J. Anim. Sci. 2005. 83:948-957
© 2005 American Society of Animal Science
Breeding for societally important traits in pigs1
E. Kanis*,2,
K. H. De Greef
,
A. Hiemstra*,3 and
J. A. M. van Arendonk*
* Animal Breeding and Genetics Group, Wageningen University, 6700 AH Wageningen, The Netherlands; and
and
Animal Sciences Group, 8200 AB Lelystad, The Netherlands
 |
Abstract
|
|---|
Pig breeding programs traditionally focus on the genetic improvement of production and reproduction traits that have a clear economic value. Because citizens and consumers increasingly attach value to traits that have little or no direct relationship to production costs or to the price of a product, pig breeding organizations want to pay more attention to societally important traits, such as the welfare and health of pigs, the ecological effects of pork production, and the healthiness and sensory quality of pork. Most societally important traits have an economic and a noneconomic value and are sufficiently heritable for effective genetic selection, although many genetic and phenotypic (co)-variances still have to be estimated. However, it often is not clear to a pig breeding organization how it may deal with the noneconomic value of breeding-goal traits. In this study, a retrospective selection-index method is proposed to obtain the proper weights for societally important traits in the breeding goal. First, the genetic-progress space for each breeding-goal trait is explored by increasing the weights, in a stepwise manner, to each societally important trait in the breeding goal, starting from zero. Subsequently, a pig breeding organization can adopt the resulting genetic-progress scenario that it considers most sustainable or most acceptable. The weights underlying the adopted scenario are considered to be the proper breeding-goal weights. The noneconomic value of each societally important breeding-goal trait is found by deducting its economic value from its obtained weight and is thereby expressed in monetary units. In addition to obtaining weights for societally important breeding-goal traits, the proposed method offers the possibility to estimate the societal costs of selecting for economic traits only, as well as the societal benefits and the economic costs of selecting for traits with a noneconomic value. The method is therefore a useful tool for the development of sustainable breeding goals. An example has been worked out for a sow-line breeding program.
Key Words: Breeding Goals • Economic Weights • Health • Noneconomic Value • Pigs • Welfare
|
Introduction
|
|---|
Traditionally, pig breeding programs have focused on the genetic improvement of production traits, such as growth rate, meat percent, feed efficiency, and piglet production, because these traits are economically important. However, this can be considered as a rather short-term goal because it is based on the current and fluctuating market economy (Olesen et al., 2000
). Moreover, there is more than just economic importance. For example, there is concern about the undesired side effects of selection for high production efficiency on animal health and welfare (Harper and Henson, 2001), which is not unjustified (Rauw et al., 1998), and there is concern about the environmental effects of pig production (Wheatley, 2003). Pig breeding organizations cannot neglect these public concerns and must become convinced that a sustainable breeding goal should not only include economic values of traits, but also noneconomic values, such as emotional and societal values (Gamborg and Sandøe, 2003; Grandinson, 2003).
In this article, traits with a noneconomic value will be called "societally important traits." Many traits have both economic and noneconomic importance, but little has been published about ways to deal with noneconomic weights in breeding goals because they are difficult to quantify. Therefore, in almost all breeding goals, the individual target traits are weighed on the basis of just their economic importance, thereby neglecting noneconomic importance of certain traits.
The purpose of this article is to clarify the concept of noneconomic value for breeding-goal traits and to introduce a method for quantifying the weights of pig breeding goal traits with a noneconomic value. Quantification of noneconomic value offers new possibilities to discuss the trade-off between genetic selection for economic and noneconomic traits. As an example, some consequences of selection for traits with economic and noneconomic importance will be demonstrated in a sow-line breeding program.
|
Pig Production and Societal Concerns
|
|---|
Modern pig production is no longer focused on just economically efficient pork production. Consumers, citizens, and governments press pig producers to also take other aspects into account, such as the safety and quality of pork, the welfare and health of pigs, and the environmental impact of the production system (e.g. Verbeke and Viaene, 2000; McInerney and Bennett, 2003). This is predominantly implemented by means of legislative regulations (e.g., restrictions on the use of medicines and feed additives, treatment and transport of pigs, regulations on minimum floor space and presence of distraction material for pigs, and various measures to limit emissions of minerals and gases from manure to soil, water, and air). Kanis and De Greef (2003) presented an overview of the most relevant legislation in the European Union.
In addition to this legislative framework, some pig producers voluntarily commit themselves to certain production schemes (e.g., organic farming) or to certain production methods (e.g., a green-label pig house with minimal emissions). Such combinations of legislative and voluntary commitments form a framework in which most pig producers aim for an economic optimization of production; however, on top of these minimum requirements, the pig industry often can and wants to do more in relation to its corporate social responsibility. In the longer term, by meeting societal concerns, the industry aims at improving its public image and/or obtaining a larger market share or even new markets.
Societal concerns can be dealt with in several ways and by several actors in the pork production chain (Kanis et al., 2003). Such actions have to be evaluated on the basis of their effectiveness, cost, and side effects. Pig breeding organizations can genetically improve those societally important traits that show sufficient genetic variation. Thus, relevant questions are which societally important traits should be genetically improved, to what extent, how fast, and at what expense.
|
The Concept of Economic and Noneconomic Values of Traits
|
|---|
A trait of an animal has an economic value if a level change of that trait results in economic benefit (e.g., lower production costs or a higher price for the product); however, apart from a possible economic benefit, there may be other, noneconomic reasons to change the level of a trait. For example, many sow keepers desire good vitality and health for their newborn piglets even if that means that litter size is somewhat smaller (Knol, 2001). Moreover, there is increasing societal pressure to improve welfare of animals, to produce animal products with less detrimental effect on the environment, and to improve the quality and safety of animal products. Usually, the industry will not be rewarded directly for such efforts.
