ABSTRACT
To assess the economic importance of breeding traits, economic values (EV) were derived for 3 German dairy cattle breeds: German Holstein (HOL), Angler (ANG), and Red and White Dual-Purpose (RDN). For that purpose, the stochastic bio-economic model SimHerd (SimHerd A/S, Viborg, Denmark) was used, which simulates the expected monetary gain in dairy herds. The EV was calculated as the alteration in average net return of the herd responding to a marginal change in the trait of interest. When deriving EV using SimHerd, economic consequences resulting from changes in the age structure of a dairy herd (i.e., structural herd effects) are considered. However, this requires the simulation of relationships between traits in the bio-economic model. To avoid double counting, the EV of a trait was corrected for effects from alterations in correlated traits using multiple regression analysis. The EV were derived for 23 traits in terms of production, conformation and workability, dairy health, calf survival, and reproduction performance. Furthermore, the relative economic importance of the breeding traits was calculated. Relative emphasis on production was between 39.9 and 44.4% in the breeds studied. Total costs per case of ketosis and metritis ranged from €167 to €196 and €173 to €182, respectively. Highest marginal EV of direct health traits were found for mastitis (€257 to €271 per case) and lameness (€270 to €310 per case). Consequently, relative emphasis on direct health traits was between 15.7 and 17.9%. The EV of reproduction performance showed largest differences among the cattle breeds. Overall relative emphasis on reproduction was 10.5% in HOL, 10.8% in ANG, and 6.5% in RDN. The relative economic importance of cow mortality ranged from 15.5 to 16.0% across the breeds. Collectively, the study showed the high economic importance of functional traits in the cattle breeds studied.
Key words
INTRODUCTION
In recent years, many national breeding goals of dairy cattle have become broader as the focus has shifted from production toward functional traits (
Miglior et al., 2005
, Miglior et al., 2017
). According to Groen et al., 1997
, functional traits might be defined as characteristics that increase efficiency by reducing costs in dairy production, such as health, reproduction performance, and calving traits. Selection index theory (Hazel, 1943
) states that traits included in the breeding objective have to be weighted based on their relative economic importance. Economic values (EV) denote the marginal utility of a trait. In dairy cattle breeding, they reflect the direction and the effect of each trait in the breeding goal on monetary profit while keeping all other traits constant.The classical approach to achieving the marginal utility of a breeding trait relies on the partial derivative of profit equations (
Brascamp et al., 1985
; Van Arendonk, 1991
; Goddard, 1998
). For highly managed livestock production systems, the partial budgeting represents an appropriate procedure to derive EV (Janssen et al., 2017
). However, a profit function is a simple accounting method that probably does not reflect all interactions and their economic effects as they exist in reality (Nielsen et al., 2014
). Therefore, bio-economic models are increasingly being used to derive EV, because they allow a production system to be described in a more realistic manner. Such simulation models connect biological and economic components within a livestock production system based on mathematical modeling (Groen et al., 1997
). As the dynamics in dairy herds can be quite complex, bio-economic models might be preferable for the derivation of EV.According to
Wolfová and Wolf, 2013
, 2 aspects are particularly important when deriving EV. First, structural herd effects caused by changes in the age structure of a dairy herd should be taken into account. For instance, an increase in mastitis rate will result in a changed age distribution of the herd toward a higher proportion of younger cows, because more animals have to be replaced by heifers. This will result in a decreased average milk production of the herd but also in decreased treatment costs as susceptibility to diseases is lower for younger cows. These economic effects are purely of structural origin and must be assigned to the marginal utility of the evaluated trait to avoid an underestimation of its EV. Second, existing genetic correlations between traits have to be considered in a proper way when deriving EV, as otherwise there is a risk of so-called double counting (Wolfová and Wolf, 2013
). Double counting means that a change in a certain trait also alters genetically correlated traits (Berry et al., 2005
; Nielsen et al., 2006
). For instance, the occurrence of mastitis has impairing effects on the milk performance of a cow. However, if milk yield is a part of the breeding goal with an EV of its own, these economic consequences must not be assigned to the EV of mastitis. When deriving EV by means of a bio-economic model, double counting is a problem if genetic correlations between traits are simulated. To address this problem, Østergaard et al., 2016
proposed to correct the calculated EV of a breeding trait using multiple regression. By doing so, the economic effects of traits simulated as genetically correlated with the trait of interest are removed from the corresponding EV.Current EV of the traits in the breeding objective of German Holstein (HOL) are based on calculations from
Lind, 2007
. These EV were derived using a deterministic herd model developed by Amer et al., 1996
and adapted by - Amer P.R.
- Kaufman A.
- Künzi N.
Breed choice and pricing system implications for farmers and political institutions from a Swiss cattle farm model.
in: Dent J.B. McGregor M.J. Sibbald A.R. Livestock Farming Systems. Research, Development, Socio-Economics and the Land Manager. Wageningen Academic Publishers,
Wageningen, the Netherlands1996: 253-258
Miesenberger, 1997
. Since then, the market situation has changed substantially (e.g., discontinuation of milk quota, shifted market prices and costs). Thus, a re-estimation of the economic importance of breeding traits is required based on updated market conditions and population parameters. Furthermore, since 2019, a genetic evaluation of direct health traits has been available for HOL. The implementation of these traits in the overall index is expected in 2021; thus, EV of health traits are needed. For other dairy cattle breeds, notably for local breeds with smaller population sizes, a separate calculation of EV has not yet been carried out. The aim of this study was to estimate the EV of production and functional traits for HOL and for the local breeds Angler (ANG) and Red and White Dual-Purpose (RDN). With regard to functional traits, direct health traits and calf mortality were included, because EV for these traits are presently not available. For this purpose, a bio-economic model (Østergaard et al., 2005
) in combination with a regression approach (Østergaard et al., 2016
) was applied. On the one hand, this procedure enables inclusion of structural herd effects in the EV of a given trait. On the other hand, the double counting of economic effects, which can occur using a bio-economic model, is avoided.MATERIALS AND METHODS
The calculation of EV for HOL, ANG, and RDN was performed in 2 steps. First, the bio-economic model SimHerd (
Østergaard et al., 2005
) was calibrated to represent a general dairy herd for each of the particular breeds and, subsequently, effects on the average annual net return of the herd were investigated, given a change in the trait of interest. A detailed description of this stochastic herd model can be found in Østergaard et al., 2005
and, therefore, in this paper the model is outlined only briefly. Second, the output of the bio-economic model (i.e., the average annual net return) was modified using the statistical procedure proposed by Østergaard et al., 2016
to avoid the double counting of effects.SimHerd
The SimHerd model (
Østergaard et al., 2005
) has been used in various studies and for different purposes, such as to investigate economic consequences of crossbreeding (Clasen et al., 2020
), to economically assess activity meters for estrus detection (Pfeiffer et al., 2020
), and to study the economic implications of reproduction strategies in dairy herds (Ettema et al., 2011
). Applying SimHerd, typical structures in dairy herds comprising lactating cows and heifers were simulated as depicted in Figure 1. The model was used to mimic state changes of dairy herds over time, simulating each animal separately. For this purpose, the following operating principles were implemented in SimHerd: (1) Every cow and heifer in the herd was dynamically described in incremental steps week by week, generating cow-specific properties such as age, parity, lactation stage, culling status, actual milk yield, body weight, reproductive status, SCC, and disease status. (2) Both the revenues related to the herd performance (e.g., milk yield, sold heifers, slaughtered animals) and the costs arising from the utilization of input factors (e.g., feed, treatment costs, insemination costs) were mechanistically determined by summing up all sources of outputs and inputs for each animal in the herd. As a result, the average annual net return at herd level was obtained. (3) The occurrence of diseases was stochastically simulated drawing random numbers from relevant probability distributions to create phenotypic variation between animals. The incidence of disease may not only have an effect on feed intake, milk yield and body weight in the current and later lactations, but also influences other variables such as reproductive ability, the risk of cow mortality, and the occurrence of associated diseases. These effects have been explicitly described in earlier studies (Østergaard et al., 2003
, Østergaard et al., 2005
; Ettema et al., 2010
). Analogously, the presence of discrete events such as heat detection and conception, fetal death, sex and viability of born calves, and cow mortality were stochastically modeled.
