycows would be expected to produce less milk and tobe less efficient if they are stressed. Several hundredstudies utilizing bST demonstrate increased milkyield and productive efficiency. Studies have alsoclearly demonstrated that bST has no effect on theenergy expended (as heat) for maintenance or forefficiency of milk synthesis.

Commercial Introduction

The FDA must approve bST before it can be soldlegally in the United States. Each company seekingFDA approval to market bST must demonstrate thatits product is effective (does what the companyclaims) and safe. The safety evaluation involvesthree areas:

1. safety of the animal-food products for humans,2. safety of the bST-supplement to the target

animals, and3. safety of using bST in the environment.

In addition, FDA requires that each company provethat its manufacturing process can produce bST toconsistent and acceptable quality standards.

FDA has determined that sufficient scientificinformation exists to indicate that the milk and meatfrom bST-supplemented cows is safe for humanconsumption, and has allowed for these animalproducts to be marketed from the test herds duringthe remainder of the investigational period.

In addition to the United States, many countriesare reviewing bST for commercial use. In allcountries where bST studies are being conducted,

6 ● U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

the appropriate regulatory agencies have completedthe human safety evaluations and without exception,have found it safe for human consumption.

Product Labeling

Some States are considering requiring all foodproducts derived from the milk of bST-supple-mented cows to be labeled as such in the market-place. The basis for labeling seems to relate to aconcern about the safety of the products for humanconsumption. At least two considerations need to beaddressed.

First, is the scientific merit or basis for labeling.If there is a valid safety concern, then the foodshould not be marketed for human consumption.Labeling is not the appropriate method for handlinga food safety concern. If the regulatory system toevaluate food safety is inadequate, then the systemshould be changed. Labeling does not excuse theinadequacy.

The second consideration is verification. Aneffective labeling program requires developmentand adoption of appropriate regulations and theestablishment and funding of a system for imple-mentation and verification. In the case of bST, noknown test or technology exist that could be used todistinguish milk from bST-supplemented cows frommilk from non-treated cows. Indeed, no change inmilk composition as a result of bST supplementationwas found in FDA human safety evaluations.

OTHER EMERGINGTECHNOLOGIES

There are a number of emerging technologies thatwill have a significant influence on the dairyindustry in the 1990s in addition to bST. Advancesin animal reproduction, animal health, and foodprocessing are occurring, and many of the newtechnologies being developed use highly sophisti-cated and complex biotechnology methods. Bycomparison, the biotechnology methods used toproduce bST are rather rudimentary; potentiallysome of these new technologies could make bSTobsolete.

Animal reproduction technologies are advancingrapidly. Researchers have significantly improved

their understanding of egg development in the ovary,how to stimulate the release of numerous eggs atonce, and how to enhance the development andfertilization of eggs outside of the cow. Embryos canbe frozen for later use. Both embryos and sperm canbe sexed. It is possible to create multiple copies ofan embryo, each of which can be transplanted into acow whose reproductive cycle has been adjusted tobe able to accept the embryo and carry it to term.These new technologies make it possible to improveherd quality more rapidly than can be achieved usingtraditional breeding methods.

It is possible to create transgenic cattle,2 however,the techniques currently used are inefficient andrequire the use of thousands of eggs to produce onetransgenic animal. These inefficiencies make it tooexpensive to produce and market transgenic live-stock commercially. However, scientific break-throughs are leading to the development of technolo-gies that will improve the efficiency of transgenicanimal production and substantially lower the costof doing so. Transgenic livestock may becomecommercially available in small numbers by the endof the decade.

BST potentially could be supplanted by thedevelopment of transgenic cattle. Dairy cows can bedeveloped to produce higher levels of bST so thatdaily injections or timed release formulations are nolonger needed. Alternatively, genes that code forchemicals that suppress bST production can bealtered in the cow such that a cow’s normal bSTproduction will increase.

New biotechnology products are also being devel-oped to improve animal health. Products include newvaccines and diagnostic kits, as well as compoundsthat enhance an animal’s ability to fight disease.

Not only are new biotechnology products beingdeveloped for use in livestock production, but theyare also being developed for use in food processing.New products will improve the production of milkproducts such as cheese and yogurt. They can also beused to detect milk contaminants.

Effective use of these new technologies will placea premium on management skills. New informationtechnologies are being developed to aid farm man-agement. These new technologies can incorporate

2Anim~5 whose her~it~ DNA hm been augmented by the addition of DNA from a source other than parental germplasm using recombinant DNAtechniques.

Chapter 1--Summary ● 7

individual farm data, with pertinent informationfrom national databases, into computer programsthat will aid farmers in the decisionmaking process.

These new technologies are in various stages ofdevelopment. Some, such as embryo transfer, re-combinant DNA vaccines, and information technol-ogies are already available commercially or will besoon. Other technologies, such as transgenic cattleand advanced reproductive technologies, will not beavailable until the end of the decade. The collectiveeffect these emerging technologies, including bST,will have on the economic and policy environmentof the 1990s is examined next.

ECONOMIC AND POLICYIMPACTS

Dairy Industry Trends

Before discussing the economic and policy im-pacts that emerging technologies discussed abovehave on the dairy industry, it is important to considerthe major economic trends already at work withinthe industry. Milk output per cow has been increas-ing at a very steady rate for many years. Output percow has grown more rapidly than milk consumptionper capita, resulting in a gradual trend towardreduced cow numbers.

Changes in output per cow vary regionally. ThePacific region’s output per cow has been about 30percent higher than the national average and 50percent higher than that of the lowest producingregion. Climatic conditions contribute to some ofthese differences, but the main factors seem to berelated to progressiveness, philosophy, and qualityof management demonstrated by different dairyfarmers. These factors directly impact technologyadoption and the size of dairy farms. Generally,larger dairy farms experience lower productioncosts. The Pacific Coast and Florida lead the Nationwith herd sizes in the 500- to 1,500-cow range. Thetraditional milk producing regions of the UpperMidwest and Northeast are typically in the 50- to150-cow range.

There is a corresponding variation in regionalprofits. The Pacific and Southeast regions realizedfavorable returns in 1988 ($1.05 per cwt and $1.94per cwt. respectively) whereas the Upper Midwestand Corn Belt regions had negative returns (–$0.62per cwt and –$0. 18 per cwt, respectively). Returns in

the early months of 1991, however, are less favora-ble in all regions. Farm milk prices have declinedsignificantly from January through March and areexpected to fall by 15 to 20 percent for the yearcompared to 1990. Dairy farms in the traditionalmilk producing regions are expected to lose consid-erable equity under these conditions. Pacific andSoutheast farms, although still profitable, are ex-pected to operate much closer to their respectivebreak-even points.

These differences have led to shifts in productionpatterns. The largest increases in milk productionhave been in the Pacific region where marketinghave risen by nearly 40 percent. The traditionalUpper Midwest and Northeast regions have eachincreased milk production about 5 percent. Thesetraditional regions produce about half of the Na-tion’s milk supply and will continue to be a majorforce in the dairy industry. But if the Upper Midwestand Northeast regions are to maintain their roles asthe “dairy States,” major changes in scale ofoperation, progressiveness in technology adoption,philosophy, and quality of management and perhapsdairy policy may be required.

Technology Adoption

When emerging technologies, suchcome available commercially it is not

as bST, be-known with

any degree of certainty how many dairy farmers willuse them or when. Farmers have been surveyed toproject expected adoption levels once bST becomesavailable. Results indicate relatively rapid adoption—50-percent adoption within the frost year and at least80-percent within 3 years.

However, these surveys may not be accurateindicators of prospective adoption. Many of the bSTsurveys were conducted prior to the availability ofwidespread information on bST. Most other dairytechnologies, moreover, have not been adoptedrapidly. Artificial insemination technology is usedonly by 70 percent of dairy farms despite beingavailable for some 40 years. Dairy Herd Improve-ment technology, available for 50 years, is used byonly 45 percent of farmers.

OTA’s statistical analysis of historical rates oftechnology adoption by dairy farmers providesanother basis for predicting bST adoption. Theanalysis found:


. A slower rate of adoption than suggested byproducer surveys of farmers on probable bSTuse (17 percent or less the first year).

. Regional variations in rates of technologyadoption in the dairy industry. Based on this,bST adoption after 5 years is forecast to be 40percent in the Pacific region, where technologyadoption is most rapid, and 25 percent in theCorn Belt. This and other traditional milkproduction regions tend to be slower to adoptnew technologies.

National Impacts

The interactions of technology adoption, dairypolicy, and consumer reaction and their effects onfuture milk supply prices and returns to dairyfarmers were captured using LIVESIM, a regionaland national computer simulation model.3 Thepolicy options analyzed included a fixed pricesupport, a price-support trigger, and a quota pro-gram. In all policy scenarios, the governmentpurchases at least 3 billion pounds of milk annuallyto satisfy food-program needs (i.e.. school lunchprograms).

Fixed Price Support

This scenario frees price at the 1989 level of$10.6O per cwt. This serves as a useful bench markfor comparing other policy options. In this case, thegovernment purchases excess milk, at the supportprice, in order to clear the market. Without bST, milkproduction would increase from the present level of144 billion pounds to 152 billion in 1995. With bST,production would increase an additional 4 to 5billion pounds over the period (see table l-l);government purchases would rise as high as 7 and 9billion pounds in any one year, and overall wouldincrease by 3 to 6 billion pounds over the minimumpurchases of 3 billion pounds for food programs (seetable 1-2).

Trigger Price Policy

This option triggers a price-support reductioneach time the level of government purchases risesabove 5 billion pounds annually. This scenario issimilar to the producer-assessment option in the1990 farm bill because the assessment will effec-tively trigger reductions in producer returns throughmilk price declines. Without bST, a single price-

support reduction is triggered to a level of $10.10 percwt in 1991. With bST, two price-support reductionsare triggered in 1991 and another in 1993 to a levelof $9.60 per cwt. These price reductions moderateproduction increases to keep government purchasesnear the 3-billion-pound minimum.

Quota Policy

A quota policy is another method to manageexcess production. It establishes a level of milkproduction for each farm and provides effectivedisincentives to the farmer if production exceeds thequota. This might be accomplished by a two-tieredpricing system or some other mechanism thatprovides disincentives for producing over quotalevels.

In the analysis, the quota policy was designed tomaintain government purchases at or near theminimum government use target of 3 billion pounds.The quota was adjusted downward any year govern-ment expenditures exceeded 3 billion pounds. Theresults show that the quota avoids the high level ofgovernment purchases that result under the fixedprice-support scenario (see table 1-2).

Demand Reduction

While claims that consuming milk and milkproducts from cows supplemented with bST or othernew technologies could adversely affect humanhealth have not been substantiated, a range of foodsafety and other considerations will affect consumerpurchases. Policy needs to be designed consideringthe full range of potential consumer response;accordingly two scenarios of reduced milk con-sumption were analyzed.

Small Demand Reduction-In this scenario,per-capita demand decreases by 10 percent in 1991,5 percent in 1992 (i.e., demand increases from 1991to 1992), and 2.5 percent annually thereafter. Gov-ernment purchases total 21.2 billion pounds in thefirst year (1991), 9.7 billion in 1992, and 8.4 billionin 1993. The support trigger decreases the price-support level to 9.10 per cwt in 1994. Even thoughgovernment purchases are high for 3 years, thetrigger mechanism seems to accommodate a tempo-rary demand reduction.

Large Demand Reduction-The second demandscenario assumes a permanent 1 O-percent annual

3A major focus of tie ~~ysis is on tie use of bST because of its effect on productivity and commercial availability ~ tie eaf~Y 1990s.

Chapter I--Summary . 9

Table l-l—Level of Milk Production, With and Without bST, Under Alternative Policy Scenarios, 1990-98(billions of pounds)

Fixed support Trigger Quota

Year With bSTa Without bST With bSTa Without bST With bSTa Without bST

1990 . . . . . . . . . . . . . . . . . 1441991 . . . . . . . . . . . . . . . . . 1461992 . . . . . . . . . . . . . . . . . 1461993 . . . . . . . . . . . . . . . . . 1531994 . . . . . . . . . . . . . . . . . 1531995 . . . . . . . . . . . . . . . . . 1561996 . . . . . . . . . . . . . . . . . 1571997 . . . . . . . . . . . . . . . . . 1601998 . . . . . . . . . . . . . . . . . 161

144144143150149152153155157

144146146153152156155159159

144144143150148152153155157

144146145148150152155157160

144144144146148150153155157

abSTisassumed tobecommercially availablein 1991.

SOURCE: Office ofTechnology Assessmen~ 1991.

Table 1-2—Level of Government Purchases, With and Without bST, Under Alternative Policy Scenarios,1990-98, Milk Equivalent (billions of pounds)


With bSTa Without bST With bSTa Without bST With bSTa Without bST

1990 . . . . . . . . . . . . . . . . . 3.01991 . . . . . . . . . . . . . . . . . 7.31992 . . . . . . . . . . . . . . . . . 4.31993 . . . . . . . . . . . . . . . . . 9.01994 . . . . . . . . . . . . . . . . . 6.01995 . . . . . . . . . . . . . . . . . 7.01996 . . . . . . . . . . . . . . . . . 4.81997 . . . . . . . . . . . . . . . . . 5.31998 . . . . . . . . . . . . . . . . . 3.6

3.05.33.05.73.03.03.03.03.0

3.07.33.06.83.03.03.03.03.0

3.05.33.03.83.03.03.03.03.0

3.07.33.53.43.13.03.03.03.0

3.05.33.03.03.03.03.03.03.0

abST is assumed to be commercially available in 1991.

SOURCE: Office of Technology Assessment, 1991.

reduction in per-capita consumption. The triggermechanism does not easily adjust the industry underthis scenario. The support price must be lowered to$7.60 in 1997 to bring government purchases below4 billion pounds (see figure l-l). Such a low supportprice would make it difficult (impossible) for eventhe best managed dairy farms to avoid economiclosses. A quota program or termination program (aone-time government buy-out of dairy herds) wouldbe needed to bring government purchases back to the3-billion-pound minimum. However, as this studyshows, termination programs do not result in perma-nent reductions in supply. Quota programs caneffectively reduce supply over a period of time. Butwith either program, approximately 1 million cowswould need to be slaughtered causing beef prices todecline by 4 to 6 percent.

Conclusions

A mechanism such as the trigger price policy orproducer assessments, which allow producer returnsto decline as government purchases increase, could

effectively adjust supply without excessively largeinventory accumulations. However, if sharp reduc-tions in demand accompany the introduction of bST,production quotas may be required. A quota policy,however, has some potentially harmful effects,including:

. higher production costs,

. elimination of dynamic adjustment within theindustry,

. negative impact on beef cattle prices,

. difficulty of discontinuing, once initiated, and

. the capitalization of benefits into the quota.

Farm Level Impacts

The effects of emerging technology, dairy poli-cies, and consumer demand can be more easilyvisualized by analyzing the impacts on representa-tive dairy farms. The farm-level impacts of the threepolicy scenarios-fixed price support, trigger, andquota---over a 10-year period were analyzed usingFLIPSIM, a farm level simulation model.

IO ● U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

Figure l-l—Projected Impact of a 10-PercentPermanent Demand Reduction on U.S. Government

Milk Purchases Under Alternative Dairy Policies,1990-98

30 ; : : : : : :

~ 2 4 7c ‘“o.— m. . . . . . . . . . . . . . :

~ 1 8 -alU-J

Qx=

o ~ ~ ; ; ~ ~ :

90 91 92 93 94 95 96 97 98Year

— Trigger Price Fixed Support-- Dairy Termination Q u o t a


Adoption Incentives

Once bST becomes available, strong incentiveswill exist to adopt the technology. Payoffs from bSTadoption are substantial, regardless of region (seetable 1-3). Nonadopters of bST will have moreproblems surviving and will be more likely to exitthe industry. Likewise, dairy farmers located inStates that have a moratorium on adoption will beplaced at a substantial disadvantage relative to thosein States where a moratorium does not exist.

Regional Competitiveness

Several reasons for regional shifts in milk produc-tion patterns can be seen in tables 1-3 and 1-4. UpperMidwest farms have problems realizing sufficientearnings to achieve a reasonable return on equity,compete, and survive. While Northeast farms per-form better, they too were found to be at adisadvantage relative to Pacific and Southeast farms.In all regions, adoption of bST increases thepotential to survive, especially for larger farms.

Policy Impacts

The fixed price-support policy, with its higherearnings, increases the probability of farm survivaland the chances of earning a 5-percent return oninitial equity (see table 1-5). While Upper Midwestdairies are able to maintain cash flow, net worthcontinues to erode on the 125-cow Upper Midwest

dairy due to the relatively high investments in fixedassets (buildings, equipment, etc.).

From the producer standpoint, the quota programdoes not perform as well as the trigger price or freedprice-support programs. This is because the quotaprice objective is the same as the fixed price support($10.60) and because restrictions on output curbexpansion and raise production costs. To maintaindairy farm income under a quota system, the priceobjective must be sufficiently high to offset theeffects of lower production—and this will result inhigher prices to consumers.

The economic payoff from bST adoption is aboutthe same for a trigger price policy and a freedprice-support policy. However, all the representativefarms experienced at least a 20- to 40-percentdecrease in economic payoff under a quota com-pared to the trigger price policy. Adoption of bSTwould be slowed by imposing a quota as opposed tothe trigger price policy.

Even with reduced demand, strong incentiveswould exist f-or all farms in all regions to adopt bST.With the continuation of the current trigger policy,a 52-cow Upper Midwest dairy’s probability ofsurvival declines under a small decrease in demand,but is relatively enhanced by adopting bST (see table1-6). The same is true for the larger dairies. If a majordecrease in demand occurs, small and large dairyfarms in the Upper Midwest will be most vulnerable.

Increased Pressure on Traditional Farms

A major controversy concerning bST is that it willforce many dairy farms out of the industry, espe-cially in traditional milk-producing regions. BSTalone, however, will not force these traditional farmsout of existence. As discussed earlier, the trendtoward fewer total cows and larger farms has beenunderway for many decades. This trend is a result ofthe combination of emerging technology, industryeconomics, and policy. The trend will no doubtaccelerate in the 1990s as the result of the combina-tion of bST and other cost-reducing technologiesand a more market-oriented dairy policy. As hasbeen the case for years, such changes inherently putsincreased pressure on traditional dairy farms. Thesepressures are not new, although they are accentuatedby technological change.

If policymakers decide to change or at least slowthis trend toward fewer but larger farms, changes inpolicy will be needed. First, to reduce the magnitude

Table l-3—impacts of bST Adoption on the Economic Viability of Moderate-Size Representative Farms, Assuming No Changein Demand for Milk Due to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

52-cow 52-cow 350-COW 200-COWUpper Midwest Northeast Southwest Southeast

Non- bST Non- bST Non- bST Non- bSTMeasure of impact adopter adopter adopter adopter adopter adopter adopter adopter

Probability of survivala . . . . . . . . . . . . . . . . . . . . . . 58% 74% 10070 100% — 950/0 97% 100% 100%Probability of earning 5-percent return

on equity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 74 100 100 95 97 100 100Probability of increasing equityb . . . . . . . . . . . . . . . 0 0 3 3 60 79 13 24Present value of ending net worth as percent

of beginning net worthc . . . . . . . . . . . . . . . . . . . . 16 29 72 77 109 128 76 89%hance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.bchance that the individual farm wilt increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.


Table l-4—impacts of bST Adoption on the Economic Viability of Large Representative Farms, Assuming No Change in Demand for MilkDue to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

125-cow 200-COW 1 ,500-cow 1 ,500-COWUpper Midwest Northeast Pacific Southeast


Probability of survivala . . . . . . . . . . . . . . . . . . . . 9570 99% 100% 100?40 10070 100?40 100% 1 00%Probability of earning 5-percent return

on equity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 95 99 100 100 100 100 100Probability of increasing equityb . . . . . . . . . . . . . . . 8 12 43 53 100 100 88 99Present value of ending net worth as percent

of beginning net worthc . . . . . . . . . . . . . . . . . . . . 57 69 92 102 195 214 129 147

Whance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.bchance that the individual farm will increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.

SOURCE: Office of Technology Assessment. 1991.

12 . U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

Table l-5—impacts of bST Adoption on the Economic Viability of Representative Large (1 25-cow)Upper Midwest Farms Under Alternative Dairy Policies, Assuming No Change

in Demand for Milk, 1989-98 (in percent)

Trigger price Fixed price support QuotaNon- bST Non- bST Non- bST

Measure of impact adopter adopter adopter adopter adopter adopter

Probability of survivala. . . . . . . . . . . . . . . . . 95% 99% 99% 100% 85% 920/.Probability y of earning 5-percent

return on equity . . . . . . . . . . . . . . . . . . . . 90 95 95 98 67 78Probability of increasing equityb. . . . . . . . . . 8 12 11 18 2 3Present value of ending net worth as

percent of beginning net worthc . . . . . . . . 57 69 67 78 37 46%hance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.bchance that the individual farm will increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.SOURCE: Office of Technology Assessment, 1991.

of adjustment the rate of technology adoption bytraditional farms would need to be increased. Thiswould require additional expenditures on publicresearch and extension with the specific goal ofenhancing the survivability of these farms. Duringtimes of rapid technological change, research andtechnology adoption strategies for traditional farmsneed to be developed and implemented by USDA,land-grant universities, and dairy cooperatives ifthese farms are to survive.

Second, dairy policy may need to change back toa fixed price-support policy. As seen in the previousanalysis, a fixed support policy enhances the tradi-tional farm’s probability of survival compared toother policies. It is, however, significantly morecostly to the government than current policy.

It is possible to at least slow the trend towardfewer total cows and larger dairy farms. However,such change may be costly. As noted, to keep lessprogressive traditional farms in the industry willrequire increased expenditures for research andextension to improve technology adoption andincreased funds to support the price of milk at a levelthat will allow these farms to compete. Policymakerswill need to weigh the benefits of traditional farmswith these costs in determining the policy path tofollow in the 1990s. This is particularly the case fordairy farms outside the areas with a comparativeadvantage in dairying where a large share of feedsupplies are purchased.

Science Policy and Emerging Technology

The controversy surrounding biotechnology—including bST—raises questions concerning whatsocial needs are being met by these technologies and

the appropriateness of public sector investment intheir development. The questions raised point to theneed for broadbased, ex ante information concerningnew technologies. Presently, little information aboutnew technologies is available prior to commerciali-zation. There is no institution within the agriculturalscience policy community that develops informationon the benefits and risks of any technology ex ante.There also is no formal structure that provides inputto decisionmakers from all affected parties (farmers,marketers, researchers, consumers, etc.). Thus thereis no comprehensive information about the benefitsand risks of a new technology prior to commer-cialization and, therefore, no inclusive criteria todetermine how public research resources should beallocated.

The development of an institutional framework toprovide and act on such information is needed. Hadsuch an institution been in existence a decade ago, itis possible that the bST controversy could have beenavoided or minimized. Consideration of the costsand benefits of new technologies and input from awider range of clientele could lead decisionmakersto consciously choose a different allocation of publicsector research funding than that which occurs in theabsence of such information.

Clearly, all benefits and risks of new technologydevelopment cannot be determined a priori, andovercentralization of research decisionmaking raiseslegitimate concerns. Care must be taken in establish-ing such an institution. However, a broad baseddiscussion of issues involving all relevant users ofnew technologies can point to potential problems,determin e further research needs, and provide infor-mation about the relative social benefits to be gained

Table l-6—impacts of bST Adoption on the Economic Viability of Moderate-Size Representative Farms, Assuming Small Decrease inDemand for Milk Due to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

52-cow 52-cow 350-COW 200-cowUpper Midwest Northeast Southwest Southeast


Probability of survivala . . . . . . . . . . . . . . . . . . . . . . 40% 480/0 1009!0 1009!0 88% 94% 99% 100%Probability of earning 5-percent return



Whance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.bchance that the individual farm will increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.



by investments in competing technologies. Theseeds of such a framework can be found in the firstreport of this series, Agricultural Research andTechnology Transfer Policies for the 1990s. Con-gress has subsequently taken the first step byauthorizing an Agricultural Science and TechnologyReview Board in the 1990 farm bill that will beginto develop ex ante information on selected technol-ogies. It is a beginning, but much more work will beneeded in the future.

CONCLUSIONSEmerging technologies, such as bST, industry

economics, and public policy will play critical rolesin shaping the U.S. dairy industry in the decade ofthe 90s, Advances in health, reproduction, andinformation technology all will affect the industry.The most dramatic impact will be due to bST.Claims have been made that bST is unsafe inconsumer food products, an unsafe technology forcows, and a technology that will economicallydestroy many traditional farms. This report con-cludes just the opposite. It is a technology that, basedon today’s research findings, poses no additionalrisk to consumers, one that does not produce adversehealth effects to cows, and one that alone will noteconomically disadvantage the traditional farm op-erator. Emerging technologies (including bST),industry economics, and current dairy policy willmerely accelerate an existing trend-the pressure ontraditional farms to grow or exit the industry.Changes in: rate of technology adoption, researchand extension policy, and perhaps dairy policy maybe required to reverse this trend.

A national dairy policy that provides a mechanismfor allowing producer returns to decline as govern-

ment purchases increase, such as the trigger price-support policy or producer assessments as providedfor in the 1990 farm bill, could effectively adjustsupply without excessively large inventory accumu-lations. However, if demand for dairy productsdeclines sharply with the introduction of bST,supply-management programs such as productionquotas or termination programs may be required,Termination programs are costly and do not effec-tively reduce supply over a period of time. Produc-tion quotas can effectively control supply. However,they also freeze regional production shifts and(because the quota has an economic value) make itmore costly for new entrants into the industry.Because of costs and rigidities associated with quotaprograms, consideration might be given to observinggovernment purchases over a 2-year, as opposed toa l-year, period before implementing such a pro-gram. This would permit a more accurate assessmentof whether the demand reduction is temporary orpermanent.

The introduction of bST has caused considerablecontroversy. Little, if any, information was availableearly in its development to foresee the biological,economic, social, and political impacts of its poten-tial adoption. Lack of such information establishesa clear need to consider the benefits and risks of newtechnology more seriously and to use that informa-tion in allocation of public sector research funds. Aninstitutional framework needs to be developed toprovide this information and involve all the relevantusers of new technology.

Chapter 2

Overview of the Dairy Industry

ContentsPage

FORCES DRIVING THE INDUSTRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17Technological Change +... . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . .. . . . . . . . . . . . . . . . . . 17Economies of Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18Costs of production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18Dairy Receipts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Net Income . . . . . . . . . . . . . .. .. +. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Regional Production Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Demand Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

MAJOR DAIRY POLICY ISSUES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,.,. 23Butterfat Surplus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,. . . . . . . . . . . . ● . . . . . . . . 23

Milk Price Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25Price Instability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26Milk Production Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27Federal Milk Marketing Orders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Basic Formula Price . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...,*. 28Emerging Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

CHAPTER 2 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 29

FiguresFigure Page

2-1. Annual Milk Outputter COW, 1965-89 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172-2. Average Number of Milk Cows on Farms and Total Milk Production, 1980-89 . . . 172-3. USDA Farm Production Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182-4. Milk Outputter Cow by Region, 1989 . . . . . . . . . . .,...,. . . . . . . . . . . . . . . . . . . . . . . 192-5. Milk Production Costs Related to Herd Size, for 10 States, 1985 . . . . . . . . . . . . . . . . 202-6. Regional Differences in Costs of Producing Milk, 1988 . . . . . . . . . . . . . . . . . . . . . . . . 212-7. Regional Differences in Receipts From Dairy Farm Sales, 1988, ....,..,..,,.. 222-8. Regional Differences in Net Income (profit), 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232-9. Change in Milk Production by Region, 1980-89 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

2-10. Share of Milk Production by Region, 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .,. 252-11. Change in Share of Milk Production by Region, 1980-89 .. .. .. .. .. ... .,. ....,. 262-12+ Actual and Projected Consumption of Lowfat Milk, 1950-93 . . . . . . . . . . . . . . . . . . . 272-13. Market Share of Lowfat Milk as a Percent of Total Fluid Milk Consumption ...,. 272-14. Actual and Projected Consumption of American Cheese, 1950-93 . . . . . . . . . . . . . . . 272-15. Actual and Projected Consumption of Other Cheese . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272-16. Uncommitted Government Dairy Product Inventories, 1985-90 . . . . . . . . . . . . . . . . . . 282-17. Minnesota-Wisconsin (M-W) Price and Milk Price-Support Level,

January 1987 to April 1990 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 28

Chapter 2

Overview of the Dairy Industry

The dairy industry is large, dynamic, and drivenby a number of forces. Dairy products account forabout 13 percent of total cash receipts from all farmcommodities. In 1989, cash receipts from dairyproducts totaled $19.3 billion; only cattle and calvesbrought greater returns. Although milk is producedand processed in every State, two-thirds of the total1989 milk supply was produced in 10 States (l). Atleast half of the total 1989 U.S. milk productioncame from Wisconsin, California, New York, Min-nesota, and Pennsylvania.

A central feature of the dairy industry is therelatively constant 1.5 to 2.0 percent annual increasein output per cow (see figure 2-l). Exceptions haveoccurred only after major weather disruptions (lead-ing to sharp increases in feed prices in the early1970s and in 1989) and with changes in governmentpolicy (namely, the 1983-84 milk diversion pro-gram). This chapter describes the supply, demand,and regulatory forces driving the industry and theresulting policy issues for the 1990s.

FORCES DRIVING THEINDUSTRY

Technological Change

Increases in milk output per cow have not comeautomatically; they reflect the continuous adoptionby farmers of artificial insemination, Dairy HerdImprovement Association (DHIA) recordkeeping,

Figure 2-l—Annual Milk Output Per Cow, 1965-89

15,000

11,2503$g 7,500—

3,750

0 l-l1965 1970 1975 1980 1985 1989

Year

SOURCE: U.S. Department of Agriculture, Economic Research Service,Dairy Situation and Outlook Report, various years, 1965-1990.

three-times-a-day milking, automated feeding, for-age testing, and other technologies that periodicallyenter the market as products of public- and private-sector research and development. Productivity hasalso risen because of constant improvement in thequality of management, which in turn partiallyreflects improved packaging of technology to in-crease output per cow.

With milk consumption per capita growing lessrapidly than output per cow, there has been a gradualnational trend toward reduced cow numbers (seefigure 2-2). This trend was interrupted in the early1980s by government policy that supported the priceof milk at 80 percent of parity in the face of decliningfeed prices.

The U.S. Department of Agriculture (USDA), inreporting production statistics, divides the UnitedStates into 10 farm-production regions (see figure2-3). Changes in output per cow have not beenuniform nationally (see figure 2-4). For example,USDA’s Pacific region had a milk output per cow of18,389 pounds in 1988, whereas the U.S. averagewas 29 percent lower (14,213 pounds). The Pacificregion’s output per cow is 51 percent higher thanthat of the Appalachian region. While climaticconditions contribute to some of these differences,the main factors seem to be progressiveness, philos-ophy, and quality of management—factors that alsoare believed most directly to impact the adoption of

Figure 2-2—Average Number of Milk Cows on Farmsand Total-Milk Production, 1980-89

,., .,, .,,.

. . .. ’

“.

1 0 ~ 1 0 01965 1970 1975 1980 1985 1989

Year— Milk cows ... , Milk production


–17–

18 ● (U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

Figure 2-3—USDA Farm Production Regions

r )WA

- -

rMN

1 . ““ liiii’1

PA

n

● Pacific Regionincludes Alaska and Ha

K - M ( -Southern Plains

SOURCE: U.S. Department of Agriculture, Economic Research Service, 1990.

technologies. These factors in turn are related to theavailability of extension education services, consult-ants, and the infrastructure of input and technologysuppliers. Success in technology adoption will beone of the major factors determining future milkproduction patterns. Other major factors impactingthese patterns include dairy policies, environmentalpolicies, water availability, population pressures,climate, and resource availability.

Economies of Size

Larger dairy farms, as a general rule, experiencelower per-unit production costs. Studies currently inprogress suggest that in traditional milk productionregions, such as the Upper Midwest and the North-east, economies of size (reduced per-unit productioncosts associated with increased farm size) have ledto the establishment of several larger size dairyoperations. And current research suggests that thesedairy operations have the potential to realize evenlarger economies of size.

ind qStatek

SoutheastLA

FL

Regional differences in dairy herd size are associ-ated with different economies of size (see figure2-5). The Pacific coast and Florida lead the Nationwith herd sizes typically in the 500- to 1,500-cowrange and enjoy the lowest production costs per unitoutput. In traditional milk production regions of theUpper Midwest and Northeast, dairies are typicallyin the 50-to 150-cow range, and production costs arerelatively high.

Costs of Production

Substantial regional differences in costs of pro-ducing milk reflect regional differences in output percow as well as in herd sizel (see figure 2-6). TheEconomic Research Service/USDA has estimatedthe cash costs and total economic costs of productionsince 1974. Cash costs depend primarily on the share

l~e USDA cost of production I@OnS & not include the same States as the production regions indicated in figure 2-3. With minor exceptions, tieUpper Midwest cost of production region is equivalent to the Lake States production region.

Chapter 2----Overview of the Dairy Industry ● 19

Figure 2-4—Milk Output Per Cow by Region, 1989

20,000

15,000

5,000

0

LS NE

Key: LS - Lake StatesNE - NortheastPA - Pacific

SOURCE: U.S. Department of Agriculture,

of inputs purchased and milk

PA MT SP NP SE DL CB AP usRegion

MT - Mountain SE - Southeast AP - AppalachianSP - Southern Plains DL - Delta States US - United StatesNP - Northern Plains CB - Corn Belt (national average)

Economic Research Service, 1990.

output per cow.2 For moderate cash cost of $9.07 per cwt. Regions thatexample, in the Southeast, where dairies purchase grow much of their feed have the lowest cash costs.most of their feed inputs (about 50 percent ofproduction costs) and where average output per cow The level of cash costs is significant from a policyis low (12,604 pounds, see figure 2-4), the cash cost perspective. Farmers who are not covering cashaveraged $11.63 per hundredweight (cwt) (see fig- costs have strong economic incentives to shut downure 2-6). Dairies in the Pacific region also purchase their operations. They are either building debt ora high percentage of their inputs but have the highest eroding equity on virtually a daily basis. Notoutput per cow (17,527 pounds), thus yielding a surprisingly, farmers in high-cash-cost regions likely

Wash costs reflect the minimum break-even prices needed to produce in the short run. Subtracting cash costs from the gross value of production leavesnet cash available before replacement of depreciable assets. It excludes income taxes and principal payments.


Figure 2-5—Milk Production Costs Related toHerd Size, for 10 States, 1985

2018 \\1716 I FL

WA/ \

ID

o 200 400 600 800 1,0001,2001,400

Herd size (adult cows)


are the first to complain about milk prices being toolow.

Total economic costs include all fixed and cashcosts of production.3 Fixed costs are highest inregions that grow a large percentage of their feed,have substantial investments in housing and feedstorage (silos), and/or are expanding rapidly (mean-ing high depreciation costs). Thus, while the UpperMidwest and Northeast regions have among thelowest cash costs, they have relatively high totaleconomic costs due to their large housing andfeed-storage investments.

Total economic costs are meaningful from apolicy perspective because they influence the long-run economic viability of a region. If these as wellas cash costs are not offset by high milk prices,farmers will have no incentive to invest. This is thecase in the Upper Midwest and Corn Belt regionswhere the dairy industry is in relative economicstagnation and even decline.

Dairy Receipts

Dairies obtain most of their receipts from milk(about 90 percent) and from the sale of cows (nomore than 10 percent). The price of milk isdetermined nationally and regionally by the interac-tion of government policy, consumer demand, andthe supply of milk. Government dairy programsinclude the Federal and State milk marketing orderprograms and the Federal dairy price support pro-gram.

The milk marketing order programs regulate theprice of milk eligible for fluid consumption: proces-sors are required to pay minimum class prices basedon how the milk is used. The lowest prices are forClass III uses (milk used to manufacture butter,cheese, and nonfat dry milk). Milk used for softproducts (ice cream and yogurt) receives a slightlyhigher Class II minimum price, and milk used forfluid consumption receives a substantially higherClass I minimum price. Dairy producers receive theaverage (blend) of the three class prices weighted bythe share of milk used in each class. Class II and IIIminimum prices are fixed at the average of themarket prices paid by manufacturers in Wisconsinand Minnesota. Class I prices are determined b y t h eMinnesota-Wisconsin prices plus a differential thatincreases with increasing distance from Eau Claire,WI. Thus, Federal milk marketing order Class I(fluid use) prices increase from the Upper Midwestto the South and East.

The Federal Government purchases cheese, but-ter, and nonfat dry milk in quantities sufficient tomaintain market price at a minimum level, estab-lished by the Federal price support program. Reduc-tions in the price of milk may occur for severalreasons: because the government lowers the pricesupport, because the share of milk used for fluidpurposes in a marketing order declines (due to anincrease in supply or a fall in demand), or becausepremiums over minimum order prices (based onsupply and demand conditions) paid by processorsdecline.

This combination of government-administeredand market-determined price relationships is impor-tant because it has a marked impact on the regionaldistribution of milk receipts (see figure 2-7). Re-ceipts are highest in the Southeast (14.87 per cwt)and lowest on the Pacific coast (11.13 per cwt). TheUpper Midwest has slightly higher milk receipts($11 .92 per cwt) than the Pacific region.

Net Income

Profits vary regionally with receipts and costs.The combination of high costs and low prices in theUpper Midwest in 1988 led to a negative cashincome (–$0. 11 per cwt) and an even lower return tomanagement (–$0.62 per cwt) (see figure 2-8).4 The

sTot~ economic costs do not include a return to management.Wash income is the difference betvveen gross value of production and cash expenses and capital replacement.

Chapter 2-----Overview of the Dairy Industry ● 21

Figure 2-6—Regional Differences in Costs of Producing Milk, 1988

UM NE PA SP SE CB AP

= C a s h = T o t a l e c o n o m i c

Key: UM - Upper Midwest PA - Pacific CB - Corn Belt

NE - Northeast SP - South Plains AP - Appalachia

SE - Southeast

SOURCE: U.S. Department of Agriculture, Economic Research Service, Economic Indicators of the Farm Sector, 1989.

Corn Belt also experienced a loss in returns tomanagement (of $0.27 per cwt). The highest returnregions were the Southeast and Appalachia. Despitehigh costs, the Southeast realized higher returns dueto favorable treatment under the Federal milkmarketing order system. The Pacific region realizeda favorable return despite having the lowest receipts.This reflects the overall efficiency of relativelyindustrialized production on the west coast.

The income situation improved for dairy farms inthe traditional milk producing regions in 1989 and1990 when farm milk prices increased. These farmsexperienced positive returns in those years. How-ever, in the early months of 1991 prices declinedsignificantly and are expected to fall by 15 to 20percent for the year compared to 1990. Dairy farmsin the traditional milk producing regions are ex-pected to lose equity under these conditions. Farmsin the nontraditional areas, such as the Pacific

region, are expected to operate much closer to theirbreak-even point.

Regional Production Changes

Sustained regional differences in profit lead toshifts in the geography of production. The largestincreases in milk production have been in the Westand Southwest, where marketing have risen bynearly 40 percent since 1980 (see figure 2-9). Milkproduction in the traditional dairy areas of the LakeStates and Northeast has increased by 6.5 and 4.0percent, respectively. The Corn Belt, which consist-ently has had the lowest net returns, experienced aproduction increase of only 4.9 percent.