One example of a trait that has both economic and noneconomic value is the leg condition of pigs. Poor leg condition increases production costs due to higher veterinary costs and a lower feed efficiency. In addition to those economic aspects, for emotional and ethical reasons, pig farmers, transporters, slaughterers, and the public do not like to see (or know of) pigs that have difficulty walking. Therefore, in addition to its economic value, leg condition also represents a noneconomic value. In Table 1, for various characteristics of pigs, five points (pluses) are arbitrarily allocated to their economic and noneconomic value to indicate the relative importance of both values for each trait. Economic values can usually be derived from bioeconomic models that describe relations between production levels and production costs over the whole production chain. It is obvious that assigning noneconomic values to traits is quite subjective.
View this table:
[in this window]
[in a new window]
|
Table 1. Arbitrary example of allocating relative economic and noneconomic values to various characteristics of pigs
|
|
|
Including Traits with Noneconomic Value in a Breeding Goal
|
|---|
Pig breeding can only effectively address those concerns that are linked to the genetic ability of pigs. Kanis et al. (2003) gave an overview of various possible societal concerns in pork production. From those concerns, the following four groups of traits can be derived as animal-related and can therefore be considered for genetic selection: 1) traits related to pig welfare and health; 2) traits related to ecological effects and use of natural resources; 3) healthiness and sensory quality of pork; and 4) traits mainly related to the production costs of pork. Which traits actually will be included in a sustainable breeding goal predominantly depends on their genetic aspects (e.g., (co)variances and particularly heritabilities) and on their relative importance (economic and noneconomic value). Both of these criteria are variable, depending on the husbandry conditions in which the pigs have to be kept and on the vision and the market situation of the particular pig breeding organization. For example, if crop farmers have a high demand for pig manure as natural fertilizer, then pig manure will not be considered as a burden to the environment, as long as maximum manure-application or odor norms are not exceeded. In that case, high feed efficiency is only important for its economic (production) aspects and not for its noneconomic (environmental) aspects.
Genetic Aspects
Most societally important traits seem sufficiently heritable to be included effectively in selection programs. For example, in a large dataset of growing pigs, Henryon et al. (2001) found heritabilities (h2) of 0.1 to 0.2 for resistance to various clinical and subclinical diseases, and Hemsworth et al. (1990) estimated h2 = 0.37 for fear of humans in gilts. Hermesch et al. (2000) found h2 = 0.13 to 0.27 for some performance traits of growing pigs that are related to excretion of minerals (including h2 = 0.15 for feed efficiency) and h2 = 0.14 to 0.35 for various meat quality traits. In some other species, behavioral traits have been found to be heritable (e.g., feather pecking in chickens, Kjaer et al., 2001; and stereotypies in striped mice, Schwaibold and Pillay, 2001).
However, some societally important traits are not yet accurately defined (e.g., welfare-related traits) and under prevailing pig-husbandry conditions, heritabilities for those traits are often not known precisely enough for application in breeding programs. The same holds for their genetic correlations with other (economic and noneconomic) traits. Both heritabilities and genetic correlations can only be estimated accurately in large datasets with a sufficient pedigree structure. It takes time, however, to decide on the measurement of a new societally important trait and to collect sufficient data for genetic analyses. Another point to consider is whether major genes are involved and whether accurate DNA markers are, or will soon become, available to test for favorable alleles of some of these traits. A well-known and successful example is selection for decreased stress susceptibility in pigs by means of a DNA test (De Vries et al., 2000).
As long as DNA tests are unavailable and the necessary collection and analyses of phenotypic data have not been carried out, the relevant genetic parameters must be guessed, and the derived breeding values will be inaccurate. Pig breeding organizations are therefore reluctant to include traits with noneconomic value in their breeding programs.
Importance
The second condition for traits with noneconomic value to be included in the breeding goal is that their importance is considered high enough and can be quantified. Traits with economic and noneconomic value have to be combined in one breeding goal, and the economic value is usually expressed in monetary units; thus, noneconomic values also should be expressed in monetary units. Several semieconomic approaches have been applied, mainly with respect to meat quality traits (see also Olesen et al., 1999): 1) To some traits, like sensory pork-quality traits, no economic value can be assigned because the market does not pay for sensory quality. Moreover, these traits are difficult to measure quickly and cheaply and there is no general consensus about the best quality (e.g., color, taste). Therefore, either restricted selection indexes are applied that do not change sensory pork quality, or economic weights are used, based on an expected economic surplus value of pork with a better sensory quality (Knap et al., 2002). 2) Another approach is to find out how much more consumers will pay (willingness to pay) for pork from pigs with better welfare or health, for pork that is produced in an environmental friendly way, fed feed without additives, etc., compared with regular pigs (e.g. Olesen et al., 2000; Meuwissen et al., 2003). Such information is often based on questionnaires presented to consumers face to face or by telephone or Internet. However, the additional amount that consumers actually are willing to pay in practice is usually much less than what they say they are willing to pay (Harper and Henson, 2001). 3) Consumers can be asked for their preferences for certain visible pork characteristics that are not part of a payment system (e.g. fatness, color) or their purchasing behavior with respect to these characteristics can be analyzed. Based on such preferences, a producer can try to enlarge its market share. The value of an enlarged market share can be expressed in monetary units. As with the previous approach, this approach also may easily result in estimates of selling figures that are too optimistic for alternative and more expensive pork. 4) Better quality may result in more of the product in a high price class. For example, regarding pork quality traits with an optimum, such as pH, color, and i.m. fat content, Hovenier et al. (1993a) assumed that pork with these characteristics within certain boundaries could be used for fresh meat products rather than for processed meat. The economic value for a pork quality trait can then be derived from the assumed higher profit of fresh meat compared with processed meat and depends on its mean and distribution in the pig population. Von Rohr et al. (1999) extended this method by assuming three or more pork quality classes with different prices.