Figure 1Schematic representation of the stochastic bio-economic model SimHerd.
In SimHerd, there is a fundamental distinction between culling and cow mortality with different economic effects associated. Culling was modeled as voluntary and involuntary culling, both generating income from slaughtered cows. Voluntary culling occurred for nonpregnant cows that produced less than 15 kg of ECM/d. Furthermore, lowest-yielding cow on the culling list was designated for culling when a pregnant heifer was available to enter the herd. Culling for other reasons than production (i.e., involuntary culling), was represented with a base risk of 0.9% in first week of milk that declined linearly to 0.079% in wk 29. In the remaining lactation, the base risk stayed constant at this level. Additionally, the cow's individual risk of involuntary culling was affected by the occurrence of diseases and the failure of conception within a specified breeding period. The breeding period was bounded by the voluntary waiting period (VWP) and the end of breeding. The VWP, defined as the time period postpartum after which cows will be inseminated, differed for the breeds studied (Table 3). Although VWP for primiparous cows was set to 56 d postpartum for all breeds, VWP for multiparous cows was reduced by 7 d for the breeds ANG and RDN. An extended VWP of 56 d for multiparous cows in HOL was applied, because the probability for silent estrus and, thus, a failure of heat observation increases in high-yielding cows (
Crowe, 2008
). The breeding period terminated at wk 50 and 44 for high-yielding primiparous and high-yielding multiparous cows, respectively. Both voluntary and involuntary culling resulted in economic revenues from slaughter cows. By contrast, cow mortality was modeled as the on-farm death of a cow without generating any salvage slaughter value.Model Input
The SimHerd model can be controlled by various input parameters such as levels of phenotypic traits and economic key figures. As the breeds ANG and RDN are mainly located in the Northern part of Germany, all input variables are typical for the current production conditions in the federal state of Schleswig-Holstein. The milk performance data were provided by the state control association Landeskontrollverband Schleswig-Holstein (
LKV SH (Landeskontrollverband Schleswig-Holstein), 2017
) and are shown in Table 1. As information about health traits from official recordings in Germany is lacking, incidence rates were adopted for most diseases from Danish Holstein for HOL and from VikingRed for ANG and RDN (Table 2). Phenotypic profiles for the reproductive performance of the breeds are given in Table 3. The key figures for calving interval, replacement rate, stillbirth risk, and calf mortality were taken from official annual reports (- LKV SH (Landeskontrollverband Schleswig-Holstein)
State Control Association Schleswig-Holstein (Landeskontrollverband Schleswig-Holstein). Annual report 2017.
https://www.lkv-sh.de/images/Download/pdf/Jahresbericht_2017.pdf
Date: 2017
Date accessed: June 4, 2018
VIT, 2017
; - VIT
Vereinigte Informationssysteme Tierhaltung w.V. Annual report 2017.
https://www.vit.de/fileadmin/Wir-sind-vit/Jahresberichte/vit-JB2017-gesamt.pdf
Date: 2017
Date accessed: June 4, 2018
LKV SH (Landeskontrollverband Schleswig-Holstein), 2017
). Prices and costs used for the calculation of EV are shown in Table 4. Compared with HOL and ANG, higher prices per kg liveweight, as well as higher sale prices of springing heifers, nonpregnant heifers, and bull calves, were assumed for RDN. Furthermore, prices for semen differed among the breeds. Due to a lower milk yield of RDN compared with HOL and ANG, a better utilization of roughage was assumed and, therefore, a lower requirement of concentrates. The additional labor time for farmers caused by the occurrence of diseases was set to 1.5 h for each case of mastitis and lameness; 1 h for ketosis, milk fever, and metritis; and 3 h for dystocia. Effort for stillbirth, calf mortality, and cow mortality was assumed to be 0.25 h per case. The personnel costs were set to €20.00/h.- LKV SH (Landeskontrollverband Schleswig-Holstein)
State Control Association Schleswig-Holstein (Landeskontrollverband Schleswig-Holstein). Annual report 2017.
https://www.lkv-sh.de/images/Download/pdf/Jahresbericht_2017.pdf
Date: 2017
Date accessed: June 4, 2018
Table 1Mean phenotypes for 305-d ECM, fat, and protein (kg) for the breeds Holstein (HOL), Angler (ANG), and Red and White Dual-Purpose (RDN)
| Period | HOL | ANG | RDN | ||||||
|---|---|---|---|---|---|---|---|---|---|
| ECM | Fat | Protein | ECM | Fat | Protein | ECM | Fat | Protein | |
| First lactation | 7,728 | 311 | 259 | 7,151 | 303 | 235 | 6,339 | 262 | 210 |
| Second lactation | 8,994 | 363 | 303 | 8,438 | 360 | 279 | 7,181 | 298 | 239 |
| Third lactation | 9,379 | 379 | 312 | 8,773 | 374 | 288 | 7,644 | 316 | 254 |
| >Third lactation | 8,776 | 357 | 287 | 8,417 | 357 | 276 | 6,992 | 290 | 229 |
Table 2Number of treatments (per 100 cow-years) for the respective diseases and the breeds Holstein (HOL), Angler (ANG) and Red and White Dual-Purpose (RDN)
| Disease | Breed | ||
|---|---|---|---|
| HOL | ANG | RDN | |
| Mastitis | 32.0 | 26.0 | 26.0 |
| Lameness | 24.0 | 19.0 | 19.0 |
| Ketosis | 5.0 | 4.4 | 4.4 |
| Milk fever | 4.0 | 3.5 | 3.5 |
| Metritis | 8.0 | 7.0 | 7.0 |
| Dystocia | 2.0 | 1.3 | 1.6 |
1
Sørensen et al., 2018
.- Sørensen L.P.
- Pedersen J.
- Kargo M.
- Nielsen U.S.
- Fikse F.
- Eriksson J.A.
- Pösö J.
- Stephansen R.S.
- Pedersen Aamand G.
Review of Nordic Total Merit Index, Full Report, November 2018. Nordisk Avlsvaerdi Vurdering.
https://www.nordicebv.info/wp-content/uploads/2018/11/2018.11.06-NTM-2018-report-Full.pdf
Date: 2018
Date accessed: June 4, 2018
2 Annual report,
VIT, 2017
.- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Annual report 2017.
https://www.vit.de/fileadmin/Wir-sind-vit/Jahresberichte/vit-JB2017-gesamt.pdf
Date: 2017
Date accessed: June 4, 2018
Table 3Mean values of reproduction traits assumed for the breeds Holstein (HOL), Angler (ANG), and Red and White Dual-Purpose (RDN)
| Trait | Unit | Breed | ||
|---|---|---|---|---|
| HOL | ANG | RDN | ||
| Calving interval | days | 411 | 400 | 390 |
| Replacement rate | % | 33.8 | 35.5 | 36.0 |
| Conception rate heifers | % | 49 | 51 | 53 |
| Insemination rate heifers | % | 44 | 46 | 48 |
| Conception rate cows | % | 42 | 44 | 46 |
| Insemination rate cows | % | 38 | 41 | 44 |
| Stillbirth risk | % | 5.1 | 4.8 | 4.8 |
| Calf mortality | % | 6.0 | 6.3 | 8.0 |
| VWP primiparous cows | days | 56 | 56 | 56 |
| VWP multiparous cows | days | 56 | 49 | 49 |
1 Stillbirth reflects the proportion of dead calves within 48 h postpartum as an average of both primiparous and multiparous cows. Calf mortality is the probability of a calf dying in the period from 3 to 458 d postpartum. VWP = voluntary waiting period.