However, no region is homogeneous. During the1980s, for example, centers of rapidly increasingproduction developed within regions (i.e., centralTexas and southern Georgia). In the late 1980s,persistent production declines occurred in Minne-


Figure 2-7—RegionaI Differences in Receipts From Dairy Farm Sales, 1988

16.14

Paoific South Plains Appalachian Upper Midwest Corn Belt Southeast Northeast

Region

m cows = T o t a l

SOURCE: U.S. Department of Agriculture, Economic Research Service, Dairy Situation and Outlook Report, 1989.

sota—traditionally one of the largest dairy States,Because of the higher cost conditions associatedwith smaller dairies (see figure 2-5), and low profits(see figure 2-8), significant sections of the LakeStates may have lost its comparative advantage toother regions. Questions arise as to whether tradi-tional Federal order pricing institutions, which relyon Minnesota and Wisconsin as the base point forpricing milk (the M-W price), are appropriate intoday’s milk industry. The answers to these ques-tions are complex and merit further study and debate(2, 3).

While the national share of milk production in theLake States and the Northeast has declined by atleast 2 percent since 1980, these two regions stillproduced almost half of the Nation’s milk supply in1989 (see figures 2-10 and 2-1 1). If these regions areto maintain their role as “dairy States, ” majorchanges in scale of operation, levels of technologyadoption, support for dairy research and extension,and, perhaps, dairy policy may be required.

Demand Changes

Changes in demand may be as important aschanges in supply in determining the future courseof the dairy industry. Shifts in population toward theWest and South have favored increased milk produc-tion in these regions.

There have also been major changes in demandfor individual dairy products, such as sharplyreduced butter consumption, increased lowfat (2.0percent butterfat or less) milk consumption (seefigures 2-12 and 2-13), and increased cheese con-sumption. The shifts from whole milk to lowfat milkand the rapidly rising cheese consumption have beenparticularly dramatic. The trend away from consum-ing whole milk likely reflects increasing consumerconcerns about calories, fat, and cholesterol con-sumption.

Cheese is overwhelmingly the bright spot in termsof dairy product demand. While American-stylecheeses (predominantly cheddar) have experiencedsubstantial growth (see figure 2-14), the demand for

Chapter 2-Overview of the Dairy Industry ● 23

Figure 2-8—Regional Differences in Net Income (profit), 1988

3

2

g ,&

o

-1

.

Upper ’Midwest Northeast Pacific South Plains SoutheastI

Corn ‘Belt Appalachian

~ C a s h m M a n a g e m e n t

SOURCE: U.S. Department of Agriculture, Economic Research Service, Dairy Situation and Ouf/ook Report, 1989.

other cheeses (predominantly Italian-style) has growneven more rapidly and consistently (see figure 2-15).Italian-style cheese demand is largely a result of therapidly growing convenience and fast food (pizza)market. Therefore, cheeses have capitalized onconsumer trends toward microwave convenienceand eating out. Other dairy products have not bene-fited as much from these changing market trends.

MAJOR DAIRY POLICY ISSUESThe milk industry may well be the most highly

regulated of any in the United States. A complexsystem of health, food safety, and labeling regula-tions exist at the Federal, State, and local levels.These regulations reflect the perishability of theproducts, which are ideal media for the growth ofmicroorganisms; the potential for their adulteration;the potential for drugs and/or chemical residuesrelated to milk production processes; and the poten-tial for variations in the nutritional value of products.Over time, more stringent water quality standards

. . . . . . . ,.,. . .have been placed on dames, affecting the manage-ment of animal wastes and runoff. In some instances,air pollution regulations have also been imposed.Overlying the EPA- and FDA-oriented regulations isan extensive set of Federal and State milk-pricingregulations. These include a network of Federal milkmarketing orders, State milk marketing orders, andthe milk price-support programs discussed earlier.The following section summarizes some of themajor issues of dairy policy and how policy consid-erations and regulatory mechanisms may interact toshape the industry’s future.

Butterfat Surplus

Nutrition- and diet-conscious consumers increas-ingly have shunned higher butterfat products. (In-creased consumption of premium ice cream is one ofthe few exceptions to this trend.) As a result ofdeclining demand, a butterfat surplus has developed,although overall, milk supply and demand have beenin relative balance during the late 1980s and early


Figure 2-9—Change in Milk Production by Region, 1980-89

LS NE PA MT SP NP SE DL CB AP USRegion

Key: LS - Lake States MT - MountainNE - Northeast SP - Southern PlainsPA - Pacific NP - Northern Plains


1990. Commodity Credit Corporation (CCC) price-support purchases of butterfat have continued evenas USDA has had to enter the commercial market tosatisfy cheese demand for its child nutrition pro-grams. USDA has attempted to remedy the butterfatsurplus problem by consistently lowering the priceof butterfat to stimulate consumer demand whileholding the cheese price constant. This strategy hasbeen only partially successful. Surpluses in cheeseand nonfat dry milk (NFDM) have completelydisappeared since 1988 and butterfat surplusescontinue (see figure 2-16).

Future trends could further complicate the butter-fat surplus problem. Lowfat cheeses and lowfat icecream are capturing a larger part of that market. Fatsubstitutes are also being developed for use in icecream, and perhaps other dairy products. Thesetrends represent a two-edged sword; they could

SE - Southeast AP - AppalachianDL - Delta States US - United StatesCB - Corn Belt (national average)

result in increased total demand for dairy products,yet further reduce butterfat demand.

Research progress in removing cholesterol frombutterfat may be the solution to the problem.However, for reasons of diet and health, consumersare concerned over a range of issues: total fatconsumption, calorie intake, saturated fat consump-tion, and cholesterol intake. Fat substitutes aggra-vate the butterfat problem whether or not cholesterolis effectively removed from butterfat.

From the above analysis, it seems that thesolution to the butterfat problem may be to reducebutterfat production. This can be partially accom-plished by changing feeding practices in the shortrun and by breeding for reduced butterfat in the longrun. Pricing incentives must exist for either of thesepotential solutions to occur. Milk currently is priced,to a large extent, on the basis of butterfat content.

Chapter 2----Overview of the Dairy Industry ● 25

Figure 2-10—Share of Milk Production by Region, 1989

40

30

20

10

0

27.4

LS NE PA MT SP NP SE DL CB AP

Region

Key: LS - Lake States MT - Mountain

NE - Northeast SP - Southern PlainsPA - Pacific NP - Northern Plains


Shifts to pricing on the basis of protein or nonfatsolids are possible and practiced in some markets.Industry initiatives and USDA leadership are re-quired to obtain widespread adoption of suchinnovative pricing alternatives.

Milk Price Support

Related to the butterfat issue is the mechanism foradjusting the milk price-support level. From 1949through 1981, the milk price-support level was set asa percent of parity (a price that will give a farmer thesame purchasing power he/she had in abase period).Generally, the Secretary of Agriculture was given adiscretionary range of 75 to 90 percent of paritywithin which to set the milk price-support level. In1981, a trigger mechanism relating changes in theprice support to the level of government purchaseswas adopted. Under the 1985 farm bill, the milkprice support was raised in $0.50 per cwt increments

SE - Southeast AP - AppalachianDL - Delta States US - United StatesCB - Corn Belt (national average)

when CCC purchases of dairy products were pro-jected for the following year to be less than 2.5billion pounds and decreased at the same rate whensuch purchases were projected to be greater than 5.0billion pounds. The pounds were measured on abutterfat-milk-equivalent basis. The butterfat basisbecame an issue when CCC cheese and dry milkpurchases ended in 1989, and as butterfat purchasesincreased.

The 1990 farm bill dairy policy provisions frozethe price support at $ 10.10 per cwt through 1995. Fordeficit reduction purposes, it assesses $0.05 per cwtfor all milk produced in 1991 and $0.1125 in1992-1995. This assessment is refunded on proofthat the farmer’s milk production was not increasedover the previous year. The Secretary is required toprepare a report with recommendations to Congresson how it plans to limit growth of CCC purchases ofdairy products by August 31, 1991 with exclusions


5I

-2.5

Key:

Figure 2-1 l—Change in Share of Milk Production by Region, 1980-89

4.1

LS NE PA MT SP NP SE DL CB AP

Region

LS - Lake States MT - Mountain SE - Southeast AP - AppalachianNE - Northeast SP - Southern Plains DL - Delta States US - United StatesPA - Pacific NP - Northern Plains CB - Corn Belt (national average)


of cow slaughter and price-support reduction op-tions. If Congress fails to enact dairy legislation by1992 and CCC purchases are expected to exceed 7billion pounds, an assessment covering the full costof the CCC purchases over 7 billion pounds isauthorized.

As a result of these decisions, the following policyissues are pending as the industry enters the 1990s:

the potential for effective demand expansionprograms-the 1990 farm bill authorizes aprocessor-funded demand expansion check-offif approved by referendum,

the potential for developing effective tempo-rary supply and/or management systems that donot lead to industry inefficiencies and rigidityin production patterns,

what to do about declining demand for butterand the accumulation of butter stocks, and

how much discretion the Secretary of Agricul-ture should have in determining the provisionsof dairy policy.

Price Instability

As the price-support level has declined, the priceof milk and manufactured dairy products has be-come more variable. This instability is the result ofthe interaction of an inelastic supply and demand formilk. With lower price supports, instability isparticularly evident in autumn when milk suppliesare often relatively short. For example, in 1989, theMinnesota-Wisconsin (M-W) price rose from a lowof $11.20 per cwt in May to $15.10 in December. Itthen fell back to $12.20 in March 1990 (see figure2-17). By September 1990, the M-W was approach-ing the milk price-support level of $10.10 per cwt.

From an industry perspective, one of the benefitsof dairy policy has been the stability provided by theprice support and Federal order program. Currentpredictions regarding M-W prices are widely varia-ble. While some argue that the milk price supportwill once again determine the price of milk under the1990 farm bill, others suggest that significantsegments of the milk industry cannot survive suchlow prices. Planning for the future has becomeexceedingly difficult for producers, processors, USDA,and Congress.

Chapter 2-Overview of the Dairy Industry ● 27

Figure 2-12—Actual and Projected Consumption ofLowfat Milk, 1950-93

Figure 2-1 3—Market Share of Lowfat MiIk as a Percentof Total Fluid Milk Consumption

60 m l40, . ~ : : : :m,,! ,.

: , . ”,..

L, . ”

.“ :

I45

150 ~ ~ ~ ~ : : : :

1950 1955 1960 1965 1970 1975 1980 1985 19901993Year o

-- Projected — Actual 1970 1975 1980 1985 ‘1 990Year


Figure 2-15—Actual and Projected Consumption of

SOURCE: Reginald Adamus and Emerson Babb, “Projections of U.S.Dairy Product Consumption, 1989 -1993,” Food and ResourceEconomics Department, University of Florida, 1990.

Figure 2-14—Actual and Projected Consumption ofAmerican Cheese, 1950-93 -

Other Cheese3.0

r2.5 –

I I I I I I I 11950 1955 1960 1965 1970 1975 1980 1985 19901993 1950 1955 1960 1965 1970 1975 1980 1985 19901993

Year-- Projected — Actual


Year--- Projected — Actual


Milk Production Controls

During the 1980s, a dairy diversion program anda dairy termination (buyout) program were imple-mented as means of reducing milk production—inaddition, the price-support level was lowered from ahigh of $13.10 per cwt to $10.10 per cwt. Thetermination program was considerably more effec-tive than the diversion program but also morecontroversial because of its negative impact on theprice of beef.

The combination of the termination program andlower price supports has brought milk productioninto relative balance with consumption. With thisaccomplished, attention has turned to the potential

need for production controls should surpluses onceagain accumulate. This could happen if milk supplyincreases sharply due to rapid adoption of a newtechnology, and/or if demand falls due to negativeconsumer reaction to the same technology. Afterpassage of the 1990 farm bill, options for productioncontrols (or inventory management) would appear toinclude:

utilization of a combination assessment andproduction control system;implementation of some type of quota system;andin concert with either assessments and/or quo-tas, an aggressive program to expand domesticand foreign demand.

●

●

●

292-860 O - 91 - 2


Figure 2-16—Uncommitted Government DairyProduct Inventories, 1985-90

1,200

1,000

g 800

200}-Q- -. .-------

0 1 i i - i”” ‘II

1985 1986 1987 1988 1989 1990— Butter . . . . Cheese ---- Non-fat dry milk

SOURCE: U.S. Department of Agriculture, Economic Research Service,Dairy Situation and Out/ook Report, various issues, 1985-1991.

Federal Milk Marketing Orderss

Historically, decisions regarding Federal milkmarketing orders have been left largely to theSecretary of Agriculture. However, the 1985 farmbill increased Class I milk prices, emphasizingmarkets distant from the Upper Midwest. In 1988,the General Accounting Office (GAO) published astudy indicating that Federal orders tended to favorregions to the South and East at the expense ofproducers in the Upper Midwest. The GAO reportrecommended a gradual but progressive successionof steps to reduce the level of regulation in Federalorders.

In light of these developments, the Secretary ofAgriculture has initiated a series of national hear-ings, which were to be completed by the end of 1990.The 1990 farm bill mandates decisions on Federalorders by 1992. Major issues identified for thehearings include:

●

●

●

●

●

the level of class prices for milk,the number of classes and products included,the geographic structure of prices including thepotential for multiple basing points,the need for uniformity in order provisions, andthe appropriate basis for new Federal orderclass prices (as opposed to the Minnesota-Wisconsin series).

Figure 2-17—Minnesota-Wisconsin (M-W) Price andMilk Price-Support Level, January 1987 to April 1990

20

zv&

15

10

5

0\ i i i ‘I i

1/87 6187 1/88 6188 1/89 6/89 1/90 4/Month/year

— M - W . . . Price support

SOURCE: U.S. Department of Agriculture, Economic Research Service,Dairy Situation and Outlook Report, various issues, 1988-1991.

Basic Formula Price

Since the 1960s, the Minnesota-Wisconsin priceseries has served as the basic formula price used tomove or change the level of all Federal order milkprices. The M-W price has also served as the guidefor determining if the milk price support objectiveset by Congress or the Secretary of Agriculture hasbeen realized.

The M-W price series is the average price paid forGrade B milk (which can only be used for processingdairy products such as cheese and butter) byMinnesota and Wisconsin processing plants. How-ever, the volume of Grade B milk produced inMinnesota and Wisconsin has declined to where theM-W price series may not reliably reflect the forcesof supply and demand for this milk

In 1989, the GAO concluded that alternatives tothe M-W price series should be evaluated andimplemented. It recommended two options:

. a product formula based on the market pricesfor butter, NFDM, and/or cheese; and

. a competitive price for Grade A and Grade Bmilk.

5A ~e~ation issued by tie Secre[q of Agl-icul~e Specifying minimum prices and conditions under which mi~ can be bought and sold within aspecified geographic area.

Chapter 2---Overview of the Dairy Industry ● 29

Emerging Technology

The dairy industry will be among the first to adoptmany of the newly emerging technologies from thebiotechnology era. Considerable controversy sur-rounds the potential use of these technologies. Thisis especially true of bovine somatotropin (bST), nowundergoing review by the Food and Drug Adminis-tration (FDA). Concerns about food safety, animalsafety, the manufacturing process to produce thetechnology, and the economic impacts on the dairyindustry have all coalesced to make bST the focus ofa controversy of unprecedented magnitude in thedairy industry.

A number of actions have already been taken toslow or stop the use of bovine somatotropin. TwoStates have declared a moratorium on the use of thetechnology if approved by FDA. Four States haveenacted or are considering enacting labeling require-ments on dairy products produced from bST-supplemented milk. And consumer groups have

successfully pressured some large retail food storesnot to market dairy products produced with thistechnology.

The next chapter provides information and ananalysis of the issues relevant to this technology.Subsequent chapters discuss other major emergingtechnologies that should become available in thisdecade for the dairy industry.

1.

2.

3.

CHAPTER 2 REFERENCESU.S. Department of Agriculture, Dairy: Backgroundfor 1990 Farm Legislation, Economic Research Serv-ice, staff paper AGES9020 (Washington, DC: 1990).U.S. General Accounting Office, Milk MarketingOrders: Options for Change, GAO/RCED-88-9 (Wash-ington, DC: March 1988).U.S. General Accounting Office, Milk Pricing: NewMethod for Setting Farm Milk Prices Needs To BeDeveloped, GAO/RCED-90-8 (Washington, DC: No-vember 1989).

Chapter 3

An emerging Technology:bovine somatotropin

ContentsPage

STATE OF THE ART . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33PRODUCTION RESPONSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35NUTRIENT REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37BOVINE REPRODUCTIVE PERFORMANCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37MILK AND MEAT COMPOSITION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38MECHANISMS OF ACTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40BOVINE HEALTH AND STRESS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41ONFARM ENVIRONMENTAL POLLUTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... ...43INSTITUTIONAL INVOLVEMENT IN bST RESEARCH . . . . . . . . . . .. .. ... ... ......43TIMING OF COMMERCIAL INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44PRODUCT LABELING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..45CHAPTER PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

FiguresFigure Page3-1. Effect of Quality of Management on Milk Response of Dairy Cows Receiving

bovine Somatotropin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ,... 353-2. Temporal Pattern of Milk Yield, Net Energy Intake, and Net Energy Balance

During a Lactation Cycle in Dairy Cows.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363-3. Efficiency Gains by Reduction in the Proportion of Nutrients

Used for Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

TablesTable Page3-1. Impact of bovine Somatotropin on Animal Numbers, Feed Requirements, and

Waste Production of Dairy Cows to Achieve 1988 U.S. Milk Production . . . . . . . . . . 353-2. Effect of bovine Somatotropin on Specific Tissues and Physiological Processes

in Lactating COWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413-3. Comparison of Theoretical Gains in Milk Yield Per Cow for

. . . . . . . ,.

Different Dairy Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Chapter 3

An Emerging Technology: bovine Somatotropin

The U.S. dairy industry has a rich history oftechnological advances underpinningg steady andsignificant increases in the efficiency and economicreturns of milk production. Progress in biotechnol-ogy and information technology will carry this trendinto the 1990s. Through the application of biotech-nology, farmers will be able to manipulate to anunprecedented degree herd reproduction, genetics,and the physiological variables that affect theiranimals’ productive efficiency.

Genetically superior (more productive) animalsare distinguished by their regulation of nutrients;new concepts of how this occurs have recently beenestablished (6,10,1 1). Potentially one of the mostsignificant of the new biotechnologies is bovinesomatotropin (bST). Recent work has demonstratedthat somatotropin exerts a key control over nutrientuse. When administered exogenously, it markedlyimproves the productive efficiency (milk per unit offeed) in lactating cows. It does so by coordinatingthe metabolism of body tissues such that morenutrients are used for milk synthesis.

However, a number of questions involving humansafety, animal safety, economic concerns, and ethicssurround bST. Lay articles focusing on human andanimal safety issues of bST imply that little researchhas been done. In fact, the scientific literature on bSTcontains at least 1,000 studies involving some10,000 dairy cows.l This chapter provides anoverview of bST technology and the concernssurrounding it. It assesses the validity of theseconcerns based on currenton dairy industry trends.

STATE OF

scientific knowledge and

THE ART

Somatotropin, a hormone, was discovered about50 years ago. Initial investigations showed thatwhen rats were injected with a crude pituitaryextract, growth rate was increased (only later didscientists discover that milk yield of lactatinganimals increased as well). This extract factor wascalled somatotropin from the Greek derivationmeaning “tissue growth. ” Based on this derivation,

somatotropin is sometimes referred to as growthhormone or GH.

Somatotropin is produced by the anterior pituitarygland, a small gland located at the base of the brain.Like any hormone, it is transported in the blood-stream to the various body organs where it exerts itsbiological effects. In effect, it acts as a chemical linkbetween different cells and organs of the body. Theterm “hormone” has taken on negative connota-tions in recent times, primarily due to steroid abuseby athletes. However, the chemistry of hormones isas diverse as their biological functions. For example,vitamin D (with which pasteurized milk is fortified)is a steroid hormone.

Somatotropin is also a protein, unlike steroids,which are nonprotein hormones. Like all proteins, itis composed of amino acids. (There are 20 differentamino acids, which combine in specific sequences toform some 10,000 different proteins in the body.Amino acids are analogous to letters in an alphabet,which combine to form a diverse vocabulary ofwords.) The amino acid sequence of somatotropin isknown for many species, including cattle (41,79).Bovine somatotropin can be either 190 or 191 aminoacids long and either of two different amino acids(leucine or valine) occupy position number 126 inthe sequence (80). Thus, four different variants ofbST are produced naturally. Typically, the pituitaryproduces equal amounts of the 190 and 191 amino-acid bST. About two-thirds of the total bST pro-duced has leucine at position 126, while the remain-ing one-third has the amino acid valine at position126.

Artificially introduced somatotropin must beinjected to be biologically active. If somatotropin isgiven orally, it is broken down by digestive trackenzymes to amino acids for absorption. This is truefor all dietary proteins and large protein hormones inall species. Just as human diabetics must take insulininjections (insulin, another protein hormone, is alsoinactive if taken orally), humans deficient in soma-totropin must take injections of human somatotropin(hST). Studies have demonstrated that when fed torats, bST was inactive even at a daily dose (units/

IBecause of tie l~ge quanti~ of res~ch on bST, this chapter will mainly cite review articles of these studies.

–33–


body weight) equivalent to 2.3 million times what ahuman would be exposed to in five 8-ounce glassesof milk (38,67).

Somatotropin is referred to as “species limited”in the scientific literature. This means that there aredifferences in the ability of somatotropin from onespecies to elicit biological effects when injected inother species. To have a biological effect, a proteinhormone must first bind to a specific receptorlocated on the cell surface. The amino acid sequenceof somatotropin gives it a unique three-dimensionalshape, which determines whether the protein will beable to bind to tissue somatotropin receptors andelicit a biological response.

Some 25 years ago it was discovered that certaintypes of human dwarfism were due to an inadequatepituitary production of somatotropin. Because hSTwas scarce, physicians conducted an extensive seriesof studies in which patients were treated withinjections of bST. These clinical studies uniformlydemonstrated that bST does not elicit any of itsnormal biological actions in humans even if injected(20,38,41,79). Somatotropin isolated from the pitui-tary glands of sheep, pigs, and whales was alsoineffective in humans, Biological activity in humansis only observed if somatotropin from primates isused.

The reason for bST’s lack of effect in humansbecame clear when its amino acid sequence wasidentified; the sequence of bST, which gives it itsthree-dimensional shape, differs by about 35 percentfrom that of hST (79). Thus, bST is not able to bindto the somatotropin receptors of human tissues(36,38). In contrast, ovine somatotropin and bSTdiffer in only one amino acid position so bST isbiologically active if given to sheep.

Recombinantly derived bST products differ slightlyfrom the bST produced by the pituitary gland in thata few extra amino acids may become attached to theend of the bST molecule in the manufacturingprocess. The number of extra amino acids variesfrom 1 to 8 depending on the particular manufactur-ing process (38). For some manufacturing processes,no additional amino acids are produced. Some claimthat for processes that produce extra amino acids,and the fact that recombinantly derived bST isproduced by bacterial ribosomes, it renders the

product hyperpotent to cows and dangerous tohumans (23,46,64). However, the additional fewamino acids on the end of the protein do not alter thebiological activity of bST in dairy cows or the lackof activity of bST in humans because the three-dimensional shape of the active part of the moleculeis not changed (36,79). This shape, moreover, isdetermin ed by the sequence of amino acids, not bywhether or not bacterial ribosomes were used for thesynthesis.

The first bST research with lactating cows wasreported in 1937 by Russian scientists (l). Withadvances in protein chemistry, somatotropin prepa-rations gradually improved in purity and severaldozen studies have since been conducted with dairycows. Particularly significant were a series of studiesin the 1940s by scientists in the United Kingdom(81) and later studies by Brumby and Hancock (13)and Machlin (49).

Prior to the 1980s, bST research progress re-mained slow for two reasons. First, bST availabilitywas limited to what could be extracted from thepituitary glands of slaughtered animals. Thus, onlya small number of cows could be treated forshort-term studies. Second, because bST w a sthought to act by acutely stimulating the use of bodyfat reserves, scientists believed it would work onlyin fat cows with a low milk yield. Thus, only lowproducing cows (generally less than 7 kilogram (kg)milk/day) were studied. It was assumed that bSTwould cause ketosis2 and adverse health effects inhigh producing cows.

Scientists at the National Institute for Research inDairying in England, and others at Cornell Univer-sity, in the late 1970s began to work on bST. Bothgroups concluded that the physiological basis forgenetically superior cows (i.e., those more efficientin milk production) was better use of absorbednutrients (10,12). Based on new concepts of howanimals regulate the use of nutrients (7), both groupshypothesized that somatotropin could play a key rolein nutrient regulation and that the previously pro-posed mechanism of action (acutely stimulated useof body fat) was wrong. Over the last decade theseconcepts have been applied to research with soma-totropin and the biology of nutrient use duringgrowth, pregnancy, and lactation for many species.

2A me~bolic disorder that OCCUrS when production of ketones exceeds the ability of the body to use them. Occurs in dafi cows when the need forglucose exceeds the production of glucose.

Chapter 3--An Emerging Technology: bovine Somatotropin ● 35

SOURCE: D.E. Bauman, “Bovine Somatotropin: The Cornell Experience,”Proceedings of the National Invitational Workshop on BovineSomatotropin, pp. 46-56, USDA Extension Service, Washing-ton, DC, 1987.

Initial investigations with cows used pituitary-derived bST. After landmark breakthroughs inbiotechnology, in 1982, the Cornell scientists con-ducted the first study with dairy cows using recom-binantly derived bST produced by Monsanto Co.and Genetech Co. (8). Since that time the quantityand scope of research with bST has increasedexponentially.

PRODUCTION RESPONSEQuality of management will be the major factor

affecting the magnitude of lactation response to bST(4,16,17,57). This concept is illustrated qualitativelyin figure 3-1. Across 45 field trial studies conductedin the United States, Europe, and Africa a correlationof 0.58 was observed between pretreatment groupmilk yield (an indication of management quality)and kg milk response to bST (57). A similarrelationship between management quality and gainsrealized is observed for other technologies, such asartificial insemination (AI) with semen from supe-rior sires.

Facets that contribute to the quality of the overallmanagement program include the herd health pro-gram, milking practices, nutrition program, andenvironmental conditions. Inadequate managementcan result in a near zero response to bST supplement(51,52,57). McCutheon et al. (51) studied bST’seffects as the quality of herd nutritional managementvaried over the course of the 26-week treatment

period. Cows were fed only pasture, and milkresponses to bST were greatest (+18 percent) in thespring when pasture supply was adequate, declinedto zero during the summer drought, and were againsignificant during the fall. As this illustrates, bovinesomatotropin is not magic. If cows are given aninadequate amount of feed or fed a diet that is notnutrient-balanced, the magnitude of the response tobST will decrease accordingly (see figure 3-l).

While the milk response to bST on an individualfarm will vary according to quality of management,a reasonable expectation is that successful adopterswould experience an average gain in productivity of12 percent. This gain in productivity could lead tosubstantial savings in dairy cattle feed nationally.Assuming 100-percent adoption and milk produc-tion at 1988 levels (26), then decreased cow numbers(10.7 percent) and increased productive efficiencywould translate into annual savings in dietary energyequivalent energetically to 2.5 x 109 kg of corn grain.Annual savings in dietary protein supplementswould be equivalent to 5.6 x 107 kg of soybean meal(see table 3-l). These savings in feedstuffs represent

Table 3-1—impact of bST on Animal Numbers, FeedRequirements, and Waste Production of Dairy Cows

To Achieve 1988 U.S. Milk Productiona

Variable Impact of bSTb

Animals:Cow numbers . . . . . . . . . . . . . . . .Milk yield per cow . . . . . . . . . . . . .

Feed: c

Energy equivalent ascorn grain . . . . . . . . . . . . . . . . . .

Protein supplement equivalentas 44% soybean oil meal . . . . .

Waste:Manured. . . . . . . . . . . . . . . . . . . . .Urinee . . . . . . . . . . . . . . . . . . . . . .Urinary nitrogene. . . . . . . . . . . . . .Methanef . . . . . . . . . . . . . . . . . . . .

Decrease by 10.70/.Increase by 12.0?4.

Decrease by 2.5 x 109 kg

Decrease by 5.6 x 107 kg

Decrease by 6 x 109 kgDecrease by 8 x 109 LDecrease by 8 x 107 kgDecrease by 8 x 1010 L

aU. S. 1988 milk production values were 10.24x 10G cows, 6,460 kg milk percow, and 66 x 109 kg total milk production (26).

bAssumed I oo.percent adoption and that use would increase avera9eannual milk yield per cow by 12 percent. If commercially approved,expected impact would be less because technology rarely achieves100-percent adoption.

cBased on nutrient requirements for dairy cows averaging 650 kg bodyweight and producing milk of 3.5-percent fat content (54).

dBaSed on an average diet composition of 1.62 Meal net energy/kg, a dietdigestibility of 65 percent, and fecal dry matter of 16 percent (72).

eBased on a daily urine production of 20 L per cow with 1-percent nitrogenin urine (72).

fAssumed that methane production represents 5 percent of gross ener9Yintake (72).

.SOURCE: D.E. Bauman, “Bovine, Somatotropin: Review of an EmergingAnimal Technology,” commissioned background paper pre-pared for the Office of Technology Assessment, Washington,DC, 1990.


maximal estimates because commercial adoption ofnew technology is rarely 100 percent (see ch. 5).

Milk yield gradually increases the first few daysof bST treatment, peaking at about the sixth day(37,56). A maximum milk response is achieved at abST dose (daily injection) of about 30 to 40milligrams (mg) per day; no further increase occurseven at doses several-fold higher. Most productiontrials have used a bST dose between 10 to 50 mg perday. Exogenous bST must be introduced every day(by injection or via a prolonged-release formulation)in order to maintain an augmented milk response.This is because bST is cleared rapidly from thebloodstream and is not stored in the body. Removalis by protein breakdown to amino acids—a normalbody function. Several prolonged-release formula-tions have been developed and are administered bysubcutaneous injection at intervals ranging from 2 to4 weeks (17).

Response to bST varies according to stage oflactation (see figure 3-2). In general, the milkresponse is small or negligible when bST is adminis-tered in early lactation (the interval immediatelypostpartum, prior to peak milk yield) (4). Thebiological basis for this low response relates to thenutrition/endocrine status of the animal during thisinterval. In contrast, substantial increases in milkyield occur when bST is administered after peakyield of milk is attained. Lactational responses tobST have been reported for all dairy breeds exam-ined, including North American and Europeanbreeds as well as Murrah buffalo, and for animals ofdifferent parity (lactation number) and geneticpotential (16,17,56).

Marked improvement in persistency of lactationoccurs in cows receiving bST (16,17,56). Thegreater overall milk yield with bST supplementationoccurs in part because of an immediate increase inmilk yield, but mainly because of a reduction in thenormal decline in milk yield that occurs as lactationprogresses.

Cows treated with bST show a range of responses.In a few instances this has been cited as evidence ofindividual variation in response. However, this ismisleading. All studies with bST have shown thatthe yield variations within bST-supplemented

groups is similar to that of untreated groups (5,57).Thus, to a large extent, all cows in a herd respond tobST in a fairly similar manner. The bST-supple-

Figure 3-2—Temporal Pattern of Milk Yield, NetEnergy Intake, and Net Energy Balance During

a Lactation Cycle in Dairy Cows

Panel A-Milk yield (4% fat-corrected milk)

15 I\

35 r Panel B-Net energy intake>

151

8 Panel C-Net energy balance

I~ z - - - ~

, < ‘ .$ 4

,/ ‘\\u If

h o -----------------Q

-12 L

o 8 16 24 32 40Week of lactation

Cows averaged 9,534 kg milk (21,000 Ibs) over the first 305 daysof lactation. Typically, daily milk yield peaks during the first monthafter parturition (birth of the calf) and then progressively de-creases through the remainder of the lactation cycle (Panel A).The rate of this decline in daily milk yield is referred to as thepersistency of lactation. Typically, voluntary feed intake increasesgradually over the first few weeks of lactation and peaks about 6to 10 weeks after parturition (Panel B). Dairy cows are generallyin negative energy balance during the first portion of the lactationcycle (Panel C). During the first month of lactation for these cows,the body reserves being utilized (i.e., net energy deficit) wereenergetically equal to about one-third of the milk produced. Undernormal management conditions the daily cow is overfed (positiveenergy balance) during the last third of lactation to allow forreplenishment of body energy stores needed to support the nextlactation cycle (dashed line, Panel C).SOURCE: D.E. Bauman and W.B. Currie, “Partitioning of Nutrients During

Pregnancy and Lactation: A Review of Mechanisms InvolvingHomeostasis and Homeorhesis,” J. Da”ry %“. 63:1514-1529,1900.


mented cow that ‘‘appears’ to be a low respondersimply matches the low producing control cow.

NUTRIENT REQUIREMENTS

Nutrient requirement tables are unchanged forbST-supplemented dairy cows(11, 16). The basis forthis is two-fold. First, digestibilities of dry matter,carbon, nitrogen, and energy are not altered whenlactating cows are receiving bST. Second, bioener-getic studies have demonstrated that bST does notalter energy expenditure for maintenance, or for thesynthesis of a unit of milk. Nutritional needs formaintenance, milk production, pregnancy, and toreplenish body reserves over a lactation cycle for acow producing 10,000 kg milk per year are the sameregardless of whether she received bST or not.

No special diets or unusual feed ingredients areneeded to obtain a milk response to bST: substantialmilk responses have been observed on diets rangingfrom pasture to forage concentrates.3 Overall, how-ever, the dairy cow receiving bST has a greater totalnutrient requirement because she is producing moremilk. She has a higher productive efficiency becausea larger proportion of her total nutrient intake is usedto make milk (see figure 3-3).

Voluntary intake of feed increases in bST-sup-plemented dairy cows, beginning after a few weeksof bST supplementation, and persisting throughoutthe interval of bST use. This has been consistentlyobserved across a wide range of diets (16,17,56).Overall, cows supplemented with bST adjust theirvoluntary intake in a predictable manner related tothe extra nutrients required for the increased produc-tion of milk. The magnitude of increase in feedintake is dependent on the response in milk yield andthe energy density of the diet (17). It is expected withcurrent feed costs that use of bST for dairy cows will,on average, lead to a predictable increase in theenergy density of the diets used (increased ratio ofconcentrate to forage). This is because income overfeed cost increases with level of milk productioneven though the cost for higher energy ingredients isgreater. However, as noted, bST increases milk yieldeven when pasture is the only dietary ingredient.

Figure 3-3-Efficiency Gains by Reduction in theProportion of Nutrients Used for Maintenancea

❑ Milk

❑ Maintenance30

20

10

0Control bST

aFor this example, the hypothetical control cow Produced 6,818 kg (15,000lb) of milk in a’lactation-and use of bST increased milk yield by 20 percent

SOURCE: D.E. Bauman, “Biologyof Bovine Somatotropin in Dairy Cattle,”Advanced Technologies Facing the Da”ry lndustry:bST, CornellCooperative Extension Animal Science Mimeograph Series#133 (Ithaca, NY: Cornell University, 1989), pp. 1-8.

BOVINE REPRODUCTIVEPERFORMANCE

Of special interest are the effects of bST onreproductive variables such as conception rate (serv-ices per conception), pregnancy rate (proportion ofcows becoming pregnant), and days open (days fromparturition to conception). Normally, variation inreproduction variables is large. Effects of bSTsupplementation are small enough that large datasets are needed to allow definitive conclusions.Several reviews have summarized many of thestudies on reproduction variables (28,55,59). Ingeneral, these summaries indicate that bST supple-

slJse of somatotropin for growing animals will require modification of the maintenance and growth components of the nutrient requirement tablesbecause of tbe shift in the type of growth (increased prote@ decreased lipid).

38 ● U.S. Dairy Industry at a crossroad: Biotechnology and Policy Choices

mentation results in a decrease in pregnancy rate butthat conception rate is not altered. For example,Ferguson and Skidmore (28) found that pregnancyrate (n >3,000 cows) was 89.2 percent for controlsand 81.2 percent for bST-supplemented cows (bSTdose ranged from 5 to >200 mg per day). Days openincreased a few days in bST-supplemented cowsaccording to most studies.

The studies that identified changes in pregnancyrate and days open for bST-supplemented cowsfollowed management practices geared to achieve a12- to 13-month calving interval. Thus, period ofbreeding (commencing 50 to 60 days postpartum)would generally have coincided with the earlyperiod of bST supplementation when milk yield hadincreased but before voluntary feed intake hadincreased. It is well established that decreasedpregnancy rate and increased days open are associ-ated with increases in milk yield (15). This isbecause of the inverse relationship between level ofmilk production and energy balance that typifies theearly stage of lactation. Ferguson and Skidmore(28), for example, found the decrease in pregnancyrate to be related to the increase in milk yield ratherthan the dose of bST when they analyzed theirmultistudy data by controlling for confoundingfactors. Hard et al. (35) summarized a series ofstudies that had a similar design and found that daysopen increased by 5 days in the bST-supplementedgroup; however, when data were stratified by levelof milk production, days open did not differ betweencontrols and bST-supplemented cows (35). Thus,effects of high milk yield on reproductive perform-ance are the same whether or not the high yield wasdue to the use of bST

Calving interval for optimum economic return forU.S. dairy farms will probably increase with bSTsupplementation. Although conventional wisdomhas been that a 12- to 13-month calving intervalmaximizes profit, many managers of high producingdairy herds indicate that their calving interval islonger (6). In the case of bST supplementation, notonly does milk yield increase, but persistency oflactation is also improved. Thus, it is logical that thecalving interval for optimum economic return maybe substantially increased when bST is used. Ferry(29) modeled the effects of a 12- and 14-monthcalving interval on a herd basis and concluded thatwith bST use, income over feed cost was consider-ably increased with a 14-month calving interval.More extensive modeling, which included factors

such as veterinary costs and replacement values,yielded a similar conclusion (70).

Extending the calving interval also has somebenefits from the standpoint of the physiology of thecow. In the case of reproduction, increasing thecalving interval improves conception rate (71),probably as a consequence of the nutritional statusof the cows as discussed earlier. In addition, themajority of health problems and veterinary costs fordairy cows occur during the first 45 days postpartum(24). Thus, increasing the calving interval reduceshealth problems and costs on an annual basis for aherd and over an individual animal’s lifetime.

Optimum calving interval with bST use probablywill be different in the United States than in othercountries, particularly countries that have a seasonalsupply of feedstuffs and a beef industry largelybased on the offspring of dairy cows. Thus, theactual calving interval that optimizes economicreturn will vary according to a number of manage-ment and economic factors; still, a major determi-nant will be the magnitude of milk response and theincreased lactation persistency that occurs with bSTuse.

Genetic evaluation of sires might be affected byuse of bST if an interaction between genotype andthe milk response to bST occurs, or if bST isinappropriately used to manipulate sire proofs.Several studies have concluded that no evidenceexists of a genotype-response interaction in bST-supplemented cows (32). Sire evaluations involvethe comparison of the performance of a bull’sdaughters with the performance of their contempo-rary herdmates. A bias in sire evaluations can occur(and has) if a farmer gives preferential treatment tothe daughters of a particular sire that he hopes tomarket commercially (25). Potential for bias fromthe use of bST is similar and can be handled byproperly coded records and/or use of AI-proven sires(25,32), With proper coding of records, bST-treateddaughters could be compared only with herdmatesthat had also received bST supplement. Similarly,when the sire is evaluated with AI records, manymore daughters in a large number of herds will beinvolved; thus the chance for bias is negligible.