All four approaches described above can be considered "semieconomic" because they all assume a higher income (or lower production costs) from a changed trait level, though this economic profit cannot be derived unambiguously by using common bioeconomic models. However, many societally important traits have no direct relationship to economic profit, and other approaches to quantify their importance must be applied.
|
Desired-Gains Approaches
|
|---|
In a common selection index the correlation between index value and aggregated breeding-goal traits is maximized. Therefore, all traits in the breeding goal should be expressed on the same scale. This is usually done by multiplying each trait by its proper economic weight which implies that all traits be expressed in monetary units. Economic weights are usually estimated irrespective of the predicted genetic gain. See Goddard (1998) for further considerations regarding economic weights.
In contrast with a common index, a desired-gains index aims to maximize the correlation with a predefined (desired) linear combination of relative genetic gains for some or all breeding-goal traits. The (implicit) weights for the breeding-goal traits that actually match the desired genetic gains can be calculated afterward. If a breeding goal consists of traits with economic weights and traits with relative desired gains, then the desired proportion of the maximal possible genetic progress for the desired gain traits should also be predefined. See Brascamp (1984) for further details.
Determination of appropriate desired gains and matching weights for (some) breeding goal traits is often based on subjective elements. Some approaches are given: 1) Zero gains. Some traits might be already at their optimal level or for some traits, it might be desired that they do not deteriorate any further as a consequence of selection for other traits. Feed intake capacity has been considered as an example (e.g., Brandt, 1987). Kempthorne and Nordskog (1959) already presented selection indexes with zero changes for one or more breeding goal traits. 2) Competitive position. For competitive reasons (to maintain or enlarge market share), a breeding organization might temporarily want to gain more genetic progress for one or a few traits, for which it is "weaker" than its competitors. Often this is achieved by subjectively increasing the (economic) weights of the "weak traits" such that their estimated genetic progress is at the desired level. De Vries (1989) introduced a more objective method to calculate the weight for a weak trait in the breeding goal, depending on the difference between performance levels of competitors for this trait, the degree by which the weak trait can be compensated by a "strong trait," and the standard deviation of the frequency distribution of the purchasers at various acceptance levels of the weak trait. 3) Retrospective index. For a retrospective index, theoretically, all possible combinations of estimated genetic progress (the progress space) are compared (given all (co)variances) by varying the weights of the breeding-goal traits. The set of weights that results in the most desired scenario of estimated responses is then selected (Walsh and Lynch, 2000). This approach is similar to a "usual" desired-gains index, except that instead of using predefined desired gains, iteration on the estimated trait responses is required until the desired scenario of responses has been obtained.
For a breeding goal consisting of economically and societally important traits, the retrospective index is probably a good tool because it offers all options for the pig breeder to select pigs for future husbandry conditions in a future society and for a future market. Furthermore, communicating a breeding program on the basis of estimated responses is easier than on the basis of (economic) weights in the breeding goal. Inevitably, however, some subjective elements are involved in determining the most desired scenario of genetic responses.
|
Example
|
|---|
To illustrate dealing with societally important traits in a retrospective index, an example is given for a sow line that is selected for growth performance (growth rate and lean meat percentage of slaughter pigs) and reproductive performance (litter size, sow feed intake during first lactation, and interval from weaning to estrus). Ultrasonic back fat thickness at 110 kg live weight is measured as predictor of meat percent in live pigs. In addition to these economic traits, one trait with both economic and noneconomic value (leg quality of sows during gestation and lactation) and one trait with only noneconomic value (sow behavior during gestation) are included in the breeding goal. Group housing of sows during gestation is assumed and individual housing during lactation. The computer program SelAction (Rutten et al., 2002) was used to calculate estimated genetic progress per generation.