2 Annual report,
VIT, 2017
.- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Annual report 2017.
https://www.vit.de/fileadmin/Wir-sind-vit/Jahresberichte/vit-JB2017-gesamt.pdf
Date: 2017
Date accessed: June 4, 2018
3 Annual report,
LKV SH (Landeskontrollverband Schleswig-Holstein), 2017
.- LKV SH (Landeskontrollverband Schleswig-Holstein)
State Control Association Schleswig-Holstein (Landeskontrollverband Schleswig-Holstein). Annual report 2017.
https://www.lkv-sh.de/images/Download/pdf/Jahresbericht_2017.pdf
Date: 2017
Date accessed: June 4, 2018
Table 4Levels of prices and costs (in Euro) used for simulations and the breeds Holstein (HOL), Angler (ANG), and Red and White Dual-Purpose (RDN)
| Item | Unit | HOL | ANG | RDN |
|---|---|---|---|---|
| Milk + livestock | ||||
| Energy-corrected milk | kg | 0.32 | 0.32 | 0.32 |
| Bonus for milk fat | kg | 2.33 | 2.33 | 2.33 |
| Bonus for milk protein | kg | 4.68 | 4.68 | 4.68 |
| Liveweight | kg | 1.60 | 1.60 | 1.90 |
| Destruction costs dead cow | piece | 50 | 50 | 50 |
| Springing heifer | piece | 1,400 | 1,500 | 1,500 |
| Nonpregnant heifer | piece | 700 | 750 | 750 |
| Bull calf | piece | 130 | 110 | 220 |
| Feed | ||||
| Total mixed ration | kg | 0.19 | 0.19 | 0.19 |
| Concentrate | kg | 0.25 | 0.25 | 0.25 |
| Roughage | kg | 0.16 | 0.16 | 0.16 |
| Milk replacer | kg | 2.20 | 2.20 | 2.20 |
| Veterinary costs | ||||
| Mastitis | treatment | 110 | 110 | 110 |
| Lameness | treatment | 80 | 80 | 80 |
| Ketosis | treatment | 50 | 50 | 50 |
| Milk fever | treatment | 80 | 80 | 80 |
| Metritis | treatment | 70 | 70 | 70 |
| Dystocia | treatment | 180 | 180 | 180 |
| Reproduction | ||||
| Semen, proven bull | portion | 16.00 | 14.50 | 12.60 |
| Insemination costs | treatment | 30 | 30 | 30 |
1 Average milk price in 2017 (
Chamber of Agriculture Schleswig-Holstein, 2017
).2 The bonus payment for additional milk fat/protein was calculated based on information from dairies in Schleswig-Holstein.
3 M. Leisen (Rinderzucht Schleswig-Holstein, Neumünster, Germany), personal communication.
4
German fee guideline of veterinarians, 2017
.Simulation Process
The EV of a trait was derived by investigating alterations in herd profit when changing the respective trait. For each breed, a default scenario was simulated based on the input parameters shown in Tables 1 to 4. Afterward, for each of the evaluated traits, 2 additional scenarios were simulated, representing the given trait at a lower and a higher level, respectively. The difference in the average annual net return of scenario low and scenario high illustrated the economic consequences when the given trait was changed in the dairy herd. All scenarios were simulated over a time horizon of 40 yr to establish the herd dynamics required for the quantification of economic effects. However, the results of yr 1 to 10 were rejected, as they were strongly influenced by the initial starting point of the simulation. Additionally, each scenario was replicated 1,000 times to ensure precise estimates.
Statistical Analysis
The simulation process described above takes into account correlations between traits, which allows the structural herd effects to be modeled and, thus, the associated economic effects to be included in the EV of a given trait. The structural herd effect comprises all economic consequences owing to a shift in the age structure of the dairy herd. For example, more cows in higher-producing age groups will increase the average milk yield of the herd but will also increase feeding costs. These economic effects purely result from structural interactions in the herd and are strictly separated from any changes in genetically determined traits. For each trait evaluated in the study, the economic consequences due to the structural herd effect were consistently assigned to its EV. In this way, the underestimation of the economic importance of a trait is avoided (
where NetReturnijkl is the average annual net return, resulting from ith simulated replicate (i = 1, …, 1,000) for jth simulated level (j = 1,2) of the trait xij; βa denotes the corresponding regression coefficient and, at the same time, represents the estimate of the EV of trait xij. As μ denotes the fixed intercept, βbk is the regression coefficient of the mediator variable mk, where each of the n mediator variables have to be considered in the regression. Because SimHerd provides stochastic simulation elements, x_diffij is included with its regression coefficient βc to account for independent random variation within the simulated risk level, and εijkl is the random residual error. This regression approach is explained in detail by
Wolfová and Wolf, 2013
). In general; however, the EV for each trait must be derived independently of other breeding goal traits. Because this is not possible in SimHerd, simulated data were corrected afterward to avoid double counting. The correlations between the traits can be understood as indirect pathways (referred to as mediator effects) from the trait of interest to the simulated outcome (i.e., the average annual net return). Fairchild and MacKinnon, 2009
proposed a multiple regression approach including mediator variables to eliminate such mediator effects. Østergaard et al., 2016
adopted this approach to ensure independent trait changes when deriving EV, and thus, to avoid double counting. For the inclusion of a trait in the regression as a mediator variable, 2 criteria had to be fulfilled: (1) the mediator variable had to be modeled as correlated with the trait of interest in SimHerd, and (2) the mediator variable also had to be a part of the breeding goal with an EV of its own. An overview of mediator variables used in the regression analyses of disease traits and dystocia is given in Table 5. As the occurrence of diseases affected milk production, milk yield was used as a mediator variable for all disorders. Lameness, ketosis, and metritis caused alterations in the reproductive performance of the cows. Therefore, fertility was applied as another mediator variable in the regression analyses of these disorders. Additionally, the EV of mastitis and lameness were corrected for changes in cow mortality. Milk fever was modeled as a predisposing disease for metritis, dystocia, and mastitis with corresponding effects on these traits. Therefore, the EV of milk fever was corrected using metritis, dystocia, and mastitis as mediator variables. Furthermore, ketosis was used as a mediator variable in the regression analysis of metritis. The occurrence of dystocia was simulated to be associated with both raised cow mortality and stillbirth. To avoid double counting, economic effects of changes in these traits have been removed from the EV of dystocia. The general regression model for the statistical analysis can be written as:
where NetReturnijkl is the average annual net return, resulting from ith simulated replicate (i = 1, …, 1,000) for jth simulated level (j = 1,2) of the trait xij; βa denotes the corresponding regression coefficient and, at the same time, represents the estimate of the EV of trait xij. As μ denotes the fixed intercept, βbk is the regression coefficient of the mediator variable mk, where each of the n mediator variables have to be considered in the regression. Because SimHerd provides stochastic simulation elements, x_diffij is included with its regression coefficient βc to account for independent random variation within the simulated risk level, and εijkl is the random residual error. This regression approach is explained in detail by
Østergaard et al., 2016
.Table 5Mediator variables used in the regression analyses for health traits and dystocia based on the simulated relationships in the SimHerd model
| Mediator variable | Disease trait | |||||
|---|---|---|---|---|---|---|
| Mastitis | Lameness | Ketosis | Milk fever | Metritis | Dystocia | |
| Milk yield | X | X | X | X | X | |
| Fertility | X | X | X | |||
| Cow mortality | X | X | X | |||
| Metritis | X | |||||
| Dystocia | X | |||||
| Mastitis | X | |||||
| Ketosis | X | |||||
| Stillbirth | X | |||||
The derived EV of a given trait was expressed as change in profit per cow-year when the mean value of that trait was increased by one marginal unit. To allow a comparison of the economic importance across traits, relative economic values (REV) were provided. These REV were calculated as the product of the absolute marginal EV and the genetic standard deviation and expressed in percentage of the sum of standardized EV over all traits. Because estimates of genetic standard deviations were not available for the local breeds ANG and RDN, those for HOL were used.
Description of Traits
Economic values were derived for 23 breeding traits grouped in 5 different trait categories.