MILK AND MEAT COMPOSITIONGross composition of milk (fat, protein, and

lactose content) and meat has been examined inmostbST production trials, and is not substantially altered

Chapter 3-An Emerging Technology: bovine Somatotropin ● 39

by bST supplementation (3,16,50,56,75). There canbe minor changes, primarily in fat content of milk,during the first few weeks of bST supplementationas the cow’s metabolism and voluntary feed intakeadjust. However, these changes are temporary andminor when compared with variations that normallyoccur over a lactation cycle. Whereas the lactosecontent of milk is relatively constant, the content offat, and to a lesser extent protein, normally varieswidely due to many factors including genetics,breed, stage of lactation, age, diet composition,nutritional status, environment, and season (48).These factors affect the fat and protein content ofmilk in the same manner in bST-supplemented andnontreated cows (3,16,56). The meat derived fromtreated cows has a lower fat content but is otherwiseidentical (53).

The temporary shift in milk fat that can occurduring the first few weeks of bST supplementationrelates to nutritional status (11,16,56). Cows innegative energy balance produce milk with a higherfat content due to a greater reliance on lipidsmobilized from body fat stores. Milk fat content ismost likely to increase when bST supplementationis initiated in the first 100 days postpartum, whencows are typically in a lower energy balance.However, the negative energy balance typical of thisperiod (and especially of the first 8 weeks oflactation) is far in excess of that associated with bSTsupplementation.

The lipid composition affects the milk’s nutritivevalue, flavor characteristics, and manufacturingproperties. Studies demonstrate that fatty acid com-position and cholesterol content of milk are notaltered by bST (3). Cows that are in negative energybalance (as occurs early in lactation) shift milk fatcomposition toward longer chain, unsaturated fattyacids whether or not they receive bST supplementa-tion. In addition, the same fatty acid compositionchanges are observed in untreated and bST-supplemented cows as lactation progresses.

Composition of milk proteins has been examinedin at least a dozen studies because of the impact onfunctional properties of milk used in the manufactur-ing of dairy products (3,75). These studies havedemonstrated that the content and composition ofcasein (et-casein, ß-casein, k-casein) are not alteredby bST supplementation; and that casein as a percentof true protein is either unchanged with use of bSTor shows a small numerical decrease (often nonsig-

nificant). One short-term study reports a smallincrease in ct-lactalbumin content of milk in bST-supplemented cows but this was not observed inlong-term studies. The nonprotein nitrogen (NPN)content of milk from bST-supplemented cowsshould vary with nutritional status just as it does inuntreated cows. Some countries routinely test NPNlevels in bulk milk as a management tool for farmersto evaluate the protein adequacy of their nutritionalprogram (61).

Mineral content of milk from bST-supplementedcows has been examined in short- and long-termstudies involving large numbers of animals (3,75).Results have uniformly demonstrated that bST doesnot alter ash (total mineral content) or any nutrition-ally important mineral. Only one published reportexists on the vitamin content of milk, but milk frombST-supplemented cows did not differ in content ofany vitamin (vitamin A, thiamine, riboflavin, pyri-doxine, vitamin B12, pantothenic acid, and choline)except for biotin, which showed a slight increase(75). The increase in biotin content of milk is toosmall to be considered a benefit; biotin, a member ofthe b-vitamin family, is widely distributed in plantand animal food products and is also synthesized inthe intestine of humans.

Manufacturing characteristics of milk have beeninvestigated in a smaller number of studies butresults have consistently demonstrated that milkfrom bST-supplemented cows does not differ frommilk of untreated cows (3,75). Characteristics stud-ied include freezing point, pH, alcohol stability,thermal properties, proteases, lipases, susceptibilityto oxidation, and sensory characteristics, includingflavor. Similarly, no differences were observed incheese-making properties, including starter culturegrowth, coagulation, acidification, and syneresis, orin the yield, composition, or sensory properties ofthe various cheeses.

Minor constituents of milk include hormonessuch as estrogen, progesterone, glucocorticoids,thyroid hormones, prolactin, and growth factors.Trace concentrations of bST also normally occur inmilk and meat but this concentration is not apprecia-bly altered when cows receive exogenous bST(38,53,65,75). The level of bST in milk is only asmall fraction of the blood concentration. Only whenblood levels are increased about 30-fold by asubstantial dose of bST is there a small, butsignificant, increase in milk concentrations of bST

40 U.S. Dairy Industry at a crossroad: Biotechnology and Policy Choices

(65). This lack of an appreciable change in milkconcentration of bST when exogenous bST isadministered is expected, given that mammaryepithelial cells do not have receptors for soma-totropin (19). Pasteurization of milk destroys 85 to

90 percent of immunoreactive bST (33).

Some part of the biological actions of soma-totropin may be mediated by insulin-like growthfactor I (IGF-I). IGF-I, a protein hormone andmember of the somatomedin family, normally oc-curs in trace levels in milk. The concentration ofIGF-I is higher in cows’ milk (3 to 10 parts perbillion) than in human milk (1 to 3 parts per billion).Administration of bST to dairy cows results in anincrease in the amount of IGF-I in milk (by 2 to 5parts per billion), but the levels are still within therange typically observed in early lactation of un-treated cows (31,38,65,75). There is approximatelytwice as much IGF-I in meat of treated cows (2).

Studies with laboratory animal models havedemonstrated that IGF-I, like bST, has no biologicalactivity if administered orally (38). It is digested intoits amino acid, di- and tripeptide constituents by gutenzymes. Similarly, no evidence exists that frag-ments of IGF-I are biologically active in humans,nor is there evidence of systemic biological effectsin humans from any IGF-I absorbed intact. Theamounts of IGF-I that might potentially be ingestedin food products from treated cows are orders ofmagnitude less than those required to produce sucheffects (53).

The amount of IGF-I ingested in one liter of milkfrom bST treated cows approximates the amount ofIGF-I in human saliva swallowed daily by adults(31). Young children and infants already ingestIGF-I in regular cows’ milk. The importance of theadditional amount of IGF-I in milk from bST-treatedcows—whether it has a local effect on the esopha-gus, stomach, or intestine of infants-is unknown(53). However, most infants are either breast fed orfed commercially prepared infant formulas; the heattreatment used in the manufacture of these formulasinactivates approximately 90 percent of IGF-I.

MECHANISMS OF ACTIONSomatotropin regulates the use of absorbed nutri-

ents. When milk production is increased, extranutrients are needed by the mammary glands toprovide the raw materials and energy to make milk.Somatotropin coordinates the metabolism of various

body organs and tissues in a manner that supports theincreased nutrient use by the mammary g l a n d s .These coordinated adjustments in tissue metabolisminvolve all nutrient classes--carbohydrates, lipids,proteins, and minerals (see table 3-2), and are due tothe direct action of somatotropin on tissues (e.g.,liver and adipose). The adjustments made arecharacteristic of metabolic changes needed to sup-port lactation in all mammals (11,56,77).

Glucose metabolism illustrates the coordinatedmanner in which bST alters tissue processes. Glu-cose is a carbohydrate used as an energy source bymany tissues and as a raw material for milk synthesis(primarily for production of milk sugar). Nearly allof a cow’s daily glucose requirement is made by theliver and the mammary glands typically use about 85percent of the total. With bST-supplementation, theuptake of glucose by the mammary glands increasesin a manner parallel to the increases in milkproduction. This increased use of glucose for milksynthesis is accommodated by whole-body adjust-ments which include increase glucose production bythe liver and reduced glucose use for energy by otherbody tissues. In part, these adjustments occurbecause bST alters the response of tissues to acutesignals (e.g., insulin), thereby allowing a greaterallocation of glucose for milk synthesis while stillmaintaining normal body functions. Without suchadjustments in metabolism, initiation of bST supple-mentation would cause the glucose use to exceedthat which is available, resulting in ketosis anddeath. Ketosis from bST supplementation has notbeen observed in the hundreds of studies performed,even in tests where bST resulted in increased milkproduction of 40 percent or more.

Lipid metabolism provides another example ofthe coordination which occurs with bST supplemen-tation. The adjustments in tissue lipid metabolismdepend on the nutritional status of the cow at thetime bST-supplementation is initiated. Normally, ifa cow’s nutrient intake is greater than her require-ments, the excess nutrients are used to make bodyfat. BST administration causes adipose tissue toreduce its use of nutrients to synthesize body fat andallows for reallocation of these nutrients to supportincreased milk production. A different metabolicadjustment occurs if the cow’s nutrient intake isequal to or less than her requirements. In thisinstance, somatotropin directs adipose tissue tomobilize deposits of body fat so that these energyreserves can be used to support increased milk

Chapter 3-- An Emerging Technology: bovine Somatotropin ● 41

Table 3-2—Effect of bST on Specific Tissues andPhysiological Processes in Lactating Cowsa

Process affected during first few days andTissue weeks of supplementMammary T’

‘rT

Liver To

Adipose ~

J’

o

Muscle ~

Pancreas 0

K i d n e yb ~

Intestine b T

‘r

T

Whole -Jbody

T

-J

0

0

‘T

‘rT

secretory activity and maintenance of mammaryglandsblood flow and nutrient uptakesynthesis of milk with normal composition

production of glucoseresponse to acute signals (e.g., insulin) that allowfor greater glucose production

mobilization of fat stores to meet needs forincreased milk production if nutrient intake isinadequateuse of nutrients for fat storage so that they can beused for increased milk production if nutrient intakeis adequateresponse to acute signals (e.g., insulin and otherhormones that affect lipid metabolism) that allowsfor synthesis and breakdown of body fat reserves tobe coordinated with changes in use and availability yof nutrients

uptake of glucose

insulin and glucagon secretion reponse to changingglucose levels

production of 1,25 vitamin D3

absorption of Ca, P and other minerals required formilkability of 1,25 vitamin D3to stimulate calcium bindingproteincalcium binding protein

use of glucose by some organs so more can be usedfor milk synthesisuse of fat stores for energy if nutrient supply isinadequateuse of nutrients to make body fat if nutrient supply isadequateinsulin and glucagon clearance ratesenergy expenditure for maintenanceenergy expenditure consistent with increase in milkyield (i.e., heat per unit of milk not changed)cardiac output consistent with increases in milk yieldproductive efficiency (milk per unit of energy intake)

~hanges (’?=increased, ~=decreased, a=no change, o=change) thatoccur in initial period of bST supplement when metabolic adjustmentsoccur to match the increased use of nutrients for milk. With longer termtreatment, voluntary intake increases to match nutrient requirements.

bDemonstrat~ in nonlactating animals and consistent with observedperformance in lactating cows.

SOURCE: D.E. Bauman, “Bovine Somatotropin: Review of an EmergingAnimal Technology,” ~mmissioned background paper for theoffice of Technology Assessment, Washington, DC, 1990.

production. In both situations, these adjustmentsinvolve alterations in adipose tissue response toacute signals (e.g., insulin and other hormones thataffect lipid metabolism) thereby allowing the use ofbody fat reserves to be coordinated with changes inthe animal’s need for, and availability of nutrients.Over time, bST teated cows gradually increase their

feed intake so that stores of body fat are replenishedduring a lactation cycle. This replenishment occursunder a wide range of dietary conditions (11,16,17).If these adjustments in lipid metabolism did notoccur, cows would become emaciated, decreasetheir milk production, and be less efficient in theiruse of feed for milk production. These effects havenot been observed; cows administered bST havedemonstrated increased milk production and im-proved feed efficiency.

In addition to the direct metabolic effects that bSTcoordinates, somatotropin indirectly affects the mam-mary gland via its impact on other controllingcompounds (e.g., somatomedins such as IGF-1).Effects include increased cellular rates of milksynthesis and improved maintenance of secretorycells (i.e., slower rate of cell loss). The net result isthat bST-supplemented cows have higher daily milkyields and produce higher levels of milk throughoutthe lactation cycle. Over the years, selection ofhigher producing dairy cows has resulted in the sameimprovements. Thus, it is not surprising that cowsthat produce high levels of milk have highercirculating levels of naturally produced somatotropinthan do cows that produce low levels of milk (6).

BOVINE HEALTH AND STRESSCatastrophic health effects have been postulated

to occur with bST supplementation of dairy cows.Ketosis, fatty liver, and chronic wasting have allbeen proposed as possible side-effects of bST use(12,43). Crippling lameness, milk fever (a feverishdisorder following parturition), mastitis (inflamma-tion of the udder), infertility, heat intolerance,sickness, suffering and death are recent additions tothe list of adverse health claims (23,30,45,62,63,64).These postulated catastrophic effects were not basedon actual data but rather on the presumption that bSTcaused the mobilization of lipids from body fatreserves and/or overtly caused stress.

Metabolic disorders, if they occurred, would mostlikely manifest themselves during the first few daysof bST use. None of the above catastrophic healtheffects have been observed in any of the short-termor long-term studies with dairy cows going back tothe frost bST study in 1937. They were not observedin chronic toxicity studies (22,55), or in acutetoxicity studies where dairy cows were given 30,000mg of bST over a 2-week period, an amountapproximately equal to what would be administered


in four lactations (about 40 months) (78). Nor wereadverse effects observed in studies where inadequa-cies in the overall quality of the managementprogram resulted in negligible milk responses tobST supplementation. An increase in ketones (anindicator of subclinical ketosis) was reported in oneearlier study involving two cows given bST for 9 or10 days (42). However, that pituitary-derived prepa-ration was contaminated with other hormones (42,44),and this work has not been verified in acute orchronic studies using larger numbers of cows treatedfor longer periods with a wider dose range ofpurified bST (1 1,16,59,78).

It also has been postulated that somatotropin willreduce resistance to infectious and contagious dis-eases and thereby increase sickness and suffering indairy cows (23,30,63,64). Incidence of disease isgenerally very low in dairy cows and a thoroughevaluation of these claims will require extensivesummarization across studies to obtain a large dataset. However, none of the hundreds of bST studiesreported lower milk yield or decreased productiveefficiency, both of which are associated with anyincrease in sickness and suffering. On the contrary,somatotropin plays a key role in several aspects ofmaintaining immune competence (39). Immunityand disease resistance are compromised in somato-tropin-deficient laboratory animals and humans andsomatotropin supplementation enhances immunecompetence in both groups. Such studies have notyet been extended to lactating cows, but it has beendemonstrated that bST supplementation had a bene-ficial effect on cows’ recovery from experimentallyinduced E. coli mastitis (14).

Stress is more difficult to evaluate, but severalindices exist that demonstrate no stress effects dueto bST. Dairy cows that are stressed produce lessmilk less efficiently and expend more energy as heatthan expected. All of the several hundred studies ofbST in the scientific literature report increased milkyield and productive efficiency. The duration of bSTuse in these studies has ranged from a few weeks toat least four successive lactations. While numerousphysiological variables have been monitored toassess stress and have been shown not to change,nothing illustrates the normalcy of bST-supple-mented cows as effectively as the persistent gains inmilk yield and productive efficiency throughout thetreatment period. Recent studies spanning positive-to-negative energy and nitrogen balances, moreover,have clearly demonstrated that bST has no effect on

the energy expended for maintenance or for thesynthesis of a unit of milk (40,68,73).

Subtle health effects require examination of‘‘large numbers of animals treated under a range ofenvironmental and management conditions” (21). Acomplete summary of individual studies done through-out the world is beyond the scope of this review.Many have appeared as abstracts in the last 2 yearsand have not yet been published as full-lengthpapers. However, these summarizations are requiredof companies seeking the Food and Drug Adminis-tration (FDA) approval and are used by regulatoryagencies in their evaluation. Phipps (59) summa-rized a substantial portion of the studies conductedthrough 1988. This review showed that the indicesof animal health for bST-supplemented animalswere similar to these for controls and were consistentwith values reported in the literature for untreatedcows at a similar level of milk production. Variablesexamined included physical examinations, boneradiography, blood chemistry, metabolic disorders,subclinical ketosis, udder health, and welfare of thetreated cows as well as the health, growth, andperformance of their offspring.

Subtle effects on the incidence and duration ofmastitis are of special interest. Major factors affect-ing the incidence of mastitis include milking man-agement and herd health programs. However, theincidence of mastitis and milk somatic cell countsare also positively correlated with milk yield (58,60,69).Effects are quite small and amount to an annualincrease of approximately 0.4 cases/cow for each1,000 kg genetic gain in milk yield. Thus, it will takevery large numbers of cows to detect and evaluatewhether subtle effects, independent of milk-yieldresponse, occur with the use of bST. Phipps (59)summarized the incidence of clinical mastitis acrossstudies totaling over 1,300 cows and found that therelative incidence of mastitis was not affected bybST supplementation. Ferguson (27) likewise sum-marized eight studies reporting mastitis and foundthat there was no indication that bST was associatedwith increased mastitis infections.

Concern has been raised that even the smallincreased incidence of mastitis from higher produc-ing animals will increase the use of antibiotics incows. However, Burvenich et al. (14), reported thatrecovery time from experimentally induced mastitisis reduced in cows receiving bST supplement; it willbe of interest to learn whether the same beneficial

Chapter 3---An Emerging Technology: bovine Somatotropin ● 43

effects of bST supplement are observed for naturallyoccurring cases of mastitis under field conditions.

ONFARM ENVIRONMENTALPOLLUTION

Environmental pollution is reduced with bST useas a result of the gain it yields in productiveefficiency. Substantially less feed is required toproduce the same quantity of milk. This correspond-ingly reduces the use of fertilizer and other inputsassociated with producing, harvesting, processing,and storing of dairy feedstuffs.

Total U.S. animal fecal waste could also bereduced by as much as 6 x 109 kg assuming 100percent adoption and production of milk at 1988levels (see table 3-l). Similarly, the productiveefficiency gains with use of bST supplement couldresult in an annual reduction of 8 x 109 liters urineand 8 x 107 kg of urinary nitrogen for the total U.S.dairy herd (see table 3-l).

Ruminants also produce methane, a gas having astrong greenhouse effect. Ruminants and animalwaste account for about one-fifth of total worldwidemethane emissions (18), with cattle accounting forabout three-fourths of the livestock methane emis-sions or about 15 percent of total global methaneemissions (47). Because of the gain in productiveefficiency when bST is used, methane production bydairy cows could be reduced by as much as 5.5percent per unit of milk produced. For milk produc-tion at 1988 levels, this amounts to an annualreduction of 8 x 1010 liters of methane in the UnitedStates (see table 3-l).

It has been suggested that even though totalenvironmental pollution may be reduced, the morerelevant concern is whether animal wastes (manureand urine) are dispersed widely or concentrated (34).On large feed-lot farms, located primarily in theWest and South, most feed is not grown on site andanimal wastes are collected and stored. Such opera-tions may represent point-sources of surface andgroundwater contamination. On diversified farmslocated in the Upper Midwest and Northeast mostfeed is grown on site; animals have access to pasture,and wastes are left in pasture fields and/or recycledonto the fields. Some argue that diversified farms areless polluting than confined operations becausewastes are spread over a more extensive area. Aconcern exists that bST will provide the economic

incentive to create more confined operations andthereby increase pollution of ground and surfacewater.

These concerns are questionable. Diversified pas-ture operations are potential nonpoint sources ofpollution in ground and surface water. If notproperly managed, they could be significant sourcesof pollution. By the same token, handling practicesand environmental regulations can minimize thethreat of pollution from confined operations. Effec-tive in-use handling practices include: 1) flush,lagoon irrigation systems; and 2) mechanical scrap-ers, storage pit, and tank wagon transport systems. Insome areas, there is also a market for wastes for useas fertilizer, feed stuff, or fuel. In addition, the U.S.Environmental Protection Agency has providedregulations for confined livestock and poultry opera-tions for surface-water protection and several Statesand local entities have stringent groundwater protec-tion requirements for these operations (74). It isexpected that these requirements will become quitecommon throughout the United States and eventu-ally be applicable to most farming operations. Manysmall, diversified dairy operations will be at aneconomic disadvantage compared to larger opera-tions in meeting these environmental requirements(66).

INSTITUTIONAL INVOLVEMENTIN bST RESEARCH

Research in the technology of bovine soma-totropin has involved scientists and financial supportfrom Federal agencies (NSF, NIH, USDA), Stateagricultural experiment stations, and private indus-try worldwide. This extensive collaboration hasbeen of tremendous value in developing an under-standing of the biology of somatotropin and oflactation. The number of publications on soma-totropin to date is probably unprecedented for ananimal technology not yet approved for commercialuse. The bST literature is substantially larger thanthat for many dairy technologies in current use.

Some claim that extensive cooperation has totallycompromised the quality and value of the researchwith bST. Kronfeld (44,45,46) has claimed thatacademic and government scientists are ‘ ‘inden-tured’ and ‘‘biased’ because of this association.Rifkin (63,64) and Epstein (23) have quoted Kron-feld and echoed these claims, repeatedly suggestingthat the reporting of data has involved exclusion of


sick cows and the suppression and deletion ofadverse or negative results observed with bSTsupplementation of lactating cows. While theseindividuals offer no specific documentation ofscientific fraud, such claims are not to be takenlightly. Current events demonstrate that researchfraud is possible. However, a distinguishing featureof science is that research results are examined andrepeated by others. This mechanism helps to identifyinaccurate research. Published studies on bST haveinvolved at least 10,000 dairy cows and results havebeen verified not only by numerous groups of U.S.scientists but by many other scientists throughoutthe world. The claims of Kronfeld, Rifkin, andEpstein imply a worldwide conspiracy involving atleast 1,000 animal scientists in academia, govern-ment, and industry and hundreds of dairy farmersinvolved in the bST experiments. The possibility ofsuch a conspiracy seems remote.

TIMING OF’ COMMERCIALINTRODUCTION

Commercial use of bST requires approval by FDAand until this occurs, bST cannot be sold legally.Currently, bST is under review by the FDA andFederal law prohibits the agency from disclosingproprietary information on a drug under review.However, companies interested in bST have beenrelatively open about the fact that they are seekingapproval and have published a considerable quantityof their own proprietary research. In addition, FDAhas published an article using the companies’ data todemonstrate the safety of bST for human consump-tion (38). This extensive disclosure of informationon a drug under review is rare.

Each company wishing to market bST must provethat its product is effective (does what the companyclaims) and safe in order to secure FDA approval(67). The safety evaluation involves three areas:safety of the animal food products for humans;safety of the bST supplement to the target animals;and safety of using bST to the environment. Inaddition, each company must prove to the FDA thattheir manufacturing process can produce bST toconsistent and acceptable quality standards.

In 1984, the FDA had sufficient scientific infor-mation from extensive published literature andthen-unpublished studies (38,67) to make the deter-mination that the milk and meat from bST-

supplemented cows were safe for human consump-tion. Specific conclusions were:

1. bST is a protein that is digested enzymaticallylike any food protein when consumed orally.

2. bST has no biological activity in humans evenif injected.

3. A trace level of bST naturally occurs in milkand meat but this level is not appreciablyaltered in cows receiving bST supplement.

4. The overall composition of milk is not altereddue to bST treatment. The minor changes thatcan occur in the first few weeks of treatmentdue to shifts in nutrient balance are temporaryand well within the normal variation encoun-tered over the course of a lactation.

Thus, animal products were allowed to be mar-keted for the remainder of the investigational period.

In all other countries where bST is under review,the appropriate regulatory agencies have also com-pleted the human safety evaluations and withoutexception, allow the milk to be used for humanconsumption. By 1990, a few countries completedall facets of their review and registered bST forcommercial use by one or more companies (SovietUnion, Czechoslovakia, Bulgaria, Mexico, Brazil,and South Africa). In March 1991, the Committeefor Veterinary Medicinal Products of the Commis-sion of the European Communities released afavorable opinion on the application by one com-pany for the marketing of bST.

PRODUCT LABELINGSome States are seriously considering mandatory

labeling of all food products derived from milk ofcows supplemented with bST. The basis for labelingseems to be a concern about the safety of theproducts for human consumption. At least twoconsiderations need to be addressed.

First is the scientific merit or basis for labeling. Ifthere is a valid safety concern then the food shouldnot be marketed for human consumption. Labelingis not the appropriate method for handling a validconcern for consumer safety. If the regulatorysystem to evaluate food safety is inadequate, then thesystem should be changed. Labeling does not excusethe inadequacy. As just discussed, food safetyconcerns have been addressed and the conclusionreached is that food produced from bST-treatedcows is safe for human consumption (38).


The second consideration is verification. Aneffective labeling program requires developmentand adoption of appropriate regulations and theestablishment and funding of a system for imple-mentation and verification. In the case of bST, thereis no known test or technology that could be used todistinguish milk from bST-supplemented cows (38,67). Indeed, no change in milk composition as aresult of bST supplementation was found in humansafety evaluations by FDA or analogous agencies inother countries.

CONCLUSIONSOTA concludes that recombinant bovine soma-

totropin has no known adverse health effects on thecows receiving bST supplements or on humansdrinking milk or consuming milk products fromthese cows. Recombinant bovine somatotropin is thefirst major biotechnology developed for agriculture.It will have potentially significant impacts on thedairy industry, based mainly on the fact that it canproduce an average gain in milk per cow of 12percent per year. However, this technology is notmagic. It is distinguished only by the unprecedentedmagnitude of the productivity gains it yields. Forexample, the gain in productive efficiency obtainedwith bST supplementation would take 10 to 20 yearsto achieve using a combination of artificial insemi-nation (using superior sires) and embryo transfertechniques (see table 3-3). Only the eradication ofmastitis could increase milk yield per cow and

Table 3-3—Comparison of Theoretical Gains in MilkYield Per Cow for Different Dairy Technologies

Theoretical annual gain inTechnology milk per Cowa

Artificial insemination (Al) . . . . . . . . . . . . . . . 100 kgb

Al plus sexed semen . . . . . . . . . . . . . . . . . . . . 115 kgb

Al plus embryo transfer . . . . . . . . . . . . . . . . . . 135 kgb

Bovine somatotropin . . . . . . . . . . . . . . . . . . . . . >1,000 kac

aActuaI observed gain would average less because of variation in quality ofmanagement and other factors. For example, observed gain from usingartificial insemination andsuperiorsires is approximately 50 percent of thetheoretical gain.

bFrom Van V[eck (76). Gain would be cumulative for successive genera-tions so long as variation exists in the population.

cFrom Bauman et al. (9)

SOURCES: L.D. Van Vleck, “Potential Genetic Impact of Artificial insemi-nation, Sex Selection, Embryo Transfer, Cloning and Selfingin Dairy Cattle,” in: New T&nologies in Animal Breeding,B.G. Brackett, G.E. Seidel, Jr., and S.M. Seidel (eds.) (NewYork, NY: Academic Press, Inc., 1981), pp. 221-241; D.E.Bauman, P.J. Eppard, M.J. DeGeeter, and G.M. Lanza,“Response of High Producing Cows to Long-Term TreatmentWith Pituitary-and Recombinant-Somatotropin,” J. DairyScience, 68:1352-1362, 1985.

productive efficiency as much as or more than theuse of bST (6).

For bST to be effective, dairy farmers must beexpert managers. Poorly managed farms, whereanimals are stressed, underfed, and/or sick, will havenegligible milk response to bST. In this respect, thebST-supplemented cow presents the same chal-lenges as any high producing cow—the ultimategains to be captured depend not on the technologyper se, but on the management skills of its adopters.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

CHAPTER 3 REFERENCESAsimov, G.J. and Krouze, N. K., “The LactogenicPreparations From the Anterior Pituitary and theIncrease of Milk Yield in Cows,” J. Dairy Sci.20:289-306, 1937.Baile, C., “Quality of Meat from bST-Treatedcows,’ ‘ NIH Technology Assessment ConferenceAbstracts (Bethesda, MD: National Institutes ofHealth, 1990), pp. 51-63.Barbano, D.M. and Lynch, J. M., “Milk From bST-Treated Cows: Composition and Manufacturing Prop-erties, ’ Advanced Technologies Facing the DairyIndustry: bST, Cornell Cooperative Extension Ani-mal Science Mimeograph Series #133 (Ithaca, NY:Cornell University, 1989), pp. 9-18.Bauman, D. E., “Bovine Somatotropin: The CornellExperience,’ Proceedings of the National Invita-tional Workshop on Bovine Somatotropin, USDAExtension Service, Washington, DC, 1987, pp.46-56.Bauman, D. E., “Biology of Bovine Somatotropin inDairy Cattle,” Advanced Technologies Facing theDairy Industry: bST, Cornell Cooperative ExtensionAnimal Science Mimeograph Series #133 (Ithaca,NY: Cornell University, 1989), pp. 1-8.Bauman, D. E., ‘‘Bovine Somatotropin: Review of anEmerging Animal Technology,” OTA commis-sioned background paper, Washington, DC, 1990.Bauman, D.E. and Currie, W. B., “Partitioning ofNutrients During Pregnancy and Lactation: A Re-view of Mechanisms Involving Homeostasis andHomeorhesis,” J. Dairy Sci. 63:1514-1529, 1980.Bauman, D.E. et al., “Effect of RecombinantlyDerived Bovine Growth Hormone (bGH) on Lacta-tional Performance of High Yielding Dairy Cows, ”J. Dairy Sci. 65(Suppl. 1):121, 1982.Bauman, D.E. et al., “Response of High ProducingDairy Cows to Long Term Treatment with Pituitary-and Recombinant-Somatotropin,’ J. Dairy Sci. 68:1352-1362, 1985.Bauman, D.E. et al., “Sources of Wriation andProspects for Improvement of Productive Eftlciency


in the Dairy Cow: A Review, ’ J. Anim. Sci.60:583-592, 1985.

11. Bauman, D.E. et al., “Regulation of Nutrient Parti-tioning: Homeostasis, Homeorhesis and ExogenousSomatotropin,’ keynote lecture in: Seventh Interna-tional Conference on Production Disease in FarmAnimals, F.A. Kallfelz (cd.) (Ithaca, NY: CornellUniversity, 1989), pp. 306-323.

12. Bines, J.A. and Hart, I. C., “MetabolicLimits to MilkProduction, Especially Roles of Growth Hormoneand Insulin,’ J. Dairy Sci. 65:1375-1389, 1982.

13. Brumby, P.J. and Hancock, J., “The GalactopoeticRole of Growth Hormone in Dairy Cattle,” NewZealand J. Sci. Tech. 36A:417-436, 1955.

14. Burvenich, C. et al., ‘‘Effect of Bovine Somatotropinon Milk Yield and Milk Composition in Penparturi-ent Cows Experimentally Injected with escherichiaCoIi, ’ Use of Sornutotropin in Livestock Production,K. Sejrsen, M. Vestergaard, and A. Neimann-Sorensen (eds.) (New York, NY: Elsevier AppliedScience, 1989), pp. 277-280.

15. Butler, W.R. and Smith, R.D., “InterrelationshipsBetween Energy Balance and Postpartum Reproduc-tive Function in Dairy Cattle, ” J. Dairy Sci. 72:767-783, 1989.

16. Chalupa, W. and Galligan, D.T., “Nutritional Impli-cations of Somatotropin for Lactating Cows, ’ J.Dairy Sci. 72:2510-2524, 1989.

17. Chilliard, Y., “Ixmg-Term Effects of recombinantBovine Somatotropin (rbST) on Dairy Cow Perform-ances: A Review,’ Use of Sornatotropin in LivestockProduction, K. Sejrsen, M. Vestergaard, and A.Neimann-Sorensen (eds,) (New York NY: ElsevierApplied Science, 1989), pp. 61-87.

18. Cicerone, R.J. and Oremland, R. S., “Biogeocherni-cal Aspects of Atmospheric Methane, ” GlobalBiogeochemical Cycles 2(4):299-327, 1988.

19. Collier et al., ‘‘Changes in Insulin and SomatomedinReceptors and Uptake of Insulin, IGF-I and IGF-11During Mammary Growth, Lactogenesis and Lacta-tion, ’ Biotechnology in Growth Regulation, R.B.Heap, C.G. Presser, and G.E. Lamming (eds.)(Butterworths, London: 1989), pp. 153-163.

20. Daughaday, W.H. and Barbano, D. M., “BovineSomatotropin Supplementation of Dairy Cows: Is theMilk Safe?” J. Am. Med. Assoc. 264:1003-1005,1990.

21. Eppard, P.J. et al., “Effect of 188-Day TreatmentWith Somatotropin on Health and ReproductivePerformance of Lactating Cows,” J. Dairy Sci.70:582-591, 1987.

22. Eppard, P.J. et al., “Effect of Bovine Somatotropinon Animal Health,” Proc. Maryland Nutr. Conf.,1989, Pp. 74-79.

23. Epstein, S. S., Report sent to the Food and DrugAdministration entitled “Potential Public Health

Hazards of Biosynthetic Milk Hormones,” Jul. 19,1989.

24. Erb, H.N. et al., ‘‘Rates of Diagnosis of Six Diseasesof Holstein Cows During 15-Day and 21-Day Inter-vals,” Am. J. Vet. Res. 45:333-335, 1984.

25. Everett, R.W. et al., “Dairy Genetics in a bSTEnvironment,” Advanced Technologies Facing theDairy Industry: MT, Cornell Cooperative ExtensionAnimal Science Mimeograph Series #133 (Ithaca,NY: Cornell University, 1989), pp. 81-83.

26. Fallert, R.F. and Liebrand, C. B., “Economic Impli-cations of bST for the U.S. Dairy Industry,’ J. DairySci. (in press), 1991.

27. Ferguson, James D., “Bovine Somatotropin—Reproduction and Health,” NIH Technology Assess-ment Conference Abstracts (Bethesda, MD: NationalInstitutes of Health, 1990), pp. 65-81.

28. Ferguson, J.D. and Skidmore, A., “Bovine Somato-tropin-Reproduction and Health,” Advanced Tech-nologies Facing the Dairy Industry: bST, CornellCooperative Extension Animal Science MimeographSeries #133 (Ithaca, NY: Cornell University, 1989),pp. 57-69.

29. Ferry, J. W., “Managing the bST Treated Herd—AVeterinarian Perspective, ” Meeting the Challengesof New Technology, Monsanto Technical Sympo-sium Preceding the Cornell Nutrition Conference,Animal Sciences Division, Monsanto AgriculturalCo., St. Louis, MO, 1989, pp. 45-54.

30. Fox, M.W., “Why bST Must Be Opposed,” paperpresented at the Athene Trust International Confer-ence entitled ‘‘Action Alert: The Bio-Revolution—Cornucopia or Pandora’s Box?” (London, England:October 1988).

31. Gluckman, Peter D. et al., “TheEffects of Growth onLactation and Performance in Ruminants and Hu-mans: Mechanisms of Action and Effects on MilkHormone Composition, ” NE-I Technology Assess-ment Conference Abstracts (Bethesda, MD: NationalInstitutes of Health, 1990), pp. 41-45.

32. Gravert, H. O., “Influences of Somatotropin onEvaluation of Genetic Merit for Milk Production, ”Use of Somatotropin in Livestock Production, K.Sejrsen, M. Vestergaard, and A. Neimann-Sorensen(eds.) (New York, NY: Elsevier Applied Science,1989), Pp. 120-131.

33. Groenewegen, P.P. et al., “Bioactivity of Milk FrombST-Treated Cows,” J. Nutr. 120:514-520, 1990.

34. Hanson, Michael K., Biotechnology and Milk—Benefit or Threat? An Analysis of Issues Related tobGH/bST Use in the Dairy Industry, ConsumerPolicy Institute, Consumers Union, Mount Vernon,NY, 1990.

35. Hard, D.L. etal., “Effect of hng Term Sometribove,USAN (Recombinant Methionyl Bovine Somatotropin),Treatment in a Prolonged Release System on Milk


36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

Yield, Animal Health and Reproductive Performance-Pooled Across Four Sites,” J, Dairy Sci. 71( Suppl.1):210, 1988.Hocquette, J.F. et al., “The Human Liver GrowthHormone Receptor, “ Endo. 125:2167-2174, 1989.Johnsson, I.D. and Hart, I. C., “Manipulation of MilkYield with Growth Hormone, ’’Recent Advances inAnimal Nutrition, W. Haresign and D.J.A. Cole(eds.), Butterworths, London, 1986, pp. 105-123.Juskevich, J.C. and Guyer, C. G., “Bovine GrowthHormone: Human Food Safety Evaluation,’ Science249:875-884, 1990.Kelley, K.W. et al., “Hormonal Regulation of theAge-Associated Decline in Immune Function,’ Neu-roimmunornodulation, B.D. Jankovic, B.M. Markovicand N.H. Spector (eds.) (New York NY: The NewYork Academy of Sciences, 1987), pp. 91-97.Kirchgessner, M., Schwab, W., and Muller, H. L.,“Effect of Bovine Growth Hormone on EnergyMetabolism of Lactating Cows in Long-Term Ad-ministration,’ Energy Metabolism of Farm Animals,Y van der Honing and W.H. Close (eds.), PudocWageningen, Netherlands, 1989, pp. 143-146.Kostyo, J.L. and Reagan, R. C., “The Biology ofSomatotropin,’ Pharmac. Ther. B. 2:591, 1976.Kronfeld, D. S., ‘‘Growth Hormone-Induced Ketosisin the Cow,” J. Dairy Sci. 48:342-346, 1965.Kronfeld, D. S., “Major Metabolic Determinants ofMilk Volume, Mammary Efficiency and Spontane-ous Ketosis in Dairy Cows, ’ J. Dairy Sci. 65:2204-2212, 1982.Kronfeld, D., “Concerns About Use of GrowthHormone in Dairy Cows,’ Food and Drug Adminis-tration Seminar Notes of 9/10/86, Center for Veteri-nary Medicine, Food and Drug Administration,Rockville, MD, 1986.Kronfeld, D., “Health Risks in Dairy Cows GivenBiosynthetic Somatotropin,” Proc. Nutrition Insti-tute, NFIA, West Des Moines, IA, 1987.Kronfeld, D., Letter to editor entitled: BST MilkSafety, J. Am. Vet. Med. Assoc. 195:288-289, 1989.Lxxner, J., Matthews, E., and Fung, I., “MethaneEmissions from Animals: A Global Resolution DataBase,” Global Biogeochemical Cycles 2(2):139-156,June 1988.Linn, J. G., “Factors Affecting the Composition ofMilk from Dairy CowS,” Designing Foods-AnimalProduct Options in the Marketplace (Washington,DC: National Academy Press, 1988), pp. 224-241.Machlin, L. J., ‘‘Effect of Growth Hormone on MilkProduction and Feed Utilization in Dairy Cows,” J.Dairy Sci. 56:575-580, 1973.McBride, B. W., Burton, J. L., and Burton, J. H., “TheInfluence of Bovine Growth Hormone (somatotropin)on Animals and Their Products, ’ Res. Dev. Agric.5:1-21, 1988.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63,

64.

65.

McCutcheon, S.N. et al., “Application of BovineSomatotropin (bST) Technology to Pastoral DairyFarming Systems,” Seventh International Confer-ence on Production Disease in Farm Animals, F.A.Kallfelz (cd.) (Ithaca, NY: Cornell University, 1989),pp. 332-335.Mollett, T.A. et al., “Biosynthetic or PituitaryExtracted Bovine Growth Hormone Induced Galac-topoiesis in Dairy Cows,” J. Dairy Sci. 69(Suppl.1):118, 1986.National Institutes of Health, “Technology Assess-ment Conference Statement on Bovine Somatotropin,’Bethesda, MD, December 1990.National Research Council, “Nutrient Requirementsof Dairy Cattle, ’ 6th revised edition (Washington,DC: National Academy of Science, 1988).Noakes, D. E., “Use of Somatotropin in the DairyCow with Regard to Cow Safety,” J. Dairy Sci. (inpress), 1991.Peel, C.J. and Bauman, D. E., “Somatotropin andLactation,” J. Dairy Sci. 70:474-486, 1987.Peel, C.J. et al., “Bovine Somatotropin: Mechanismof Action and Experimental Results from DifferentWorld Areas,” Meeting the Challenges of NewTechnology, Monsanto Technical Symposium Pre-ceding the Cornell Nutrition Conference, AnimalSciences Division, Monsanto Agricultural Co., St.Iais, MO, 1989, pp. 9-18.Philipsson, J. et al., Genetic Aspects on Breeding forMastitis Resistance, Symposium on Bovine Mastitis,Munich, West Germany, 1978, pp. 1-14.Phipps, R. H., “A Review of the Influence ofSomatotropin on Health, Reproduction and Welfarein Lactating Dairy Cows,’ Use of Sornatotropin inLivestock Production, K. Sejrsen, M. Vestergaardand A. Neimam-Sorensen (eds.) (New York NY:Elsevier Applied Science, 1989), pp. 88-119.Poutrel, B., “Susceptibility to Mastitis: A Review ofFactors Related to the Cow,” Ann. Rech. Vet.13:85-99, 1982.Refsdal, A. O., Baevre, L., and Bruflot, R., “UreaConcentration in Bulk Milk As An Indicator of theProtein Supply at the Herd Level,” Acta Vet. Scand.26:153-163, 1985.Rifkin, J., Petition to Secretary of Health and HumanServices and the Food and Drug Administration andIts Division of Veterinary Medicine, Apr. 1, 1986.Rifkin, J., petition to the Food and Drug Administra-tion, Sept. 16, 1987.Rifkin, J., petition to the Food and Drug Administra-tion, Aug. 23, 1989.Schams, D., ‘‘Somatotropin and Related Peptides inMilk,” Use of Sornatotropin in Livestock Produc-tion, K. Sejrsen, M. Vestergaard, and A. Neimann-Sorensen (eds.) (New York, NY: Elsevier AppliedScience, 1989), pp. 192-200.