Parameters
Table 2 shows all eight traits involved, including their assumed heritabilities, genetic and phenotypic correlations, standard deviations, and economic values. Many of these heritabilities and correlations are based on a literature review by Eon (2002). Economic values for ADG, lean meat percent, litter size, and interval from weaning to estrus are based on average costs of feed, housing, piglets, etc., and on the Dutch payment system for carcass quality to the farmer, taking into account that the sow line passes only 50% of its genes to the slaughter pigs. A higher feed intake (capacity) of the sow during first lactation is considered more important in future, despite the higher feed costs, to support milk production by the sow and thereby support weight gain of the steadily growing number of piglets per litter. Moreover, sow feed intake is important to support the physical condition of the sow to enable her to produce more litters at lower replacement and veterinary costs. The economic value of 
1.50 per slaughter pig for each extra kilogram of feed eaten daily during lactation is based on feed costs, assumed profits from increased sow longevity, and on regressions of litter growth on sow feed intake during lactation. These regressions are obtained from Eissen (2000) and E. F. Knol (Inst. for Pig Genetics, Beuningen, The Netherlands, unpublished results). The standard deviation of feed intake during lactation is also obtained from Eissen (2000), whereas the other parameters for this trait are mainly based on relationships of feed intake in growing pigs with other traits. The economic value of leg condition (1.50) was assumed to be the decrease in production costs per slaughter pig if leg condition of the sow increases by 1 point on a linear scale from 1 to 5. This cost reduction is due to lower veterinary and replacement costs in the sows and lower veterinary costs in the slaughter pigs. In addition to this economic value, leg condition is supposed to have a noneconomic value because "society" does not accept pigs with leg problems. For behavior of sows, only a noneconomic value is considered because similar to leg problems, society does not want sows with abnormal behavior, such as stereotypic and aggressive behavior, including heavy fighting and vulva, ear, and tail biting. It is assumed that inspectors of the breeding organization assess leg condition of all available sows on fixed time intervals and that the farmers record behavior of their sows on a more or less permanent basis during gestation and express it in a score. Despite the same linear scale from 1 to 5, the standard deviation of behavior is slightly smaller than that of leg condition because inspectors are supposed to apply more variation in their scores than farmers do. In general, mildly unfavorable correlations are assumed between leg condition and behavior on the one hand and (re)productive traits on the other hand. The rg of –0.30 between behavior and feed intake during lactation is due to the assumption that sows with a higher feed intake capacity still get an average, but restricted, quantity of feed during pregnancy and are therefore hungrier and more sensitive to abnormal behavior.
View this table:
[in this window]
[in a new window]
|
Table 2. Traits involved in the sow-line breeding program and their parameters. Assumed heritabilities on diagonal (bold), genetic correlations above the diagonal, and phenotypic correlations below the diagonala
|
|
Population Characteristics
A sow-line nucleus population consisting of 400 sows and 40 boars is assumed. Each boar is mated to 10 sows, resulting in eight selection candidates per sow (four males, four females), which are all tested for ADG and backfat thickness during a growth period from 25 to 110 kg live weight. The new generation of 40 sires is selected out of 1,600 candidates based on performance for ADG and backfat thickness from 25 to 110 kg live weight, performance of seven full sibs and 72 half sibs for the same traits, and on pedigree information (BLUP breeding values) for all traits. Of the 1,600 female candidates, 800 are selected in the first stage on the same basis as the boars and mated to boar-line boars. In the second selection stage on a subnucleus, out of these 800 sows,the new generation of 400 dams is selected, based on the information in the first stage and their own performance for feed intake during lactation, interval from weaning to estrus, litter size, leg condition and behavior, performance of one full sib and 18 half sibs for the same traits, and pedigree information for all traits.
Results
Given the above set of parameters and population characteristics, with neither selection for leg condition nor for behavior (weights in breeding goal are both zero), these two traits slightly deteriorate with about –0.08 points per generation each (approximately –0.20 and –0.25 genetic standard deviations, respectively). These results match the evidence (Rauw et al., 1998) that, because of ongoing selection for economic traits, traits like pig health and welfare decrease. Figure 1 shows the total economic selection response and the responses for leg condition and behavior with increasing weight on leg condition and zero-weight on behavior. With only the economic value of leg condition (1.50; no noneconomic value) as its weight in the breeding goal, the deterioration of leg condition is approximately halved (–0.10 genetic standard deviations), whereas a weight of about 3.00 is needed to stop this decrease. As expected, the total economic response (progress per trait multiplied by its economic value, summed over all traits) reaches a maximum (approximately 1.25 per slaughter pig including 0.06 loss in leg condition) with a weight on leg condition of 1.50. However, if the weight on leg condition is doubled, the loss in total economic response is small. A further increase in weight on leg condition results in a significant decrease of total economic response, despite the strong improvement of leg condition itself. As shown in Figure 2, most of the total economic response originates from ADG and litter size, be it that the response for the first trait strongly decreases with increasing weight on leg condition, whereas the response for litter size is relatively constant. Similar to ADG, the response for lean meat percent also shows a rapid decrease with increasing weight on leg condition, with a negative response if weight on leg condition is more than approximately 5.0. Economic responses for feed intake during lactation and interval from weaning to estrus are low, even slightly negative, for interval from weaning to estrus, but rather independent on the weight for leg condition.
View larger version (25K):
[in this window]
[in a new window]
|
Figure 1. Total economic response and responses for leg condition and behavior with increasing weight on leg condition. No weight placed on behavior.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Figure 2. The economic responses for the economic traits, with increasing weight on leg condition. No weight placed on behavior.
|
|
Due to the small positive correlation with leg condition, behavior shows some improvement (or less impairment) with increasing weight on leg condition (see Figure 1
). Improvement in behavior is larger, however, if extra weight is put on it in the breeding goal. Figure 3 shows that with weights of 3.0 on both leg condition and behavior, the trait behavior will hardly continue to worsen (–0.01 
g), whereas leg condition improves (+0.07 g). The total economic response then decreases by approximately 13% from approximately 1.25 at its maximum in Figure 1 to approximately 1.08 with a weight of 3.0 on both leg condition and on behavior (including 0.04 economic gain in leg condition). More weight on behavior as well as more weight on leg condition results in a stronger decrease of total economic response. A slight increase in both societally important traits without too much loss in the economic traits can be obtained with a weight on behavior of approximately 3.5 and on leg condition of approximately 2.5. The response patterns for the individual economic traits with a weight of 3.0 on behavior are similar to Figure 2 (no weight on behavior), except at a somewhat lower level.