Production Traits
Milk production traits included ECM yield (kg), fat yield (kg), and protein yield (kg). The EV for fat and protein were calculated based on average payments a farmer would receive from dairies in Schleswig-Holstein for milk with higher milk components. For beef production, the traits average daily gain and EUROP form score (where E = excellent, U = very good, R = good, O = fair, and P = poor carcass quality) were considered. The proportions of animals in the individual grading classes differed among breeds in the EUROP form score. For HOL and ANG, the majority of animals was assigned to the classes P and O. In contrast, RDN provided better meat quality and a higher proportion of animals was represented in quality classes O and R.
Conformation and Workability
Regarding conformation and workability, EV were calculated for the traits feet and legs and udder conformation as well as for milkability and temperament, respectively. These traits were assessed in monetary terms according to the additional working time caused by a deviation of 1 scoring point from their respective optimum. A detailed description of the exact procedure can be found in the Nordic Total Merit Index report (
Sørensen et al., 2018
) and - Sørensen L.P.
- Pedersen J.
- Kargo M.
- Nielsen U.S.
- Fikse F.
- Eriksson J.A.
- Pösö J.
- Stephansen R.S.
- Pedersen Aamand G.
Review of Nordic Total Merit Index, Full Report, November 2018. Nordisk Avlsvaerdi Vurdering.
https://www.nordicebv.info/wp-content/uploads/2018/11/2018.11.06-NTM-2018-report-Full.pdf
Date: 2018
Date accessed: June 4, 2018
Kargo et al., 2014
.Health Traits
Economic values were derived for the following diseases: mastitis associated with udder health, lameness in relation to the health of claws and legs, both ketosis and milk fever associated with metabolic health, and metritis representing a reproductive disorder in dairy cattle.
Calving Traits, Calf Survival and Cow Mortality
Calving traits were represented by dystocia and stillbirth. Dystocia was defined as the probability of a difficult calving with veterinary assistance. Stillbirth reflected the proportion of dead calves within 48h postpartum as an average of both primiparous and multiparous cows. For calf survival, the traits early and late calf mortality were considered, which indicated the probability of a calf dying in the period from 3 to 14 d postpartum and 189 to 458 d postpartum, respectively. Furthermore, EV for cow mortality was derived. The trait cow mortality was defined as the probability of a cow dying accidentally on the farm without it being the effect of a health or fertility problem. Thus, the EV of cow mortality only includes economic consequences that have not already been accounted for in the EV of direct health traits and fertility traits. In this study, cow mortality represents the total loss of a cow from the herd (i.e., there is no economic return from slaughtering).
Fertility
Regarding fertility, EV were calculated for conception rate and insemination rate of both cows and heifers. Conception rate was defined as the probability of a cow or heifer to become pregnant after insemination. Insemination rate was defined as the percentage of cows or heifers in which estrus was observed and subsequently inseminated.
Sensitivity Analysis
As in many other studies (
De Vries, 2006
; Nielsen et al., 2006
), sensitivity analyses were performed to assess the effects of alterations in economic circumstances on EV. In the present study, changes of ±10% and ±20% with respect to the original values of prices for milk and beef as well as the costs for feed and veterinary treatments were investigated. Each parameter was changed separately while the remaining parameters were kept constant.RESULTS AND DISCUSSION
In this study, EV were derived for breeding goal traits of 3 German dairy breeds. These EV can be considered optimal from an economic point of view for the breeds studied and the economic and biological assumptions that were made. A comparison of EV across studies might create certain problems due to the following aspects: (1) EV are breed specific, (2) EV are strongly influenced by assumptions about the economic and production circumstances, (3) EV are affected by the methodology used for computation, and (4) the definition of breeding traits can differ across studies (
Wolfová et al., 2007
). In the following, the results of the present study are discussed in the context of some general statements from the literature, with emphasis on functional traits. The computed marginal EV and REV for all traits and breeds are presented in Table 6.Table 6Marginal economic values (in € per change in trait unit and cow-year) and relative economic values (in %) for traits and studied breeds
| Traits (unit) | Marginal economic value | Relative economic value | ||||
|---|---|---|---|---|---|---|
| Breed | Breed | |||||
| HOL | ANG | RDN | HOL | ANG | RDN | |
| Production | ||||||
| ECM (kg) | 0.16 | 0.16 | 0.17 | 21.7 | 21.6 | 22.3 |
| Fat (kg) | 1.16 | 1.16 | 1.19 | 6.0 | 6.0 | 5.9 |
| Protein (kg) | 3.17 | 3.17 | 3.18 | 12.3 | 12.3 | 11.9 |
| ADG (kg/d) | 0.31 | 0.38 | 0.59 | — | — | 3.7 |
| EUROP form score (point) | 12.88 | 12.85 | 12.50 | — | — | 0.6 |
| Conformation | ||||||
| Feet and legs (point) | 15.97 | 15.97 | 15.97 | 1.1 | 1.1 | 1.1 |
| Udder (point) | 23.04 | 23.04 | 23.04 | 4.1 | 4.1 | 3.9 |
| Workability | ||||||
| Milkability (point) | 15.97 | 15.97 | 15.97 | 0.5 | 0.5 | 0.5 |
| Temperament (point) | 7.10 | 7.10 | 7.10 | 0.2 | 0.2 | 0.2 |
| Health | ||||||
| Mastitis (%) | −2.71 | −2.69 | −2.57 | 8.2 | 8.1 | 7.5 |
| Lameness (%) | −3.10 | −3.05 | −2.70 | 5.3 | 5.2 | 4.5 |
| Ketosis (%) | −1.96 | −1.87 | −1.67 | 1.3 | 1.2 | 1.0 |
| Milk fever (%) | −2.23 | −2.14 | −1.98 | 1.9 | 1.8 | 1.6 |
| Metritis (%) | −1.82 | −1.74 | −1.73 | 1.2 | 1.1 | 1.1 |
| Fertility | ||||||
| Conception rate heifers (%) | 1.51 | 1.30 | 0.84 | 2.6 | 2.2 | 1.4 |
| Conception rate cows (%) | 2.21 | 2.49 | 1.42 | 3.3 | 3.7 | 2.1 |
| Insemination rate heifers (%) | 1.15 | 0.96 | 0.73 | 2.0 | 1.7 | 1.3 |
| Insemination rate cows (%) | 1.70 | 2.12 | 1.15 | 2.6 | 3.2 | 1.7 |
| Calving difficulty, calf survival, cow mortality | ||||||
| Dystocia (%) | −3.41 | −3.50 | −3.97 | 4.3 | 4.5 | 5.0 |
| Stillbirth (%) | −2.14 | −1.92 | −2.59 | 2.6 | 2.1 | 2.7 |
| Early calf mortality (%) | −1.43 | −1.76 | −1.78 | 0.9 | 1.1 | 1.1 |
| Late calf mortality (%) | −3.78 | −3.50 | −5.03 | 2.4 | 2.3 | 3.1 |
| Cow mortality (%) | −14.39 | −14.90 | −15.20 | 15.5 | 16.0 | 15.8 |
1 Relative economic values were calculated as marginal economic values multiplied by the genetic standard deviation of each trait and expressed as percentage of the sum of all marginal economic values. HOL = Holstein; ANG = Angler; RDN = Red and White Dual-Purpose.