66.

6’7.

68.

69.

70.

71.

72.

73.

74.

Schwart, B. et al., “Size and Waste ManagementCosts,” Balanced Dairying Economics, vol. 11, No.1, Texas A&M University, College Station, Texas,1991.Sechen, S., ‘‘Review of Bovine Somatotropin by theFood and Drug Administration, Advanced Technol-ogies Facing the Dairy Industry: bST, CornellCooperative Extension Animal Science MimeographSeries #133, Cornell University, Ithaca, NY, 1989,pp. 101-110.Sechen, S.J. et al., “Effect of Somatotropin onKinetics of Nonesterified Fatty Acids and Partition ofEnergy, Carbon and Nitrogen in Lactating Cows,’ J.Dairy Sci. 72:59-67, 1989.Shook, G.E., Genetic Aspects of Mastitis, Proc. 25thAnnual Meeting National Mastitis Council, Inc.,1986, pp. 68-77.Skidmore, A. L., Development of a Simulation Modelto Evaluate Effectiveness of Dairy Herd Manage-ment, Ph.D. thesis, Cornell University, Ithaca, NY,1990.Trimberger, G. W., “Conception Rates in DairyCattle From Services at Various Intervals AfterParturition,” J. Dairy Sci. 37:1042-1049, 1954.Tyrrell, H. F., Ruminant Nutrition Laboratory, U.S.Department of Agriculture, Beltsville, MD, personalcommunication, 1990.Tyrrell, H.F. et al., “Effect of Bovine Somatotropinon Metabolism of Lactating Dairy Cows: Energy andNitrogen Utilization as Determined by RespirationCalorimetry, “J. Nutr. 118:1024-1030, 1988.U.S. Congress, Office of Technology Assessment,Beneath the Bottom Line: Agricultural Approaches

75.

76.

77.

78.

79.

80.

81.

to Reduce Agrichemical Contamination of Ground-water, OTA-F-4 18 (Washington, DC: U.S. Gover-nment Printing Office, November 1990).van den Berg, G., ‘‘Milk from bST-Treated Cows: ItsQuality and Suitability for Processing,” Use ofSomatotropin in Livestock Production, K. Sejrsen,M. Vestergaard, and A. Neimann-Sorensen (eds.)(New York, NY: Elsevier Applied Science, 1989),Pp. 178-191.Van Vlec~ L. D., “Potential Genetic Impact ofArtificial Insemination, Sex Selection, Embryo Trans-fer, Cloning and Selfing in Dairy Cattle,” NewTechnologies in Animal Breeding, B.G. Brackett,G.E. Seidel. Jr., and S.M. Seidel (eds.) (New YorkNY: Academic Press, Inc., 1981), pp. 221-241.Vernon, R. G., ‘‘Influence of Somatotropin on Metab-olism, ” Use of Somatotropin in Livestock Produc-tion, K. Sejrsen, M. Vestergaard, and A. Neimann-Sorensen (eds.) (New York, NY: Elsevier AppliedScience, 1989), pp. 31-50.Vicini, J.L. et al., “Effect of Acute Challenge Withan Extreme Dose of Somatotropin in a Prolonged-Release Formulation on Milk Production and Healthof Dairy Cattle, ” J. Dairy Sci. 73:2093-2102, 1990.Wallis, M., “The Molecular Evolution of PituitaryHormones,” Bio2. Rev. 50:35-98, 1975.Wood, D.C. et al., ‘‘Purification and Characterizationof Pituitary Bovine Somatotropin, ’ J. BioZ. Chem.264:14741-14747, 1989.Young, F. G., “Experimental Stimulation (galacto-poiesis) of Lactation, ” Brit. Med. Bull. 5:155-160,1947.

Chapter 4

Emerging Technologiesin the Dairy Industry

ContentsPage

BIOTECHNOLOGY AND THE DAIRY INDUSTRY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52Reproductive Technologies and Transgenic Animals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52Animal Health Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61Food Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65

KNOWLEDGE-BASED INFORMATION SYSTEMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65Expert Systems and Other Computer-Based Decision Aids.... . . . . . . . . . . . . . . . . . . . . 66Application of New Information Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

CONCLUSIONS .. .. . .. .. .. ... ... ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....67CHAPTER PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Box4-A. Definitions

Figure

of Commonly

BoxPage

Used Terms . . . . . . . . . . . . . . . . . . . . .. . . . ....... . . . . . . 52

FiguresPage

4-1. Reproductive Technologies Used To Produce Transgenic Animals . . . . . . . . . . . . . . . 544-2. Identification and Isolation of Desired Gene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554-3. Process of Producing a Transgenic Mouse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574-4. Gene Transfer Using Embryo Stem Cell Culture . . . . . . . . . . . . . . . . . . . . . . . . . .4-5. Nuclear Transplantation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4-6. Preparation of Monoclinal Antibodies . . . . . . . . . . . . . . . . . . . . + . . . . . . . . . . . . . . .

TablesTable4-1. Predicted and Actual Sex Ratios of Offspring After Intrauterine Insemination

of Sorted X and Y Chromosome-Bearing Rabbit Sperm... . . . . . . . . . . . . . . . . .4-2.Monoclinal Antibodies for Diagnostic Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

, . . . . . 5 8... ...60... ...64

Page

. . . . . . 62

. . . . . . 65

Chapter 4

Emerging Technologies in the Dairy Industry

Bovine somatotropin (bST) is the first majorbiotechnology product developed for the dairyindustry. This product has been controversial andhas raised many scientific and socioeconomic ques-tions. BST, however, is not the only new technologythat will affect the dairy industry. Advances inanimal reproduction, animal health, and food proc-essing are occurring, and many of the new technolo-gies being developed use highly sophisticated andcomplex biotechnology methods. By comparison,the biotechnology methods used to produce bST areactually rather rudimentary; potentially some ofthese new technologies could make bST obsolete.These new technologies will increase the level ofmanagement skills needed to use them effectively;new information technologies will aid the decision-making process. Several conclusions can be drawnconcerning the development of these new technolo-gies:

The dairy industry is on the verge of atechnological revolution. Biotechnology meth-ods that are more advanced than those used toproduce bST are currently under development.The impact of these technologies, in conjunc-tion with new information technologies in thenot too distant future may rival that of bST.Assessment and analysis, similar to that of bSTmay be warranted for many of these technolo-gies.The field of animal reproduction is advancingrapidly. Researchers ‘have significantly im-proved their understanding of egg developmentin the ovary, how to stimulate the release ofnumerous eggs at once, and how to achievefertilization and development of eggs outsideof the cow. Embryos can be frozen for later use.Both embryos and sperm can be sexed. It is alsopossible to create multiple copies of an embryo,each of which can be transplanted into a cowwhose reproductive cycle has been adjusted tobe able to accept the embryo and carry it toterm. These new technologies make it possibleto improve herd quality more rapidly than canbe achieved using traditional breeding meth-ods.

●

●

●

●

●

[t is now possible to create transgenicl cattle,however, the techniques currently used areinefficient and require the use of thousands ofeggs to produce just one transgenic animal.These inefficiencies make it too expensive tocommercially produce and market transgeniclivestock. However, scientific breakthroughsare leading to the development of technologiesthat will improve the efficiency of transgenicanimal production and substantially lower thecost of doing so. Transgenic livestock maybecome commercially available in small num-bers by the end of the decade.BST potentially could be supplanted by thedevelopment of transgenic cattle. Dairy cowscan be developed to produce higher levels ofbST so that daily injections or timed releaseformulations are no longer needed. Alterna-tively, genes that code for chemicals thatsuppress bST production can be altered in thecow such that a cow’s normal bST productionwill increase.New biotechnology products are also beingdeveloped to improve animal health. Productsinclude new vaccines and diagnostic kits, aswell as compounds that enhance an animal’sability to fight disease.Not only are new biotechnology products beingdeveloped for use in livestock production, butthey are also being developed for use in foodprocessing. New products will improve theproduction of milk products such as cheese andyogurt. They can also be used to detect milkcontaminants.Effective use of these new technologies willplace a premium on management skill. Newinformation technologies are being developedto aid farm management. These new technolo-gies can incorporate individual farm data, withpertinent information from national databases,into computer programs that will aid farmers inthe decisionmaking process.

This chapter describes information systems, andbiotechnology methods and products that have beendeveloped, or are expected to be developed for use

lm~5 whose heredi~ DNA has been augmented by the addition of DNA from a source other than parental germplasm uStig recombmt DNAtechniques.

-51–


in the dairy industry in the decade of the 1990s.Many of these technologies are highly sophisticated,cutting-edge technologies, and as such have not beenextensively discussed in the lay literature. However,it is important that the potential of these technolo-gies be understood. Given the nature of these newtechnologies, the following descriptions are, bynecessity, somewhat technical.

BIOTECHNOLOGY AND THEDAIRY INDUSTRY

The term biotechnology refers to a wide array oftechniques that use “living organisms (or parts oforganisms) to make or modify products, to improveplants or animals, or to develop microorganisms forspecific uses’ (46). Under this broad definition,biotechnology includes many long-practiced dairy

technologies such as animal breeding and cheese-making. New biotechnologies include recombinantDNA techniques, cell culture, and monoclinalantibody (hybridoma) methods (see box 4-A). Theapplication of these new methods to the dairyindustry has already generated a number of productsfor improving milk production, animal health, andfood processing, and will continue to do so. Productsnow emerging range from rapid diagnostic tests for

contaminants in dairy products to cows geneticallyengineered to produce high-value pharmaceuticals.

Reproductive Technologies andTransgenic Animals

Animal scientists generally agree that the mostimportant cause of economic loss in the animalindustries results from reproductive inefficiency

Box 4-A—Definitions of Commonly Used Terms

Antibody: Proteins produced by specific white blood cells (i.e., B lymphocytes) that bind specifically toforeign antigens in the body.

Antigen: Any substance that elicits a defensive (immune) response.Cell culture: The growth and maintenance of cells derived from multicellular organisms under controlled

laboratory conditions.Chromosome: A thread-like structure composed primarily of DNA, that carries the genes which convey

hereditary characteristics; in mammals chromosomes are contained in the nucleus and the X chromosome conveysfemale characteristics and the Y chromosome the male characteristics.

DNA (deoxyribonucleic acid): The molecule that is the repository of genetic information in all organisms(with the exception of a small number of viruses in which the hereditary material is ribonucleic acid-RNA).

Estrus: The period during which a female mammal is receptive to sexual activity.Estrous cycle: The period of time needed for the reproductive cycle that includes egg maturation and ovulation

in the ovary and preparation of the uterus to receive fertilized eggs. This cycle is under hormonal control and extendsfrom the beginning of one period of estrus to the beginning of the next.

Hybridoma: A new cell resulting from the fusion of a particular type of immortal tumor cell line, a myeloma,with an antibody-producing B lymphocyte. Cultures of such cells are capable of continuous growth and specificantibody production

In vivo: Within the living organism.In vitro: Outside the living organism and in an artificial environment, e.g., test tube.Monoclinal antibodies: Identical antibodies that recognize a single, specific antigen and are produced by a

clone of specialized cells.Recombinant DNA: A broad range of techniques involving the manipulation of the genetic material of

organisms; term is often used synonymously with genetic engineering; term also used to describe a DNA moleculeconstructed by genetic engineering techniques composed of DNA from different individuals or species.

Transgenic animals: Animals whose hereditary DNA has been augmented by the addition of DNA from asource other than parental gerrnplasm using recombinant DNA techniques.

Uteri: The plural for uterus— the organ of the female mammal for containing and nourishing the young duringdevelopment prior to birth.SOURCE: Office of Technology Assessment, 1991.

Chapter 4--Emerging Technologies in the Dairy Industry ● 53—

(i.e., low conception rates and embryo mortality).The field of animal reproduction is currently under-going a scientific revolution. In the 1986 reportTechnology, Public Policy, and the Changing Struc-ture of American Agriculture, OTA predicted that,beginning about the year 2000, eggs matured andfertilized in vitro and transplanted to a recipientanimal (artificial inembryonation) would in part,replace natural and artificial insemination in theanimal breeding system, and embryos altered byrecombinant DNA techniques (transgenics) wouldbe commercially available. In fact, embryos pro-duced by new reproductive methods are currentlybeing marketed, although at present no transgenicembryos are commercially available. However, sig-nificant new advances are occurring, and the direc-tion and timetable of developments will almostcertainly be affected. This section focuses on recentadvances in reproduction and recombinant DNAtechniques and their application to the dairy industryduring the decade of the 1990s.

It is now possible to select genetically superiorfemales and induce them to shed large numbers ofeggs (superovulation) that can be matured in vitroand fertilized with sperm from males selected fortheir desirable traits. The resulting embryos can besexed, split to make duplicate copies, and storedfrozen until needed. An embryo can then be trans-ferred by nonsurgical techniques into the uterus of arecipient animal whose estrous cycle has beensynchronized with the stage of development of theembryo. These new reproductive technologies can,and are being used to improve the quality oflivestock herds more rapidly than could be achievedwith traditional breeding, although currently manyof these technologies are still relatively inefficient.

The ultimate goal of many workers in the field ofanimal biotechnology, however, is to produce ani-mals whose hereditary DNA has been augmented bythe addition of DNA from a source other thanparental germplasm, using recombinant DNA tech-niques (transgenic animals) (47). Transgenic ani-mals can be created that possess traits of economicimportance including improved disease resistance,growth, lactation, or reproduction.

Transgenic livestock may also prove to be effec-tive factories for the production of high-valuepharmaceuticals, a development particularly perti-nent for the dairy industry (9). Genes that code foranimal proteins (e.g., tissue plasminogen activator-

TPA) can be linked to regulatory sequences thatdirect high levels of gene expression in the mam-mary gland. This enhanced expression results inincreased secretion of the desired animal protein intothe milk of lactating females. This protein could thenbe extracted and purified from the milk. Transgeniccows producing pharmaceuticals have not yet beenreported, but these animals are under development ina number of public and private laboratories. Highlevels of milk production coupled with the ease ofmilk collection may make this production methodmore cost effective than the cell culture systemscurrently used in the production of certain pharma-ceutical proteins.

The production of transgenic animals is inextrica-bly linked to development of new reproductivetechnologies. It is impossible to produce animalscontaining foreign DNA in their germlines withoutfirst manipulating the embryo and transferring it toa recipient animal. New reproductive technologies .such as superovulation, in vitro egg maturation andfertilization, nuclear transplantation, and embryosexing can be used to upgrade livestock herds.However, when they are combined with recombi-nant DNA technologies (the identification, isolation,and transfer of selected genes), they provide oppor-tunities to efficiently and cost effectively producetransgenic animals, and to rapidly improve livestockquality (see figure 4-l). Therefore, advances inreproductive technologies will be discussed withinthe context of their applicability to the production oftransgenic animals, recognizing that they are in theirown right powerful tools for livestock improve-ments.

Steps in the Development of Transgenic Animals

The production of transgenic animals is a com-plex process, and involves four major steps. First,the desired gene must be identified, isolated, andprepared for insertion into a fertilized egg. Second,the host cell must be obtained and prepared for geneinsertion. In livestock, the host cells used aregenerally fertilized eggs or early stage embryos.Third, the desired gene must be transferred to thehost cell. And fourth, the resulting recombinantembryo is duplicated, and the resulting multipleembryos are transplanted into recipient cows thathave had their estrous cycles synchronized to receivethe embryo. This duplication process allows for theproduction of multiple offspring of geneticallysuperior animals. Advances in each stage, discussed

Figure 4-1—Reproductive Technologies Used To Produce Transgenic Animals

kIdenttflcatlon and Isolatlon

r I

E\ .+;~’’’’’’’nisrandom”superovulation combined with

eggI

Host cell preparation using1

superovulation combined within vitro egg maturation andfertilization 1

I I 12 to 4 + 2 to 4embryos Embryo cell embryos

L J multiplicationI J

by splitting

1

Embryo multiplication bynuclear transplantationand recloning

Egg cell with nucleus removedEarly stage embryo(less than 64 cells) -

Gene transfer

I Implantation into cowwhose estrous cycle has~ m

~ been synchronized to JY receive the embryo

E + I i s t a r g e t e d”by gene identificationand isolation.

SOURCE: Office of Technology Assessment, adapted from J.P. Simons and R.B. Land, “Transgenic Livestock,” J. Reprod Fert. Supp/. 34:237-250, 1987.

I

Chapter 4--Emerging Technologies in the Dairy Industry .55

below, are improving the efficiency of transgenicanimal production.

Step 1: Gene Identification and Isolation--Be-fore a foreign gene can be transferred into thegenome of a host cell to create a transgenicorganism, the gene must first be identified andpurified. This is done by a process called cloning.The tools used to clone DNA include specialenzymes that cut and paste DNA, nucleic acidfragments that can be used to identify specific DNAsequences (probes), vehicles used to carry a foreigngene into a host cell (vectors), and host cells that canbe used to produce multiple copies of the gene. Thehost cells used to multiply gene numbers are usuallybacteria. The vectors most commonly used arebacterial plasmids, circular pieces of DNA that areeasily inserted into bacterial cells and are capable ofindependent replication inside the bacteria (seefigure 4-2).

Isolating a single gene is complicated by the factthat a DNA sample may contain millions of genes.Researchers must be able to separate the one gene ofinterest from all of the other genes. To do this, theDNA sample containing the desired gene is cut intopieces with special enzymes (restriction endonu-cleases). The bacterial plasmid (vector) is also cutwith the same enzyme. The pieces of sample DNA,including any pieces carrying the desired gene, areinserted into vectors and the loose ends of the sampleDNA and bacterial plasmid are pasted together(using the enzyme DNA ligase). These recombinantDNA vectors are then inserted into bacterial cells,and the bacterial cells are grown. At this point thebacterial cells containing the plasmid carrying thedesired gene must be identified and isolated from therest. This is done by using a probe, a nucleic acidsequence that recognizes the desired gene. Once thebacterial cells containing the desired gene are found,those cells can be grown in isolation to producemillions of copies of the vector containing thedesired gene. The vector can be removed from thebacterial cells, and the desired gene isolated. Thisprocedure can yield millions of copies of the desiredgene that are free from contamination by other genes(48).

The purified gene can then be combined withappropriate regulatory genes that control the circum-stances under which the gene is turned on and off.The purified gene and its regulatory sequences canbe inserted into a vector such as a virus, for example,

Figure 4-2—identification and Isolationof Desired Gene

DNA containing gene to be isolated

1 Use restriction enzymesto cut the vector and to

icut the DNA containing

.$: the desired gene into~ pieces I I

f9,

\ - .

F %8, $

\ /’%

- - m ’ c7~; {k/’ ‘$ &,..1

,,&f@

%, Joining of Vector

\ /DNA Fragments With

$“ DNA Fragments to beq, c’ Cloned Using the% k’-%.?$:2:.... .;” Enzyme DNA Ligase‘ “ ’ $7”

Recombinant DNA Molecules

Introduction ~into Bacteria tj;~.,;

tiim Vector with

( 2Cloned DNA

Bacterial FragmentChromosome

@3

j ,

@@’@@DesiredMultiplication of Bacteria Containing the

Gene To Yield Many Identical Copies of Fragments



that will carry these genes into an animal cell andincorporate them into the genome (see gene transfertechnologies). The tools used to purify genes arewell developed. The major challenge is determiningwhich genes are to be purified.

Step 2: Preparation of the Host Cell—Becausecurrent technologies used to transfer a gene into ahost cell are inefficient, to get just one transgenicanimal requires the use of thousands of fertilizedeggs. To attain such large numbers of embryos, cowsmust be induced to shed large numbers of eggs(superovulation) that can be matured and fertilizedin vitro. Increased understanding of the control ofovarian functions is improving the efficiency ofobtainin g sufficient numbers of fertilized eggs.

Control of Ovarian Functions-The lack ofhighly repeatable, efficient means for inducingsuperovulation is a major constraint. Induction ofsuperovulation requires detailed knowledge of thehormonal factors that control the development of theegg in the ovary. The process of egg developmenthas been subjected to intense investigation and anumber of significant advances have been madeduring the past 5 years. Studies that have exploredthe basic mechanisms controlling egg growth andmaturation, and corpus luteum2 function, havepaved the way for developing even more precisemethods for regulating the estrous cycle, producingsuperovulation, and reducing the heavy losses due toearly embryo deaths.

Perhaps the most important development in ovar-ian physiology in recent years is the discovery of theovarian hormone inhibin, which decreases the ovu-lation rate.3 Some breeds of animals with excep-tionally high ovulation rates, such as the Booroolastrain of Merino sheep in Australia, are known tohave low levels of circulating inhibin (4). Cattleimmunized against inhibin have lower circulatinglevels in their blood and show increased ovulationrates (21). The genes controlling inhibin productionhave been cloned and the potential exists forproducing transgenic animals in which these genesare repressed or deleted.

Progress has also been made in understanding thecontrol mechanisms that regulate corpus luteumfunction and its production of progesterone, ahormone that regulates the length of the estrous

cycle and helps maintain pregnancy. This under-standing paves the way for development of moreprecise methods for regulating the estrous cycle,which is needed to synchronize surrogate mothers,and for producing superovulation. Superovulationtreatments are initiated when the ovaries are underthe influence of progesterone. Currently, super-ovulation treatments use highly purified hormonesproduced by recombinant DNA technology andproduce, on average, about 10 viable eggs pertreatment (compared to the 1 egg a cow normallyproduces per ovulation) (21). As the new knowledgeof the factors controlling egg development andcorpus luteum function is applied, the number ofviable embryos produced by each superovulationtreatment is expected to increase.

In Vitro Maturation and Fertilization of Eggs—Invitro maturation and fertilization of eggs recoveredby superovulation provides a means of overcomingthe problem of livestock reproductive inefficiency.Normally, a bovine ovary contains about 50,000immature eggs at puberty, however, on average only3 to 4 of these eggs will result in the births of livecalves during the animal’s lifetime. Using presentsuperovulation methods, about 10 viable eggs can beharvested from the ovaries of 1 treated cow andabout half of these develop into embryos suitable fortransfer. Improved superovulation technology maylead to the recovery of more eggs suitable for in vitrofertilization such that the number of live birthsresulting from a superior animal could be quite high.

In vitro fertilization occurs only when a capaci-tated sperm (i.e., a sperm specially prepared topenetrate the egg cell membrane) encounters an eggat an optimal maturation state. Great progress hasbeen made in understanding the factors involved inegg maturation and sperm capacitation in livestock.As a result, offspring have been produced in cattle,swine, sheep, and goats following in vitro fertiliza-tion (16) and attempts to market embryos producedwith these techniques are already underway.

Step 3: Gene Transfer Technologies—Achiev-ing the goal of transgenic animal production requiresthe development of efficient and cost effective waysto transfer the selected genes to an embryo. Micewere the first transgenic mammals created (seefigure 4-3), and were produced by microinjecting

z~e CO~u.S lute~ is a temp~~ endocrine organ that is produced at the site of ovulation dtig each W@Ous cYcle.

%hibin decreases owlation rates by suppressing the secretion of follicle stimulating hormone (FSH), a hormone produced by tie Pitiw gl~d.

Chapter 4--Emerging Technologies in the Dairy Industry ● 57

Figure 4-3-Process of Producing aTransgenic Mouse

Y Implant injected embryosinto pseudopregnant

Q

4’Fetal Developmental regulationexpression / . Live birth

\test for transgene

eFounder animal

1 Breed transgenics

6 3A & ~ 2Fetal/neonatal pup

Adult

SOURCE: Sally A. Camper, Fox Chase Cancer Center.

cloned DNA into a fertilized egg4 (32). Alterna-tively, viral vectors can be used to transfer clonedDNA into a host embryo (14). The embryo is thentransferred to a recipient animal whose estrous cyclehas been synchronized to accept and carry to term,the developing embryo.

A number of transgenic cattle, pigs, sheep, andchickens have been produced by these techniques,however, they are of limited use because of the highcost and low efficiency of microinjection tech-niques, and the absence of appropriate viral vectorsfor use in most livestock species. Additionally, DNAtransferred by these methods is inserted randomlyinto the host genome, resulting in a lack of controlof the gene transfer process (20).

Because of the deficiencies encountered withusing viral vectors or microinjection methods tocreate transgenic livestock, alternative methods arebeing sought. A promising new method for generat-ing transgenic animals has recently been developedin mice and may be applicable to other mammals.This new technique uses stem cells derived from anembryo. Stem cells are cells that are normallyundifferentiated, that is, they do not become special-ized tissue cells such as muscle, brain, liver cells,etc. However, stem cells retain their ability tobecome specialized cells when given the properstimuli (i.e., they are pluripotent).5 These stem cellscan be used as vectors to introduce selected genesinto a host embryo. This method has severalsignificant advantages over microinjection methods,the most profound of which is that for the first time,it is possible to insert DNA at specific, predeter-mined sites within the genome of the stem cells (8).Targeted insertion is possible because stem cellshave an intrinsic ability to recombine similar (ho-mologous) DNA sequences, which results in thereplacement of the endogenous gene with the desiredgene.

Stem cells must first be isolated (see figure 4-4).An early stage embryo is cultured on a thin layer ofspecially prepared cells. The proliferating embryocells are recultured until individual stems cells canbe isolated. These individual stem cells can then becultured indefinitely. At this stage, DNA sequencescontaining desired genes can be inserted into thestem cells.6 A genetically transformed stem cell isthen microinjected into an immature embryo toproduce a chimera, an organism that contains cellsfrom more than one source. If the stem cells areincorporated into the germ lines of these chimericanimals, then these animals can be interbred toobtain offspring that are homozygous for the desiredtrait (8).

Application of the gene targeting method makespossible a broad range of phenotypes for transgenicanimals that could not be produced economicallyusing direct microinjection or viral vectors. Targetedgene insertion allows endogenous genes to be

4SPcific~Iy, tie DNA is ~ject~ into tie male Pronucleus of the fertilized egg. The pronucleii are the egg ~d sPerm nucleii Present after tie ‘Pmpenetrates the egg membrane.

Splmipotency help me stem cells attractive vectors of DNA transfer. While in tissue culture, DNA Cm emily be tiefied ~to stem ce~s. menstem cells are injected into an early stage embryo, the conditiom for tissue specialization are present, and stern cdls undergo the normal tissuedevelopment that occurs as the embryo develops during pregnancy. Thus, using stem cells provides an efficient means to transfer DNA.

6Me~ods US~ inc]ude viral infmtion and use of an electric pulse to make cell membranes leaky (electropomtion).

58 ● U.S. DairyIndustry at a Crossroad: Biotechnology and Policy Choices

Figure 4-4—Gene Transfer Using Embryo Stem Cell Culturea

A-,: I - t-

- 1 -

–7 1-1--1 , :1- . , . . “ ;:I - -. ‘ +–,—, - .(

, . .

t

A mouse embryo (blastocyst)is donated.

The embryo is culturedonto a thin layer ofspecially prepared(feeder) cells.

The embryo attaches tothe cell layer, and theinner cells of the embryobegin to proliferate.

Groups of these differentiatingembryo cells are separated,recultured and single coloniesof embryonic stem cells areidentified and transferred.

Single cells are reseededonto the feeder cells. Foreign

K “ ’ ’ ’ O th e ~ <l < l

F r

#+-’-JTJ

A second mouse embryo is donatedand is injected with a culturedembryonic stem cell. This embryois transplanted into a surrogatemouse, which gives birth to achimera (a mouse with cells fromboth parent embryos). Mating twochimeras gives rise to offspringwith the desired traits.

. ----Y’.. .( - 3‘“* :

“

aThis technique is currently developed only for mice and hamsters and possibly rabbits and swine. It is not yet developed for cattle.

SOURCE: M.R. Capecchi, “The New Mouse Gene!ics: Altering the Genome by Gene Targeting,” Trends in Genetics 5:70-76, 1989.


inactivated or replaced with modified forms of thegene, such as one that is expressed at a higher level,has anew pattern of tissue specific expression, or hasa modified biological activity.

Perhaps the most significant advantage of thegene targeting approach is that it allows for en-dogenous genes to be effectively removed. Genescan be inactivated by targeting insertion into anessential region of the gene. This fact is of particularinterest to the livestock industry, because inactiva-tion of genes that have inhibitory physiologicaleffects is likely to result in improvement in a numberof productive traits. For example, bovine somato-statin is a hormone that inhibits bovine somatotropinproduction; inactivation of this gene would result inincreased endogenous somatotropin secretion and,presumably increased milk production and moreefficient growth. If successful, this technologywould supplant the need to administer bST ex-ogenously to increase milk production (see ch. 3).The genes controlling the production of inhibin, theovarian hormone that reduces ovulation rate, is yetanother example. The ability to inactivate genes alsoprovides a powerful research tool that allows scien-tists to study the function of genes in vivo. Theabsence of a method to produce embryo stem celllines from cattle is currently preventing the use ofthe gene targeting approach in the dairy industry,7

however, a number of laboratories are trying todevelop this technology.

Step 4: Embryo Multiplication-The productionof multiple copies of a genetically engineeredembryo is the final step in the development oftransgenic animals. Efficient and inexpensive multi-plication technologies are tremendously importantfor improving the efficient development of transgenicanimals. Multiple copies of a mammalian embryowere first produced by physically splitting an earlyembryo into halves, giving rise to identical twins(21). If the embryo is divided more than twice,however, few offspring survive. Thus, no more thanfour identical animals can be produced by splittingand generally only two embryos are produced by thismethod. This procedure is already used in the cattleembryo transfer industry, nearly doubling the num-ber of offspring produced.

A more efficient and promising method of produc-ing multiple copies of an embryo is by a technique

called nuclear transplantation. Basically, the proce-dure involves the transfer of a nucleus from a donorembryo into an immature egg cell whose ownnucleus has been removed. The recipient egg cell isactivated by exposure to an electric pulse, allowed todevelop into a multicelled embryo, and then used asa donor in subsequent nuclear transplantations togenerate multiple clones. This procedure (outlinedin figure 4-5) has been used successfully with cattle(7,35), sheep (42,49), and swine (36). Using thistechnique, hundreds of pregnancies have alreadybeen produced in cattle and recloning has beenperformed successfully resulting in as many as eightcalves from one embryo (28).

The value of this technique to the dairy industryis enhanced by the ability to successfully transfernuclei from frozen embryos into eggs whose nucleihave been removed. Conception rates obtained aftertransfer of embryos produced by nuclear transplan-tation are variable, but rates as high as 50 percenthave been obtained. However, embryo losses aftertransfer are higher than normal, resulting in actualpregnancy rates ranging from 15 to 33 percent (7).Combining the techniques of in vitro fertilization,embryo cloning, and artificial estrous cycle regula-tion will likely result in major changes in dairy cattlebreeding and in the rates of genetic improvement.

While significant advances in transgenic animalproduction have been made, it is unlikely thattransgenic animals will be commercially availablebefore the end of the 1990s at the earliest. The abilityto produce transgenic cattle possessing traits ofeconomic value is currently limited by the absenceof embryo stem cell technology and the lack ofknowledge about the relationship between the ex-pression of a specific gene and the physiologicalconsequences. While the techniques for isolatingand sequencing bovine genes are now straightfor-ward, understanding of the functions of the geneshas lagged. Analysis of gene function is complicatedby the fact that many traits are controlled by multiplegenes. Thus, manipulation of such traits will requiredetailed understanding of these genes and theirinteractions. Ultimately identifying and understand-ing the physiology of the major genes controllinglactation, reproduction, and disease and stress resis-tance in dairy cattle is needed. An active genomemapping program could help enhance these develop-ments.

7~e stem ce~ tec~o]on has been developed for mice, titers, ~d swine.

292-860 0 - 91 - 3


Figure 4-5—Nuclear Transplantation

— 1 — , — , — , —,

An embryo is nonsurgically removedfrom a donor cow or is produced byin vitro fertilization.

Individual embryoare removed.

cells

Each embryo cell is injected intoa specially prepared egg cell thathas had its nucleus removed. Anelectric pulse is administered tocause fusion.

Each egg cell is grown to amulticell embryo at which pointthe cloning procedure can be repeated

or

the embryo can be transplanted toa cow that eventually gives birth.

SOURCE: ~’fice of Technology Assessment, adapted from R.S. Prather and N.L. First, “Cloning Embryos by Nuclear Transfer,” Genetic Engineering ofAnima/s, W. Hansel and B.J. Weir (eds.), Journa/ of Reproduction and Ferti/ify~td, Cambridge, UK, 1990, pp. 125-134.

Chapter 4--Emerging Technologies in the Dairy Industry . 61

Sexing of the Offspring

The availability of a technique to preselect the sexof the progeny is of great economic potential for thedairy industry where females are the major incomeproducers and artificial insemination is alreadywidely used. Until recently, none of the methodsresulted in the degree of separation needed forcommercial use. However, recent advances in theseparation of the X and Y sperm, and sexing of theembryo have been made.

Separation of the X- and Y-Beating Sperm—Ithas long been a goal of mammalian p h y s i o l o g i s t s t odevelop a method for effectively separating X and Ychromosome-bearing sperm to control the sex of theoffspring. Most sperm separation techniques arebased on potential differences in the size and densityof the two sperm types. 8 These methods, however,have met with limited success (41).

Development of cell sorting techniques based onthe differences in sperm size and fluorescence ofsperm DNA (flow cytometric measurements) hasprovided the first effective method to sort the spermcells. Johnson et al. (24) recently reported successfulseparation of intact viable X and Y chromosome-bearing sperm using this method (see table 4-l).Although the difference in DNA contents of the Xand Y chromosome-bearing sperm in rabbits amountsto only about 3 percent, 94 percent of the rabbits(does) inseminated with X-bearing sorted spermproduced females and 81 percent of the doesinseminated with Y-bearing sorted sperm producedmales. Commercial use of this process is limited, atpresent, by the number of sperm that can be sortedper hour and by increased embryo mortality ob-served in the embryos produced after inseminationwith the sorted sperm. Neither of these factors isthought to represent an insurmountable difficulty.

Embryo Sexing—The most accurate method forsexing embryos is to create a picture of the number,size, and shape of the chromosomes contained in theembryonic cells (karyotyping). However, this methodrequires removal of about half of the cells ofearly-stage embryos, which decreases embryo via-bility and limits the number of embryos that can betransferred. Another method uses antibodies todetect proteins (antigens) unique to male embryos.

This method is not damaging to the embryos andencouraging results have been obtained in onelaboratory (2), however, the technique yields varia-ble results and has not been widely adopted.

More recently, the sex of bovine embryos hasbeen determined by using fragments of DNA that arecontained only on Y chromosomes as a means ofidentifying the same DNA fragments in the embryo(6). Due to its chemical structure, a fragment ofDNA will combine with a second DNA fragmentthat has a corresponding nucleic acid sequence.Therefore, a fragment of DNA that is specific tomales can be used as a probe to identify male DNAfragments in the embryo. Combined with technolo-gies that multiply the number of copies of the DNAfragments, this method determines the sex of theembryo using only a few cells. It is rapid (about 6hrs) and extremely accurate, but may be overtakenby the rapidly developing technology, describedabove, for separating X and Y chromosome-bearingsperm.

Animal Health Technologies

Biotechnology is rapidly acquiring a prominentplace in veterinary medical research. Initially ap-plied to vaccine development, it most recently hascontributed to efforts to develop diagnostic proce-dures and to improve detoxification systems.

Vaccines

Vaccines are agents that stimulate an effectiveimmune response but do not cause disease. Tradi-tional methods of vaccine development involvedkilling or modifying the pathogenic organism toreduce the potential for disease while preserving itsability to induce an immune response. Recombinantvaccine development involves either deletion orinactivation of genes necessary for disease, orinsertion of immunizing genes into nonpathogenicvectors.

Gene deletion technology has been successfullyused to develop both viral and bacterial vaccines. Anaturally occurring mutant of E. coli, for example,has been shown to provide protection againstgram-negative bacterial infections in cattle andswine (15,18). Live Salmonella modified to prevent

gMethods used me differenti~ s~imentation techniques including differential velocity sedimentation free-flow electrophoresis, and Convectioncounter streaming galvanization.

%e antibodies are attached (labeled) to a fluorescent compound to allow for detection.

Table 4-1—Predicted and Actual Sex Ratios of Offspring After Intrauterine Insemination of Sorted X and Y Chromosome-Bearing Rabbit Sperm

Percentage and number of offspring

Number of does Total of Predicted a Actualb

Treatment of sperm Inseminated Kindled young born Males Females Males (N) Females (N).—Sorted Y . . . . . . . . . . . . . . . . . . . . . . . . . 16 5 21 81 19 81 (17) 19 (4)Sorted X . . . . . . . . . . . . . . . . . . . . . . . . . 14 3 16 14 86 6 (1) 94 (15)Recombined X and Y . . . . . . . . . . . . . . . 17 5 14 50 50 43 (6) 57 (8)

Total ... , . . . . . . . . . . . . . . . . . . . . . . 47 13 51 — — 47 (24) 53 (27)aRepresents the results of reanalysis for relative DNA content of aliquots of sorted X- and Y-bearing sperm populations.bRepresents actual biflhs.

SOURCE: L.A. Johnson, J.P. Flook, and H.W. Hawk, “Sex Pre-Selection in Rabbits: Live Births From X and Y Sperm Separated by DNA and Cell Sorting,” Bio/. Reprcd. 41 :199-203, 1989.


reproduction in vivo have also proven to be effectivevaccines for cattle (12).

The first gene deletion viral vaccine to beapproved and released for commercial use was thepseudorabies virus vaccine for swine (26,30). Ini-tially, a single gene deletion reduced the virulence ofthe virus. Since then, other genes have been deletedwith a continuing reduction of virulence.

Vaccines can also be created by inserting intopathogens, genes that produce protective antigensthat reduce the ability of the pathogen to causedisease. 10 Some of these vaccines, however, willhave to be carefully tested because they have a slightpotential to cause human infection (31). Others arein the early developmental stages (notably herpesvirus vaccine and adenovirus vaccine).

Immunomodulators

Immunomodulators are chemical compounds thatboost or accentuate immune response. Several suchcompounds (e.g., interleukins and interferon) havebeen identified in mammals, and the genes encodingsome of these compounds have been isolated andcloned into bacteria (31). Mechanisms by whichthese regulatory proteins modulate immune re-sponse is now being investigated in domesticanimals (l). Biotechnology is being used to identifyand replicate these compounds so that their functioncan be investigated.11

Interleukin genes and genes for compounds thatcause immune responses in animals (antigens) arebeing inserted together into viral or bacterial vac-cines. This combination could possibly enhance theimmune response of the animal and lead to increasedprotection against the antigen. The recombinantinterleukins produced in bacteria or other expressionvectors may also be used therapeutically to assist inovercoming certain infections.