View larger version (23K):
[in this window]
[in a new window]
|
Figure 3. Total economic response and responses for leg condition and behavior with increasing weight on leg condition. The weight on behavior in breeding goal was 3.0.
|
|
Suppose that from all possible genetic-progress scenarios, the one with a weight of 3.0 on both leg condition and behavior is considered most acceptable or sustainable. This implies that the noneconomic values for leg condition and behavior are supposed to be 1.5 and 3.0 per scale point, respectively. Given those weights, Table 3 shows the estimated genetic gain with the economic-selection scenario and with the sustainable-selection scenario. With economic selection, the economic progress is maximal, but this is accompanied by a large loss in noneconomic (societal) value. Sustainable selection leads to a 16% higher total response (1.11 vs. 0.96).
View this table:
[in this window]
[in a new window]
|
Table 3. Estimated genetic gain (/slaughter pig) with economic selection and sustainable selection, with respective breeding-goal weights on leg condition and behavior shown in parentheses
|
|
In practice, the problem is that with economic selection, the breeding industry gains 1.25 per slaughter pig, whereas there is a gain of just 1.08 with sustainable selection (the economic gain). However, in the first case, society actually loses 1.25 – 0.96 = 0.29, whereas in the second case, society gains 1.11 – 1.08 = 0.03. Therefore, to stimulate sustainable selection, society should compensate the industry by paying it at least 1.25 – 1.08 = 0.17 more per slaughter pig. This amount is still 40% (0.17 vs. 0.29) less than the societal cost with just economic selection considered.
The above example illustrates the dilemma faced by a pig-breeding organization dealing with selection for societally important traits. Assuming that correlations between traits with economic value and those with noneconomic value are generally negative, (extra) selection emphasis on traits with noneconomic value will always result in less than maximal progress in economic traits or, in other words, in less reduction in pork production costs by breeding. As shown, for any combination of weights on noneconomic traits in the breeding goal, the resulting genetic progress for these traits can be predicted, as can the loss in economic progress. In practice, the breeding organizations should find some balance between both phenomena. Therefore, additional aspects should be considered, such as the extra costs that are involved in the measurement of societally important traits and the risk that those traits may be of more or less importance in the future. Further, the competitive position of the breeding organization for each breeding goal trait on the various (exports) markets is important in making the right decision.
|
Candidate Breeding Goal Traits with Noneconomic Value
|
|---|
Pig Welfare and Health
Traits related to animal welfare and health can be classified into the following categories: 1) temperament, which includes internally driven traits, such as aggressiveness, activity, curiosity, and docility; 2) stress resistance, including mental resistance against environmental influences and expressed in traits such as fearfulness, nervousness, and flexibility; and 3) robustness, including bodily soundness and fitness and disease resistance, which give pigs the capability to adapt successfully to changing environmental management and health conditions, thereby better tolerating those conditions.
There is ample evidence that in general these traits have sufficient genetic variation for successful selection (for overviews, see Faure and Mills, 1998; Simianer and König, 2002; Kanis et al., 2004), and that their economic and/or noneconomic values are high enough to make such selection worthwhile (Harper and Henson, 2001; Meuwissen et al., 2003). The challenge, however, is to clearly define a limited number of representative welfare and health traits (and environments to test them) for which individual quantitative information can be obtained routinely. At present, many pig breeding programs pay attention to some welfare-related traits, such as stress susceptibility, quality of legs, longevity of sows, vitality of piglets, and heritable defects, mainly because of economic reasons. Breeding for societal aspects would mean that weights of the aforementioned traits in the breeding goal should be increased by their noneconomic value and that some behavioral traits, such as the incidence of deviant behavior (e.g., aggression, ear and vulva biting, some stereotypies), should be included.
Ecological Effects and Natural Resources
There are many possible animal effects on the ecology and use of natural resources regarding pork production. Examples are as follows: 1) Excretion of minerals (in particular N and P) and heavy metals in manure, which, among other factors, is influenced by feed efficiency (depending on digestion, maintenance requirements, growth rate, body composition, etc.), number of growing and finishing pigs, and number of sows to be maintained per kilogram of pork produced. 2) Use of fossil energy, which can be decreased by breeding pigs that can resist cold (and heat), and which is thus related to animal welfare. 3) Use of various organic residuals and waste from the human food chain as feed for pigs, and consequently breeding of pigs that produce well on high quantities of waste (pigs as waste converters), and thus saving resources for human applications. 4) Use of medicines and feed additives that either directly (by spilling) or indirectly (excretion of residuals) can pollute the environment. By not using those substances, and by breeding pigs that produce well without them (healthy and natural pigs), these risks can be decreased, which also meets another societal concern: safety of pork for consumption. 5) Maintenance of genetic diversity is considered to be increasingly important. Although at present the focus is on conservation of rare breeds, it may give rise to restrictions on selection within breeds and lines in the future.
We focus here on concerns related to manure production as they are currently considered to be most important, especially in regions with high pig density, because manure surpluses can severely affect the quality of air, land, and water in various ways (Jongbloed and Lenis, 1998). The fact that the European Union has developed a "nitrate directive," allowing a yearly application of 170 kg of N/ha of cultivated land to keep or get the N concentration of surface waters below 50 mg of nitrate/L, indicates that a manure surplus (in particular N surplus) is seen as an important threat to water quality.