Economic Values for Production, Conformation, and Workability
The EV of ECM is equal for the breeds HOL and ANG (€0.16/kg) and slightly higher for RDN (€0.17/kg), reflecting lower feed costs for RDN owing to a better utilization of roughage. The same is observed for the EV of fat and protein. The relative emphasis of milk component traits is higher for protein (11.9–12.3%) and lower for fat (5.9–6.0%) in all breeds. Thus, the results indicate an economically optimal fat-to-protein ratio of 1:2, which matches the current weighting of milk ingredients in the total merit indices of the studied breeds (
VIT, 2020
). The EV of average daily gain is markedly higher for RDN (€0.59/kg) compared with HOL (€0.31/kg) and ANG (€0.38/kg). Because RDN offers better meat quality, an improvement in daily gain would result in higher economic returns. In contrast, the EV of EUROP form score is slightly lower for RDN than for the other breeds. This is caused by the fact that the majority of RDN bulls was assigned to higher carcass quality classes (O and R) compared with ANG and HOL (P and O). As the price difference to the next better quality class becomes smaller, an improvement in carcass quality causes a larger effect in HOL and ANG than in RDN. However, the breeding objectives of HOL and ANG do not currently include beef production traits. Accordingly, the relative importance of these traits is only given for RDN, which is 3.7% for average daily gain and 0.6% for EUROP form score. Summing up the relative importance of all production traits considered, the total emphasis for production traits is 40.0, 39.9, and 44.4% for HOL, ANG, and RDN, respectively. By comparison, the overall emphasis on production in the current total merit indices of the studied breeds is 45% for HOL and RDN and 40% for ANG (- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Beschreibung der Zuchtwertschätzung für Milchleistungsmerkmale, Zellzahl, Exterieurmerkmale, Nutzungsdauer und Zuchtleistungsmerkmale.
https://www.vit.de/fileadmin/DE/Zuchtwertschaetzung/Zws_Bes_deu.pdf
Date: 2020
Date accessed: February 7, 2020
VIT, 2020
).- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Beschreibung der Zuchtwertschätzung für Milchleistungsmerkmale, Zellzahl, Exterieurmerkmale, Nutzungsdauer und Zuchtleistungsmerkmale.
https://www.vit.de/fileadmin/DE/Zuchtwertschaetzung/Zws_Bes_deu.pdf
Date: 2020
Date accessed: February 7, 2020
For conformation traits, EV are found to the amount of €15.97 per scoring point for feet and legs and €23.04 per scoring point for udders. The relative importance of conformation is about 5% in all breeds. Currently, linear type traits take a relative emphasis of 15% for HOL and RDN, and 20% for ANG in the overall indices (
VIT, 2020
). Based on the results of this study, these weightings cannot be considered economically optimal. For workability traits, marginal EV of €15.97 per scoring point for milkability and €7.10 per scoring point for temperament are found, which corresponds to a relative weighting of 0.7% in all breeds.- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Beschreibung der Zuchtwertschätzung für Milchleistungsmerkmale, Zellzahl, Exterieurmerkmale, Nutzungsdauer und Zuchtleistungsmerkmale.
https://www.vit.de/fileadmin/DE/Zuchtwertschaetzung/Zws_Bes_deu.pdf
Date: 2020
Date accessed: February 7, 2020
Economic Values of Health Traits
In Holstein Friesian, more and more breeding goals comprise direct health traits, as improvements in dairy cattle health are unequivocally associated with higher economic net returns (
Miglior et al., 2017
). In Germany, direct health traits were officially introduced in the national genetic evaluation for HOL in 2019. However, these traits have not yet been implemented in the overall index. Therefore, we provide reliable estimates of EV (Table 6) for the diseases mastitis, lameness, ketosis, milk fever, and metritis. These estimates are expressed as the expected change in profit per cow-year when the mean value of the incidence rate is increased by 1 percentage point. Multiplying the declared EV by 100 corresponds to the average economic loss caused by 1 case of the respective disorder. In the results, the EV of all health traits are highest for HOL, slightly lower for ANG, and lowest for RDN on both the marginal and relative scales. Across the diseases, highest EV are seen for mastitis and lameness in all breeds. Mastitis is considered one of the most prevalent and costly diseases in dairy herds (Santos et al., 2004
; Halasa et al., 2007
; Koeck et al., 2012
). The present study reveals total economic costs of €257 to €271 per case of mastitis (Table 6). This includes direct costs such as veterinary treatment, economic losses owing to withdrawal milk, and additional labor costs. Moreover, the EV of mastitis takes into account the economic consequences attributable to the structural herd effect. In SimHerd, dependencies between the occurrence of mastitis and a reduced production performance are illustrated along with an increased cow mortality. To avoid double counting, economic effects caused by changes in the traits milk production and cow mortality, which are modeled as genetically correlated with mastitis, must not be considered in the EV of mastitis. Therefore, these correlated effects were removed from the EV of mastitis. The relative importance of mastitis is between 7.5 and 8.2% in the breeds studied. The EV for mastitis given in the literature differs notably. Krupová et al., 2016
and Wolfová et al., 2007
found costs to the amount of €70.65 and €64.19 per case of mastitis for a Slovakian dual-purpose breed and Czech Holstein cattle, respectively. In both studies, the milk price as well as the treatment costs were lower compared with the price levels assumed in our study. Furthermore, the average milk yield of these breeds was lower, resulting in fewer monetary losses due to withdrawal milk. In the study of Hietala et al., 2014
, costs of €389.40 per case of mastitis were found for Finnish Ayrshire. A markedly higher milk price as well as higher treatment costs were the main reasons for the greater economic importance compared with the findings of our study. Moreover, all studies mentioned above computed EV by partial derivation of profit functions, which makes it difficult to compare the results directly.Besides mastitis, lameness is a frequently appearing disease of cows in Germany (
VIT, 2017
). The present study reveals the highest EV for lameness compared with all other health traits (Table 6). However, the relative importance of lameness (4.5–5.3%) is below that of mastitis, owing to a lower genetic variation. Principally, the economic consequences of lameness result from veterinary treatment costs, discarded milk and additional labor time. Additionally, in the present study, financial effects due to the structural herd effect are allocated to the EV of lameness. It is well known that lameness is associated with reduced production performance (- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Annual report 2017.
https://www.vit.de/fileadmin/Wir-sind-vit/Jahresberichte/vit-JB2017-gesamt.pdf
Date: 2017
Date accessed: June 4, 2018
Amory et al., 2008
) and compromised fertility (Melendez et al., 2003
; Morris et al., 2011
). Thus, lame cows have a higher risk of leaving the herd prematurely (Booth et al., 2004
). However, the economic effects of lower production performance, impaired fertility as well as increased cow mortality must not be included in the EV of lameness, as otherwise double counting occurs (Wolfová et al., 2007
). To overcome this problem, these economic effects were removed from the marginal utility of lameness.In this study, we provide EV for milk fever and ketosis as disorders related to metabolic health in dairy cattle. Financial losses attributable to these diseases include the veterinary treatment and costs for farmers´ additional labor time as well as economic consequences due to the structural herd effect. The higher marginal EV of milk fever in all breeds compared with ketosis are mainly caused by higher costs for veterinary treatments. In SimHerd, milk fever has an impairing effect on milk production. Furthermore, milk fever is modeled as a predisposing disease for dystocia, mastitis, and ketosis. Thus, an increase in the incidence rate of milk fever causes changes in these traits. To avoid double counting, the EV of milk fever was corrected by associated changes in milk production and correlated diseases. The results presented are in accordance with findings from
Liang et al., 2017
, who calculated the costs per case of milk fever to range from €156 to €238 ($172 to $263). For ketosis, the economic effects of lowered milk production as well as impaired fertility were removed from its marginal utility to avoid double counting. The results for EV of ketosis in this study are between estimates reported by Gohary et al., 2016
and McArt et al., 2015
, which found average costs of €142 (CAN$203) and €268 (US$289) per case of ketosis in Canada and the US, respectively. In the present study, the overall REV for metabolic diseases is between 2.6 and 3.2% across breeds.For metritis, marginal EV range between €1.73 and €1.82 per increase in mean incidence rate by 1 percentage point and cow-year, corresponding to a relative importance of 1.1 to 1.2% across breeds (Table 6). The EV of metritis includes direct costs attributable to veterinarian treatment, additional labor time and withdrawal milk. Additionally, a change in metritis influences the herd demography, thus, resulting economic effects owing to the shift in herd structure must be assigned to the EV of metritis. As in the SimHerd model metritis is simulated to be genetically correlated with milk production, conception rate and ketosis, these traits were used as mediator variables to remove economic effects caused by changes in these correlated traits from the EV. In the literature, slightly lower costs of €147 (US$162) per case of metritis were found for Iranian Holstein cattle (
Mahnani et al., 2015
). In contrast, considerably higher average costs of €342 (US$377) per case of metritis were reported by Lima et al., 2019
. Summing up the REV of all health-related traits, relative weightings are 17.9, 17.4, and 15.7% in HOL, ANG, and RDN, respectively.Economic Values of Calving Traits, Calf Survival and Cow Mortality
For calving traits, EV and REV differ markedly among breeds (Table 6). Although the EV of both dystocia and stillbirth are highest in RDN, lower EV are found in HOL and ANG, respectively. This is mainly attributable to higher sale prices for heifers and bull calves in RDN. Furthermore, a higher beef price causes greater financial losses when dystocia or stillbirth occurs in RDN. To avoid double counting, the EV of dystocia was corrected for associated changes in cow mortality and stillbirth, as correlations between these traits were simulated in the bio-economic model. The marginal utility of dystocia, therefore, comprises costs associated with veterinary treatment and additional labor, as well as economic effects owing to the structural herd effect. For stillbirth, the lowest EV is found for ANG (€1.92 per cow-year given an increase in stillbirth rate by 1 percentage point) and highest EV is calculated for RDN (€2.59 per cow-year given an increase in stillbirth rate by 1 percentage point). This is caused by a higher price for calves, which is twice as high for RDN than for ANG. The REV of stillbirth is also highest for RDN (2.7%), slightly lower for HOL (2.6%) and lowest for ANG (2.1%). In total, the relative importance of both dystocia and stillbirth is between 6.6 and 7.7% across the breeds. This is much larger than the current weighting of these traits in the overall breeding objectives. Although the relative weighting of calving traits is presently 3% in the total merit indices of HOL and RDN, these traits are not included in the breeding objective of ANG (
VIT, 2020
). Due to the large economic effect of calving traits, the inclusion of these traits with an appropriate weighting in the breeding goals can be highly recommended.- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Beschreibung der Zuchtwertschätzung für Milchleistungsmerkmale, Zellzahl, Exterieurmerkmale, Nutzungsdauer und Zuchtleistungsmerkmale.