Diagnostics

Safe, accurate, rapid, inexpensive, and easy-to-use diagnostic procedures are critical to the dairyindustry at virtually all points in the production

process. Examples of diagnostic tests include preg-nancy tests and assays for pathogenic organisms.Many of the currently used diagnostic tests arecostly, time consuming, and labor intensive, andsome still require the use of animal assay systems.Monoclinal antibodies and nucleic acid hybridiza-tion probes can be used to produce simpler, easilyautomated, and highly sensitive and specific diag-nostic procedures.

Antibodies are proteins produced by the body inresponse to foreign chemical substances. Monoclo-nal antibodies are produced by a cell line expressingonly a single antibody type (see figure 4-6). Theycan be used to prevent disease,12 and are the primarytools for biotechnology-based diagnostics. At least15 different rapid diagnostic tests are on the marketor will be soon. These tests are highly specific andmost lend themselves to automation, potentiallyallowing their application in mass screening systemsfor disease surveillance and control. Some of thetests have been adapted to field use and can be usedby veterinarians or producers. The rapid commer-cialization of these products is having a significantimpact on animal health management and diseasecontrol.

Monoclinal antibodies are also being used inenzyme-linked-immunoabsorbent-assay (ELISA)systems to provide sensitive, quantitative bloodassays (see table 4-2) of toxins, hormones, chemicals(e.g., pesticide and antibiotic residues), and a varietyof antigens including microbial agents (19,29,34,38).Many of these tests are commercially available. Insome instances monoclinal antibody diagnosticshave been used to replace bioassays such as mouseinoculation tests.

“Nucleic acid hybridization’ can also be used todiagnose the presence of microbes and parasites.Such assays rely on the bonding of a specific DNAor RNA segment to complementary RNA or DNAfragments in a test sample. Specific segments(probes) are available to detect viruses such as

lo~e first veterinq recombinant viral vaccine was the vaccinia vectored vesicular stomatitis vaccine (27). This has been followd by the vaccfiiavectored rabies and rinderpest vaccines (3,5,50).

1 lclon~ ~terle~n genes, such as bovine ~p~, beta and ~mm intefieron, bov~e interlefin.2 ~L.2) etc. have been s~died both h vitro ~d hvivo.

12A comerci~ Prepmation of monocloMI antibodies directed to the K-99 pilus antigens of pathogenic E. cOIi, for examP1e, Prevents ~~hea hnewborn calves (31).


Figure 4-6—Preparation of Monoclinal Antibodies

*

* Mouse isimmunized With

A

a foreignsubstance or“antigen”

. . - - :. - Spleen is.1 removed and

minced torelease antibodyProducing cells(B lymphocytes)

a“. ,..,”. >

. ..’:.--,..t.t.

Myeloma cellsare mixed andfused withB lymphocytes

1

SOURCE: Office of Technology Assessment. 1988.

bluetongue, bovine virus diarrhea, and foot andmouth disease as well as many parasites andbacterial diseases. The major limitation of thistechniqu e is the small amount of target nucleic acidpresent in some samples. Also, the most reliablemethods use radioactively labeled probes, and re-

Y Mouse myelomaa

- % %

(tumor) cellsare removed

t and placed inu tissue culture

0 “..,,. . . , ~.:... s:,.:

Cells divide inliquid medium

Ju

The products of thisfusion are grown in aselective medium. Onlythose fusion productswhich are both “immor.tal ” and contain genesfrom the antibody-pro.ducing cells survive.These are called“hybridomas.”

Hybridomas are clonedand the resulting cellsare screened for anti-body production. Thosefew cells that producethe antibodies beingsought are grown inlarge quantities forproduction of mono-clonal antibodies,

quire expensive equipment and trained technicians,thus precluding their use in the field. Alternativecalorimetric techniques currently in developmentwill replace the radioactively labeled probes andmake the use of this technology more commerciallyattractive.


Table 4-2—MonoclinalAntibodies for Diagnostic Tests

E. coli K99 antigenPseudorabies virus antibodyAvian Ieukosis virus antigenEquine infectious anemia antibodyAvian reovirus antibodyBluetongue virus antibodyBluetongue virus antigenBrucella abortus antibodyAvian encephalomyelitis antibodyBovine progesteroneSulfa methazine in milkCrytosporidiumBovine leukemia virusBovine herpes virusAflatoxin BSulfamethazine-swine

SOURCE: B.1. Osburn, “Animal Health Technologies,” commissionedbackground paper prepared for the Office of TechnologyAssessment, Washington, DC, 1990.

Food Applications

Dairy products will be among the first foodproducts to be impacted by biotechnology. Forexample, during the next decade, the geneticallyengineered version of the enzyme rennet, recentlyapproved by FDA for use in cheese manufacturingsystems, will replace the enzyme preparation nor-mally extracted from the forestomach of calves.Other enzymes, which are added to the curd toaccelerate ripening, or to produce dairy productsacceptable for digestion by lactose-intolerant indi-viduals, will also be produced more economically byengineered microorganisms (22).

Dairy starter cultures are living microorganismsused for the production of fermented dairy productsincluding cheese, yogurt, butter, buttermilk, andsour cream. They have been safely consumed byhumans for centuries and serve as ideal hosts for theproduction of these natural foods. The metabolicproperties of these organisms directly affect theproperties of the food product including flavor andnutritional content. In order to improve variousproperties of food products, food microbiologistsattempt to manipulate the traits of the microorgan-isms, primarily through mutation and selection. Thecloning and gene transfer systems developed in the1980s are being used to construct strains withimproved metabolic properties more rapidly andprecisely than is possible with traditional methods.The development in this decade of new strains withprecise biochemical traits will have an impact onseveral aspects of dairy fermentation, including

production economics, shelf-life, safety, nutritionalcontent, consumer acceptance, and waste manage-ment (22).

Although much of the current work in new straindevelopment has focused on the use of E. coli andother nonfood microorganisms, there are distinctadvantages to engineering starter cultures for pro-ducing high-value foods. For example, constructionof cultures resistant to attack by viral infection willimpact processing costs by eliminating waste. Clon-ing of the genes responsible for ripening of agedcheeses will decrease storage costs by acceleratingripening. Production of natural preservatives such asnisin, effective in inhibiting foodborne pathogensand spoilage organisms, will help ensure the safetyand extend the shelf-life of fermented dairy prod-ucts. Cloning of the gene(s) responsible for enzy-matic reduction of cholesterol or modification of thedegree of saturation of milk fat will improve thenutritional quality of fermented dairy products. Theability to engineer strains capable of producingenhanced flavors or natural stabilizers will influenceconsumer acceptance of fermented dairy foods.

Engineered yeast strains capable of fermentingthe lactose in whey to value-added products, such asvitamin C, biofuels such as ethanol and methanol, orpharmaceuticals, will facilitate management of thiswaste product. Whey protein could potentially beused to produce specialty chemicals with biotech-nology.

Nucleic acid probes and monoclinal antibodiescan be used to analyze raw materials, ingredients,and finished products for pathogenic organisms,bacterial or fungal toxins, chemical contaminants(i.e., pesticides, heavy metals), and biological con-taminants (i.e., hormones, enzymes). Animal cellcultures may partially replace whole animal systemsto test for acute toxicity. Biosensors maybe used tomonitor food processing, packaging, transportation,and storage (22).

KNOWLEDGE-BASEDINFORMATION SYSTEMS

The economic vitality of an animal enterprise isdependent on expert managers who formulate,implement, and continually fine-tune relevant plansand goals in order to optimize resource use andoutput (10). However, producers may have difficultyevaluating the many interrelated factors that go into


such planning (17). Even a relatively simple animaloperation requires that complex decisions be made,based on simultaneous consideration of dynamicallychanging factors related to risk, efficiency, disease,milk production, gestation status, and weather. Withtechnology changing so rapidly, it has becomealmost impossible for agricultural managers tobalance all of the facets of milk production nowunder their control (45).

Many producers rely on consultants and experts tosift through the data and information needed forinformed management decisions. In addition, com-puters have come to play a significant role indairying, an industry that historically has usedrecords to make management decisions. Initially,data resided in mainframe computers. This allowedfor professional maintenance of the database, but theultimate user-producer had limited access to thatdatabase.

Examples of database access via telecommunica-tion systems include the Direct Access to Records byTelephone (DART) system, run by the Dairy Rec-ords Processing Center at Raleigh, North Carolina;and the Remote Management System (RMS) avail-able through Northeast Dairy Herd Improvement(DHI). Although dairy records processing centers(DRPCs) like the one at Raleigh have not developedsoftware for onfarm data calculation and informa-tion storage, the private sector has. Pollock andFredericks (33), for example, offer a microcomputer-based diagnostic program with which producers canavoid the time, recurring costs, and problems ofphone access to a distant mainframe.

As computer languages have evolved and micro-computers decreased in price and gained computingcapacity, database accessibility has increased. Mi-crocomputers provide for direct, rapid delivery ofmanagement data, as well as more efficient datahandling and user interfaces. They have, accord-ingly, revolutionized production record-keeping,and made possible onsite data manipulation andfarm-level processing of information (45).

Expert Systems and Other Computer-BasedDecision Aids

Management decisions rest on a knowledge baseconsisting of two kinds of information: that which iswidely shared and generally publicly available(domain information); and rules-of-thumb judg-ments and sometimes educated guesses (heuristics),

which typically characterize human decisionmak-ing. Both kinds of information are fundamental tocomputer-based expert systems (11), the objectiveof which is to raise the performance of the averageproducer to the expert level (39). Expert systemseffectively and rationally integrate numeric, judg-mental or preferential, and uncertain information, allof which come into play in the biologically based,weather-influenced production systems that typifyanimal agriculture (23). Another promising newinformation technology is the management-infor-mation system, with which managers can test theoutcome of various management alternatives. De-cision-support systems also hold high promise ofenabling managers to balance production inputs in away that maximizes response (output).

Expert systems, knowledge-based systems, ordecision-support systems offer the potential ofbringing the consultant to the farm through themicrocomputer. An expert system provides a flexi-ble yet structured approach to many problems thatExtension specialists now solve relatively routinely(43). Interest in these systems is beginning to emergeas a field of research and development in agriculture,reflecting both industry awareness and appreciationof new information-management technologies (13).With the widespread introduction of specializeddevelopment tools, expert system construction hasaccelerated (25,37). For example, development ofexpert systems was, until recently, restricted toexpensive LISP (a computer language) processingmachines and mainframe computers. Recent ad-vances in hardware and software have made possiblethe development of reasonably sized expert systemson microcomputers. Newly released expert systemshells have removed the necessity to program inLISP (44). While conventional computer programsmanipulate data (11), expert systems manipulateknowledge and help determine which data are usefulto the decisionmaker. They are not competitors butextensions to conventional computer programs.

Application of New Information Technology

Pressure for the management expertise offered bythese new information technologies will grow in the1990s as farms increase in size, new technologiesemerge, prices fluctuate, and consumer concernabout food safety and diet increases. Problemsolving and successful adoption of new agriculturaltechnologies like bST will be facilitated if theknowledge acquired from research and the expertise

Chapter 4-Emerging Technologies in the Dairy Industry . 67

acquired in practice are combined and made readilyavailable in easy-to-use forms.

Indeed, the possibility of fusing expert knowledgefrom different domains (extension, research, producer/managers) into a cohesive, accessible structuremight be the most promising advantage of the newinformation technologies (11). This will allowmanagement opportunities to be maximized, a widergroup of individuals to be reached, and specialists toallocate more time to new areas of concern. Expertsystems, for example, will provide farmers withonline access to needed knowledge: the humanexpert farmers would otherwise rely on gains timefor research and for expanding his or her expertise(40).

New information technologies will also revolu-tionize dairy record-keeping. For example, milk dataare typically recorded once during a 30-day intervaland extrapolated to predict total milk for that period.With new automatic metering devices, milk weightscould be recorded from each milking, For a 305-daylactation, this would increase the data points from12 to 710, if the cow is milked twice a day. Withappropriate data-handling tools, this informationcould be tied to other information, such as measuresof milk conductivity and temperature, and profilesdeveloped to monitor cows for estrus. This wouldallow increased reproductive efficiency, while re-ducing labor requirements, and decrease the need forvisual observation.

CONCLUSIONSAdvances in biotechnology and information tech-

nology will revolutionize the dairy industry. Atten-tion by farm groups, consumers, and policymakershas focused on the first major biotechnology productfrom this new era—bST. In the future bST will besurpassed by more advanced biotechnology meth-ods in animal reproduction, transgenic animal pro-duction, and animal health technologies. The moreadvanced technologies, for example, will increase acow’s endogenous bST production and milk synthe-sis by inactivating genes that inhibit bST production,eliminating the need to administer bST exogenously.Similar advanced technologies will produce higherquality cows, improve disease prevention and man-agement, and allow for the production of high-valuepharmaceuticals in milk.

These advanced biotechnologies will requiresophisticated management capability to use them

effectively. Knowledge-based information systemswill assist in providing this management capability.Expert systems, for example, can help farmersintegrate information for decisionmaking. To effec-tively use these systems farmers will need access tosoftware that is specific to their individual situationand feasible for use in a variety of economic andpolicy situations.

The technologies from this new era are in variousstages of development. Some of these technologies,such as embryo transfer, recombinant DNA vac-cines, and information systems, are already commer-cially available or will be soon. Other technologies,such as transgenic cattle and advanced reproductivetechnologies, will not be available until the end ofthe decade. The next chapter examines the collectiveeffect these emerging technologies, including bST,will have on the dairy industry in the economic andpolicy environment of the 1990s.

1.

2.

3.

4.

5.

6.

7.

8.

9.

CHAPTER 4 REFERENCESAltman, A. and Dixon, F.J., “Irnmunomodifiers inVaccines,” Adv. in Vet. Sci. and Comp. Med. 33:301,1989.Anderson, G. B., “Identification of Embryonic Sexby Detection of H-Y Antigen,” Theriogenology27:81-97, 1987.Belshaw, G.J. et al., ‘‘Immune Response and Protec-tion of Cattle and Pigs Generated by a Vaccinia VirusRecombinant Expressing the F Protein of RinderpestVirus,” Vet. Rec. 124:655, 1989.Bidon, B.M. and Piper, L., “Reproductive Biology ofProlific Sheep Genotypes,” Oxford Review ofReproductive Biology, vol. 8:414-457, J.R. Chuk(cd.) (Oxford, England: Oxford University Press,1986).Blaneon, J. et al., “Safety and Efficacy of anAntirabies Vaccine Consisting of Recombinant Vac-cina — Rabies Virus Administered to Fox, Dog andCat,” Anal. Researches Veterin 20:195, 1989.Bondioli, K.R. et a l . , “The Use of Male-SpecificChromosomal DNA Fragments to Determine the Sexof Bovine Preimplantation Embryos, ” Theriogeno2-ogy 31:95-104, 1989.Bondioli, K. R., Westhusin, M. E., and honey, C. R.,“Production of Identical Bovine Offspring by Nu-clear Transfer,” Theriogenology 33:165-174, 1990.Capecchi, M. R., “The New Mouse Genetics: Alter-ing the Genome by Gene Targeting,” Trends inGenetics 5:70-76, 1989.Cladq A.J. et al., ‘‘Pharmaceuticals From TransgenicLivestock” Trends in Biotechnology 6:179-183,1987.


10.

11.

12.

13.

14.

15.

16.

17,

18,

19.

20.

21,

22.

23.

Conhn, J., “Emerging Technology for Dairy Man-agement,” Proceedings of the American Farm Bu-reau Federation, Orlando, FL, 1990.Doluschitz, R. and Schrnisseur, W. E., “Expert SyS-tems: Applications to Agriculture and Farm Manage-ment,” Computers and Electronics in Agriculture2:173-182, 1988.Dougan, G., Smith, L., and Heffron, F., “LiveBacterial Vaccines and Their Application to Carriersfor Foreign Antigens,” Advances in VeterinaryScience and Comparative Medicine, J.L. Bittle andF.A. Murphy (eds.), Academy Press, San Diego, CA,33:271, 1989.Eastwood, B., program leader, U.S. Department ofAgriculture, Washington, DC, personal communica-tion, 1989.Egletis, M.A. et al., ‘‘Gene Expression in Mice AfterHigh Efficiency Retroviral-Mediated Gene Trans-fer,” Science 230:1395-1398, 1985.Fenwick, B.W. et al., “Mechanisms Involved inProtection Provided by Immunization Against CoreLipopolysaccharides of Eschevieria coli J5 fromLethal Hemophilus pleuropneumoniae Infections inSwine,” Inj and Immune. 53:298, 1986.First, N.L. and Parrish, J.J., “Sperm Maturation andIn Vitro Fertilization, ” Proc. II Internat. Congr.Animal Reprod. and Artif Insemin. 5: 160-168,1988.Gaultney, L. D., The potential for expert systems inagricultural systems management, American Societyof Agricultural Engineers Summer Meeting, EastLansing, MI, 1985.Gonzalez, R.N. et al., “Prevention of Clinical Coli-form Mastitis in Dairy Cows by a Mutant Escherichiacoli Vaccine, ’ Canad. J. Vet. Res. 53:301, 1989.Haines, D.M., Clark E. G., and Chelak, B.J., “TheDetection of Bovine Respiratory Syncytial Virus inForrnalin Fixed Bovine Lung with CommerciallyAvailable Monoclinal Antibodies and Avidin BiotinComplex Irnmunohistochernistry,” Canud. J. Vet.Res. 53:366, 1989.Hansel, W., “Emerging Technology in AnimalReproduction,’ U.S. Congress, Office of Technol-ogy Assessment, Technology, Public Policy and theChanging Structure of American Agriculture, OTA-F-285 (Springfield, VA: National Technical Informa-tion Service, March 1986).Hansel, W., ‘‘Reproduction and Embryo Transfer,”OTA commissioned background paper (Washington,DC: 1990).Harlander, S., “Biotechnology in Food Processing inthe 1990s,” OTA commissioned background paper(Washington, DC: 1990).Higgins, L., Barrett, J. R., and Jones, D. D., “ExpertSystems: Concepts and Opportunities,” Agri. Eng.,Jan/Feb:21-23, 1986.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

Johnson, L. A., Flook, J. P., and Hawk, H.W., “SexPre-Selection in Rabbits: Live Births From X and YSperm Separated by DNA and Cell Sorting,” Biol.Reprod. 41:199-203, 1989.Kinnucan, P., ‘‘Software Tbols Speed Expert SystemDevelopment,” High Technology, Mar. 16-20,1985.Kit, S., Kit, M., and Pirtte, E. C., “AttenuatedProperties of Thymidine Kinase-Negative DeletionMutant of Pseudorabies Virus,” Am. J. Vet. Res.46:1359, 1985.Mackett, M. et al., “Vaccinia Virus Recombinant:Expression of VSV Genes and Protective Immuniza-tion of Mice and Cattle,” Science 227:433, 1985.Massey, J. M., “Animal Production Industry in theYear 2000,” Genetic Engineering of Animuls, W,Hansel and B.J. Weir (eds.), J. Reprod. Fert. Ltd.,1990, pp. 199-208.Nantulya, V.M. et al., ‘‘Monoclinal Antibodies ThatDistinguish Trypanosoma Congolese, T. Vivax, andT Brucii,” Parasit. Immunol. 9:421, 1987.Norby, E., “Modern Approaches to Live VirusVaccines,” Advances in Vet. Sci. and Comp. Med.,J.L. Bittle and F.A. Murphy (eds.), Academy Press,Inc., San Diego 33:249, 1989.Osburn, B. I., ‘‘Animal Health,” OTA commissionedbackground paper (Washington, DC: 1990).Palmiter, R.D. et al., “Metallothionin-Human GHFusion Genes Stimulate Growth in Mice,” Science222,809-814, 1983.Pollock, R.V.H. and Fredericks, T.A., “Provides: AComplete Veterinary Medical Information System,’Can. Vet. J. 29:265-270, 1988.Portelle, D., Marnmerick, M., Burny, A., “Use ofTwo Monoclinal Antibodies in an ELISA Test forthe Detection of Antibodies to Bovine I_akaerniaVirus Envelope Protein,” J. Viro/. Methods 23:211,1989.Prather, R.S. and First, N.L., “Cloning Embryos byNuclear Transfer,’ Genetic Engineering of Animals,W. Hansel and B.J. Weir (eds.) J. Reprod. Fert. Ltd.,1990, pp. 125-134.Prather, R. S., Sims, M. M., and First, N. L., “NuclearTransplantation in Early Pig Embryos.” BioZ. Re-prod. 41:414-419, 1989.Richer, M. H., “An Evaluation of Expert SystemDevelopment Tools,” Expert Systems 3:166-183,1986.Rodriguez, M. et al., “Detection of Bovine Herpes-virus Type 1 via an Irnmunoperoxidase Method,Using Monoclinal Antibodies,” Am. J. Vet. Res.50:619, 1989.Santarelli, M. B., “Artificial Intelligence: Impact onBusiness,” Software News 4(1)30-31, 36, 1984.Schreinemakers, J.F. et al., “The Introduction ofExpert Systems in Animal Husbandry,’ The Veteri-nary Quarterly 10(4), 1988.

Chapter 4---Emerging Technologies in the Dairy Industry . 69

41.

42.

43.

44.

45.

Seidel, G.E., “SexingM ammalianS permandErn- 4 6 .bryos,” Proc. llth Internat. Congr., Animal Reprod.and Arti$ Insemin. 5:136-145, 1988.Smith, L.C. and Wilmut, I., “Factors Affecting theViability of Nuclear Transplanted Embryos,” Theri- 47.ogenology 33:153-164, 1990.Smith, T. R., “The Potential Applications of ExpertSystems in Dairy Extension Education,” J. DairySci. 72:2760-2766, 1989. 48.Spahr, S. L., Jones, L. R., and Dill, D. E., Symposium:Expert Systems —Their Use in Dairy Herd Manage- 49.ment, J Dairy Sci. 71:3, 1988.Tomaszewski, M. A., “Use of Expert Systems in 50.Animal Agriculture,” OTA commissioned back-ground paper (Washington, DC: 1990).

U.S. Congress, Office of Technology Assessment,Commercial Biotechnology: An International Analy-sis, OTA-BA-218 (Springfield, VA: National Tech-nical Information Service, 1984).U.S. Congress, Office of Technology Assessment,New Developments in Biotechnology: Patenting Life— Special Report, OTA-BA-370 (Washington, DC:U.S. Government Printing Office, April 1989).Wessells, N.K. and Hopson, J. L., Biology, (NewYork NY: Random House, Inc., 1988), pp. 321-324.Willadsen, S. M., “Nuclear Transplantation in SheepEmbryos, “ Nature 320:63-65, 1986.Yilma, T, “Adjuvant Genes for Infectious VirusRecombinant Vaccines: Use of Lymphokines asAdjutants,” Biotechnology U.SA., 1987:238.

Chapter 5

Economic and Policy Impacts ofEmerging Technologies on

the U.S. Dairy Industry

TECHNOLOGY ADDITION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .NATIONAL AND REGIONAL IMPACTS UNDER ALTERNATIVE

DAIRY POLICIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ● . . . . .Fixed Price Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..Support Price Adjusted by Trigger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +.... . . . . . +.Milk Production Quota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .Regional Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

NATIONAL IMPACTS UNDER ALTERNATIVE DEMAND ANDSUPPLY SCENARIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Small Demand Reduction . . . . . . .. ....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Large Demand Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . .Large Supply Increase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

FARM LEVEL IMPACTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Alternative Dairy Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Alternative Demand and Supply Scenarios . . . . . . . . .... . . . . . . . . . . . . . . . . . . . . . .Enhancing the Survival of Traditional Farms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

BENEFICIARIES OF TECHNOLOGICAL CHANGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .INTERNATIONAL TRADE PROSPECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .POLICY IMPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. .. .. .. .. .. ..+. . . . . . . . . .CHAPTER PREFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .+ . . . . . . . . . . . . . . . . . . . .

73

7474747575

767676777878808282848485

FiguresFigure Page5-1. Comparative bST Adoption Curves Projected for the Pacific, Lake States,

and Northeast Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735-2. Projected Impact of a Temporary Demand Reduction on Government Milk

Purchases Under Trigger Price Policy, 1990-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 765-3. Projected Impact of a 10-Percent Permanent Demand Reduction on Government

Purchases Under Alternative Dairy Policies, 1990-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775-4. Dairy Price Indexes at Three Market Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 835-5. Farm and Retail Prices of Beverage Milk per Half Gallon . .. . .......+. ., ...., ..+ 835-6. Farm, Wholesale, and Retail Prices of Cheese . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

Table5-1.5-2.

5-3.

5-4<

5-5.

5-6.

5-7.

5-8.

5-9.

TablesPage

Forecasted Adoption Rates for bST Technology, Selected Years, 1991 -2000 . . . . . 73Level of Milk Production, With and Without bST, Under Alternative PolicyScenarios, 1990-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74Level of CCC Purchases, With and Without bST, Under Alternative PolicyScenarios, 1990-98, Milk Equivalent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .74

Summary Characteristics of Representative Moderate- and Large-SizeDairy Farms, by Region . . . . . . . . . . . . . . . ... , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79Impacts of bST Adoption on the Economic Viability of Moderate-SizeRepresentative Farms, Assuming No Change in Demand for Milk Due to bST,Trigger Price Policy, by Region, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79Impacts of bST Adoption on the Economic Viability of Large-SizeRepresentative Farms, Assuming No Change in Demand for Milk Due to bST,Trigger Price Policy,by Region, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79Impacts of bST Adoption on the Economic Viability of RepresentativeLarge-Size (125-COW) Upper Midwest Farms Under Alternative Dairy Policies,Assuming No Change in Demand for Milk, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80Comparison of Average Annual Economic Payoffs from bST Adoption for EightRepresentative Dairy Farms Under Three Alternative Dairy Policies, AssumingNo Change in Milk Demand, 1989-1998 . . . . . . +... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80Impacts of bST Adoption on the Economic Viability of Moderate-SizeRepresentative Farms, Assuming Small Decrease in Demand for MilkDue to bST, Trigger Price Policy, by Region, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . 81

5-10. Impacts of bST Adoption on the Economic Viability of Large-Sizeepresentative Farms, Assuming Small Decrease in Demand for MilkDue to bST, Trigger Price Policy, by Region, 1989-98 . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Chapter 5

Economic and Policy Impacts of EmergingTechnologies on the U.S. Dairy Industry

Technologies from the biotechnology era willplay an important role in sustaining or acceleratingthe historical trend of constantly increasing milkoutput per cow. The new technology likely to havethe most to do with this growth is bovine somato-tropin (bST). In the following analysis of theeconomic and policy implications of emergingtechnologies, special emphasis will be given to bST.As with any analyses, the conclusions are only asaccurate as the assumptions made. Of specialinterest and importance is the assumption regardingthe adoption of bST by farmers.

TECHNOLOGY ADOPTIONIt is not known when and how many dairy farmers

will adopt new technologies, such as bST, once theybecome commercially available. Several studies ofbST either directly address the issue of adoption ormake assumptions regarding adoption rates andpatterns. In a survey of dairy farmers, Lesser et al. (6)found that about 50 percent of respondents wouldadopt bST within the frost year of its commercialavailability, and that over 80 percent would within3 years. Most analysts, relying heavily on suchstudies, have tended to assume relatively rapidadoption of bST (1,4).

However, the use of surveys to indicate prospec-tive adoption rates of a technology that is not yetavailable is problematic. For example, informationregarding the technology is incomplete. Most of thebST surveys were done several years ago when therewas little negative reaction from public interestgroups. Moreover, new dairy technologies, as ageneral rule, have not tended to be adopted rapidly.For example, despite having been available com-mercially for over 40 years, artificial inseminationtechnology is used only by 65 to 70 percent of dairyfarmers. Likewise, Dairy Herd Improvement (DHI)technology, available for 50 years, is used by only45 percent of farmers (13). In addition, regionallyvariable patterns of use are associated with bothtechnologies.

This report considers the history of technologyadoption by farmers for insight into potential rates ofbST adoption. Statistical analyses indicate that the

- .

variables most closely (and positively) related tofarmer adoption of new technologies (e.g., auto-matic grain-feeding systems, automatic milking-unitremoval, three-times-a-day milking, DHI, and artifi-cial insemination) were milk output per cow and sizeof herd. Efficiency in the utilization of capital, labor,and feed were also found to be significantly relatedto technology adoption in particular regions (7).

Using this information, and assuming that adop-tion of bST would closely parallel that of othertechnologies, bST technology adoption curves werederived (see figure 5-l). (See app. A for details.)Comparative regional information on the level ofadoption after 1,5, and 10 years is contained in table5-1. The results reflect:

. the tendency of the dairy industry to adopttechnologies at different rates regionally;

. the progressiveness of the Pacific region dairyindustry compared with that of other regions,including traditional milk production regions;and

● a slower rate of adoption than is indicated byproducer surveys of probable bST use, and onethat is more typical of past dairy industrytechnology adoption patterns.

Figure 5-l-Comparative bST Adoption CurvesProjected for the Pacific, Lake States,

and Northeast Regions

80

01990 1992 1994 1996 1998 2000

Year. . . Pacific — Lake States -. N o r t h e a s t


- 7 3 -


Table 5-l—Forecasted Adoption Rates forbST Technology, Selected Years, 1991-2000a

Percent of farmers adopting

Region 1991 1995 2000

Pacific . . . . . . . . . . . . . 17% 46% 67%Lake States . . . . . . . . . 15 35 46Northeast. . . . . . . . . . . 15 31 43Appalachia . . . . . . . . . 15 32 46Southeast . . . . . . . . . . 15 29 39Southern Plains . . . . . 13 34 42Corn Belt . . . . . . . . . . . 13 25 31%ST is assumed to be commercially available in 1991.

—


OTA’s analysis indicates that during the first yearthat bST is commercially available, no more than 17percent of farmers will use it. After 5 years, bSTadoption is forecast to range from 25 percent in theCorn Belt to 46 percent in the Pacific region. After10 years, bST adoption is forecast to range from31percent in the Corn Belt to 67 percent in the Pacificregion.

NATIONAL AND REGIONALIMPACTS UNDER ALTERNATIVE

DAIRY POLICIESFuture milk-supply prices and dairy farmer re-

turns are determined by the interactions of technol-ogy adoption, consumer demand, and dairy policy,as established in the 1990 farm bill. These inter-actions were captured using a national computersimulation model referred to as LIVESIM with thefollowing assumptions:l

*

a

o

●

*o

●

regional adoption curves as indicated in thepreceding section;output per cow increases 1.5 percent per year inbase scenario without bST;output per cow increases 1,320 pounds annu-ally with use of bST;2

bST is injected for 150 days annually;cost of bST use is $0.30 per cow per day;feed efficiency increases by 5 percent duemainly to bST; andthe minimum level of government purchases bythe Commodity Credit Corporation to satisfy

food program needs (i.e., school lunch pro-gram, etc.) is 3.0 billion pounds annually.

The policy options analyzed included a fixed pricesupport, a price support trigger, and a quota program.It is important to keep in mind that this analysisbegins with the industry in relative supply-demandbalance and in the absence of strong incentives foreither increased or reduced production (10).

Fixed Price Support

This option fixes the price support level at $10.60per cwt ($0.50 per hundredweight (cwt) higher thanthe level authorized by the 1990 farm bill) for allyears and serves as a useful bench mark for policyoption comparisons. In this scenario, the govern-ment purchases excess milk, at the support price, inorder to clear the market. Without bST, milkproduction would increase progressively under thisscenario from a projected 144 billion pounds in 1990to 152 billion in 1995 (see table 5-2). With bST,production would increase an additional 4 to 5billion pounds over the period 1994 to 1998; annualgovernment purchases for food programs would riseas high as 9.0 billion pounds, but generally increaseby 3 to 4 billion pounds over the minimum (3 billionpounds) (see table 5-3).

Support Price Adjusted by Trigger

This option, similar to policy from 1985-1990,triggers a price support reduction each time the levelof government purchases rises above 5.0 billionpounds annually. This option resembles the assess-ment option in the 1990 farm bill that effectively willtrigger reductions in producer returns as milk pricedeclines. The simulation period begins with a milksupport price of $10.60 per cwt. This is adjusteddownward in $0.50 per cwt increments in any yearthat expected government purchases are greater than5 billion pounds. Without bST, a single price supportreduction brings the support price to $10.10 per cwtin 1991. With bST, two price support reductions aretriggered; one in 1991 and another in 1993, reducingthe price support level to $9.60 per cwt. These pricereductions moderate production increases to keep

~L,IVESIM was &VelOped by D.S. Peel, Department of Agricultural Economics, Oklahoma State University. App. B provides a description of tiemodel and detailed results of the analysis.

Z’rhe ticreme in output per cow in a given herd tends to be closer to an absolute number of pounds of milk than to a perCentage ticrease. neEfO%approximately the same increase in pounds of milk produced pcr cow might be expected in comparably managed herds with cows each producing 12,000to 20,000 pounds of milk per year.

Chapter 5-Economic and Policy Impacts of Emerging Technologies on the U.S. Dairy Industry ● 75

Table 5-2—Level of Milk Production, With and Without bST, Under Alternative Policy Scenarios, 1990-98(billions of pounds)

Policy scenarios



1990 . . . . . . . . . . . . . . . . . 144 144 144 144 144 1441991 . . . . . . . . . . . . . . . . . 146 144 146 144 146 1441992 . . . . . . . . . . . . . . . . . 146 143 146 143 145 1441993 . . . . . . . . . . . . . . . . . 153 150 153 150 148 1461994 . . . . . . . . . . . . . . . . . 153 149 152 148 150 1481995 . . . . . . . . . . . . . . . . . 156 152 156 152 152 1501996 . . . . . . . . . . . . . . . . . 157 153 155 153 155 1531997 . . . . . . . . . . . . . . . . . 160 155 159 155 157 1551998 . . . . . . . . . . . . . . . . . 161 157 159 157 160 157

abST is assumed to becommercially available in1991.

SOURCE: Office of TechnoloW Assessmen~ 1991.

Table 5-3-Level of Government Purchases, With and Without bST, Under Alternative Policy Scenarios,1990-98, Milk Equivalent (billions of pounds)

Policy scenarios



1990 . . . . . . . . . . . . . . . . .1991 . . . . . . . . . . . . . . . . .1992 . . . . . . . . . . . . . . . . .1993 . . . . . . . . . . . . . . . . .1994 . . . . . . . . . . . . . . . . .1995 . . . . . . . . . . . . . . . . .1996 . . . . . . . . . . . . . . . . .1997 . . . . . . . . . . . . . . . . .1998 . . . . . . . . . . . . . . . . .

3.07.34.39.06.07.04.85.33.6

3.05.33.05.73.03.03.03.03.0

3.07.33.06.83.03.03.03.03.0

3.05.33.03.83.03.03.03.03.0

3.07.3 “3.53.43.13.03.03.03.0

3.05.33.03.03.03.03.03.03.0

abST is assum~ to be commercially available in 1991.


Commodity Credit Corporation (CCC) purchasesnear the 3.0-billion-pound minimum (see table 5-3).

Milk Production Quota

Several proposals have been made to improvesupply control in Federal dairy policy. Quotasystems utilized in California and Canada, forexample, have been suggested for use nationally inthe United States. While these systems differ in theirimplementation, each results in a much strongeropportunity for management of excess dairy produc-tion. In the simulation model, control of milkproduction is accomplished by reducing the numberof cows in a herd. In practice, these reductions mightbe triggered by a two-tiered pricing system or someother mechanism that provides disincentives forproducing over quota levels.

The quota policy is designed to maintain govern-ment purchases at or near the minimum government

use target of 3.0 billion pounds. The quota isadjusted downward any year CCC purchases exceed3.0 billion pounds. The price support remains at$10.6O per cwt; however, the market price is allowedto adjust as under the other options. The quota yieldsa much stabler market price, one that is generallyhigher than that under the trigger option. However,with bST a tendency still exists for the price to restat the support level. The quota avoids the high levelof government purchases necessary under the fixedprice support scenario (see table 5-3).

Regional Impacts

Substantial controversy has arisen over the poten-tial regional impacts of bST and other emergingtechnologies. The results of this analysis suggest acontinuation of current trends toward greater con-centration of production in the Pacific region and thelargest decline in the Corn Belt. Shifts in futuremarket shares will be largely a function of differ-


ences in rates of adoption of bST and othertechnologies. The more rapid rate of bST adoptionpredicted for the Pacific region will increase itsmarket share even faster than the historical trend.

Regional shifts in production patterns could bemoderated by changes in farm policy. The marketmechanism, as reflected in the trigger price supportmechanism, places the greatest pressure on highercost producers and regions. The freed price supportblunts the price declines associated with supplyincreases, thus providing a degree of protection tohigher cost regions. Quotas tend to freeze productionpatterns. Thus, the regional impacts of bST and othertechnologies could be reduced through the adoptionof a quota policy. However, by freezing productionpatterns, quotas discourage efficiency. The benefitsof the quota tend to be capitalized in fixed-assetvalues, thus raising costs, particularly to new en-trants to the industry (i.e., new entrants must buyquota from current dairy operators). And, becausedairy farmers would not want to see a valuable asset(quota) lose its value, it would also be difficult todiscontinue a quota policy.

NATIONAL IMPACTS UNDERALTERNATIVE DEMAND AND

SUPPLY SCENARIOSMany claims have been made concerning the

potential adverse health impacts of milk producedwith bST. While these claims remain unsubstanti-ated, consumer perceptions can be more importantthan reality in determining market demand. Asindicated previously, initiatives to label milk pro-duced by cows receiving bST could create aperception that consumption of this milk may not bedesirable. Since policy needs to be designed consid-ering the full range of potential developments, twoscenarios regarding reduced milk consumption wereanalyzed. One of these involved a substantial buttemporary reduction in demand followed by recov-ery to a smaller long-term reduction. The secondinvolved a large permanent reduction.

Small Demand Reduction

The small reduced-demand scenario drops per-capita demand by 10 percent (about 55 pounds) in1991,5 percent in 1992 (i.e., demand increases from1991 to 1992), and 2.5 percent permanently there-after. The effects of these reductions are CCCpurchases totaling 21.2 billion pounds (14.5 percent

Figure 5-2—Projected Impact of a Temporary DemandReduction on Government Milk Purchases Under

Trigger Price Policy, 1990-98

21.2 IA

1990 1992 1994 1996 1998

Year


of production) in 1991, 9.7 billion pounds in 1992,and 8.4 billion pounds in 1993 (see figure 5-2). Theanalysis assumed a continuation of the triggermilk-price support policy. The support trigger de-creases the price support level down to $9.10 per cwtin 1994. In 1994, the industry begins to stabilize atthe 3.0-billion-pound minimum purchase level. Eventhough government purchases are exceedingly highfor 3 years, the trigger mechanism seems to accom-modate a small demand reduction quite well.

Large Demand Reduction

The second reduced-demand scenario assumes apermanent 10-percent reduction in per-capita con-sumption. If the price support is sustained at $10.60per cwt under the fixed support scenario, CCCpurchases continue at a level that approaches orexceeds 20 billion pounds of production through1998 (see figure 5-3). This would exact a high costto the government.

While the trigger mechanism copes reasonablywell with a small permanent consumption reduction,the industry has difficulty adjusting to a largepermanent demand reduction scenario with thismechanism. The support price must be triggereddown to $7.60 in 1997 in order to bring CCCpurchases to below 4 billion pounds. Such a lowsupport price would make it difficult (impossible)for even the best managed dairy farms to avoideconomic losses. As indicated in figure 5-3, for eachof the years 1991 to 1995, the CCC is purchasing atleast 12 billion pounds (at least 8 percent of the milksupply).

