Sire Selection to Maximize Profits

The Total Economic Value Index


Dr. E.B. "Ted" Burnside

Nova Scotia Agricultural College,
Truro, NS, B2N 5E3, Canada
Email: tburnside@cadmin.nsac.ns.ca

Staying in the Black

Canadian dairy farmers are very fortunate to have substantial sire progeny proving systems in Canada that are well supervised and executed. Canada also has excellent milk recording and conformation evaluation systems, along with effective and expanding farmer-recorded evaluations on utility traits, and very accurate sire and cow evaluation systems that are built on top of the accurate record systems. This has not happened by chance nor over a very short period of time. As one who has spent a lifetime working on this system, helping breeders derive more accuracy and profitability from their selection decisions, I am very pleased with what has evolved in Canada. There can be little doubt that we also have some of the very best germ plasm in the world when one considers profitability. What components in the dairy cow's performance contribute to profitability, and which are the most important?

All studies point to the supremacy of production as an indicator of profitability. I can well remember an era in Canada when sires were selected mainly on show ring winnings. However, those days are gone and, while a few breeders continue to use sires that will not at all improve production, in hopes of developing a "Royal Grand Champion," this aspect of sire selection has greatly reduced in recent times. It will, no doubt, continue to reduce as more and more dairy farmers recognize that commercial profitability is not at all related to show ring winnings.

There are two profitability indexes used in Canada and while they have had quite different histories in development, they rank sires similarly, although there is considerable re-ranking. Both indices are based primarily on production and conformation ratings of dairy sires. The production component of the index in each case has many common features. The same traits are involved with heaviest emphasis on protein yield followed by fat yield which, in essence, places very high weight on volume of milk with positive emphasis on percentage of protein, and to a much less extent, percentage of fat. Selecting on an index that combines protein yield with fat yield does place very high emphasis on total volume of milk and positive emphasis on the component percentages. The weights have shifted steadily to heavier and heavier emphasis on protein over the last five to ten years and this is not surprising as marketing shifts away from fat yield and towards protein yield. Clearly, there are some little differences in milk market values for protein and fat across the country and around the world. However, dairy sires that have very high milk proofs combined with reasonable component percentages will do very well indeed. So the production index is one that puts high emphasis on protein yield (kg) and medium or lower emphasis on fat yield (kg).

Profitability Indexes Have Very Different Backgrounds

Two profitability indexes are used in Canada. The so-called "LPI" or "Lifetime Profitability Index" was the brainchild of AI and purebred leaders and based not on science, but on "seat-of-the-pants waiting for conformation and production". Weights on production and conformation have shifted from time to time over the last decade and traits that are included in the index for conformation and production have also changed. Nevertheless, it is quite consistent in giving highest emphasis to production traits at roughly two-thirds of the total and heavy emphasis on the udder, feet, and legs, and final classification for the other third. This index ranks sires effectively, and I sometimes call it the index for senior breeders. By that I mean that the emphasis here will lead to very high rankings for bulls that are popular because of high combinations of production and conformation. This index ranks dairy sires about the way many leading breeders would assume they should rank, sorting out bulls with very high combinations of protein and fat yield that also have extremely high ratings for conformation, chiefly the udder and final classification, as well as feet and legs.

Table 1 shows the formula for the LPI as well as the formula for the second index, Total Economic Evaluation (TEV). The LPI is very similar to an index published in the U.S.A. called the Total Production Index (TPI). The TPI is an arbitrary index that has been established by breed associations and AI people in the U.S.A., and it is very similar to the LPI in its weights on conformation vs. production. In fact, the emphasis in recent times on production has been slightly lower in the TPI than in the LPI. This is surprising when one considers that Canadian breeders and dairy farmers, generally speaking, put more emphasis on conformation than do U.S. breeders.

The TEV index was developed by researchers at the University of Guelph under the leadership of Dr. Jack Dekkers, who has made a solid contribution to livestock improvement in Canada. Dekkers developed this index over a period of years, building on earlier research that had attempted to rank dairy cows on profitability using production records and other pieces of information such as conformation. Much of the earlier work was done by graduate students, under my direction, in collaboration with economists in the Department of Agricultural Economics at the University of Guelph, and with Dr. Jim Wilton. Dr. Gerald Jensen, Dr. Poloa Rozzi, and a masters degree student, Helen Wilson, all made contributions towards profitability evaluation in their graduate studies. Helen Wilson's work utilized an economic method described as dynamic modelling to evaluate the profitability of cows of different sizes, weights, and production levels, and this work led to additional studies at the University for Agriculture in the Netherlands under the leadership of Dr. Johan Van Arrendonk, with Dekkers work built on Van Arrendonk's study.