https://www.vit.de/fileadmin/DE/Zuchtwertschaetzung/Zws_Bes_deu.pdf
Date: 2020
Date accessed: February 7, 2020
The EV of early calf mortality hardly differ for RDN and ANG. In contrast, a lower EV is found for HOL caused by lower prices for both nonpregnant and springing heifers. The relative emphasis for early calf mortality ranges from 0.9% in HOL to 1.1% in ANG and RDN. For the trait late calf mortality, a markedly higher EV is determined for RDN compared with HOL and ANG. As the meat price is assumed to be considerably larger for RDN, one case of late calf mortality leads to greater financial losses in this breed. The REV of late calf mortality is between 2.3 and 3.1% across breeds. These results are in accordance with findings from
Hietala et al., 2014
, who reported a relative weighting of calf mortality up to 3.6% in the overall index for Finnish Ayrshire.For the trait cow mortality, highest marginal EV is calculated in RDN, whereas EV in HOL and ANG are slightly lower. Cow mortality was modeled as the on-farm death and, thus, as the total loss of a cow from the herd. Costs such as production losses and treatment costs that preceded the death of the cow are not considered. Economic effects included in the EV of cow mortality are lost profits as animals are not sold for slaughter, costs for replacement heifers and destruction costs for handling the carcass. Moreover, economic consequences resulting from alterations in the age structure of the herd are allocated to the EV of cow mortality. The present study indicates relative importance for cow mortality of 15.5%, 16.0% and 15.8% in HOL, ANG and RDN, respectively. Regarding the practical application of the calculated EV, it is important to mention that cow mortality is not yet explicitly recorded on German dairy farms. A separate recording of whether animals were culled or died on the farm would possibly allow the trait cow mortality to be processed by breeding in the future. In this study, we assume a strict definition of the trait cow mortality and its associated economic consequences. All economic effects resulting from a cow leaving the herd are either allocated to the EV of health and fertility traits by including the structural herd effects, or are captured in the EV of cow mortality. However, there might be some residual economic effects that have not been considered. For example, culling decisions may be influenced by management practices on a particular farm and the farmers personal preferences. This type of involuntary culling with associated economic effects is not considered in the present study, which represents a weak point. This residual part, however, is likely to be marginal.
In Germany, functional longevity is currently included with a relative weighting of 20% in the overall indices of the breeds studied (
VIT, 2020
). Functional longevity; that is, adjusted for voluntary culling due to low milk production, is presently seen as a proxy for healthy and fertile cows. The prospective inclusion of health traits in the total merit index allows these traits to be directly improved. It has to be noted that data recording for direct health traits has only been available in Germany for a few years. Nevertheless, a continuously growing amount of data will enable the estimation of more accurate breeding values and genetic parameters. As a consequence of including direct health traits in the overall index, emphasis will shift from functional longevity as a proxy toward direct measurements. Thus, the relative importance of longevity in the selection index will be reduced. However, even if the relative emphasis of longevity will decrease in the overall index, it is likely that longevity is genetically improved due to the direct selection on traits such as mastitis and lameness. It is of fundamental importance that this is appropriately communicated and explained to cattle breeders.- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Beschreibung der Zuchtwertschätzung für Milchleistungsmerkmale, Zellzahl, Exterieurmerkmale, Nutzungsdauer und Zuchtleistungsmerkmale.
https://www.vit.de/fileadmin/DE/Zuchtwertschaetzung/Zws_Bes_deu.pdf
Date: 2020
Date accessed: February 7, 2020
Economic Values of Reproduction Traits
In Germany, poor fertility is a common reason for the premature departure of cows from the herd (
VIT, 2017
). Compromised reproduction ability results in an increased culling rate so that older cows in higher parities are often replaced by heifers, which leads in total to a lower average milk performance of the dairy herd (- VIT
Vereinigte Informationssysteme Tierhaltung w.V. Annual report 2017.
https://www.vit.de/fileadmin/Wir-sind-vit/Jahresberichte/vit-JB2017-gesamt.pdf
Date: 2017
Date accessed: June 4, 2018
Roxström et al., 2001
). In addition, direct reproduction costs (e.g., semen, inseminations, veterinary treatments) are higher (Cabrera, 2014
). Therefore, the fertility of cows is a major determinant for financial returns from dairy farming.In the present study, EV of reproduction traits are expressed as the expected change in profit per cow-year when the mean value of the trait was increased by 1 percentage point. In general, EV and REV of both conception rate and insemination rate are markedly lower in RDN compared with HOL and ANG (Table 6). In the following, the reasons for this are discussed in detail based on the trait conception rate of cows. Economic revenues and losses associated with an increase in cow's conception rate by 1 percentage point are shown in Figure 2, Figure 2. These monetary changes differ among the breeds studied, thereby explaining the differences in the derived EV. In all breeds, economic revenues (Figure 2a) from milk sales, calf sales, and slaughtered heifers are barely affected by an increased conception rate. In contrast, large changes are found for returns from slaughtered cows and surplus breeding heifers sold. For slaughtered cows, negative economic returns (i.e., lost sales) are observed in all breeds as an improvement in the reproduction performance results in fewer cows being slaughtered. The lost sales from slaughtered cows are largest for RDN, because the beef price is higher compared with that for HOL and ANG. Economic revenues evolve in all breeds for surplus heifers sold, because fewer heifers are needed for replacement. These revenues hardly differ between the breeds. All stated revenues are summed up as total revenues, which are largest for HOL, slightly lower for ANG and smallest for RDN.
Figure 2(A) Economic revenues (in Euro) caused by an increase in cow conception rate of 1 percentage point for the breeds German Holstein (HOL), Angler (ANG) and Red and White Dual-Purpose (RDN). (B) Economic losses (in Euro) caused by an increase in cow conception rate of 1 percentage point for the dairy breeds German Holstein (HOL), Angler (ANG) and Red and White Dual-Purpose (RDN).