The quantity of minerals excreted per kilogram of pig meat produced depends largely on production and reproduction efficiencies, which are already part of all commercial pig breeding goals for economic reasons. Efficiency traits, such as feed efficiency, ADG, and carcass lean percent, are medium to highly heritable (h2 = 0.2 to 0.5) and also are favorably correlated genetically. Heritabilities of reproduction traits are much lower (0.10 to 0.15) but still high enough for successful genetic selection. The fact that efficiency traits have environmental, and therefore societal value, suggests that these traits should get more selection emphasis than if based solely on their economic aspects.
Healthiness and Sensory Quality of Pork
Healthiness of pork refers to the natural composition of fat and meat in the pig’s carcass (excluding possible contaminants and residuals from medicines or from feed additives, etc.) and their effects on the health and longevity of the pork consumer. For sensory pork quality, two groups of traits can be distinguished: 1) traits concerning appearance, such as leanness, color, marbling, and water-holding capacity, which affect the attractiveness of pork for the consumer at purchase; and 2) eating quality, including shrinkage, tenderness, and taste, which affects the buying consistency of the consumer.
Although consumers find these traits to be important, at present most of them do not pay substantially more for better pork quality, and producers usually will not be rewarded for improving the quality of pork (Knap et al., 2002). Therefore, if pig breeders decide to select for meat quality, it should be treated predominantly as a trait with noneconomic value.
Pork has often been perceived as relatively fatty and therefore as unhealthy. However, the i.m. fat content in pork is lower on average than in beef and lamb (Verbeke et al., 1999), and the proportion of fat in pig carcasses has decreased tremendously during the last few decades due to management and breeding. Nonetheless, studies on consumer behavior regarding preference for pork consistently show that still most consumers prefer pork with little fat (Grunert, 2002; Ngapo et al., 2004). Recently, fatty acid composition, in particular the ratio between different types of polyunsaturated fatty acids, has been considered important to decreasing the risk of cancers and coronary heart disease (Wood et al., 2004).
Genetic and nutritional effects on fatty acid composition, of which nutritional effects seem most important, have been found (Cameron et al., 2000). Many sensory quality aspects of pork are heritable and can thus be changed by genetic selection (for reviews, see Hovenier et al., 1993b; Rosenvold and Andersen, 2003). Despite the fact that there is usually no big economic incentive, there is general agreement that pork quality is important, and many pig breeding programs already pay attention to various pork quality aspects, often by using physiological or DNA markers (Knap et al., 2002).
|
Discussion
|
|---|
As a consequence of increased corporate social responsibility and the desire for so called "balanced breeding programs," breeding organizations increasingly deal with animal characteristics that also (or only) have a noneconomic value. By incorporating those traits into the breeding goal, weighed with just their economic value, their importance is underestimated, resulting in a lower realized genetic gain than actually desired. Noneconomic values are difficult to combine with economic values, however, because the former are expressed in hedonistic units and the latter in monetary units. Because breeders already use economic values for most breeding goal traits, they prefer to express noneconomic values in monetary units as well. Also, the literature suggests that changes in traits with noneconomic value should be valued in economic units (e.g., Olesen et al., 2000). One example of this was presented by Wickham in 1979 (cited by Goddard, 1998), who derived an economic weight on temperament in dairy cows based on the regression of survival in the herd on milk yield (w) and temperament (t) with regression coefficients bw and bt, respectively, where bt usually has a negative value. The ratio bt:bw reflects to some extent how important temperament is for the dairy farmer relative to milk yield, and according to this approach, the (economic) weight on temperament is bt:bw multiplied by the economic weight on milk yield. The common unit for both traits here is their effect on survival in the herd, which, from the farmer’s perspective is basically an economic trait. There is reason for a negative weight on temperament because less temperament usually means a lower susceptibility for stress, and better animal welfare and survival in the herd under the actual farm conditions, which is usually appreciated by society. However, putting a weight on a trait in the breeding goal is not a guarantee that this trait will change in the desired direction. With negatively correlated traits, the total financial merit may be larger with a small undesired change in a particular trait (like temperament), compensated by a (large) desired change in other breeding goal traits (like milk yield). The retrospective index (Walsh and Lynch, 2000), as applied here, deals with the societal wish that some traits of animals should change in a particular direction. With the retrospective index, all possible combinations of changes in all breeding goal traits (economic and noneconomic) can be evaluated, and the index resulting in the most sustainable combination can be adopted and easily communicated. In practice, retrospective indices are probably already used quite often to study the effect of modified economic weights, even for selection indices containing just economic traits.
Implementing Selection for Societally Important Traits
Finding the proper balance between genetic progress of all breeding-goal traits requires foresight regarding the future importance of these various traits. This vision can be based on market and societal trends with respect to consumer preferences and citizen’s concerns, for example. However, breeding organizations will only modify weights or adopt new breeding goal traits if they are convinced that it concerns strong and durable wishes. For example, little willingness by consumers to pay a premium for pork produced according to societal requirements will not stimulate genetic selection for societally important traits.
A breeding organization must have good contacts with the best and most advanced producers and with society to get the right impressions of the future production environments and the possible problems that pigs might have in future environments. As illustrated in Figures 1 and 3, the dilemma is often that extra noneconomic progress costs in terms of economic progress and, in turn, money. Moreover, routinely obtaining extra selection information usually results in higher testing costs, particularly if special testing conditions have to be created. Therefore, pig-breeding organizations usually first take the (expected) economic values of traits as a basis for their weights in the breeding goal. Competitive position may give rise to changes in those weights (De Vries, 1989) and then, noneconomic (e.g., societal) considerations may be a further reason to modify the weights.