Table 1. Official Canadian Indexes for sires.


LPI = [6 (9P + 2F) +4 (5MS + 4F&L + Conf + Cap)] X 7

SD SD SD SD SD SD

TEV = 26 (10 Production + 4 HL - 1.5 SCS)

Where:

Production = [ 9 (Protein - 13 ) + 2 ( Fat - 15 )]

SD SD

11

Herd Life = ( Herd Life - 3.0 )

SD

Somatic Cell Score = ( SCS - 3.0 )

SD

SD = Standard Deviation

SD Protein = 22 kg

SD Fat = 30 kg

SD Type Traits = 5.2

SD Herd Life = 0.24 Lactations

SD Somatic Cell Score = 0.24


How do we rank dairy sires and cows on profitability? Well obviously, one must develop some kind of profit equation for the dairy cow, and then attempt to relate sire proofs to this profitability equation. This was the methodology Dr. Dekkers used. It involves development of profit equations for the dairy cows and then mathematical manipulation of these equations to derive the appropriate weights for different traits that contribute in a major way to profitability.

How might we describe profitability in the dairy cow? It might involve the following traits: Profits = Milk Sold + Beef Sold and Value of Calves - Feed Costs - Labour - Disease Costs - AI and Semen Costs - Other Variable Costs (e.g., hoof trimming). This must also reflect the cow's lifetime in the herd, which varies. Many profit equations have been studied for the dairy cow and results are similar. The longer the cow lives and produces at a profitable level, the more money she will make the dairy farmer.

Dekkers AI sire profitability equation is based on the value of milk and components with consideration of herd life and an udder health index. While this is not an all-inclusive index, we can rationalize the exclusion of beef sales, which have very little impact on profitability in a North American context. As well, costs such as hoof trimming will vary relatively little from cow to cow and do not affect overall profitability.

Finally, we must recognize that the biggest blank in the equation is the cost of feed. Extensive research has been carried out on the relationship between feed intake and milk production. A recent study of Canadian data collected under the DHAS system by Dr. Robert Moore of Quebec, who finished his studies at the University of Guelph, indicates that feed intake is very closely related to production of milk. This relationship is automatically enhanced because of the method of measuring feed intake under field conditions in the DHAS system. Generally, dairy farmers write down how much feed they offer to each cow and do not consider the amount of feed that is refused. Generally, as well, we would appreciate that the cow with the greatest milk production will get the most feed. Therefore, we have what is termed an auto-correlation between feed intake and production, and it was not at all surprising that Moore (2) did find very high relationships between these two traits. Accurate measurement of actual feed intake, and therefore efficiency of utilization of feed under field conditions, is still not a reality. It will not become a reality unless we can come up with a mechanism for measuring feed intake very readily under field conditions. Research has not yet turned in this direction, but this could constitute a breakthrough as it would be feasible for farmers to actually spot the cows that are most profitable from the standpoint of feed utilization in their herds. In addition, these field data could be collected just as we collect milk yield records, and also centralized and analyzed to result in feed efficiency ratings for dairy sires based on their daughters.

Now let's briefly treat the various components in the total economic value index. Firstly, the value of product is based on the market value of milk fat and protein, as stated above, and varies depending on yield of protein, fat, volume of milk, and percentages of fat and protein. In general, the greatest weight is attached to the yield of protein as indicated in the equation for total economic value. The herd life index is greatly affected by the number of daughters of a sire that survive for first, second, third, and later lactations. This is derived directly from data available on sire evaluation, and for sires with large numbers of daughters that have had the potential to be around for three or more lactations, the evaluations are reasonably accurate. They would be enhanced in accuracy if first progeny tests were based on 100 or more daughters as we demonstrated in a recently reported study based on an extensive simulation model developed by Dr. Dekkers in his own Ph.D. studies. Gordon Vander Voort, a student currently completing Ph.D. studies at Guelph, remodelled this simulation to Canadian conditions. His M.Sc. clearly demonstrated that dairy sires will be very accurately ranked in their first evaluations if these are based on 100 or more daughters (3). This is particularly true for the proofs of sires for traits such as herd life and udder health indexes, which are relatively low in heritability.