The economic costs associated with an increased conception rate (Figure 2b) are attributable to feed for cows and heifers, the insemination of cows and heifers and veterinary treatments. It is noticeable that an increased conception rate of cows results in the largest changes in all cost items for the breed HOL. In contrast, the smallest changes are observed for RDN, which can be explained by the law of diminishing returns. The HOL breed is characterized by a lower reproduction performance compared with RDN. Because of a nonlinear relationship between the reproductive performance of a dairy herd and the expected financial outcome, an increase in conception rate by one marginal unit will have larger effects in a herd with lower fertility performance (
Cabrera, 2014
). Feed costs are directly related to the changes in milk yield associated with a better reproduction performance. Therefore, the economic costs for cows' feed are seen as largest for HOL but slightly negative for ANG and RDN (Figure 2b). Regarding the costs for heifers' feed, the highest costs are observed for HOL and ANG. In comparison, the costs for RDN are negative, which means additional financial returns. Although an increase in conception rate results in more heifers in the herds HOL and ANG, the total number of heifers in the RDN herd decreases (Table 7), affecting the corresponding feed costs. Only minor changes in the insemination costs of heifers are observed, because the insemination ability of heifers is barely influenced by improved conception rate of cows (Table 7). In contrast, additional savings for insemination costs of cows are accrued, as fewer inseminations are required to achieve pregnancy (Table 7). The differences in monetary savings across the breeds are caused by different semen prices. Further, an improvement in the conception rate results in increased treatment costs for all breeds, which are largest for HOL. Improved fertility causes a shift in the age structure of the dairy herd so that the proportion of older cows becomes higher. Because older cows tend to be more susceptible to diseases, the overall costs of veterinary treatments increase. These changes in the demography of dairy cattle herds and the associated economic effects are referred to as structural herd effects, which are taken into account in deriving EV by means of the SimHerd model. Summing up all cost items, total costs are largest for HOL and markedly lower for ANG. For RDN, negative total costs are observed, which means additional revenues.Table 7Simulated herd structures for the default scenario (default) as well as for scenarios representing the trait conception rate of cows on a lower (low) and a higher (high) level
| Item | HOL | ANG | RDN | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Low | Default | High | Low | Default | High | Low | Default | High | |
| Total number of heifers | 101.65 | 101.99 | 102.29 | 122.48 | 122.81 | 122.84 | 131.06 | 130.96 | 130.84 |
| Calving interval | 413.95 | 411.88 | 410.07 | 402.56 | 400.81 | 399.14 | 391.22 | 389.67 | 388.25 |
| Inseminations per cow | 1.87 | 1.81 | 1.76 | 1.86 | 1.80 | 1.74 | 1.84 | 1.77 | 1.71 |
| Inseminations per heifer | 0.89 | 0.89 | 0.89 | 0.72 | 0.72 | 0.72 | 0.64 | 0.64 | 0.64 |
| Number of calves per cow | 1.028 | 1.032 | 1.034 | 1.076 | 1.078 | 1.079 | 1.112 | 1.112 | 1.111 |
| Total number of calves | 90.02 | 90.34 | 90.67 | 94.72 | 94.99 | 95.10 | 96.85 | 96.81 | 96.80 |
| Average number of dry days per cow | 37.26 | 38.07 | 38.76 | 39.73 | 40.50 | 41.18 | 47.00 | 47.83 | 48.58 |
1 HOL = Holstein; ANG = Angler; RDN = Red and White Dual-Purpose.
Improvements in reproductive performance are usually accompanied by increased economic net profits of the dairy herd. These results are consistent with findings of other studies (
Inchaisri et al., 2010
; Calsamiglia et al., 2018
). Moreover, Inchaisri et al., 2010
found that the economic gain achieved by improved fertility was larger for cows in higher lactation numbers compared with heifers. This is also reflected in the results of our study, because the derived EV of both insemination and conception rates are much higher for lactating cows than for heifers (Table 6).Sensitivity Analysis
In general, the sensitivity analysis confirms the robustness of the derived EV, as they are hardly affected by changes in levels for prices and costs. Due to the large amount of data, only the effects of an increase in milk price by 20% on the EV of reproduction traits are presented (Figure 3) and discussed in detail.
Figure 3Results of sensitivity analyses for reproduction traits. The economic changes caused by a 20% increase in milk price are shown. HOL = German Holstein; ANG = Angler; RDN = Red and White Dual-Purpose.
Alterations in the EV for reproduction traits caused by an increased milk price vary among the breeds as follows: (1) the EV of both cow and heifer fertility increase in HOL; (2) in ANG, the EV of cow fertility decreases, whereas the EV of heifer fertility increases; and (3) the EV for both cow and heifer fertility decrease in RDN. The simulations performed allow to investigate the effects of improved fertility within a dairy herd. Commonly, an increase in reproductive performance has various effects on the total amount of milk produced by the cows. First, the proportion of older cows in the herd increases, resulting in higher average milk yield per cow. Second, the average calving interval of a dairy herd is reduced by improved fertility (Table 7) and, thus, the distribution of lactation stages in the herd shifts. Therefore, the proportion of cows in early lactation is higher, which has a driving effect on the average milk yield too. On the other hand, there are mainly 2 aspects with negative effects on milk yield in the dairy herd owing to improved fertility. First, better reproduction performance results in fewer heifer calvings. Because the milk yield of a heifer in early lactation is higher compared with an older cow in later lactation (
Kuhn et al., 2007
), this effect has a downgrading effect on the average milk yield per cow. Second, better reproductive performance increases the average number of dry days per cow (Table 7), which lowers the average milk production per cow in the herd.The varying effects of a 20% increase in milk price on the EV of fertility traits for the breeds result from different weightings of the principles mentioned above. For HOL, the effects with a driving effect on milk yield predominate. More precisely, these effects have a greater effect because the average level of fertility is lower in the simulated HOL herd. In contrast, the RDN breed is generally characterized by good reproductive performance and, thus, improved fertility has downgrading effects on average milk yield per cow. Consequently, an increase in milk price results in a higher EV of reproduction traits for HOL, but in a lower EV for RDN.
Implications
This study provides reliable estimates of EV for different production and functional traits and 3 German dairy breeds. For HOL, the latest official derivation of EV in Germany was more than 10 years ago. A specific calculation of EV for the local breeds ANG and RDN has not been performed so far. The results of this study can be considered as a first step toward new selection indices, which are optimal from an economic point of view. However, to make use of the computed EV, more genetic parameters are required, such as genetic standard deviations and heritabilities of the respective traits. In this way, EV can be transformed into index weights and, subsequently, breeding goals can be redefined. In addition, further research is needed to examine the implications of the updated indices on the ranking of bulls. It will also be important to investigate the expected genetic response in all traits and for the breeds studied. However, this is beyond the scope of the present study.
CONCLUSIONS
With respect to the derivation of EV for functional traits, bio-economic simulation models represent a suitable method, as economic effects of structural interactions in the herd demography may be taken into account. However, an appropriate statistical method is needed when genetic correlations are considered in this type of model to avoid double counting. Ultimately, the derived economic weightings presented in this study indicate that the official breeding objectives of the breeds HOL, ANG, and RDN should be redefined. The calculated EV can provide a basis for this purpose.
ACKNOWLEDGMENTS
This work was financially supported by the German Federal Ministry of Food and Agriculture (BMEL; Bonn, Germany) through the Federal Office for Agriculture and Food (BLE), grant number 2817ERA11D. The project has received funding from the European Union Horizon 2020 Research and Innovation Program under grant agreement no. 696231 – ReDiverse (Biodiversity within and between European Red Dairy Breeds). The authors have not stated any conflicts of interest.