Chapter 5--Economic and Policy Impacts of Emerging Technologies on the U.S. Dairy Industry ● 77

Figure 5-3—Projected Impact of a 10-PercentPermanent Demand Reduction on Government

Purchases Under Alternative Dairy Policies,1990-98

301

.~. ... ~o*. .o .“ ● : . ● “ = “ “ “ . . . .

. - - - - - - - - - - - — - - - -

1990 1992 1994 1996 1998

Year— Trigger Price s ■ I Fixed Support

-- Dairy Termination - - Quota


Under this reduced demand scenario, a dairytermination program might be considered as analternative to the severe price support reductiondiscussed above. A termination program involves aone-time buyout of dairy cows, to be implementedwhen government purchases reach a certain level. Inthe model, the level was established at 15 billionpounds annually. When government purchases reachthis level, enough dairy cows would be liquated thefollowing year to eliminate the excess production.

Such a termination policy would be triggered in1992 because at least 21 billion pounds of CCCpurchases would have occurred in 1991 (see figure5-3). The herd kill to bring CCC purchases down tothe minimum 3.0 billion pounds would be 1.3million cows (13.1 percent of the herd). In theprocess, cow prices would fall by $6.11 per cwt (12percent) with a 6. l-percent drop in beef cattle prices.(See app. B.)

Once the termination is completed, milk produc-tion bounces back and CCC purchases exceed 14billion pounds in the next year (1993). This result issimilar to that of the 1986 Dairy Termin ationProgram. The lowest producing cows on average areliquated from the industry. The higher producingcows remain, providing the industry with the capa-bility of responding to increased prices. Anothertermination probably would not be feasible becauseof the high cost associated with the program and the

tendency of farmers to bid up the cost of selling out.However, the support price still would decline to$7.60 per cwt in 1998—a year later than under thetrigger option without the termination program, onceagain verifying that termination programs do notresult in permanent reductions in supply.

If a quota were imposed in 1992 with the objectiveof bringing CCC purchases back down to theminimum 3.0 billion pounds, 12.2 percent of thedairy herd (1.2 million cows) still would be sent tomarket (slaughtered) in order to reduce the herd toabout 8.8 million cows. This compares with 1.3million cows slaughtered under the termin ationprogram. Under the quota, the dairy cow price falls8.1 percent (compared with 12.0 percent under thetermination program) while the beef cattle price fallsby 4.3 percent (compared with 6.1 percent under thetermination program). Perhaps more important, thequota program effectively controls milk supply. (Seeapp. figure B-13.)

This analysis suggests that if a large permanentreduction in demand occurred, changes in dairypolicy would most likely be needed. A fixed supportprice policy would be too costly and a trigger pricepolicy or producer assessments would be too severeto producers. The policy alternatives are a termina-tion program or a quota. A termination program iscostly and does not result in permanent reductions insupply. A quota program does effectively controlsupply and, compared to the termination program, isless costly. Benefits of a quota tend to be capitalizedinto fixed asset values, thus raising the costs to newentrants and making it difficult to discontinue thequota policy. Thus, consideration should be given toobserving CCC purchases over a 2-year period, asopposed to 1 year, before a quota is implemented.This would help to determine whether demandreduction is permanent or temporary.

Large Supply Increase

Previous survey-based analyses of the impact ofbST typically assume a considerably higher rate ofadoption than this study predicts, based on pastadoption patterns. If bST results in a 15-percentincrease in the milk supply in the first year, insteadof the 5-percent increase used in the above analysis,CCC purchases rise to at least 20 billion pounds.Large supply increases could be realized not onlythrough rapid adoption of bST, but also as firms that


participated in the 1986 Dairy Termination programreenter the market beginning in 1991.

The impact of a 15-percent supply increase wouldbe similar to that of a 10-percent permanent demandreduction, i.e., CCC purchases equal 20 billionpounds in the first year (see app. figure B-12). Inboth instances, it takes 5 years for the price supporttrigger mechanism-even with a price support aslow as $7.60--to bring CCC purchases down from20 billion pounds to no more than 10 billion pounds.The problems of managing large government pur-chases over such a long period suggests the need toconsider production management options. Hereagain, the termination program only reduces produc-tion temporarily with substantial negative impactson beef prices. Quotas are effective at controllingproduction but also negatively affect beef prices,although not to the same degree as a singletermination option.

FARM LEVEL IMPACTSThe combined impacts of emerging technologies,

dairy policies, regional differences in productioncosts, and long-term industry trends can be moreeasily visualized at the level of individual dairyfarms. Representative farms are briefly described intable 5-4. The parameters of representative farmswere originally developed for OTA in 1985 and havesince been continuously updated by Agricultural andFood Policy Center faculty and staff at Texas A&MUniversity. The farms are simulated with andwithout bST adoption utilizing the FLIPSIM model.3

The farm level impacts of the three policyscenarios-fixed support price, trigger, and quota—were analyzed over the period 1989 to 1998. Itwas assumed that the same farm program provisionsoperated over the 10-year period. The initial analy-ses were conducted assuming no change in demand.Subsequently, alternative demand assumptions wereanalyzed (1 1).

Alternative Dairy Policies

The analysis indicates that once bST becomesavailable, there will be strong incentives to adopt thetechnology. Regardless of the region, the payoffsfrom bST adoption are substantial. For example,with the trigger price policy, the 52-cow Upper

Midwest dairy, a typical, moderate-size dairy farmin this region, enhances its chance of survival(probability that the farm will remain solventthrough 1998) from 58 to 74 percent by adoptingbST once it becomes available (see table 5-5). Thesame is true for large dairies (see table 5-6).Nonadopters of bST have more problems survivingand, therefore, are more likely to exit the industry.

Tables 5-5 and 5-6 provide insight into competi-tive conditions in the dairy industry and the reasonsfor regional shifts in milk production patterns.Regardless of size, Upper Midwest farms haveproblems realizing sufficient earnings to achieve areasonable return on equity, compete, and survive.While Northeast farms perform better, they too werefound to be at a disadvantage relative to the Pacificand Southeast farms.

These results raise questions about the advisabil-ity of State laws placing a moratorium on the use ofbST. Dairy farmers located in States that have put amoratorium on adoption will be placed at a substan-tial disadvantage relative to those in unrestrictedStates. If moratoriums are imposed in regions wherefarm survival probabilities are already low (relativeto other regions), the impact of a moratorium can beparticularly severe.

Policies and the choice between bST adoption ornonadoption operate together to impact survival ina number of ways (see table 5-7). Higher earningsresulting from the fixed price support increase theprobability of survival for a 125-cow Upper Mid-west dairy and the chances of a 5-percent return oninitial equity. However, even with adopting bST, networth continues to erode.

Surprisingly, perhaps, the quota program per-forms worse than either the trigger price or the freedprice support. This is because the quota priceobjective is the same as the freed price support($1O.6O) and because restrictions on output curb-bexpansion and raise costs per cwt. Thus, if a quota isto be imposed, the price objective must be suffi-ciently high to offset the effects of lower production(higher production costs per cwt) or producers couldbe worse off.

The absolute economic payoff from bST adoptionis about the same under a trigger price policy and a

3~~s1M ~&~ deve.oped by J.w, ~c~d~~n, Dep~tment of A@cul~~ ~onomics and C.J. Nixon, rlep~rnent of Accounting, Texas A&MUniversity (12). App. C provides a description of the model and detailed results of the analysis.

iTable 5-4—Summary Characteristics of Representative Moderate-Size and Large Dairy Farms, by Region

Upper Midwest Northeast Southwest a Southeast

Characteristic Moderate Large Moderate Large Moderate Large Moderate Large

Cow numbers . . . . . . . . . . . . . . . . . . 52 125 52 200 350 1,500 200 1,500Output/cow (pounds) . . . . . . . . . . . . 16,850 16,850 17,940 17,830 18,590 19,690 15,340 15,310Total asset value ($000) . . . . . . . . . . 470 940 608 1,395 1,097 3,858 1,569 7,723Land value ($000) . . . . . . . . . . . . . . . 133 295 274 640 118 492 813 4,591Percent of feed raked . . . . . . . . . . . 63 60 50 46 0 0 25 2a[ncl~es farms from both the Pacific and Mountain USDA production regions


Table 5-5—Impacts of bST Adoption on t he Economic Viability of Moderate-Size Representative Farms, Assuming No Changein Demand for Milk Due to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

52-cow 52-cow 350-COW 200-COWUpper Midwest Northeast Southwest Southeast—

Non- bST Non- bST — Non- bST Non- bSTMeasure of impact adopter adopter adopter adopter adopter adopter adopter adopter

Probability of survivala . . . . . . . . . . . . . . . . . . . . . . 58% 74% 100% 100% 95% 97% 100% 100%Probability of earning 5-percent return

on equity . . . . . . . . . . . . . . . . . . . . ., , . . . . . . . . . 58 74 100 100 95 97 100 100Probability of increasing equityb . . . . . . . . . . . . . . . 0 0 3 3 60 79 13 24Present value of ending net worth as percent


Whance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.%hance that the individual farm will increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.


Table 5-6—impacts of bST Adoption on the Economic Viability of Large Representative Farms, Assuming No Change in Demand for MilkDue to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

125-cow 200-COW 1 ,500-COW 1,500-cowUpper Midwest Northeast Pacific Southeast


Probability of survivala . . . . . . . . . . . . . . . . . . . . . . 95% 99% 100% 100% 100% 100% 100% 100%Probability y of earning 5-percent return

on equity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 95 99 100 100 100 100 100Probability of increasing equityb . . . , . . . . . . . . . . . 8 12 43 53 100 100 88 99Present value of ending net worth as percent


W2hance that the individual farm will remain solvent through 1998, i.e., maintain more than a Io-peroent equity in the farm.%hance that the individual farm will increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars,

SOURCE: Office of Technology Assessment, 1991,

I


Table 5-7—impacts of bST Adoption on the Economic Viability of Representative Large (1 25-cow)Upper Midwest Farms Under Alternative Dairy Policies, Assuming No Change

in Demand for Milk, 1989-98 (in percent)

Trigger price Fixed price support Quota

Non- bST Non- bST Non- bSTMeasure of impact adopter adopter adopter adopter adopter adopter

Probability of survivala. . . . . . . . . . . . . . . . . 95% 99%. 99% 100% 85%. 92%.

Probability of earning 5-percentreturn on equity . . . . . . . . . . . . . . . . . . . . 90 95 95 98 67 78

Probability of increasing equityb . . . . . . . . . 8 12 11 18 2 3Present value of ending net worth as

percent of beginning net worthc. . . . . . . . 57 69 67 78 37 46%hance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.bchance that the individual farm will increase its net worth in real 1989 dollars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.


freed support price policy for the representativedairy farms (see table 5-8). Increasing the price ofmilk by maintaining the milk support price at itscurrent level does not greatly increase the economicincentive to adopt bST, but that incentive is signifi-cantly lower if a quota is in effect. This suggests thatthe rate of bST adoption would be slowed byimposing a quota rather than continuing the triggerprice policy.

Alternative Demand and Supply Scenarios

Potential reductions in demand due to consumerconcern over bST would reduce the economicpayoffs from using this technology. The mostsignificant result of such demand reduction isreduced economic viability of all dairy farms, andparticularly of those in the Midwest. For example,the economic payoff for bST adoption is $10,300 forthe 125-cow dairy in the Upper Midwest if there isno decrease in milk demand. If demand decreasedslightly, the economic payoff falls to $9,200 and ifthe demand decrease is large, the economic payoffdeclines to $6,900. Thus, the incentive to adopt andthe rate of adoption would be reduced if milkdemand declines.

The adverse impacts of reduced demand could becountered by either a termination program (in theevent of a small reduction in demand) or by a quota(if larger reductions in demand occurred).4 How-ever, even with reduced demand, there would bestrong incentives to adopt bST for all farms in all

Table 5-8—Comparison of Average AnnualEconomic Payoffs From bST Adoption forEight Representative Dairy Farms Under

Three Alternative Dairy Policies, AssumingNo Change in Milk Demand, 1989-98a

(thousands $)

Policy scenarios

Trigger FixedRegion/size price support Quota

Lake States:Moderate . . . . . .Large . . . . . . . . .

Northeast:Moderate . . . . . .Large . . . . . . . . .

Southwest:Moderate . . . . . .Large . . . . . . . . .

Southeast:Moderate . . . . . .Large . . . . . . . . .

3.910.3

3.415.8

26.590.5

21.9166.4

4.110.9

3.616.6

26.691.7

22.8166.3

2.47.0

1.08.8

18.361.2

17.2132.0

aEconomic payoffs from bST are the average annual change in net cashfarm income between a nonadopter and a bST adopter over the 1989 to1998 planning horizon. The payoff is net of the cost of bST, the addedtransportation costs for milk, and the additional feed.


regions. For example, with a continuation of thetrigger policy, a 52-cow Upper Midwest dairy’sprobability of survival declines to 40 percent undera small decrease in demand; adopting bST boostssurvival probability under this scenario to 48 percent(see table 5-9). Similar trends hold true for the largerdairies (see table 5-10). Thus, the economic payofffor bST adoption is positive; those who adopt bSTwill experience greater probabilities of survival and

4Sm~1 and lmge dem~d r~uctiom are the same as explained in the previous section. A small demand reduction assumes tit milk dmand woulddecrease 10 percent in 1991, 5 percent in 1992 (i.e., demand increases from 1991 to 1992), and 2.5 percent each year in 1993-1998. A large demandreduction assumes that milk demand wotdd decrease 10 percent in each year 1991-1998,

II

Table 5-9—impacts of bST Adoption on the Economic Viability of Moderate-Size Representative Farms, Assuming Small Decrease inDemand for Milk Due to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

52-cow 52-cow 350-COW 200-COWUpper Midwest Northeast Southwest Southeast


Probability of survivala . . . . . . . . . . . . . . . . . . . . . . 40% 48% 100% 100% 88% 94% 99% 100%Probability of earning 5-percent return


of beginning net worthc . . . . . . . . . . . . . . . . . . . . 3 10 65 70 79 99 58 71Whance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.bchana that the individual farm will increase its net worth in real 1989 dolIars through 1998.cPresent value of ending net worth divided by initial net worth indicates whether the farm increased (decreased) net worth in real dollars.


Table 5-10—Impacts of bST Adoption on the Economic Viability of Large Representative Farms, Assuming Small Decrease in Demandfor Milk Due to bST, Trigger Price Policy, by Region, 1989-98 (in percent)

125-cow 200-COW 1 ,500-COW 1 ,500-COWUpper Midwest Northeast Pacific Southeast


Probability of survivala . . . . . . . . . . . . . . . . . . . . . . 85% 91% 100% 100% 100% 100% 100!40 100%Probability of earning 5-percent return



%hance that the individual farm will remain solvent through 1998, i.e., maintain more than a 10-percent equity in the farm.%hance that the individual farm will increase its net worth in real 1989 dollars through 1998.CPresent value of ending net worth divided by initial net worth indicates whether the farm inoreased (decreased) net worth in real dollars.

I



economic success than nonadopters. The positiveeconomic payoffs for bST adoption are greater underthe dairy termination program than under the triggerprice policy. Thus, bST adoption would be acceler-ated even with declining milk demand if a termina-tion program were introduced in lieu of the triggerprice policy.

The supply impacts estimated by the LIVESIMmodel in the previous section were based on pastadoption practices, not farmer survey results, whichindicate higher adoption for bST. If the surveyresults are accurate predictors of adoption then alarge increase in supply would occur. Unless supplycontrols such as a dairy termination program or aquota are imposed, adverse impacts on economicviability would be substantial (see app. tables C-12to C-17).

Enhancing the Survival ofTraditional Farms

Results of the preceding analysis indicate thatsmaller, less efficient, farms will have difficultiesrealizing sufficient earnings to achieve a reasonablereturn and survive even with the adoption of bST.Northeast farms perform better, in part, because theyreceive higher Federal milk marketing order prices.Farms that do not adopt bST will feel even moresevere impacts.

The economic viability of smaller farms may beenhanced by changes in scale of operation, progres-siveness in technology adoption, research and exten-sion, and dairy policy. The following provides abrief discussion of the importance of each item.

●

●

Scale of Operation--Generally, larger farmsexperience lower costs of production. Studiesnow in progress indicate that in the UpperMidwest and Northeast, economies of size haveresulted in the establishment of larger herdsthat have the potential to realize even moreeconomies of size involved in dairying. Farmswith herds larger than 125 cows in the UpperMidwest and 200 cows in the Northeast will bemore likely to lower their costs of productionand compete than smaller operations.Technology Adoption--A key to achieving theeconomic benefits of a new technology is toadopt it early. The traditional milk productionregions have a history of lagging behind otherregions in adopting new technology. This studyhas shown that, based on experience, the Upper

●

●

Midwest and Northeast regions will lag behindthe Pacific region by more than 20 percent inthe adoption of bST Ways must be found toencourage producers in these regions to adoptnew technology earlier to enhance their proba-bility of economic success.Research and Extension-Little, if any, empha-sis is given to conducting research and provid-ing extension services to different-size farms.Small, moderate, and large farms each havetheir own unique problems, particularly from amanagement perspective. Research is neededon developing management strategies for eachfarm size. Extension strategies also need to bedeveloped to assist farmers in technologyadoption so they can receive more of thebenefits of new technologies. Laggards intechnology adoption receive little economicpay-off.Dairy Policy--Based on this study’s analysis,a fixed support price policy provides farms inthe traditional milk-producing regions withhigher earnings that increase their probabilityof survival and the chance of earning a 5-percent return on equity. However, even withthis policy, net worth continues to erode forthese farms. Thus, the support price may needto be increased. This is, of course, more costlyin terms of government expenditures. An alter-native would be to target these farms for ahigher support price-but it still will be morecostly and administratively complex comparedto other alternatives. However, if substantialprogress were made on the items discussedabove possibly no change in policy would beneeded.

BENEFICIARIES OFTECHNOLOGICAL CHANGE

The issue of who benefits from technologicalchange is not new but is relevant to this study. Thefarm-level results indicate that bST adopters arebetter off than nonadopters. First adopters, more-over, are the greatest beneficiaries of any technolog-ical change. They receive a relatively high price fortheir product and realize the cost reductions result-ing from bST use. As more farmers adopt, the marketprice falls, which makes the consumer the ultimatebeneficiary. As the market price falls, farmers whodo not adopt may be forced out of business.

Chapter 5-Economic and Policy Impacts of Emerging Technologies on the U.S. Dairy Industry ● 83

Figure 5-4—Dairy Price Indexes atThree Market Levels (change from prior month)

6.0 —

, - - - - - - - - - - \3.0 — ,#’

m ,0g , ’o: 0.0% bc * \ 4’

-3.0 — ‘ * * - - - * ” ’ \I1I9

-6.0

-9.0

J F M A M J J A S O N D J F M

1989Month

1990

- - - R e t a i l p r i c e s , — Wholesale prices, ---- Farm prices,all dairy products all dairy products all milk

SOURCE: Compiled from U.S. Department of Agriculture dairy statisticaldata by Andrew Novakovic, DairyMarketing fWes, 1990, No. 2,Department of Agricultural Economics, Cornell University, lth-aca, NY.

Questions have been raised regarding whetherconsumer prices do, in fact, decline as producerprices fall. Some groups opposing the approval ofbST have attempted to show that retail milk pricesdo not fall as producer prices decline and, therefore,consumers do not benefit from the technology (3). Areview of the relevant data on producer pricesreceived and retail dairy prices paid do not supportthis contention.

Novakovic (8) made a comparison of the changesin dairy prices at the farm, wholesale, and retaillevels for 1989 through 1990. Figure 5-4 illustratesthe monthly changes in average aggregate farm,wholesale, and retail dairy prices converted to anindex where 1982 to 1984 = 100. The graph showsa change in each index from one month to the next.(A line on the graph below (above) 0.0 indicatesprices fell (rose) compared to the prior month.)

The data show that farm, wholesale, and retailprices did follow each other. There are, however,differences in the volatility of change. Farm prices

Figure 5-5-Farm and Retail Prices of BeverageMilk Per Half Gallon (change from prior month)

$0.05 ~

$0.04 —

$0.03 —

g $ o . 0 2 —,=m

. *= 8, 1c

s: $ 0 . 0 0 “’h,

: ,,I , , * , , 1 1 , 1 18 i,

($0.01) – ‘“., /’,***

, ,

($0.02) –“ ’ . . . . . . . . :s**9,*,

($0.03) J*,

J F M A M J J A S O N D J F M

1989 1990Month

----- Farm price, all Federal orders — Whole milk, retail price

aThis is the producer price for fluid milk (Class 1) averaged overall Federalmilk marketing orders.

SOURCE: Compiled from U.S. Department of Agriculture dairy statisticaldata by Andrew Novakovic, Dairy Marketing /Votes, 1990, No. 2,Department of Agricultural Economics, Cornell University, lth-aca, NY.

are the most volatile while retail prices are the least.Declines in farm prices are reflected in smallerdeclines at retail. However, this is true on the up sideas well. Farm prices increased the most from mid1989 to the end of 1989 and retail prices increasedthe least.

A review of actual price changes for fluid milkand manufactured products, i.e., cheese, provides amore insightful analysis (8).5 The pattern in figure5-5 is similar but not identical to figure 5-4. That isproducer and retail prices for fluid milk did followone another up and down.6 Producer prices, how-ever, decreased more than retail prices in the firsthalf of 1989 and increased less in the second half.Some buoyancy exists to retail milk prices relativeto farm prices in reflecting declines in farm prices.

Price changes in cheese markets offer a similar butmore responsive change (see figure 5-6). Note thatproducer prices lag wholesale prices by 1 month andretail prices have a 2- to 4-month lag in reflectingwholesale price changes. For example, the largest

spercentage changes me not the best way to compare farm, wholesale, and retail prices. When expressed in a common ufit of measmement (e.g.,dollars per cwt of milk), the farm price will obviously be a smaller number than the retail price. Thus taking a percentage of a smaller number is lessthan an equal percentage of a larger number. Part of the seemingly lower volatility in prices higher up the marketing channel is a result of comparingindex or percentage changes.

G~e sme result was found by outlaw et al. (9) in a more recent analysis.


increase in wholesale cheese prices begin in May.Retail price increases began to increase in Augustand peaked in October. By then wholesale priceswere increasing at a more modest rate while retailprices were increasing by the largest amount.Examining just the October data, it would bedifficult to justify a 9-cent increase in retail prices bya 2-cent increase in the wholesale price. It can,however, be justified by the 5 months of 4- to 8-centmonthly increases in wholesale prices that occurredprior to October. Also note that in the second half ofthe year, the increase in producer prices wassubstantially greater than the increase in retailprices. These results are similar to other research inthis area. Kinnucan and Forker (5) found the sameasymmetric relationship between farm and retaildairy prices. This phenomena is found in otheragricultural industries as well.7

This analysis indicates that prices of dairy prod-ucts to consumers are reflective of changes in supplyand demand factors in the market. Individual dairyproducts such as milk and cheese do respond to pricechanges differently, reflecting the specific forces atwork in each of their respective markets. Retail milkprices follow farm price increases but seem to berelatively slower in reflecting farm price declines.On the other hand, cheese prices are responsive tofarm price changes and may even start falling beforeproducer prices. Thus, technological change thatlowers farmers’ production costs will eventually bereflected in the market and the correspondingsavings passed on through lower prices to theconsumer, the ultimate beneficiary.

INTERNATIONAL TRADEPROSPECTS

Speculation exists that adoption of new technolo-gies, such as bST, will enhance the U.S. position ininternational milk markets. The U.S. dairy industryprimarily focuses its marketing efforts on thedomestic market. It has had limited success ininternational markets. This has been due to a numberof factors including: difficulties in identifying mar-kets, monetary policies, import restrictions, andpolitical uncertainty in many countries. Moreover,the world market price for dairy products is lowerthan the U.S. price—largely because of the use of

Figure 5-6-Farm, Wholesale, and Retail Pricesof Cheese (change from prior month)

$0.09- -0 ‘ - \

0$0.05- - / .,..0

/ .“””$0.01 - - - / ’ . . . ” L ” ” , #

($0.01) - - , ... ” , r \-.- ;“, ;

($().()5) -> . y ” ’ - ” ” \:; /;

‘ i !($0.09)

+t\ /

($0.17)}

J F M A M J J A S O N D J F M1989 Month 1990

---- Farm price, --- Wholesale, ~ — Retailmanufacturing milk’ cheese price cheese price

aThefarm price is the Minnesota-Wisconsin price which is also the Class Ill(manufacturing) price in Federal milk marketing orders.

~he wholesale price is the National Cheese Exchange (NCE) price for40 lb block of cheddar.

SOURCE: Compiled from U.S. Department of Agriculture dairy statisticaldata by Andrew Novakovic Dairy Marketing Notes, 1990, No. 2,Department of Agricultural Economics, Cornell University,Ithaca, NY.

subsidies to increase export sales from competingcountries.

Cost-reducing technologies, such as bST, canimprove the United States competitive position ininternational milk markets, but alone are not suffi-cient. An encompassing strategy that at a minimumidentifies promising new markets, benefits fromfavorable monetary policies, addresses export subsi-dies and import restrictions, as well as supportsresearch to provide cost-reducing technologies forthe industry will be needed.

POLICY IMPLICATIONSThe dairy industry is familiar with and has gained

strength from technological change. The constantadoption of new technology has resulted in arelatively uniform annual increase in output per cowin the range of 1.5 to 2.0 percent annually. Emergingtechnologies in the 1990s, especially bST, maytemporarily accelerate that rate of increase, puttingthe industry on a higher output-per-cow growth path.

The impact of bST on the dairy industry is heavilyweighted by the rate of adoption of the new

TI-IahtI (2), for ex~ple, fo~d that the farm, wholesale, and retail prices for beef and pork show significant evidence Of aS~e@iC price interaction.That is, prices display greater sensitivity to price-increasing shocks than to price-decreasing shocks.

Chapter 5--Economic and Policy Impacts of Emerging Technologies on the U.S. Dairy Industry ● 85

technology. Experience in adoption of dairy technol-ogies suggests slower rates of adoption than hasbeen predicted by farmer survey. However, thisanalysis still indicates substantial economic incen-tives for, and payoffs from, adoption of bST. Theanalysis also indicates that States placing a morato-rium on the use of bST run substantial risk ofdamaging the economic viability of their dairyfarmers.

The rates of adoption indicated by past technol-ogy adoption trends suggest that a mechanism thatallows producer returns to decline as CCC purchasesincrease, i.e., a trigger policy or producer assessment(as provided for in the 1990 farm bill) couldeffectively adjust supply to meet demand withoutexceedingly large inventory accumulations. How-ever, if sharp demand reductions were to accompanythe introduction of bST, supply management poli-cies such as production quotas or termination(buy-out) programs may be required. Terminationprograms, such as the one implemented in 1986, arecostly and not effective in reducing supply over aperiod of time. Production quotas can effectivelycontrol supply. However, quotas do result in freez-ing regional production shifts and since the quotahas an economic value, make it more costly for newentrants into the industry.

Regardless of farm size or region, there will bestrong economic incentives to adopt bST. However,Upper Midwest farms adopting the new technologystill will have problems realizing sufficient earningsto achieve a reasonable return on equity, compete,and survive. Northeast farms perform better but theytoo are at a disadvantage relative to the Pacific andSoutheast farms. For farms not adopting the newtechnology the dilemma will be even more severe.These results raise questions about the advisabilityof State laws, especially in the Upper Midwest, thatplace a moratorium on the use of bST. To enhancethe economic viability of farms in these regionschanges in scale of operation, progressiveness intechnology adoption, research and extension policy,and perhaps dairy policy may be required.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

CHAPTER 5 REFERENCESl?allert, l?. et d., bST and the Dairy Industry, Agri-cultural Economics Report 579 (Washington, DC:ERS/USDA, 1987).Hahn, W., “Price Transmission Asymmetry in Porkand Beef Markets, ’ Journal ofAgricultural Econom-ics Research, vol. 42, No. 4, 1990.Hanson, Michael K., Biotechnology and Milk: Bene-fzt or Threat? (Mt. Vernon, NY: Consumer PolicyInstitute, Consumers Union, 1990).Kaiser, H.M. and Tauer, L. W., “Impacts of BovineSomatotropin on U.S. Dairy Markets Under Alterna-tive Policy Options,’ North Central Journal ofAgricultural Economics 11:59-73, 1989.Kinnucan, H. and Forker, O., “Asymmetry inFarm-Retail Price Transmission for Major Dairyproducts,” American Journal of Agricultural Eco-nomics 69:285-292, 1987.Lesser, W., Magrath, W., and Kalter, R., ‘‘ProjectingAdoption Rates,’ North Central Journul of Agricul-tural Economics (July 1986).McGuckin, J. Thomas, “Adoption of bovine Soma-totropin: A National and Regional Analysis,” OTAcommissioned background paper (Washington, DC:1990).Novakovic, Andrew, Dairy Marketing Notes, No. 2(Ithaca, NY: Cornell University, 1990).Outlaw, J., Knutson, R., and Schwart, R., “TheCorrespondence of Farm and Retail Price Move-ments,’ Department of Agricultural Economics StaffPaper (College Station, TX: Texas A&M University,February 1991).Peel, Derrell S., “National and Regional Impacts ofBovine Somatotropin Adoption Under AlternativeDairy Program Policies,’ OTA commissioned back-ground paper (Washington, DC: 1990).Richardson, James W., “Farm hwel Impacts ofbovine Somatotropin Introduction and AdoptionUnder Alternative Farm Policies,” OTA commis-sioned background paper (Washington, DC: 1990).Richardson, J.W. and Nixon, C. J., “The Farm LevelIncome and Policy Simulation Model: FLIPSIM,”Texas Agriculture Experiment Station TechnicalReport No. 81-2 (College Station, TX: July 1986).U.S. Department of Agriculture, National EconomicIndicator Farm Cost and Return Survey (Washing-ton, DC: Economic Research Service, 1985).

Appendixes

Appendix A

A National and Regional Analysis of theAdoption of bovine Somatotropin

Chapter 5 presented the summary results of thenational, regional, and farm-level impacts of emergingtechnologies and public policies on the U.S. dairyindustry. A crucial assumption in the analyses is the rateof adoption of bovine somatotropin (bST). This appendixdiscusses the results of the analysis supporting OTA’sadoption rates for bST. For more detail than is providedin this appendix the reader is advised to read thebackground paper on which this analysis is based.1

Review of Previous StudiesThere have been numerous studies dealing with the

introduction of bST into the dairy industry (1,2,5,6,7,9).Lesser, Magrath, and Kalter (7) estimated the rate ofadoption of bST based on a primary survey of producerattitudes towards the technology. This estimate has beenused in simulation studies of bST (6), which predict therelatively rapid adoption of bST after it is introduced.Fifty percent of dairy herds adopt bST technology withinthe first year and over 80 percent within 3 years.

Other studies (i.e., Fallert et al. (2) and Sellschopp andKalter (9)) reviewed the impact of bST under alternativepolicy scenarios. Their conclusions were that with theintroduction of bST, inflexible support prices wouldresult in large Commodity Credit Corporation (CCC)purchases during the 1990s. To reduce governmentpurchases, the government would need to continuereducing the support price by $0.50 increments.

Using a structural model, Kaiser and Tauer alsoanalyzed the impact of bST on the national dairy marketduring the 1990s under a number of government policyoptions. Under one, CCC purchases are held stable byadjusting the U.S. herd inventory through repeatedimplementation of the dairy termination program. Theauthors note that such adjustments may be difficult toaccomplish if farmers recognize the intent of the govern-ment and raise their dairy termination program bids.Without these adjustments, CCC purchases rapidly in-crease with bST even with lower support prices. Theauthors conclude that a combination of price supportreductions and dairy termination programs would be themost effective policy for balancing the conflicting inter-ests of dairy producers, taxpayers, consumers, and beefproducers.

The previous analyses have relied on surveys andhypothetical estimates for the adoption rates of bST. The

studies have generally lacked a consistent economicfoundation for predicting adoption. This is in part becausea model that systematically explains technological changeand/or the consequences for agricultural policy has beenelusive. As Feder (3) and Just and Zilberman (4) observe,conventional economic models have not consistentlyexplained adoption patterns of agricultural innovations orwhy seemingly profitable technologies are slowly adoptedby specific classes of farms (i.e., small farms). Theanalysis here attempts to predict adoption of bST.

Technological Change

Technological change refers to change in productionprocesses that results from the application of scientificknowledge. At the firm level, technological change can berealized in several ways. It can be embodied in inputs(changes in input quality); it can be disembodied, andinvolve improved use of existing resources such that ahigher output rate per unit of input is obtained; or it canarise from entirely new processes and new inputs (e.g.,bST). A combination of these three phenomena underliesmany innovations. For example, the development ofhybrid corn varieties represented the embodiment ofscientific knowledge in corn seed. Disembodied manage-ment knowledge was then needed for its successful use,and eventually a set of new inputs in the form of pesticideswas developed for use with the hybrids.

These definitions apply to the analytical frameworkpresented here. The adoption of bST is both the conse-quence and cause of technological change. To use bSTsuccessfully, farmers must adjust to new, higher produc-tion levels by increasing technical efficiency. Farmerswith low levels of efficiency are less likely to adjust inputsto meet the requirements of higher milk production.Conversely, the higher the current productivity of thefarmer, the more likely he or she will adopt bSTtechnologies.

Operational Model of bST Adoption

In a theoretical model outlined by McGuckin (8),changes in technical efficiency (ratio of milk output tofarm resources) drive the adoption of new technologies.Milk output per cow (a productivity measure) andchanges in scale (size of the dairy unit) are strongly

l~s append~ is bas~ on tie o~ cc)mrnisslon~ backgro~d paper “Adoption of bovine Somatotropin: A National and Regional ~ysis”prepared by J. Thomas McGuck@ New Mexico State University. It is available through the National Technical Information Service.

–89–


correlated with technical efficiency and thus technologyadoption. As productivity increases over time so doesadoption of new technologies.

The linkages between productivity change, technicalefficiency, and adoption of bST are the cornerstone of thisanalysis. However, because of data limitations, onlygeneral trends in productivity over time can be obtained.Increasing farm productivity is equivalent to increasingfarm technical efficiency, which drives adoption: asproductivity increases because of improved technicalefficiency, so does the willingness to adopt new technolo-gies such as bST.

The operational model used in this analysis assumesthat a farm’s likelihood of adopting bST is a function ofits scale and technical efficiency (measured by total factorproductivity (TFP)). Because predicting adoption of bSTis ex ante, an index of adoption of previous technologies(outlined in the data section) is used as a proxy measure.The empirical model used is the following generalrepresentation:

IAf= g(Sf,TFPf), for f = 1.. Nwhere IA is an index of adoption of previous technolo-gies, S is scale or size of the dairy, TFP is total factorproductivity and f represents a cross section of dairyfarms.

Data

The analysis used cross sectional representative farmdata from the 1985 U.S. Dairy Farm Costs and ReturnSurvey. Detailed data sources and the types of informa-tion collected through the survey are reported in theUSDA report by Fallert, McGuckin, Betts, and Bruner (2).The data include dairy farm milk production, amount oflabor (both hired and family), amount of capital (con-verted to a cow capacity basis—parlor, housing, andfeeding system can berated by the number of cows milkedper day) and respective prices. As the index for thetechnological adoption, five type of technologies wereweighted according to their relevance to bST adoption.The five technologies include:

1. automatic grain feeding system (parlor or other-wise),

2. automatic milking unit takeoff,3. three times a day milking,4. herd production records (DHI), and5. artificial insemination.

The most heavily weighted (45 percent) measure was3x milking. Use of this technology indicates that a farmercan adjust feed, breeding, and herd health practices to ahigher level of production. However, 3x requires addi-tional labor while bST would not. Artificial insemination,an improved method of breeding that directly affects milkproduction, is weighted 20 percent. DHI is a management

information system, known as Dairy Herd Improvement,that is weighted 15 percent. Automatic takeoffs arerepresentative of automated milking systems (weighted15 percent); automatic grain feeders are similar toautomatic takeoffs, though not universally used byfactory style operations.

The rate of change in the index of technologicaladoption indicates the change in use of the bST technol-ogy over time and across dairy production regions of theUnited States. Because the estimated functions areindices, an initial starting point was derived. A range ofinitial adoption levels (low, medium, and high) wereidentified for each region.

Results

To predict the adoption of bST, a regression analysiswas used based on historical rates of change in capital,labor, and feed efficiency in the dairy industry. The resultsfor each region contain a low, medium, and high scenariofor the initial adoption of the technology in 1991. After1991, all regional growth in adoption is based on therelative impact of production efficiency change on thetechnological index. The results are presented by region.

The Corn Belt region is one of the lowest adoptionregions (see figure A-l). By 1995, between 20 and 35percent of herds will receive bST (low and high scenarios,respectively). By 2000, these percentages rise to 25 and45 percent, respectively. A medium scenario is 31percent.

The Southeast is also relatively slow to adopt (seefigure A-2). By 1995, between 24 and 42 percent of herdswill receive bST (low and high scenarios, respectively).

Figure A-l—Projected Adoption Rate of bSTCorn Belt Region

1990 1992 1994 1996 1998 2000Year

--- Low — Medium . . . . High


Appendix A--A National and Regional Analysis of the Adoption of bovine Somatotropin . 91

Figure A-2—Projected Adoption of bSTSoutheast Region

1, . . .1 . . . . . . .I . . . .

I . . . . . . .0.6 -/ : : : : : : : : :

l : : : : : : : ” ”0.54 : : : : : : : : ... i““ -

0.4

0.3 j

+0.2 ‘“”

0.1-

. . . . - . . . . - .. . . . . . . . .:. , .. . . ;.. . . ... :.. . .. . .

. 0 .

*● . 0. : : : : : :

/: : , : : : : :; : : : : : :

. . . . . . .

1990 1992 1994 1996 1998 2000Year


Figure A-4—Projected Adoption of bST,Lake States Region

0.8-/ : : : : : : : : :l : ” : : : : : : :

0.74 : : : : : : : : :I

1990 1992 1994 1996 1998 2Year


Figure A-6—Projected Adoption of bSTNortheast Region

Figure A-3-Projected Adoption Rate of bSTSouthern Plains Region

0.94 : : : : : : : : :

I . . . .

i0.6 : : : : : : : : :.. .0.5- ; : : ; : : : : ”. . . . . . . . .

. . . . . . - . .:. . - . . :“ “ :0.4- : : .....;.”. ““ ~ ““”-”/” - ; ; : >

: : : -------~ ---+---’, y - - - - - - - - - - - -. .0.2- j“”

. .. .. ..“ . . .

0.1- ; : : ; :

‘ : : : :

1990 1992 1994 1996 1998Year


Figure A-5-Projected Adoption Rate of bST,Appalachian Region

1 1 . . .. . . . .

I . . . . . .0.7 + : : : : : : : : :I . . . . . . .

1990 1992 1994 1996 1998 2000Year


Figure A-7—Projected Adoption Rates of bST,Pacific Region

‘1 : : : : : : : : :1 . . . . . . I I . . . .

-i0.8 : : : : : : : : :. . . .. . . . I0.74 : : : : : : : : : I

1 : : : : : : : : : 10.6~ : : : : : : : : : II . . . . . I0.5

:

0.4 –

0.3

0.2 –

0.1 –

. . . .. . . . .: ..””

. . . .

. . . . .. . . .. . . .--h--i--------? ‘ - : : {

. *. ..-..-

/ :

. . -----● ✎ ✌✜

✎● ✃ ✎

● .e :.& .● “ 4 .y. .. .

1990 1992 1994 1996 1998 2300

Year



292-860 0 - 91 - 4

1990 1992 1994 1996Year

-- Low — Medium

1998 2000

. . . High


By 2000, these percentages rise to 32 and 55 percent,respectively. A medium scenario is 39 percent. (However,confidence in these predicted rates is low, respectively.)