One practical reason that breeding organizations hesitate to implement selection for societally important traits is that many of those traits are not yet clearly defined, their heritabilities and genetic (co)variances are often unknown, and recording of those traits might be rather problematic and subjective. Multidisciplinary research programs should be developed, therefore, to provide the necessary information for successful implementation of selection for societally important traits. Nevertheless, the risk of making the wrong decisions will always remain. By quantifying the expected genetic progress for the various breeding goal traits and its monetary value with varying noneconomic weights of societally important traits, as done in the present study, this decision-making process can be supported.
|
Implications
|
|---|
Pig-breeding organizations can use the selection-index method to derive weights for breeding goal traits with noneconomic but societal importance. The method is based on comparisons of scenarios of genetic progress for traits with economic and/or noneconomic value. The resulting economic weights for societally important traits offer the possibility to estimate societal costs of selecting for just economic profit, as well as the costs for the pig production industry when applying selection for societally important traits. This method can contribute to the development of sustainable breeding goals for breeding organizations.
|
Footnotes
|
|---|
1 Comments and discussions by the Green Piggery project team of INRA-WUR on drafts of this paper have been greatly appreciated. 
3 Current address: AgriMedia, P.O. Box 42, 6700 AA Wageningen, The Netherlands.
2 Correspondence: P.O. Box 338 (phone: +31 317 482335; fax: +31 317 483929; e-mail: egbert.kanis{at}wur.nl).
Received for publication August 10, 2004.
Accepted for publication January 5, 2005.
|
Literature Cited
|
|---|
Brandt, H. 1987. Development and genetic aspects of feed intake in three breeds of pigs at German test stations and measures to prevent further deterioration. Pig News Info. 8:29–33.
Brascamp, E. W. 1984. Selection indices with constraints. Anim. Breed. Abstr. 52:645–654.
Cameron, N. D., M. Enser, G. R. Nute, F. M. Whittington, J. C. Penman, A. C. Fisken, A. M. Perry, and J. D. Wood. 2000. Genotype with nutrition interaction on fatty acid composition of intramuscular fat and the relationship with flavour of pig meat. Meat Sci. 55:187–195.
De Vries, A. G. 1989. A method to incorporate competitive position in the breeding goal. Anim. Prod. 48:221–227.
De Vries, A. G., L. Faucitano, A. Sosnicki, and G. S. Plastow. 2000. The use of gene technology for optimal development of pork meat quality. Food Chem. 69:397–405.
Eissen, J. J., E. J. Apeldoorn, E. Kanis, M. W. A. Verstegen, and K. H. De Greef. 2003. The importance of a high feed intake during lactation of primiparous sows nursing large litters. J. Anim. Sci. 81:594–603.[Abstract/Free Full Text]
Eon, S. 2002. Inclusion of welfare traits in a pig breeding program: A simulation study. M.Sc. Thesis, Animal Breeding and Genetics Group, Wageningen Univ., Wageningen, The Netherlands.
Faure, J. M., and A. D. Mills. 1998. Improving the adaptability of animals by selection. Pages 235–264 in Genetics and the Behavior of Domestic Animals. T. Grandin, ed. Academic Press, San Diego, CA.
Gamborg, C., and P. Sandøe. 2003. The making of sustainability in farm animal breeding and reproduction. Pages 89–106 in Proc. of the Final SEFABAR Workshop, Rome, Italy. Farm Anim. Ind. Platform, Beuningen, The Netherlands.
Goddard, M. E. 1998. Consensus and debate in the definition of breeding objectives. J. Dairy Sci. 81(Suppl. 2):6–18.
Grandinson, K. 2003. Genetic aspects of maternal ability in sows. Ph.D. Diss., Swedish Univ. of Agric. Sci., Uppsala.
Grunert, K. G. 2002. Current issues in the understanding of consumer food choice. Trends Food Sci. Technol. 13:275–285.
Harper, G. C., and S. J. Henson. 2001. Consumer Concerns about Animal Welfare and the Impact on Food Choice. EU FAIR CT98-3678. Centre for Food Economics Research, The Univ. of Reading, U.K.
Hemsworth, P. H., J. L. Barnett, D. Treacy, and P. Madgwick. 1990. The heritability of the trait fear of humans and the association between this trait and subsequent reproductive performance of gilts. Appl. Anim. Behav. Sci. 25:85–95.
Henryon, M., P. Berg, J. Jensen, and S. Anderson. 2001. Genetic variation for resistance to clinical and subclinical diseases exists in growing pigs. Anim. Sci. 73:375–387.
Hermesch, S., B. D. Luxford, and H.-U. Graser. 2000. Genetic parameters for lean meat yield, meat quality, reproduction and feed efficiency traits for Australian pigs. 1. Description of traits and heritability estimates. Livest. Prod. Sci. 65:239–248.
Hovenier, R., E. W. Brascamp, E. Kanis, J. H. J. Van der Werf, and A. P. A. M. Wassenberg. 1993a. Economic values of optimum traits: The example of meat quality in pigs. J. Anim. Sci. 71:1429–1433.[Abstract]
Hovenier, R., E. Kanis, Th. Van Asseldonk, and N. G. Westerink. 1993b. Breeding for pig meat quality in halothane-negative populations—A review. Pig News and Info. 14:17N–25N.