REFERENCES
- Breed choice and pricing system implications for farmers and political institutions from a Swiss cattle farm model.in: Dent J.B. McGregor M.J. Sibbald A.R. Livestock Farming Systems. Research, Development, Socio-Economics and the Land Manager. Wageningen Academic Publishers, Wageningen, the Netherlands1996: 253-258
- Associations between sole ulcer, white line disease and digital dermatitis and the milk yield of 1824 dairy cows on 30 dairy cow farms in England and Wales from February 2003–November 2004.Prev. Vet. Med. 2008; 83 (18031851): 381-391
- Economic breeding index for dairy cattle in Ireland. Irish Cattle Breeding Federation.Bandon, Cork, Ireland2005
- Effect of lameness on culling dairy cows.J. Dairy Sci. 2004; 87 (15545373): 4115-4122
- Derivation of economic weights from profit equations.Anim. Sci. 1985; 40: 175-179
- Economics of fertility in high-yielding dairy cows on confined TMR systems.Animal. 2014; 8 (24679357): 211-221
- A stochastic dynamic model of a dairy farm to evaluate the technical and economic performance under different scenarios.J. Dairy Sci. 2018; 101 (29803414): 7517-7530
- Rinderreport 2017/2018.
- Economic consequences of dairy crossbreeding in conventional and organic herd in Sweden.J. Dairy Sci. 2020; 103 (31733860): 514-528
- Resumption of ovarian cyclicity in post-partum beef and dairy cows.Reprod. Domest. Anim. 2008; 43 (19068029): 20-28
- Economic value of pregnancy in dairy cattle.J. Dairy Sci. 2006; 89 (16960063): 3876-3885
- Modelling the economic impact of three lameness causing diseases using herd and cow level evidence.Prev. Vet. Med. 2010; 95 (20371126): 64-73
- Effect of including genetic progress in milk yield on evaluating the use of sexed semen and other reproduction strategies in a dairy herd.Animal. 2011; 5 (22440464): 1887-1897
- A general model for testing mediation and moderation effects.Prev. Sci. 2009; 10 (19003535): 87-99
- Gebührenordnung für Tierärzte.
- Consensus and debate in the definition of breeding objectives.J. Dairy Sci. 1998; 81 (9777508): 6-18
- The costs of a case of subclinical ketosis in Canadian dairy herds.Can. Vet. J. 2016; 57 (27429460): 728-732
- Economic values in dairy cattle breeding, with special reference to functional traits. Report of an EAAP-working group.Livest. Prod. Sci. 1997; 49: 1-21
- Economic effects of bovine mastitis and mastitis management: A review.Vet. Q. 2007; 29 (17471788): 18-31
- The genetic basis for constructing selection indexes.Genetics. 1943; 28 (17247099): 476-490
- Economic values of production and functional traits, including residual feed intake, in Finnish milk production.J. Dairy Sci. 2014; 97 (24342692): 1092-1106
- Economic consequences of reproductive performance in dairy cattle.Theriogenology. 2010; 74 (20580069): 835-846
- Derivation of economic values for production traits in aquaculture species.Genet. Sel. Evol. 2017; 49 (28093062): 5
- Economic basis for the Nordic Total Merit Index.J. Dairy Sci. 2014; 97 (25306270): 7879-7888
- Short communication: Genetic parameters for mastitis and its predictors in Canadian Holsteins.J. Dairy Sci. 2012; 95 (23021760): 7363-7366
- Economic values for health and feed efficiency traits of dual-purpose cattle in marginal areas.J. Dairy Sci. 2016; 99 (26585480): 644-656
- Dry period length in US Jerseys: Characterization and effects on performance.J. Dairy Sci. 2007; 90 (17369250): 2069-2081
- Estimating US dairy clinical disease costs with a stochastic simulation model.J. Dairy Sci. 2017; 100 (28012631): 1472-1486
- Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.J. Dairy Sci. 2019; 102 (31178192): 7345-7358
- Ableitung der Wirtschaftlichkeitskoeffizienten und optimalen Indexgewichte des Gesamtzuchtwertes für die deutschen Milch- und Zweinutzungsrassen unter Berücksichtigung aktueller und erwarteter zukünftiger Rahmenbedingungen. PhD Diss.Georg-August-Universität, Göttingen, Germany2007
- State Control Association Schleswig-Holstein (Landeskontrollverband Schleswig-Holstein). Annual report 2017.https://www.lkv-sh.de/images/Download/pdf/Jahresbericht_2017.pdfDate: 2017Date accessed: June 4, 2018
- Consequences and economics of metritis in Iranian Holstein dairy farms.J. Dairy Sci. 2015; 98 (26117350): 6048-6057
- Hyperketonemia in early lactation dairy cattle: A deterministic estimate of component and total cost per case.J. Dairy Sci. 2015; 98 (25622874): 2043-2054
- The association between lameness, ovarian cysts and fertility in lactating dairy cows.Theriogenology. 2003; 59 (12517394): 927-937
- Zuchtzieldefinition und Indexselektion für die österreichische Rinderzucht. PhD Diss.Universität für Bodenkultur, Wien, Austria1997
- A 100-year review: Identification and genetic selection of economically important traits in dairy cattle.J. Dairy Sci. 2017; 100 (29153164): 10251-10271
- Selection indices in Holstein cattle of various countries.J. Dairy Sci. 2005; 88 (15738259): 1255-1263
- Influence of lameness on follicular growth, ovulation, reproductive hormone concentrations and estrus behavior in dairy cows.Theriogenology. 2011; 76 (21601262): 658-668
- Approaches to formulating practical breeding objectives for animal production systems.Acta Agric. Scand. A Anim. Sci. 2014; 64: 2-12
- A stochastic model for the derivation of economic values and their standard deviations for production and functional traits in dairy cattle.Acta Agric. Scand. A Anim. Sci. 2006; 56: 16-32
- A stochastic model simulating pathogen-specific mastitis control in a dairy herd.J. Dairy Sci. 2005; 88 (16291616): 4243-4257
- Avoiding double counting when deriving economic values through stochastic dairy herd simulation.Livest. Sci. 2016; 187: 114-124
- A stochastic model simulating milk fever in a dairy herd.Prev. Vet. Med. 2003; 58 (12706054): 125-143
- Evaluation of activity meters for estrus detection: A stochastic bioeconomic modeling approach.J. Dairy Sci. 2020; 103 (31733873): 492-506
- Genetic and environmental correlations among female fertility and milk production in different parities of Swedish Red and White dairy cattle.Acta Agric. Scand. A Anim. Sci. 2001; 51: 7-14
- Effect of timing of first clinical mastitis occurrence on lactational and reproductive performance of Holstein dairy cows.Anim. Reprod. Sci. 2004; 80 (15036513): 31-45
- Review of Nordic Total Merit Index, Full Report, November 2018. Nordisk Avlsvaerdi Vurdering.https://www.nordicebv.info/wp-content/uploads/2018/11/2018.11.06-NTM-2018-report-Full.pdfDate: 2018Date accessed: June 4, 2018
- Use of profit equations to determine relative economic value of dairy cattle herd life and production from field data.J. Dairy Sci. 1991; 74: 1101-1107
- Vereinigte Informationssysteme Tierhaltung w.V. Annual report 2017.https://www.vit.de/fileadmin/Wir-sind-vit/Jahresberichte/vit-JB2017-gesamt.pdfDate: 2017Date accessed: June 4, 2018
- Vereinigte Informationssysteme Tierhaltung w.V. Beschreibung der Zuchtwertschätzung für Milchleistungsmerkmale, Zellzahl, Exterieurmerkmale, Nutzungsdauer und Zuchtleistungsmerkmale.https://www.vit.de/fileadmin/DE/Zuchtwertschaetzung/Zws_Bes_deu.pdfDate: 2020Date accessed: February 7, 2020
- Strategies for defining traits when calculating economic values for livestock breeding: A review.Animal. 2013; 7 (23773679): 1401-1413
- Selection for profit in cattle: I. Economic weights for purebred dairy cattle in the Czech Republic.J. Dairy Sci. 2007; 90 (17430949): 2442-2455
Article Info
Publication History
Published online: December 23, 2020
Accepted:
October 13,
2020
Received:
September 24,
2019
Identification
Copyright
© 2020 American Dairy Science Association®.
User License
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
31061-4&id=gr1.jpg){kind=link}
31061-4&id=gr2.jpg){kind=link}
31061-4&id=gr3.jpg){kind=link}