Like the Southeast, the Southern Plains is a lowadoption region (see figure A-3). By 1995, between 28and 44 percent of herds will receive bST (low and highscenarios, respectively). By 2000, these percentages riseto 35 and 53 percent, respectively. A medium scenario is42 percent.

Adoption rates are slightly higher in the Lake Stateregion (see figure A-4). By 1995, between 26 and 46percent of herds will receive bST (low and high scenar-ios), respectively. By 2000, these percentages rise to 37and 64 percent, respectively. A medium scenario is 46percent.

Relatively high rates of adoption are also predicted forthe Appalachian region (see figure A-5). By 1995,between 27 and 48 percent of herds will receive bST (lowand high scenarios, respectively). By 2000, these percent-ages rise to 40 and 70 percent, respectively. A mediumscenario is 46 percent.

Adoption rates in the Northeast are similar to those inthe Lake States (see figure A-6). By 1995, between 25 and44 percent of herds will receive bST (low and highscenarios, respectively). By 2000, these percentages riseto 34 and 59 percent, respectively. A medium scenario is43 percent.

The adoption pattern in the Pacific Region, the fastestgrowing dairy region of the United States, is acceleratedrelative to that of all other regions (see figure A-7). By1995, between 45 and 63 percent of herds will receive bST(low and high scenarios, respectively). By 2000, thesepercentages rise to 66 and 81 percent, respectively. Amedium scenario is 67 percent. The strong coefficients ofsize and milk output per cow drive the adoption of bST inthis region at a high rate.

Overview of ResultsThe dairy industry has one of the highest rates of

productivity increases in U.S. agriculture. Yet, adoptionof existing proven technologies is not universal amongdairy producers. Though technologies such as artificialinsemination and herd record systems have existed formany years, these technologies have only been adoptedby 30 to 40 percent of producers in several major dairyregions. The most technically efficient producers (highestratio of milk output to farm resources) are the most likelyto adopt new technologies. Using regression techniques,this analysis establishes that producers with high levels ofmilk per cow and large operations are more likely to adoptnew technologies (a finding consistent with scientificliterature on adoption of new technologies).

Given that bST has similar characteristics to previousdairy technologies, improvement in productivity from anincreasing knowledge base will drive its adoption.Analysis of productivity measures in the major dairyregions suggest that between 50 and 70 percent of dairyproducers in the United States will adopt bST by the year2000. The Pacific region will lead all regions in adoption,possibly reaching 80 percent by 2000.

The projected rates of adoption in this analysis arelower than other studies based on differing methodolo-gies. Rather than basing predictions on historical trends,for example, Lesser, McGrath, and Kalter use contingentsurveys of producers and arrive at higher adoption rates.There is little to suggest that the adoption of bST will varyfrom past adoption practices by dairy operators. bST issimply a continuation of numerous other productivetechnologies in the dairy industry. The lower projectedrates of adoption are, therefore, the more realisticprojections of actual adoption rates.

1.

2.

3.

4.

5.

6.

7.

8.

9.

Appendix A References

Aradhyula, S. and Krog, D.R., ‘‘Biotechnology and the DairyHerd Buy Out Program: Implications for the U.S. Da.@Industry,” Paper presented at The American Agr. Econ.ASSOC., Lansing, ~, August 1987.

l?allert, R. et al. “bST and The Dairy lidustry: A National,Regional, and Farrn-Uvel Analysis, ” USDA AgriculturalReport 579, 1987.

Feder, G., ‘‘Farm Size, Risk Aversion and The Adoption ofNew Technology under Uncertainty,” O@ord EconomicPapers 80(32).

Just, R.A. and Zilberman, D., “Stochastic Structure, FarmSize and Technology Adoption in Developing Countries,”O@ord Economic Papers 83(35):307-28.

Kalter, R.J. et al., “Biotechnology and the Dairy Industry:Production Costs and Commercial Potential of the BovineGrowth Hormone,” A.E. Research Report No. 82-22, 1984,Department of Agricultural Economics, Cornell University.

Kaiser, H. and Tauer, L., ‘Impact of the bovine Somatotropinon U.S. Dairy Markets Under Alternative Policy Options,’North Cent. Jour. of Agri. Econ., 89(1 1).

Ixxser, W., McGrath, W., and Kalter, R., “ProjectingAdoption Rates: Application of an ex ante Procedure toBiotechnology Products,” North Cent. Jour of Agri. Econ86(8):2.McGuckin, J. T., “Adoption of bovine Somatotropin: ANational and Regional Analysis,” OTA commissionedbackground paper, Washington, DC, 1990.

Sellschopp, J. and Kalter, R., “Bovine Somatotropin: ItsImpact on the Spatial Distribution of the U.S. DairyIndustry,’ A.E. Research 89-14, Department of AgriculturalEconomics, Cornell University, 1989.

Appendix B

Detailed National and Regional Impacts of bovineSomatotropin and Other Emerging Technologies

Under Alternative Dairy Policies

The national policy evaluation.s in chapter 5 wereconducted with an econometric-simulation model of theU.S. agricultural sector (AGSIM). AGSIM is a disaggre-gate agricultural-sector model that utilizes econometricsupply and demand relationships for major crop andlivestock commodities. Figure B-1 illustrates the concep-tual framework of the simulation model. The modelcontains regional supply representations of major cropcommodities and an annual livestock supply sector. Forthis study a regional dairy supply component wasincorporated into the model to analyze regional impactsof technology adoption under alternative dairy policies.National demand relationships for all crop and livestockcommodities are utilized in the model.1

Supply relationships in the model are specified asfunctions of expected returns to production, Thus, aggre-gate supply relationships directly reflect the microlevelimpacts of policies or technological change that changerevenue components (e.g., yield) or cost components(e.g., product cost) or both. Further details of the cropportion of the model and regarding use of the model forpolicy analysis are contained in Taylor (2,3).

The livestock model (LIVESIM) utilized in the agricul-tural-sector model described above was developed byPeel (l). LIVESIM contains separate market represen-tations for fed beef, nonfed beef, pork broilers, turkey,milk, lamb, eggs, and veal. The original aggregate supplyrelationships for milk production were replaced byregional supply equations.

Of particular importance for this study is the disaggre-gation of beef and dairy sources as contributors to fed andnonfed meat supplies in the model. The indirect impactsof dairy policy alternatives on other livestock subsectorsare captured endogenously (within the model) throughchanges in fed and nonfed beef supply. Changes in dairyreturns influence not only milk production but also impactcalf crop, cow slaughter, and calf slaughter. The impor-tance of these impacts was highlighted by the controversyover the dairy termination program of 1986. That programcaused a significant decline in cattle prices.

Crop and livestock sectors are directly linked in themarket in LIVESIM. Livestock returns (which drivelivestock supply equations) are partly determined by feed

Figure B-l-Simulation Model

I Read initial values I

E = -

EReturns expectations

for crops and livestock

nunSolve for market equilibrium

prices 1

HI Crop and livestock

demands ISOURCE: D.S. Peel, “National and Regional Impacts of bovine Somato-

tropin Adoption Under Alternative Dairy Program Policies, ’’OTAcommissioned background paper, Washington, DC, 1990.

1~~ ~PP~D~ is bw~ ~~ fi~ o~ ~O~~~iO~ed baC@O~d paper ‘‘NatiO~ and R@O~ ~pacts Of ~Vine somatotir)pin Adoption UnderAltcrmtive Dairy Program Policies” prepared by Derrell S. Peel, Oklahoma State university. It is available through the National Technical InformationService.

–93–

94 U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

costs calculated internally from feed rations and cropprices. Changes in crop prices directly impact livestockreturns and thus livestock supply. In turn, total livestockproduction in part determines demands for the individualcrops and influences crop prices accordingly.

The Regional Dairy Model

For this analysis, total milk supply is determined fromregional equations for milk production per cow and dairycow inventory. Data for the econometric estimates wereaggregated from State data. Ten regions, consistent withthe standard USDA production regions (discussed in ch.2), were used in the model. Dairy returns for each of theregions is based on a USDA data series known as theregional cost of production budgets for dairy.

Market-clearing prices are calculated by balancing rawmilk production, on a per-capita basis, against per-capitamilk demand. The resulting national milk price isregionalized in the model via regressions of regional milkprice on national milk price. These regional pricerelationships implicitly capture the net effect of theclassified pricing system on regional milk prices.

Modeling Dairy Policy

The econometric-simulation model captures the pri-mary impacts of milk price support programs by calculat-ing milk and dairy returns based on the maximumequilibrium market price or on an exogenously specifiedmilk support price. Thus milk production per cow, dairyherd inventory, dairy replacement inventory, and thedairy impact on cow slaughter and calf crop all reflect theinfluence of the milk support price.

Government support of milk production is treated on araw milk equivalent (ME) basis. Since the governmentonly purchases manufactured milk products, all govern-ment purchases are made at a manufacturing milk price,which is assumed to be $1 per hundredweight (cwt) lessthan the all-milk price.

This analysis assumes that a minimum level ofgovernment milk purchases of 3 billion pounds of milkannually will be required for program needs. Governmentmay purchase more than this minimum level to balancemilk supply and demand at the prevailing support price.

Modeling Technology Adoption

The impacts of bovine Somatotropin (bST) adoptionand other emerging technologies were incorporated intothe econometric-simulation model under the followingassumptions:

1. output per cow increases 1.5 percent per year in basescenario without bST,

2. output per cow, due to bST, increases 1,320 poundsannually,

3. the daily cost of bST is $0.30 per cow,4. cows are treated for 150 days annually,5. overall feed efficiency is improved by 5 percent for

treated cows.

The model increases feed use marginally for additionalmilk production resulting from bST use. However, feedrequired per cwt of milk production is 5 percent lowerwith bST because cow maintenance requirements arespread over more units of production. The model alsoassumes that per cwt variable costs for other productionexpenses increase incrementally with bST use.

Three alternative rates of industry adoption of bST(low, medium, and high) were considered for the 10production regions of the United States. Completepresentation of the development and assumptions of thealternative adoption rates are presented in appendix A.

Results

Various combinations of the policy alternatives de-scribed above and the alternative adoption rates for bSTwere analyzed. In addition, the possibility that bSTadoption could have some exogenous impact on milkdemand was considered in several scenarios.

Impact of bST Adoption

Of primary concern in formulating dairy policy is theimpact that bST adoption will have on total milkproduction and consequently on government purchasesrelated to the dairy program. Figure B-2 shows total milkproduction under different levels of bST adoption. Thisfigure assumes an annual trigger adjustment for milksupport price. The maximum impact in terms of addi-

Figure B-2—Projected Total Milk Production WithTrigger Policy Under Alternative bST Scenarios

160● *

158-

156-

,e154-

g●

.5 152 –=—m

150-

148-. . . . .

./ ; : .~ - - - - ,

144-I : ; ~

1990 1991 1992 1993 1994 1995 1996 1997 1{

Year

---, No bST _ - Low.bST — Medium bST --- High bST


Appendix B--DetailedNational and Regional Impacts of bovine Somatotropin and Other Emerging Technologies . 95

Figure B-3-Projected Milk Production per Cow

174172-170-168- : : : : ;

142- - : : : :140 “ “ ’ : : :

1990 1991 1992 1993 1994 1995 1996 1997 1998

Year--- N o b S T = - L O W b S T — Medium bST --- High bST


tional milk production occurs in 1994, with total produc-tion of 154 billion pounds under high bST adoptioncompared with 151 billion pounds under low bSTadoption. Figure B-3 shows the impact of bST on milkcow productivity.

Differences in milk production due to alternative levelsof bST adoption would be more pronounced if the milksupport price was not triggered down (see figure B-4).With no bST, the baseline simulation of the model resultsin a single $0.50 per cwt adjustment in milk-support pricefrom $10.60 to $10.10 in 1992. Under each of the threealternative levels of bST adoption, an additional $0.50 percwt decrease to $9.60 in 1994 is required to keepgovernment purchases of milk under the 5 billion poundlevel. Figure B-5 shows the high levels of governmentpurchases of milk in 1991 and 1993 that precipitate thereductions in milk support price.

Comparison of Alternative Policies

The implications of bST adoption depend on the policyscenario under which adoption takes place. This sectionconsiders the impacts of alternative policy options onmilk production and price under the assumption of amedium level of adoption.

Figure B-6 shows total milk production under the fixedsupport price, annual trigger, and quota policies. Theimpact of the dairy termination (buyout) program is notincluded in this section because government milk pur-chases never exceed 15 billion pounds-the amountassumed to initiate a buyout program. Milk productiongenerally increases to similar levels under each of thepolicies. However, milk production is lowest for the quotaand highest for the freed support scenario for most years.The trigger policy results in milk production levels

Figure B-4—Projected All Milk Price WithTrigger Policy Under Alternative bST Scenarios

13.2 I

12.8

12.6

12.4

w i 1

● :. . . =

12.2. ● * ,

4 ,* ● .●

12.0 ● .* ● ,4 ., ● +* . .,. . . .

I \.10.6 % .

I1990 1991 1992 1993 1994 1995 1996 1997 1998

Year

= ■ = - No bST — Low bST --- Medium bST --- High bST


between those associated with the other two policies andproduction that is somewhat more variable from year toyear.

The fixed support price and quota scenarios maintain amilk support price of $10.60 (see figure B-7). The all-milkprice is $1.00 greater than the manufacturing price ofmilk. Beginning in 1995, milk price under the quotapolicy begins to rise over the support level. In contrast, thetrigger policy allows milk price to fall substantially beforeit rises again as the industry cuts production.

The impacts of the alternative policies on governmentpurchases of milk are summarized in figure B-8. Asexpected, the fixed support price policy is the mostexpensive, resulting in government purchases well abovethe minimum milk purchase level in order to maintain thesupport price. The quota and trigger policies are able tokeep government purchases much lower although thetrigger is slower to compensate for the impact of bSTadoption. Annual government purchases between 1991and 1998 average about one-third less under the triggerpolicy compared to the fixed support price.

The trigger and quota policies accomplish their goalsby different means. All of the policies result in increasedgovernment purchases for milk in 1991, the first year ofbST adoption. However, it is assumed that within a yearthe quota policy is able to reduce the size of the dairy herdto a level that limits government purchases for excessmilk and maintains the milk price at the higher supportprice ($10.60). The trigger policy reduces the supportprice in 1992 and again in 1994 before controllinggovernment purchases of excess milk.


Figure B-5—Projected Government Milk Purchases With Trigger Policy Under Alternative bST Scenarios

8

7

6

5

~g 4.—=.—m

3

2

1

0I I I

1990 1991 1992 1993 1994

_ N o b S T. . . . . . . . . .


Regional Impacts

In addition to concerns over the national impacts ofbST adoption under different policy scenarios are con-cerns about how the technology will affect the industry’sregional structure and dynamics. One way to summarizewhat the regional impacts of bST adoption might be is toanalyze changing milk production patterns across theNation.

Figure B-9 shows total production shares for the 10production regions of the country in 1990 and 1998. Thischart assumes a trigger policy for adjusting milk supportprice and a medium level of bST adoption. Trends alreadyobserved in the dairy industry continue in this simulation.Declining market shares are noted for the Corn Belt andthe Northern Plains with smaller reductions in the Delta,Appalachian, and Southeast. Largest increases in marketshare are noted in the Pacific region. The Lake States andNortheast maintain roughly their current market sharesover this period.

I I I I1995 1996 1997 1998 1999

my Medium bST _ H i g h b S T

Figure B-10 shows the impact of alternative policies onregional market shares, with medium level of bSTadoption. There is little difference between the impact ofthe fixed support price and that of trigger policies inregional market shares. The quota does not allow marketshares to change as much as the other policies. Rather, thequota is assumed to fix market shares at 1990 levels.Some change occurs because of trends in milk cowproductivity even though the dairy herd is fixed in size.

Alternative Demand Scenarios

Continued consumer concern over bST promptedconsideration of scenarios with exogenous changes inmilk demand reflecting adverse consumer reaction to bSTin milk. Three alternative demand scenarios were com-pared in the model:

Baseline: used in all previous scenarios

Temporary: large temporary demand reduction with smallpermanent demand reduction

Permanent: large permanent demand reduction

Appendix B--DetailedNational and Regional Impacts of bovine Somatotropin and Other Emerging Technologies ● 97

Figure B-6-Projected Total Milk ProductionWith Medium bST Adoption Under Alternative

Dairy Policies

Figure B-7—Projected All Milk Price WithMedium bST Adoption Under Alternative

Dairy Policies

1990 1991 1992 1993 1994 1995 1996 1997 19Year

— Fixed -- Trigger -- = Quota


Figure B-8—Projected Government MilkPurchases With Medium bST Adoption Under


8

10

9 -

1

.,-

8 -

7

G 6

g 5.-‘ 4 I

1990 1991 1992 1993 1994 1995 1996 1997 1998Year

❑ Fixed RI Trigger ■ Quota


Details of these alternative demand scenarios are pre-sented in chapter 5.

Alternative Milk Demand Under Current Policy—Figure B-n illustrates the impacts of alternative milkdemands assuming a continuation of the current triggerpolicy for adjusting milk support price and the mediumlevel of bST adoption. Changes in milk demand have

‘zp6s

10.6 ] : :

I I I1990 1991 1992 1993 1994 1995 1996 1997 1

Year— Fixed -- Trigger --- Quota


Figure B-9—Actual and Projected RegionalMilk Market Shares With Trigger Price Policy,

1990 and 1998.

30 I282624

? 22g 2 0 1

4 -CB LS NP SP DL MT PA NE AP SE

Region~ 1990 - 1998

Key: LS - Lake States SP - Southern Plains CB - Corn BeltNE - Northeast NP - Northern Plains AP - AppalachianPA - Pacific SE - SoutheastMT - Mountain DL - Delta States


large implications for milk price. While the base level ofdemand results in milk price near $12 per cwt for all years,a permanent large demand reduction would allow milkprice to fall as low as $8.60 in 1997 before beginning torise.

Figure B-12 shows the level of government milkpurchases under the different levels of demand. Reduced


Figure B-l O-Projected Change in Milk MarketShares for Alternative Dairy Policies Between

1990 and 1998

4 -

3 -

2 -

1 -

0 -

-1-

.-L ‘ —

II

I I I I I I i ICB LS NP SP DL MT PA NE AP SE

Regionm Fixed D Trigger m Quota

Key: LS - Lake States SP - Southern Plains CB - Corn BeltNE - Northeast NP - Northern Plains AP - AppalachianPA - Pacific SE - SoutheastMT - Mountain DL - Delta States

Figure B-1 l—Projected All Milk Price WithbST Adoption Under Alternative Milk Demands

15

14

13

[. . . . . . I8 1 : : ~ “ : “

1990 1991 1992 1993 1994 1995 1996 1997 1998

Year— Base ● ■ - Permanent Demand -- Temporary Demand

(no change) Reduction Reduction



milk demand, under both demand reduction scenarios,results in government milk purchases of 21 billion poundsin 1991 at a cost of about $2.5 billion. With the temporarydemand reduction, government purchases decline fairlyrapidly as the support price declines. With the permanentlarge demand reduction, however, government purchasesdecline slowly as the trigger lowers support price.

Policy Comparison With Permanently Reduced MilkDemand—In the face of large surpluses in milk produc-tion, the implications of the alternative policies are moresharply delineated. Assuming a permanent large reduc-tion in milk demand, and future excess production, thechoice of policies clearly will have much larger impactsthan it would under the baseline demand scenario.

The reduced demand scenario is useful, not because itis a likely result of bST adoption, but because similarconditions (in terms of relative supply and demand) couldprevail for a number of other reasons. For example, if bSTresults in greater average productivity increases than ishere assumed, or if adoption rates are substantially higher,then supply excesses similar to those under the reduceddemand scenario could result. This scenario thus can beviewed as a proxy for a number of supply or demandsituations that could produce large surpluses of milk.

Figure B-13 shows the impact of reduced demand onmilk production (given medium bST adoption) withalternative policies-fixed support price, trigger-adjustedsupport price, production quota, and a dairy termination

program. Differences in the time path of milk productionunder the quota and the other policies are readily apparent.The quota results in a quick downward adjustment indairy herd size necessary to avoid large governmentexpenditures while maintaining milk price at the $10.60support level. The dairy termination program (buyoutoccurs in 1992) adjusts herd size in a manner similar to thequota in 1992, but herd size and total milk productionclimb rapidly before declining again in 1998. The triggerpolicy results in an eventual but much delayed decline inmilk production. The fixed support price policy, asexpected, maintains the highest level of milk productionof the alternative policies.

The implications of the alternative policies on milkprice are likewise quite dramatic (see figure B-14). Thefixed support and quota policies maintain a milk supportprice of $10.60. Figure B-14 shows that the all-milk pricecorrespondingly is at the minimum level of $11.60 withthese policies after bST is adopted (and the demand shiftoccurs). The trigger and dairy termination programs allowmilk support price to adjust downward in the face ofexcess milk production. The trigger results in milk pricedeclines to a minimum of $8.60 in 1997. The dairytermination program also allows milk price to fall to thislevel but with a delay of 1 year compared to the triggerpolicy. This is because the dairy termination programbuyout occurs in 1992, avoiding the need for a reductionin milk support price prior to 1993.

Appendix B-Detailed National and Regional Impacts of bovine Somatotropin and Other Emerging Technologies ● 99

Figure B-15 reiterates these impacts in terms ofgovernment milk purchases. It is significant to note thatwhile the dairy termination program reduces milk pur-chases and expenditures quite successfully in 1992 (theyear that liquidation occurs), milk production quicklybounces back and milk program purchases are not muchlower than those associated with the trigger policy alone.From 1995 to 1998, purchases under the trigger policy areactually less than they are under the dairy terminationprogram.

Impacts on Other Agricultural Sectors

The adoption of bST appears to have relatively minorimpacts on agricultural sectors outside of dairy. Table B-1summarizes agricultural commodity prices over theperiod 1995-1998 for the freed support price, trigger-adjusted support price, and quota policies with andwithout bST adoption (medium level).

The adoption of bST does create a marginal increase indemand for feed in the dairy industry. However, the neteffect, when all markets adjust, is extremely small.Among all crops, impacts on the all hay price are largestwith bST adoption; average hay prices increase by $1.25to $2.50/ton depending on the policy scenario.

Impacts in the livestock sectors are limited mostly tocattle, and average price effects are minute. Interestingly,how bST adoption impacts yearling, calf, and cow pricesdepends on the policy scenario. This indicates that dairypolicy can affect the timing and magnitude of changes inthe dairy herd.

The impact of a dairy termination program on livestockprices is of particular interest. Figure B-16 shows thedynamic paths of yearling cattle price for alternative dairypolicies. (Figure B-16 also assumes a permanent milkdemand decrease in conjunction with bST adoption.)

Figure B-12—Projected Government Milk Purchases With bST Adoption Under Alternative Milk Demands

22 1

co.—=z

2120

19

I

1 8 -17–16

1 5 -

1413

12

1110

98

76

54 /m3 ,,,,:.:.:.:2 g$:.:.:.:1 – ;~::.:.:.:o x+

1990 1991 1992 1993 1994 1995 1996 1997 1998

Year

❑ B a s e ❑ Temporary Demand ■ Permanent DemandNo Change Reduction Reduction



Figure B-13—Projected Milk Production WithPermanently Reduced Milk Demand Under


162160- : ; ; : ; : :

156- : : : : ’

. 0

150-

~ 1 4 6 -= 1 4 4 -

142- I ‘ ; ● ’: ; : : ~;&@@\140- : : : : : : ;@@ ;

138- ; i ; : : : ~@:@@e : :

!136- : :’: ;@ :134- : 4 : ,.’ ; ,,- : : : :

132- :$

~ 0

130 ‘‘$- : ; ; ; :

I I I1990 1991 1992 1993 1994 1995 1996 1997 1998

Year

— Fixed Price ‘--Trigger -- Quota . . Dairy Termination Program


Figure B-14-Projected All Milk Price WithPermanently Reduced Milk Demand Under


13.5 , , , , , , ,

13.0- : : : : : : :

12.5- : : : : : : :

12.0-. . d

11.5- >..* “.* : : : : : 7+ ’

1 1.0– : ● ?*+ ‘~. : : : :- . .

10.5

10.0

9.51

● ✚☛● ✎

1990 1991 1992 1993 1994 1995 1996 1997 1998

Year— Fixed -. -Trigger - - Q u o t a = = = Dairy Termination Program


Figure B-15—Projected Government Milk Purchases With Permanently Reduced Milk Demand

24

22

20

18

16

14g

.: 12.=m

1 0

8-

6

4.

2-

0-


m;:;::.. . . . . ;:;;;.:::::.. . . . . . . . . . ;::::. . . . . . ..,.,.:::::,;::::.. . . . . .1990 199l

❑ Fixed

1992 1993 1994 1995

❑ Dairy Termination Program❑ Trigger

Year

❑ QuotaSOURCE: Office of Technology Assessment, 1991.

Appendix B--Detailed National and Regional Impacts of bovine Somatotropin and Other Emerging Technologies . 101

Table B-l—Impacts of bST Adoption on Other Agricultural Sectors, 1995-98

Policy scenarios

Fixed price support Trigger price Quota

Commodity (units) NO bST bST No bST bST No bST bST

Corn ($/bu) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.85 2.86 2.85 2.86 2.85 2.86Grain sorghum ($/bu) . . . . . . . . . . . . . . . . . . . . . 2.54 2.54 2.54 2.54 2.54 2.54Barley ($/bu) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.60 2.61 2.60 2.61 2.60 2.61Oats ($/bu) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.39 1.40 1.39 1.40 1.39 1.39Wheat ($/bu) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.09 3.09 3.09 3.09 3.09 3.09Soybeans ($/bu) . . . . . . . . . . . . . . . . . . . . . . . . 6.46 6.47 6.46 6.46 6.46 6.46Cotton ($/lb) . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.86 0.86 0.86 0.86 0.86 0.86All hay ($/ton) . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.06 105.55 102.95 104.64 102.77 104.03Yearling cattle ($/cwt) . . . . . . . . . . . . . . . . . . . . 74.87 74.94 74.92 74.79 75.08 74.97Calf ($/cwt) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.84 82.01 81.93 81.84 82.16 82.06cows ($/cwt) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56.97 57.11 57.05 56.84 57.33 57.14Hogs ($/cwt) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.08 51.09 51.06 51.11 51.01 51.04Broilers (@/lb) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61.38 61.45 61.43 61.43 61.46 61.47Turkeys (¢/lb) . . . . . . . . . . . . . . . . . . . . . . . . . . . 69.32 69.38 69.41 69.29 69.55 69.49Eggs (¢/dozen) . . . . . . . . . . . . . . . . . . . . . . . . . 81.15 81.20 81.14 81.18 81.14 81.16Lamb ($/cwt) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.46 91.63 91.45 91.62 91.41 91.51

KEY: bu==bushel; cwt=hundredweight (100 pounds); Ib= pound.

SOURCE: Office of Technology Assessment 1991.

Figure B-16--Projected Yearling Cattle PriceWith Permanently Reduced Milk Demand


76: /

7 5 - : :

7 3 -

~ 7 1 – *: e 0, : :

6 9 - ;

1990 1991 1992 1993 1994 1995 1996 1997 1998

Year— F i x e d -.-Trigger --Quota --- DairyTermination Progam

SOURCE: Office of Technology Assessment 1991.

Annual yearling cattle prices are about $4.35 per cwtlower in 1992 as a result of the dairy termination program(compared to the trigger policy). Cow prices in 1992 areover $6.00 per cwt lower with the dairy terminationprogram compared to the trigger policy. The quota policyaffects cattle price in much the same way as the dairytermination program; its impacts are slightly less inmagnitude in 1992, the year that the quota is imposed, butprices are slightly lower under the quota relative to thedairy termination program for several more years.

1.

2$

3

Appendix B ReferencesPeel, D. S., “A Dynamic Econometric-Simulation Model ofthe U.S. Livestock Industry,” Ph.D. dissertation, Departmentof Agricultural Economics, University of Illinois, 1989.Taylor, C.R., “A Description of AGSIM, an Econometric-Simulation Model of Regional Crop and National LivestockProduction in the United States,” staff paper ES89-1,Department of Agricultural Economics and Rural Sociology,Auburn University, Auburn, Alabama, January 1989.Taylor, C.R., “AGSIM User’s Manual, Version 89.5,” staffpaper ES89-11, Department of Agricultural Economics andRural Sociology, Auburn University, Auburn, AL, October1989.

Appendix C

Detailed Farm Level Impacts of bovine Somatotropinand Other Emerging Technologies UnderAlternative Policy and Demand Scenarios

The farm level impacts in chapter 5 were determined bya Monte Carlo simulation model known as the Farm LevelIncome Tax and Policy Simulation Model (FLIPSIM)developed by Richardson and Nixon (4) at Texas A&MUniversity. The model is capable of simulating represen-tative dairy farms in different regions of the United Statesunder alternative policy and technology assumptions.l

Analyzing the consequences of alternative technolo-gies on the economic viability of a representative farminvolves several steps. First, data for the representativedairy farm that is using existing technologies must bedeveloped. Second, modifications to the basic dairyfarm’s input/output coefficients must be made for eachtechnology change to be analyzed. For bST, this is doneby annually changing the milk per cow and the lactatingcow ration and by increasing variable costs per cow toreflect bST purchases. The result is a new representativefarm that has adopted bST. Third, projections for milkprices, feed prices, cattle prices, annual percentagechanges in herd size, and macroeconomic variables(interest and inflation rates) for the policy/technologyscenario being analyzed are merged with the farm’s data.Projections of regionalized milk prices and feed prices areprovided by the LIVESIM model (described in app. B),and macroeconomic variables are developed by theCOMGEM/AG-GEM model (3).

Representative Dairy Farms

The regions for analysis are the Lake States, Northeast,Southeast, and Southwest. Two different-size dairy farmsin each region—moderate and large farms-are consid-ered. In the Lake States, the representative moderate-sizefarm owns 52 cows, and the large farm owns 125 cows;both farms own 185 acres of cropland and farmstead, with155 acres devoted to the production of dairy feed (seetable C-l). These farms are most representative of dairyfarms in Minnesota. In the Northeast, the representativemoderate-size farm owns 52 cows, and the large farmowns 200 cows. The moderate-size Northeast farmdevotes 140 acres to the production of hay, corn silage,haylage, oats, corn, and pasture. The large Northeasternfarm has 450 acres of hay, corn silage, haylage, oats, andcorn, and 50 acres of pasture. The moderate-size farm is

most representative of Pennsylvania dairy farms and thelarge farm represents New York dairy farms.

The moderate and large Southwestern dairy farms,respectively, own 350 and 1,500 milk cows (see tableC-1) and only 25 acres of land. The two farms arerepresentative of dairy herds in California and Arizona.The moderate-size Southeastern dairy farm has 200 cowsand 388 acres and is most representative of farms inGeorgia. The Georgia farm has 305 acres devoted tocoastal hay and sorghum silage production and 50 acresdevoted to pasture. The large Southeastern dairy has1,500 cows and owns 873 acres, which are largely (750acres) devoted to pasture. The large Southeastern farm ismost representative of large dairies in Florida.

The initial debt-to-asset ratio was assumed to be 40percent for all of the farms. All land, machinery, andlivestock had 40-percent debt at the beginning of 1989(see table C-l). This level of debt represents a moderateinitial debt level. Each of the eight representative farmswere simulated over the 1989 to 1998 planning horizonfor alternative assumptions about the dairy farm programand the adoption of bST.

Technology Scenarios

The economic consequences of bST adoption wereanalyzed assuming bST was introduced in 1991, and thefarm either adopted it in 1991 or did not adopt bSTthroughout the planning horizon. Initial milk productionper cow was trended up at 1.5 percent per year in the basesituation without bST (see table C-2). For the bSTadoption scenarios, annual milk production per cow withbST was increased by 1,320 pounds each year from 1991to 1998. This increase in milk per cow due to the adoptionof bST assumes lactating cows are treated for 150 daysduring each lactation. All cows in the herd were assumedto be treated at an annual cost of $45 per cow.

The quantity of feed required for bST-treated cowsincreased marginally due to the increased milk productionper cow. A linear program (LP) model in FLIPSIM wasused to estimate a balanced dairy ration for the higherproducing dairy herd. Research by Chalupa and Galligan(1) indicates that the nutritional requirements for bST-treated cows are the same as they are for naturally high

lllis appm~ is basedon tie OTAcornmissioned background paper “Farm Uvelhnpacts of bovine Somatotropin Introduction and Adoption UnderAlternative Farm Policies” prepared by James W. Richardso~ Texas A&M University. It is available through the National Technical InformationService.

–lo2–

Appendix C--Detailed Farm Level Impacts of bovine Somatotropin and Other Emerging Technologies ● 103

Table C-l—Characteristics of Representative Moderate-Size and Large Dairy Farms inthe Lake States, Northeast, Southwest, and Southeast

Lake States Northeast Southwest Southeast

Moderate Large Moderate Large Moderate Large Moderate Large

Number of dairy cattle:cows . . . . . . . . . . . . . . . . . . . .Calves . . . . . . . . . . . . . . . . . . .Heifers. . . . . . . . . . . . . . . . . . .Bulls . . . . . . . . . . . . . . . . . . . . .Calves born . . . . . . . . . . . . . .

Assets ($1,000):Land . . . . . . . . . . . . . . . . . . . .Buildings and machinery . . . .Cattle . . . . . . . . . . . . . . . . . . . .

Total . . . . . . . . . . . . . . . .

Off-farm salary ($1 ,000) . . . .

Minimum family living($1,000) . . . . . . . . . . . . . . .

Labor costs ($1 ,000) . . . . . . .Milk/cow (cwt) . . . . . . . . . . . .

522125

048

133.1262.8

68.6469.5

9.8

19.812.7

168.5

1254857

0113

295.0482.6161.6

940.2

0

24.837.3

168.5

522023

047

274.2260.8

73.0

608.0

9.8

19.811.3

179.4

2007588

0178

640.1503.0251.5

1,394.6

0

30.970.0

178.3

350130152

7326

117.9467.3511.4

1,096.6

0

43.3115.8185.9

1,500495612

251,388

491.81,080.82,284.9

3,857.5

0

61.9444.2

196.9

2007588

0176

812.9487.0269.4

1,569.3

0

30.962.8

153.4

1,500432503

251,268

4,591.11,139.21,992.37,722.6

0

61.9488.1

153.1


producing cows. Thus, it was not necessary to change theinput/output coefficients in the ration-balancing LP—theincreased milk production per cow caused the LP to feedthe cow more protein, energy, and forage. In general, theration for bST cows contained 7 to 10 percent moreforage, 9 to 12 percent more grain, and 10 to 13 percentmore soybean meal (or whole cottonseed) than the rationfor control cows.

The new values for average annual milk per cow andbST costs were used to modify initial farm variables toaccount for bST adoption. All other variables for therepresentative dairy farms were assumed to remainconstant at pre-bST levels. Of 16 representative farms, 8adopt bST in 1991, and 8 do not adopt bST. The 16 farmswere simulated under alternative dairy policy scenarios toquantify the interaction between technology adoption andfarm programs.

Farm Program and Milk Demand Scenarios

Four farm programs were selected for the analysis:trigger price, fixed price support, production quota, anddairy termination program. Each of the policies wasanalyzed with a commodity-specific livestock simulationmodel, LIVESIM (discussed in app. B), under theassumption that the 1985 farm program for crops wouldcontinue through 1998 (2). The LIVESIM analyses ofthese four dairy policies were done for a no-bST scenarioand for a scenario with a medium rate of adoptionbeginning in 1991. Three different milk-demand scenar-ios were analyzed to incorporate the possibility of milkdemand changing in response to bST introduction.

The trigger-price dairy policy is similar to policy from1985 to 1990 with the milk-support price decreasing 50cents per hundredweight (cwt) each year that the Com-modity Credit Corporation (CCC) milk purchases areexpected to exceed 5 billion pounds of milk equivalent.The support price is increased 50 cents per cwt if CCCpurchases of milk are expected to fall short of 2.5 billionpounds. This policy is also similar to the producerassessment option in the 1990 farm bill because theassessment will effectively trigger reductions in producerreturns as milk price declines. The fixed support policyassumes that the dairy support price is held constant forall years of the planning horizon. The production-quotapolicy calls for the continuation of the trigger-price policywith provisions for a quota to be imposed if CCC milkpurchases exceeded 7.0 billion pounds of milk equivalent.Similarly, the dairy termination program would continuethe trigger-price policy but permit a one-time dairytermination if CCC milk equivalent purchases exceed15.0 billion pounds in 1 year. The dairy terminationprogram is analyzed only for the large demand decreasescenario, as this is the only demand situation that triggersthe termination.

The three milk-demand scenarios respectively assumeconstant demand, a slight decrease in demand, and asignificant decrease in demand after the introduction ofbST. The small demand reduction scenario assumes thatmilk demand will decrease 10 percent in 1991,5 percentin 1992 (i.e., demand increases from 1991 to 1992), and2.5 percent each year from 1993 to 1998. The largedemand reduction scenario assumes that milk demand is10 percent lower than it currently is in each year from1991 to 1998. The trigger price, fixed price support, and


Table C-2—Average Annual Production of Milk/Cow for Moderate-Size Representative Dairy Farms,in Selected Regions, With and Without bST, 1989-98 (cwt/year)

Lake States Northeast Southwest Southeast

Years No bSTa bSTb No bST bST No bST bST No bST bST

1989 . . . . . . . . .1 9 9 0 . . . . . , . . .1991 . . . . . . . . .1992 . . . . . . . . .1993 . . . . . . . . .1994 . . . . . . . . .1995 ...., . . . .1996 . . . . . . . . .1997 . . . . . . . . .1998 . . . . . . . . .

168.5171.0173.6176.2178.8181.5184.2186.9189.8192.6

168.5171.0186.8189.4192.0194.7197.4200.1203.0205.8

179.4182.0184.8187.5190.4193.2196.1199.1202.0205.1

179.4182.0198.0200.7203.6206.4209.3212.3215.2218.3

185.9188.7191.5194.4197.3200.2203.2206.3209.4212.5

185.9188.7204.7207.6210.5213.4216.4219.5222.6225.7

153.4155.7158.1160.4162.8165.3167.8170.3172.8175.4

153.4155.7171.3173.6176.0178.5181.0183.5186.0188.6

aOutput per cow increases 1.5 percent per year without bST.bST, introduced in 1991, increases output per cow 1,320 Ibs. per year.


quota policies were analyzed with and without bSTassuming no change in milk demand. The trigger pricepolicy was analyzed for the small decrease and largedecrease demand scenarios because this policy representscurrent policy under the 1990 farm bill. The dairytermination program was analyzed for the large decreasein milk demand and results were compared to those of thetrigger price policy under this demand reduction.

The adoption rates for bST in LIVESIM differ byregion and follow a sigmoid adoption function (see app.A). It is projected that 43.6 percent of the dairy farms inthe Lake States would adopt bST by 1998. In theNortheast, 39.9 percent of farms would adopt by 1998; inthe Pacific region, about 63 percent of farms would adoptbST by 1998; and 36.7 percent of those in the Southeastwould adopt by 1998. Adoption by 1998 in the remainingregions ranged from 29.7 to 42.9 percent.

For the FLIPSIM analyses, cattle and feed priceprojections from LIVESIM were regionalized usingsimple regression relationships between National- andState-level prices. Milk-price projections were regionspecific so no adjustment was necessary. The LIVESIMprojected annual changes in the dairy herd that wereregion specific; these projections were used to adjust thenumber of cows on the representative farms. It wasassumed that each farm’s herd size would changeannually (1990 to 1998) proportional to the annualpercentage change in the respective region’s total numberof dairy cows. Thus, the number of cows milked on therepresentative farms fluctuated annually with expectednet returns in the region.