Jongbloed A. W., and N. P. Lenis. 1998. Environmental concerns about animal manure. J. Anim. Sci. 76:2641–2648.[Abstract/Free Full Text]
Kanis, E., A. F. Groen, and K. H. De Greef. 2003. Societal concerns about pork and pork production and their relationships to the production system. J. Agric. Environ. Ethics 16:137–162.
Kanis, E., and K. H. De Greef. 2003. Concerns about pork production in Western Europe—Efforts to get the lights on green. Page 1–18 in Manipulating Pig Production IX. J. E. Paterson, ed. Australas. Pig Sci. Assoc., Inc., Werribee, Victoria, Australia.
Kanis, E., H. Van den Belt, A. F. Groen, and K. H. De Greef. 2004. Breeding for improved welfare in pigs: a conceptual framework and its use in practice. Anim. Sci. 78:315–329.
Kempthorne, O., and A. W. Nordskog. 1959. Restricted selection indices. Biometrics 15:10–19.
Kjaer, J. B., P. Sørensen, and G. Su. 2001. Divergent selection on feather pecking behaviour in laying hens (Gallus gallus domesticus). Appl. Anim. Behav. Sci. 71:229–239.[Medline]
Knap, P. W., A. A. Sosnicki, R. E. Klont, and A. Lacoste. 2002. Simultaneous improvement of meat quality and growth-and-carcass traits in pigs. Comm. No. 11-07 in Proc. 7th World Cong. Genet. Appl. Livest. Prod., Montpellier, France.
Knol, E. F. 2001. Genetic aspects of piglet survival. Ph.D. Diss., Wageningen Univ., Wageningen, The Netherlands.
McInerney, J., and R. Bennett. 2003. Economic aspects of sustainable European farm animal breeding and reproduction. Pages 61–88 in Proc. of the Final SEFABAR Workshop, Rome, Italy. Farm Anim. Ind. Platform, Beuningen, The Netherlands.
Meuwissen, M. P. M., K. Latouche, I. A. C. M. Van der Lans, and A. Carpentier. 2003. Consumer concerns about pork production in France and the Netherlands. Paper presented at the 82nd Eur. Sem. of the Eur. Assoc. Agric. Economists, Bonn, Germany.
Ngapo, T. M., J.-F. Martin, and E. Dransfield. 2004. Consumer choices of pork chops: results from three consumer panels in France. Food Qual. Pref. 15:349–359.
Olesen, I., B. Gjerde, and A. F. Groen. 1999. Methodology for deriving non-market trait values in animal breeding goals for sustainable production systems. Interbull Bull. 23:13–21.
Olesen, I., A. F. Groen, and B. Gjerde. 2000. Definition of animal breeding goals for sustainable production systems. J. Anim. Sci. 78:570–582.[Abstract/Free Full Text]
Rauw, W. M., E. Kanis, E. N. Noordhuizen-Stassen, and F. J. Grommers. 1998. Undesirable side effects of selection for high production efficiency in farm animals. A review. Livest. Prod. Sci. 56:15–33.
Rosenvold, K., and H. J. Andersen. 2003. Factors of significance for pork quality—A review. Meat Sci. 64:219–237.
Rutten, M. J. M., P. Bijma, J. A. Woolliams, and J. A. M. Van Arendonk. 2002. SelAction: Software to predict selection response and rate of inbreeding in livestock breeding programs. J. Hered. 93:456–458.[Free Full Text]
Schwaibold, U., and N. Pillay. 2001. Stereotypic behaviour is genetically transmitted in the African striped mouse Rhabdomys pumilio. Appl. Anim. Behav. Sci. 74:273–280.
Simianer, H., and S. König. 2002. Ist Zucht auf Krankheitsresistenz erfolgreich? Züchtungskunde 74:413–425.
Verbeke, W., M. J. Van Oeckel, N. Warnants, J. Viaene, and Ch. V. Boucqué. 1999. Consumer perception, facts and possibilities to improve acceptability of health and sensory characteristics of pork. Meat Sci. 53:77–99.
Verbeke, W. A. J., and J. Viaene. 2000. Ethical challenges for livestock production: meeting consumer concerns about meat safety and animal welfare. J. Agric. Environ. Ethics 12:141–151.
Von Rohr, P., A. Hofer, and N. Künzi. 1999. Economic values for meat quality traits in pigs. J. Anim. Sci. 77:2633–2640.[Abstract/Free Full Text]
Walsh, B., and M. Lynch. 2000. Theory of index selection. Page 434 in Evolution and Selection of Quantitative Traits. Sinauer Assoc. Inc., Sunderland, MA. Available: http://nitro.biosci.arizona.edu/zdownload/Volume2/Chapter23.pdf. Accessed May 26, 2004.
Wheatley, W. P. 2003. The natural and organic pork market: A sustainable niche for small-scale producers? A review and analysis of the evidence. Am. J. Altern. Agric. 18:18–26.
Wood, J. D., R. I. Richardson, G. R. Nute, A. V. Fisher, M. M. Campo, E. Kasapidou, P. R. Sheard, and M. Enser. 2004. Effects of fatty acids on meat quality: A review. Meat Sci. 66:21–32.
This article has been cited by other articles:
|
|
|
|
 
H. M. Nielsen, L. G. Christensen, and J. Odegard
A method to define breeding goals for sustainable dairy cattle production.
J Dairy Sci,
September 1, 2006;
89(9):
3615 - 3625.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