ResultsThe detailed results of simulating the representative

farms with and without bST are summarized in thissection. Simulation results for various scenarios arepresented in terms of three probabilities and means for the

probability distributions of four key output variables. Thevariables used for evaluating the economic impacts of thealternative scenarios are defined as follows:

● Probability of Survival--chance that the individualfarm will remain solvent through 1998, i.e., maintainmore than 10-percent equity in the farm.

. Probability of Success--chance that the individualfarm will earn a 5-percent or greater after-tax returnon initial equity.

● Probability of Increasing Equity--chance that theindividual farm will increase its net worth in real1989 dollars over the planning horizon.

● Net Present Value—present value of annual changesin net worth plus family consumption minus off-farm income.

. Present Value of Ending Net Worth (PVENW)-ending net worth for 1998 discounted to 1989dollars, assuming a 5-percent discount rate.

. PVENW as a Percent of Beginning Net Worth—PVENW divided by initial net worth indicateswhether the farm increased (or decreased) net worthin real dollars.

. Average Annual Net Cash Farm Income—total cashfarm receipts minus total cash expenses excludingfamily living, income taxes, and principle payments.

Economic Payoffs to bST Adoption forAlternative Farm Policies

Tables C-3 through C-6 summarize the simulationresults for representative dairy farms in the Lake States,Northeast, Southwest, and Southeast, respectively, as-suming bST is introduced in 1991. Results are reportedfor the bST adopter and nonadopter. The adopter isassumed to use bST on all lactating cows beginning in1991. The nonadopter does not adopt bST over the 1989to 1998 planning horizon. The economic payoffs for bSTadoption are reported for three different farm policies (seetable C-6).


Table C-3-impacts of bST Adoption on the Economic Viability of Representative Lake State Dairy FarmsUnder Alternative Dairy Policies, Assuming No Change in Milk Demand Due to bST, 1989-98

Policy scenario


Non- bST Non- bST Non- bSTadopter adopter adopter adopter adopter adopter

52-COW farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent

of beginning net worth (percent) . . . . . . . . . .Average annual net cash farm income

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125-cow farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth



($1,000) . -.. .-.. .. ... ------- . . . . . . . . . .

58.058.0

0.025.2

74.074.00.0

67.6

73.073.0

0.065.0

89.089.0

1.0105.6

41.041.00.0

–15.1

52.052.0

0.09.8

44.3 81.2 75.9 112.8 7.4 27.8

2.6 9.815.6 28.6 27.1 39.7

-2.0 1.9 1.7 5.8 -4.7 –2.3

85.067.0

2.068.7

92.078.03.0

127.7

95.090.0

8.0194.8

99.095.012.0

271.9

99.095.011.0

265.0

100.098.018.0

340.1

329.1 396.4 386.9 451.5 211.9 263.7

78.3 36.7 45.757.1 68.7 67.1

23.1 33.4 32.1 43.0 11.2 18.2SOURCE: Office of Technology Assessment, 1991.

Table C-4—impacts of bST Adoption on the Economic Viability of Representative Northeast Dairy FarmsUnder Alternative Dairy Policies, Assuming No Change in Milk Demand Due to bST, 1989-98

Policy scenario



52-COW farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth



($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .200-cow farm:Probability of survival (percent) . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1,000) . . . . . . . . . . . . . . . .Present value of ending net worth



($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

100.0100.0

3.0254.6

99.0100.0

8.0277.5

99.096.00.0

110.8

99.097.00.0

117.0

100.0100.0

3.0232.3

100.0100.0

3.0253.9

268.3 286.4 285.6 303.7 169.0 174.7

72.4 77.2 77.0 81.9 45.6 47.1

-1.9 -0.914.5 17.9 17.8 21.4

100.099.043.0

616.7

100.0100.053.0

717.6

100.0100.050.0

705.7

100.0100.066.0

812.3

88.064.0

1.0102.8

91.072.03.0

166.6

776.8 855.4 842.0 922.0 360.0 415.9

49.492.3 101.7 100.1 109.6 42.8

66.2 82.0 79.6 96.2 -1.5 7.3SOURCE: Office of Technology Assessment, 1991.


Table C-5—impacts of bST Adoption on the Economic Viability of Representative Southwest Dairy FarmsUnder Alternative Dairy Policies, Assuming No Change in Milk Demand Due to bST, 1989-98

Policy scenario



350-cow farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1,000) . . . . . . . . . . . . . . . .Present value of ending net worth



($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1,500-cow farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent) . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth



($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

97.097.056.0

715.4

95.095.060.0

739.7

97.097.079.0

885.2

99.099.081.0

903.5

99.099.089.0

1,040.0

95.095.040.0

622.9

701.2 820.7 827.1 939.5 587.1 664.1

109.5 128.1 129.1 146.7 91.7 103.7

109.6 136.1 137.2 163.8 97.1 115.4

100.0100.0100.0

4,062.8

100.0100.0100.0

4,548.7

100.0100.0100.0

4,633.6

100.0100.0100.0

5,148.7

100.0100.098.0

3,532.5

100.0100.099.0

3,853.0

4,323.0 4,751.1 4,795.4 5,249.3 3,808.5 4,091.2

194.5 213.7 215.7 236.2 171.3 184.1


Table C-6—impacts of bST Adoption on the Economic Viability of Representative Southeast Dairy FarmsUnder Alternative Dairy Policies, Assuming No Change in Milk Demand Due to bST, 1989-98

Policy scenario



200-cow farm:Probability of survival (percent).. . . . . . . . . . . . 100.0 100.0 100.0 100.0 100.0 100.0Probability of success (percent) . . . . . . . . . . . . 100.0 100.0 100.0 100.0 93.0 99.0Probability of increasing equity (percent). . . . . 13.0 24.0 23.0 44.0 5.0 9.0Net present value ($1,000) . . . . . . . . . . . . . . . . 453.3 601.5 559.9 712.0 333.3 446.7Present value of ending net worth

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 727.9 854.3 815.3 940.5 615.2 712.8Present value of ending net worth as a percent

of beginning net worth (percent) . . . . . . . . . . 75.6 88.7 84.7 97.7 63.9 74.0Average annual net cash farm income

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.3 39.2 32.5 55.3 2.6 19.81,500-cow farm:Probability of survival (percent) . . . . . . . . . . . . . 100.0 100.0 100.0 100.0 100.0 100.0Probability of success (percent) . . . . . . . . . . . . 100.0 100.0 100.0 100.0 100.0 100.0Probability of increasing equity (percent) . . . . . 88.0 99.0 97.0 100.0 75.0 91.0Net present value ($1 ,000) . . . . . . . . . . . . . . . . 4,964.9 6,165.7 5,757.5 7,000.8 4,113.7 5,062.2Present value of ending net worth

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5,901.3 6,712.4 6,415.4 7,252.3 5,261.1 5,901.2Present value of ending net worth as a percent

of beginning net worth (percent) . . . . . . . . . . 129.4 147.2 140.7 159.0 115.4 129.4Average annual net cash farm income

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609.0 775.4 714.3 880.6 481.9 613.9SOURCE: Office of Technology Assessment, 1991.

Appendix C-Detailed Farm Level Impacts of bovine Somatotropin and Other Emerging Technologies ● 107

Trigger Price—Under the trigger-price program, themilk support price is decreased 50 cents per cwt each yearthe CCC purchases 5 billion pounds of milk equivalent.This option is similar to policy from 1985 to 1990 and tothe assessment option in the 1990 farm bill-the assess-ment will effectively trigger reduction in producer returnsas milk price declines.

The average annual economic payoffs from bSTadoption (change in average annual net cash farm incomedue to adoption), given a trigger price dairy policy, rangesfrom $3,400 for a 52-cow Northeastern dairy to $166,400for a 1,500-cow dairy in the Southeast. Average annualnet cash farm income for the 52-cow Northeast dairyincreases from $14,500 to $17,900 due to bST adoption(see table C-4). The 52-cow Lake States dairy experiencesa slightly greater economic payoff from bST adoption($3,900) as net cash farm income increases from -$2,oooto $1,900 (see table C-3). The greatest economic payoffsfor bST adoption are earned by the 1,500-cow dairy farmsin the Southwest and Southeast. In the Southwest, averageannual net cash farm income increases $90,500 ($713,900to $804,400) and in the Southeast, the increase is$166,400 ($609,000 to $775,400) (see tables C-5 andC-6). Absolute increases in real net worth are also greatestfor these dairies; however, the greatest percentage in-creases are observed for the dairy farms in the Lake States,and for the moderate-size dairy in the Southwest.

Increases in average annual net cash farm income dueto adopting bST lead to greater accumulation (or slowerdecline) in net worth which, in turn, leads to greaterafter-tax net present values for bST adopters. The 52-cowLake States dairy producer who adopts bST has a $42,400greater net present value than the nonadopter, and$36,900 greater present value of ending net worth (seetable C-3). This pattern of greater net worth and netpresent values due to bST adoption is observed for alleight representative farms.

Increases in average annual net cash farm income dueto bST adoption also leads to improved probabilities ofsurvival, success, and to increases in real equity. Probabil-ity of survival increases from 58 to 74 percent for the52-cow Lake States dairy as a result of adopting bST (seetable C-3). Adopting bST increases the probability ofincreasing real net worth (equity) for five of the eightrepresentative dairy farms. The three exceptions experi-enced no change in the probability of increasing realequity due to adopting bST.

Fixed Price Support--Maintaining the milk pricesupport at the 1989 value through the 1989-1998 planninghorizon results in higher milk prices and greater averageannual net cash farm incomes than the trigger price policy(see tables C-3 to C-6). Economic payoffs from bSTadoption are only slightly greater under the fried price-support policy than under the trigger price policy for six

of the eight farms. For example, the economic payoff forthe 125-cow Lake States dairy increases only $600 from$10,300 to $10,9OO due to the policy change (see tableC-3). (The two dairy farms that experience lower eco-nomic payoffs (Southwest 350-cow dairy and 1,500-cowSoutheast dairy) experience very small reductions in theireconomic payoffs from bST adoption, $700 and $100,respectively (see tables C-5 and C-6).) These resultssuggest that the economic incentive to adopt bST wouldnot be greatly increased by increasing the price of milk,i.e., freezing the milk support price at its 1989 level.Maintaining a fixed support price would result in a greaterprobability of survival, success, and increasing real equity(i.e., increasing a farm’s economic viability) than ob-served for the trigger price scenario. For a 52-cow LakeStates dairy farm that adopts bST, probability of survivalincreases from 74 to 89 percent, and for the nonadopter,the probability increases from 58 to 73 percent (see tablec-3).

Production Quota—A quota that reduces the numberof dairy cows to maintain milk prices at levels comparableto the fixed price support policy was analyzed. Results ofthe analyses reveal that a quota reduces average annual netcash farm incomes for adopters and nonadopters, relativeto the other two dairy policies (see tables C-3 to C-6).Relative to the trigger price, the quota reduces net cashfarm income about $15,000 per year for the 125-cow LakeStates dairy farm that adopts bST, and about $11,900 forthe nonadopter. The large Southeastern dairy that adoptsbST experiences a $161,500 decrease in average annualnet cash farm income under a switch from the trigger priceto the quota policy (see table C-6). Such dramaticdecreases in net cash farm income lead to lower probabil-ities of increasing real net worth for all eight farms, andlower the probability of survival for five farms.

The economic payoffs from adopting bST while aquota policy is in effect are positive for all eight farms (seetables C3-C6). However, the absolute economic payoffsare less than under the trigger price policy. The largeSoutheastern dairy farm experiences an average annualeconomic payoff from bST of $132,000 under the quota,compared to $166,400 under the trigger price policy (seetable C-6). Similarly, the 52-cow Lake States dairyexperiences a decrease in bST economic payoffs from$3,900 to $2,400 due to the policy scenario change (seetable C-3).

The primary reason that the farms perform lessfavorably under the quota than the other two policies isthat the total milk sold is reduced while fixed costs remainthe same. Fewer cows and pounds of milk are available tospread out the fixed costs associated with the fixed plantsize. If the dairy farms were able to utilize the resultingexcess capacity for other purposes, the decrease in netcash farm income, net worth, and probabilities of survivaland success would not be as great. However, the spe-


cialized facilities associated with modern dairy farmingare not suitable for other enterprises.

Summary--Simulation results for representative dairyfarms indicate that bST adopters enjoy a greater averageannual net cash farm income than nonadopters acrossthree different types of farm policies. In addition toincreasing net cash farm income, bST adoption leads togreater real ending net worth, after-tax net present value,and probabilities of survival and success. Economicpayoffs to bST adopters are greater for larger farms thanfor smaller farms. The increased net return for largerfarms may accelerate the growth in average herd size asproducers seek to reduce fixed costs per cow, and takegreater advantage of high-level management practicesassociated with bST adoption.

The absolute economic payoff from bST adoption isabout the same under a trigger price dairy policy and afixed support price policy (see table C-7). Increasing theprice of milk by maintaining the milk support price at its1989 level does not greatly increase the economicincentive to adopt bST. On the other hand, the economicincentive to adopt bST is significantly lower if aproduction quota is in effect. All but one of the eightrepresentative farms experienced a 20- to 40-percentdecrease in the economic payoff to adopt bST under aquota. The exception (52-cow Northeast dairy) experi-enced a 70-percent decrease in the economic payoffassociated with bST adoption. These results suggest thatthe rate of bST adoption would be slowed by imposing astrict production quota rather than a trigger price policy.

Table C-7-Comparison of Average AnnualEconomic Payoffs From bST Adoption forEight Representative Dairy Farms Under

Three Alternative Dairy Policies,Assuming No Change in MilkDemand, 1989-98a (in $1,000)

Policy scenario

Trigger FixedRegion/size price support Quota

Lake States:Moderate . . . . . . . . .Large . . . . . . . . . . . .

Northeast:Moderate . . . . . . . . .Large . . . . . . . . . . . .

Southwest:Moderate . . . . . . . . .Large . . . . . . . . . . . .

Southeast:Moderate . . . . . . . . .Large . . . . . . . . . . . .

3.910.3

3.415.8

26.590.5

21.9166.4

4.110.9

3.616.6

26.691.7

22.8166.3

2.47.0

1.08.8

18.361.2

17.2132.0

aEconomic payoffs from bST are the average annual change in net cashfarm income between a nonadopter and a bST adopter over the 1989-98planning horizon. The payoff is net of the cost of bST, the addedtransportation costs for milk, and the additional feed.


Economic Payoffs to bST Adoption forAlternative Milk Demands

The introduction of bST may contribute to a change inthe demand for milk and milk products, depending on theperception of consumers. To quantify the impacts of milkdemand changes on the economic incentives to adoptbST, the eight representative dairy farms were simulatedunder three alternative milk demand scenarios, with atrigger price policy. Tables C-8 to C-n summarize thesimulation results for the following changes in milkdemand: no change, small decrease, and large decrease.For the no-change scenario, milk demand was assumed tobe the same as under the no-bST scenario. A smalldecrease in milk demand is defined as a 10-percentdecrease in 1991, a 5-percent decrease in 1992 (i.e.,demand increases from 1991 to 1992), and a 2.5-percentdecrease from 1993 to 1998. The large milk demanddecrease involves a 10-percent decrease in demandpersisting from 1991 to 1998.

Decreasing the demand for milk reduces the economicpayoffs associated with bST adoption for all eightrepresentative dairy farms (see tables C-8 to C-12). Thisresult is observed for both small and large decreases inmilk demand. For example, the economic payoff for bSTadoption is $10,300 for the 125-cow dairy in the LakeStates if there is no decrease in milk demand (see tableC-12). If demand decreases slightly, the economic payofffalls to $9,200, and if the demand decrease is large, theeconomic payoff declines to $6,900 (see table C-12).Thus, the incentive to adopt and the rate of adoptionwould be reduced if milk demand declines due toconsumers’ reaction to bST.

The probabilities of survival and economic success arereduced as well by decreases in milk demand. Theseprobabilities decline as lower milk prices lead to lower netcash farm incomes, net worths, and net present values.Examini ng the 350-cow dairy in the Southwest indicatesthat for the bST adopter, the probability of survivaldeclines slightly from 97 to 94 percent if there is a smalldecline in milk demand (see table C-10). If the milkdemand decrease is large, this farm’s probability ofsurvival falls to 69 percent. Because the economic payofffor bST adoption is positive (see table C-12), thoseproducers who adopt bST will experience greater proba-bilities of survival and economic success than nonadop-ters.

The most significant result for the demand-changescenarios is the dramatic reduction in the economicviability of dairy farms (probabilities of survival, success,and of increasing real net worth) associated with a largedecrease in milk demand (tables C-9 to C-12). AU of theregions are affected by the lower milk demand, given atrigger price dairy policy. If a large decrease in milkdemand is experienced, a dairy termination program

Appendix C-Detailed Farm Level Impacts of bovine Somatotropin and Other Emerging Technologies ● 109

Table C-8—Effects of Milk Demand Changes on the Economic Viability of Representative Dairy Farms in the LakeStates Who Adopt and Fail To Adopt bST, Assuming a Trigger Price Dairy Policy, 1989-98

Demand scenario

No change in Small demand Large demandmilk demand reduction reduction


52-COW farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125-cow farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

58.058.0

0.025.244.3

74.074.00.0

67.681.2

40.040.0

0.0–21 .8

7.5

48.048.00.04.7

27.6

13.013.00.0

-85.5-47.9

24.024.00.0

-63.4-32.2

15.6 28.6 2.6 9.7 -16.9 -11.3

-2.0 1.9 -6.8 -3.8 -12.1 -10.2

95.090.0

8.0194.8329.1

99.095.012.0

271.9396.4

85.068.0

2.078.3

227.6

91.082.0

7.0150.6290.9

46.037.00,0

-126.833.3

61.047.00.0

-48.6102.6

57.1 68.7 39.5 50.4 5.8 17.8

23.1 33.4 8.7 17.9 –10.7 -3.8SOURCE: Office of Technology Assessment, 1991.

Table C-9—Effects of Milk Demand Changes on the Economic Viability of Representative Dairy Farms in theNortheast Who Adopt and Fail To Adopt bST, Assuming a Trigger Price Dairy Policy, 1989-98

Demand scenario

No change in Small demand Large demandreduction reduction reduction


52-COW farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ZOO-COW farm:Probability of survival (percent) . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent

of beginning net worth (percent) . . . . . . . . .Average annual net cash farm income

($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

100.0100.0100.0

3.0232.3268.3

100.0100.0

3.0253.9286.4

100.0100.0

1.0199.1241.0

100.099.0

2.0218.4258.3

100.099.0

0.0149.8194.1

100.00.0

167.6210.0

72.4 77.2 65.0 69.7 52.4 56.7

14.5 17.9 9.7 12.8 2.3 5.0

100.099.043.0

616.7776.8

100.0100.053.0

717.6855.4

100.098.026.0

542.6722.8

100.099.045.0

632.7799.4

98.091.0

9.0352.2542.9

99.094.017.0

438.8618.4

92.3 101.7 85.9 95.0 64.5 73.5


11O . U.S. Dairy Industry at a Crossroad: Biotechnology and Policy Choices

Table C-l O-Effects of Milk Demand Changes on the Economic Viability of Representative Dairy Farms in theSouthwest Who Adopt and Fail To Adopt bST, Assuming a Trigger Price Dairy Policy, 1989-98

Demand scenario

No change inmilk demand

Small demandreduction

Large demandreduction

Non- bSTadopter adopter



350-cow farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent) . . . . .Net present value ($1,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1,500-cow farm:Probability of survival (percent) . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

95.095.060.0

739.7701.2

97.097.079.0

885.2820.7

88.0 94.088.0 94.035.0 51.0

506.5 655.9508.3 630.9

52.0 69.052.0 69.0

6.0 11.070.2 233.298.9 236.5

109.5 128.1 79.4 98.5 15.4 36.9

109.6 136.1 70.5 94.7 17.6 34.7

100.0100.0100.0

4,062.84,323.0

100.0100.0100.0

4,548.74,751.1

100.0 100.0100.0 100.096.0 98.0

3,230.6 3,678.33,600.8 3,992.2

98.0 100.096.0 98.053.0 70.0

1,820.4 2,268.02,278.4 2,671.1

194.5 213.7 162.0 179.6 102.5 120.2

713.9 804.4 554.5 642.5 282.7 360.2SOURCE: Office of Technology Assessment, 1991,

Table C-n-Effects of Milk Demand Changes on the Economic Viability of Representative Dairy Farms in theSoutheast Who Adopt and Fail To Adopt bST, Assuming a Trigger Price Dairy Policy, 1989-98

Demand scenario

No change in Small demand Large demandmilk demand reduction reduction


200-cow farm:Probability of survival (percent).. . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent). . . . .Net present value ($1,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

100.099.013.0

453.3727.9

100.0100.024.0

601.5854.3

99.089.0

4.0259.9562.1

100.094.0

9.0400.4685.6

88.051.00.08.0

321.5

94.080.0

1.0147.2444.2

75.6 88.7 58.4 71.2 33.4 46.1

17.3 39.2 –9.7 10.7 -42.7 -25.11,500-cow farm:Probability of survival (percent) . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . .Probability of increasing equity (percent) . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . .Present value of ending net worth as a percent


($1,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .SOURCE: Office of Technology Assessment, 1991.

100.0100.088.0

4,964.95,901.3

100.0100.099.0

6,165.76,712.4

100.0100.065.0

3,139.95,001.4

100.0100.086.0

4,032.35,772.3

100.089.019.0

1,689.13,633.2

100.099.050.0

2,562.24,390.7

129.4 147.2 109.7 126.6 79.7 96.3

609.0 775.4 453.8 615.2 200.2 343,8

Appendix C-Detailed Farm Level Impacts of bovine Somatotropin and Other Emerging Technologies . 111

Table C-12—Comparison of Average AnnualEconomic Payoffs From bST Adoption for EightRepresentative Dairy Farms Under AlternativeMilk Demand Scenarios, Assuming a Trigger

Price Dairy Policy, 1989-98a (in $1,000)

Table C-13-Comparison of Average AnnualEconomic Payoffs From bST Adoption for Eight

Representative Dairy Farms, Given a LargeReduction in Milk Demand, Assuming a TriggerPrice and Dairy Termination Program, 1989-98a

(In $1,000)

Demand reduction

Region/size No change small Large

Lake States:Moderate . . . . . .Large . . . . . . . . .

Northeast:Moderate . . . . . .Large . . . . . . . . .

Southwest:Moderate . . . . . .Large . . . . . . . . .

Southeast:Moderate . . . . . .Large . . . . . . . . .

3.910.3

3.415.8

26.590.5

21.9166.4

3.09.2

3.114.7

24.288.0

20.4161.4

1.96.9

2.712.8

17.177.5

17.6143.6



could be implemented (similar to the program in 1986) tobring production back into line with milk demand.

The LIVESIM model (app. B) analyzed a dairytermination program, given a large reduction in milkdemand. The result was higher milk prices than under thetrigger price policy, given the same demand scenario. Ifthe dairy termination program is implemented in 1991higher milk prices result from 1992 to 1998. Differencesin milk prices between the dairy termination program andthe trigger price policy declined from 50 cents per cwt in1992 to less than 10 cents per cwt in 1998 as milk supplyincreased relative to milk demand.

The farm-level impacts of the dairy termination pro-gram are summarized in tables C-13 to C-17. The triggerprice policy results assume the same milk demand (largedemand reduction) and are used as a reference policy. Thedairy termination program leads to higher probabilities ofsurvival, success, and increasing equity than the triggerprice policy for all eight representative dairy farms. Themoderate-size farms had greater increases in probabilityof survival than the large farms from the dairy termin ationprogram.

As observed for the other policy and demand scenarios,bST adopters were more profitable than nonadopters. Thisresult is summarized, in terms of the average annualeconomic payoffs for bST adoption, in table C-13. Theeconomic payoffs for bST adoption are positive for thedairy termination program and they are greater for thedairy termination program than for the trigger price

Trigger Dairy terminationRegion/size price policy program

Lake States:Moderate . . . . . . . . . . . . . . 1.9 4.1Large . . . . . . . . . . . . . . . . . 6.9 10.3

Northeast:Moderate . . . . . . . . . . . . . . 2.7 3.2Large . . . . . . . . . . . . . . . . . 12.8 15.3

Southwest:Moderate . . . . . . . . . . . . . . 17.1 25.4Large . . . . . . . . . . . . . . . . . 77.5 85.3

Southeast:Moderate . . . . . . . . . . . . . . 17.6 22.3Large . . . . . . . . . . . . . . . . . 143.6 172.7



policy. For example, the economic payoffs for a moderate-size Lake States dairy that adopts bST are $1,900 for thetrigger price policy and $4,100 for the dairy terminationprogram. A large Lakes States dairy farm had aneconomic payoff of $6,900 for the trigger price and$10,3OO for the dairy termination program (see tableC-13). In the remaining three regions, the large farmsgained more from bST adoption than the moderate-sizefarms; this differential was greater under the dairytermination program than under the trigger price policy(see table C-13). This reflects the higher milkprices underthe dairy termination program. It also suggests that bSTadoption would be accelerated even in the face ofdeclining milk demand if a dairy termination program wasintroduced.

1.

2.

3.

4.

Appendix C References

Chalupa, W. and Galligan, D.T., ‘Nutritional Implications ofSomatotropin for Lactating Cows,” J. Dairy Sci. 72(1989):2510-2524.Peel, D., ‘National and Regional Impacts of bovine Somato-tropin Adoption Under Alternative Daixy Policies, ” OTAcommissioned background paper, Washington, DC, 1990.Penson, J.B., Jr. and Chen, D.T., ‘‘Design and Application ofCOMGEM for Farm Policy Analysis,’ Large Scale Model-ing of Agriculture Sector, Iowa State University Press, 1990,ch. 8.Richardson, J.W. and Nixon, C.J., Description of FLIPSIM.V: A General Firm Level Policy Simulation Model, TexasAgricultural Experiment Station, bulletin B-1528, 1986.


Table C-14-Effects of a Large Reduction in Milk Demand on Representative Lake States Dairy Farms Who Adoptand Fail To Adopt bST, Given a Trigger Price and a Dairy Termination Program, 1989-98

Policy scenario

Trigger price policy Dairy termination program

Non- bST Non- bSTadopter adopter adopter adopter

52-COW farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth ($1,000) . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent of

beginning net worth (percent). . . . . . . . . . . . . . . . . . . . . . . . . .Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . .125-cow farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth ($1,000) . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent of

beginning net worth (percent) . . . . . . . . . . . . . . . . . . . . . . . . . .Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . .,SOURCE: : Office of Technology Assessment, 1991.

13.013.00.0

-85.5-47.9

-16.9-12.1

46.037.0

0.0-126.8

33.3

5.8-10.7

24.024.0

0.0-63.4-32.2

–11 .3-10.2

61.047.00.0

–48.6102.6

17.8–3.8

90.090.0

1.0111.9117.2

41.33.3

99.097.020.0

326.1438.0

76.035.9

96.096.0

4.0146.9147.5

51.97.4

100.099.031.0

395.6498.6

86.546.2

Table C-15--Effects of a Large Reduction in Milk Demand on Representative Northeast Dairy Farms Who Adopt andFail To Adopt bST, Given a Trigger Price and a Dairy Termination Program, 1989-98

Policy scenario



52-COW farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . . 100.0 100.0 100.0 100.0Probability of success (percent). . . . . . . . . . . . . . . . . . . . . . . . . . . 99.0 100.0 100.0 100.0Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . . 0.0 0.0 3.0 7.0Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149.8 167.6 245.5 265.1Present value of ending net worth ($1,000) . . . . . . . . . . . . . . . . . 194.1 210.0 277.6 294.4Present value of ending net worth as a percent of

beginning net worth (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . 52.4 56.7 74.9 79.4Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . . 2.3 5.0 15.4 18.6200-cow farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . . 98.0 99.0 100.0 100.0Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.0 94.0 100.0 100.0Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . . 9.0 17.0 55.0 68.0Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.2 438.8 733.8 819.5Present value of ending net worth ($1 ,000) . . . . . . . . . . . . . . . . . 542.9 618.4 871.6 944.5Present value of ending net worth as a percent of

beginning net worth (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . 64.5 73.5 103.6 112.3Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . . 27.4 40.2 82.5 97.8SOURCE: Office of Technology Assessment, 1991.


Table C-16--Effects of a Large Reduction in Milk Demand on Representative Southwest Dairy Farms WhoAdopt and Fail To Adopt bST, Given a Trigger Price and a Dairy Termination Program, 1989-98

Policy scenario



350-cow farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent of

beginning net worth (percent) . . . . . . . . . . . . . . . . . . . . . . . . . .Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . .1,500 -cow farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth ($1 ,000) . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent of

beginning net worth (percent) . . . . . . . . . . . . . . . . . . . . . . . . . .Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . .

52.052.0

6.070.298.9

69.069.011.0

233.2236.5

100.0100.091.0

1,104.0984.3

100.0100.095.0

1,094.91,094.9

15.417.6

36.934.7

153.7158.8

170.9184.2

98.096.053.0

1,820.42,278.4

100.098.070.0

2,268.02,671.1

100.0100.0100.0

5,241.55,263.1

100.0100.0100.0

5,735.85,693.9

102.5282.7

120.2360.2

236.8882.1

256.2967.4


Table C-17—Effects of a Large Reduction in Milk Demand on Representative Southeast Dairy Farms WhoAdopt and Fail To Adopt bST, Given a Trigger Price and a Dairy Termination Program, 1989-98

Policy scenario



200-cow farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth ($1,000) . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent of

beginning net worth (percent) . . . . . . . . . . . . . . . . . . . . . . . . . .Average annual net cash farm income ($1,000) . . . . . . . . . . . . .

1,500-cow farm:Probability of survival (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of success (percent) . . . . . . . . . . . . . . . . . . . . . . . . . . .Probability of increasing equity (percent). . . . . . . . . . . . . . . . . . .Net present value ($1 ,000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Present value of ending net worth ($1,000) . . . . . . . . . . . . . . . . .Present value of ending net worth as a percent of

beginning net worth (percent). . . . . . . . . . . . . . . . . . . . . . . . . .Average annual net cash farm income ($1 ,000) . . . . . . . . . . . . .

88.051.0

0.08.0

321.5

94.080.0

1.0147.2444.2

100.0100.045.0

703.1938.7

100.0100.068.0

837.41,054.1

33.4-42.7

46.1–25.1

97.550.8

109.573.1

100.089.019.0

1,689.13,633.2

100.099.050.0

2,562.24,390.7

100.0100.099.0

5,470.56,946.5

100.0100.0100.0

6,459.47,798.4

79.7200.2

96.3343.8

152.3867.7

171.01,040.4


Appendix D

Workshop Participants and Reviewers

Agricultural Research and Technology Transfer Policy for the 1990s Workshop

Norman Berg Paul GenhoSeverna Park, MD National Cattlemen’s Association

Deseret Ranches of FloridaDonald BillsResearch Leader Robert HeilAgriculture Research Service DirectorU.S. Department of Agriculture Agricultural Experiment StationBeltsville, MD Colorado State University

Patrick Borich Jim HildrethDirector Managing DirectorCooperative Extension Service Farm FoundationUniversity of Minnesota Verner HurtLucas Calpouzos DirectorProfessor Agricultural and Forestry Experiment StationCalifornia State University Mississippi State University

Mary CarterAssociate AdministratorAgricultural Research ServiceU.S. Department of Agriculture

Neville ClarkTexas Agricultural Experiment StationTexas A&M University

Arnold DentonSenior Vice-PresidentCampbell Soup Co.Camden, NJ

Chester T. DickersonDirector, Agricultural AffairsMonsanto Co.

Catherine DonnellyAssociate DirectorAgriculture Experiment StationUniversity of Vermont

Jeanne EdwardsWeston, MA

W.P. FlattDeanCollege of AgricultureUniversity of Georgia

Ray FrisbeIPM CoordinatorDepartment of EntomologyTexas A&M University

Terry KinneyYork SC

Ronald KnutsonProfessorDepartment of Agricultural EconomicsTexas A&M University

William MarshallPresidentMicrobial Genetics DivisionPioneer Hi-Bred International, Inc.

Roger MitchellDirectorAgricultural Experiment StationUniversity of Missouri

Lucinda NobleDirectorCooperative ExtensionCornell UniversityIthaca, NY

Susan OffuttOffice of Management and BudgetWashington, DC

Dave PhillipsPresidentNational Association of County AgentsLewistown, MT

–114--

Appendix D--Workshop Participants and Reviewers ● 115

Jack PincusDirector of MarketingMichigan Biotechnology Institute

Dan RagsdaleResearch DirectorNational Corn Growers Association

Roy RauschkolbResident DirectorMaricopa Agricultural CenterMaricopa, AZ

Alden ReineDeanCooperative Agricultural Research CenterPraire View A&M University

Grace Ellen RiceLegislative AffairsAmerican Farm Bureau Federation

Robert RobinsonAssociate AdministratorEconomic Research ServiceU.S. Department of Agriculture

Jerome SeibertEconomistDepartment of Agricultural EconomicsUniversity of California

Keith SmithDirector of ResearchAmerican Soybean Association

William StilesSubcommittee ConsultantSubcommittee on Department Operations,

Research, and Foreign AgricultureCommittee on AgricultureU.S. House of Representatives

William TallentAssistant AdministratorAgricultural Research ServiceU.S. Department of Agriculture

Jim TavaresAssociate Executive DirectorBoard on AgricultureNational Research Council

Luther TweetenAnderson Chair for Agricultural TradeDepartment of Agricultural

Economics and Rural SociologyOhio State University

Walter WallaDirectorExtension ServiceUniversity of Kentucky

Tim WallaceExtension EconomistDepartment of Agricultural EconomicsUniversity of California-Berkeley

Mike WehlerVice PresidentNational Pork Producers CouncilPlain, WI

John WoesteDirectorCooperative Extension ServiceUniversity of Florida

Fred WoodsAgricultural ProgramsExtension ServiceU.S. Department of Agriculture


Animal Technology Workshop

Dale BaumanProfessorDepartment of Animal ScienceCornell University

David Berkowitz .DirectorTechnology Transfer and Coordination StaffFood Safety Inspection ServiceU.S. Department of Agriculture

Roger BlobaumDirector of Americans for Safe FoodCenter for Science in the Public Interest

Thomas T. ChenAssociate DirectorCenter of Marine BiotechnologyUniversity of Maryland

Gary CromwellProfessorDepartment of Animal ScienceUniversity of Kentucky

Stanley CurtisProfessorDepartment of Animal ScienceUniversity of Illinois-Urbana

Dennis DykstraVice-President of Quality AssuranceJack Frost Golden Plump PoultrySt. Cloud. MN

Terry EthertonProfessorDepartment of Dairy and Animal SciencePennsylvania State University

Don GillRegents Prof. and Extension Animal NutritionistOklahoma State University

William HanselDistinguished ProfessorDepartment of Veterinary ScienceLouisiana State University

Michael HansenResearch AssociateConsumer Policy InstituteConsumers Union

Norman HarveyFarmerFlorence, VT

Burke HealeyFarmerSouthern Cross RanchDavis, OK

Walter HobgoodDirector of Animal Nutrition and HealthMonsanto Agricultural Co.

Maynard HogbergHeadDepartment of Animal ScienceMichigan State University

Gerald IsaacsChairmanDepartment of Agricultural EngineeringUniversity of Florida

Daniel D. JonesDeputy DirectorOffice of Agricultural BiotechnologyU.S. Department of Agriculture

Lawrence D. JonesAssistant ProfessorDepartment of Animal ScienceCornell University

James KleibensteinProfessorDepartment of EconomicsIowa State University

John KopchickProfessorEdison Animal Biotechnology CenterOhio University

Edward KorwekPartnerHogan and HartsonWashington, DC

Darold McCallaPresidentGranada Biosciences, Inc.Houston. TX

Appendix D--Workshop Participants and Reviewers ● 117

Tom McGuckinAssociate ProfessorDepartment of EconomicsNew Mexico State University

Dave MeisingerProduct ManagerNew ProductsPitmann-Moore, Inc.

Jan E. NovakofskiAssociate ProfessorDepartment of Animal ScienceUniversity of Illinois

B.I. OsburnAssociate Dean for ResearchSchool of Veterinary MedicineUniversity of California-Davis

Dennis PowersDirectorHopkins Marine StationStanford University

Rodney PrestonProfessorDepartment of Animal ScienceTexas Tech University

Vernon PurselResearch LeaderAgricultural Research ServiceU.S. Department of Agriculture

Allan RahnProfessorDepartment of Animal ScienceMichigan State University

Caird RexroadResearch LeaderAgricultural Research ServiceU.S. Department of Agriculture

Tim RoseFarmerLyons, KS

Andrew RowanAssistant DeanSchool of Veterinary MedicineTufts University

Colin ScanesProfessor and ChairmanDepartment of Animal ScienceCook College, Rutgers University

Tom SporlederProfessorDepartment of Agricultural Economics

and Rural SociologyOhio State University

Charles StrongExtension Biotechnology CoordinatorUniversity of Georgia

Michael TomaszewskiProfessor/Extension Dairy SpecialistDepartment of Animal ScienceTexas A&M University

Edward VeenhuizenManagerAnimal Science ProjectsLilly Research LaboratoriesEli Lilly and Co.

NOTE: OTA appreciates and is grateful for the valuable assistance and thoughtful critiques provided by the workshop participants. me workshopparticipants do not, however, necessarily approve, disapprove, or endorse this report. OTA assumes full responsibility for the report and theaccuracy of its contents.


Reviewers

L.G. AdamsProfessorDepartment of PathologyTexas A&M University

David BarbanoProfessorFood Science DepartmentCornell University

Kenneth BondioliResearch ScientistGranada Biosciences, Inc.

William ChalupaProfessorDepartment of Animal ScienceUniversity of Pennsylvania

Richard FallertEconomistEconomic Research ServiceU.S. Department of Agriculture

Ned FirstProfessorDepartment of Meat and Animal ScienceUniversity of Wisconsin

Harold HafsVice President-ResearchAnimal Science ResearchMerck, Sharp, Deane Research Laboratories

N.A. JorgensenProfessorUniversity of Wisconsin

Robert KalterProfessorDepartment of Agricultural EconomicsCornell University

Ronald D. KnutsonProfessorDepartment of Agricultural EconomicsTexas A&M University

James LauderdaleDirector of ResearchThe Upjohn Co.

Bernard J. LiskaProfessorDepartment of Food SciencePurdue University

Kenneth McGuffeyResearch ScientistLilly Research LaboratoriesEli Lilly and Co.

J. Thomas McGuckinProfessorDepartment of EconomicsNew Mexico State University

Travis McGuireProfessorDepartment of PathologyWashington State University

J.D. MurrayProfessorDepartment of Animal ScienceUniversity of California-Davis

Nanette NewellResearch ScientistThe Synertech Group, Inc.

C.J. PeelAnimal ScientistMonsanto Co.

Derrell S. PeelAssistant ProfessorDepartment of Agricultural EconomicsOklahoma State University

James W. RichardsonProfessorDepartment of Agricultural EconomicsTexas A&M University

William ThatcherProfessorDepartment of Dairy ScienceUniversity of Florida

H. Allen TuckerProfessorDepartment of Animal ScienceMichigan State University

Henry TyrrellResearch ScientistAgricultural Research ServiceU.S. Department of Agriculture

J. Gregory ZeikusPresidentMichigan Biotechnology Institute

NOTE: OTA appreciates and is grateful for the valuable assistance and thoughtful critiques provided by the reviewers. The reviewers do no4 however,necessarily approve, disapprove, or endorse this report. OTA assumes full responsibility for the report and the accuracy of its contents.