Evaluations for herd life have been centralized around an average of three lactations per sire progeny, which is an appropriate number for the Canadian scene as evidenced by research at Guelph. Data indicated variation between sire progeny groups in their ability to withstand the rigours of high production.

The udder health index provides an evaluation of the resistance of the sire progeny group (i.e., daughters of a particular sire) to mastitis and udder infection. This is based on two pieces of information: the sire's proof for somatic cell score, which is a direct reflection of the number of somatic cells in the cow's udder and a good indicator of disease resistance, plus an evaluation of the sire's daughters for udder depth. With deep udders, the sire's progeny will undoubtedly have more udder and teat injury. The first study, and probably one of the most thorough, in this area was done using the North Carolina State Institutional dairy herds by myself and my colleague, Ben McDaniel, under the leadership of Dr. Ed Legates, some 36 years ago when we were finishing our Ph.D. studies (1). The data were obtained by measuring actual depth of udders from the base of the rear teat to the floor. They indicated the cow's udder approaches the ground at a rate of at least one inch per lactation, and also showed that deeper udders are more prone to injury and to mastitis. The data on sire progeny for somatic cell scores and for udder depth give greater weight to the udder depth proof in the early days of the sire's progeny groups, but as more data are available on somatic cell scores the weights shift a little towards somatic cell information.

Finally, these data are pooled to create an estimate of the value of returns over variable costs for an average daughter of a particular sire. This approach enables dairy farmers to look at net returns of a sire progeny group and to make financial comparisons between sires. For example, if an average daughter of Sire A gives greater cash flow than for Sire B, and this flow is projected over a ten year period for the daughter and all of her progeny, and discounted to present value at an appropriate rate, then of course we are very close to predicting the true economic value of that sire. Table 2 illustrates the advantages and disadvantages of this particular approach. As we can see, Sire A has a TEV of +$500, Sire B +$100. If we take six doses of semen from each of these two sires to obtain one first lactation daughter, which is a rather high estimate (nevertheless equal for the two sires), it will result in a distinct advantage for Sire A, whereas Sire B whose semen price is $10 lower than that of Sire A, will yield a net loss because his daughters do not return as much for production, or are more prone to mastitis, or will be shorter lived.

Now, let's look briefly at the correlations between TEV and LPI. As shown is Table 3, LPI has a very high relationship to TEV and other traits. As we can see, from the very top correlation, the two indexes are fairly closely related at a correlation of .97, where a perfect correlation or relationship would be 1.0. Nevertheless, sires will re-rank quite substantially at this level of correlation. The big advantage of TEV is that it does provide the bright, young dairy farmer who is paying off a mortgage, a tool for selecting most profitable sires at a very reasonable rate. To illustrate, the TEV can help you to save money and make money in the sire selection phase of your dairy operation. The LPI is really most appropriate for use by farmers who have their mortgages paid down and who enjoy a little extra emphasis on conformation, perhaps take a few cows out to the local show and derive some pleasure from that sort of activity, as well as some extra profitability from sale of breeding stock that have desirable conformation. For the strictly commercial operation run by a young family who are seeking to pay off a mortgage, I would definitely choose the TEV index.

Table 2. Using TEV & semen price in sire selection.
Semen Price 6 Doses* TEV Diff.
Sire A $50 $300 +$500 +$200
Sire B $40 $240 +$100 -$140

* 6 Doses = One 1st Lactation Daughter.

Table 3. Correlations for TEV and LPI.
TEV LPI
LPI 0.97
Protein Yield 0.94 0.92
HL-COMB 0.31 0.28
HL-DIR 0.15 0.05
SCS -0.15 -0.06
Final Score 0.35 0.47
Capacity 0.17 0.29
Rump 0.13 0.19
Feet & Legs 0.13 0.19
Mammary System 0.34 0.43
Udder Depth -0.17 -0.15

Table 4. Top 50 LPI Holstein bulls in Canada.
www Milk Fat Protein Conf LPI TEV
1

2

3

4

5

6

7

7

8

9

10

11

12

12

13

14

15

16

17

18

STARTMORE RUDOLPH

MAUGHLIN STORM

OLIVEHOLME AEROLINE

MEADOW BRIDGE MEGABUCK

MADAWASKA AEROSTAR

GRANDUC REPUTATION

KARNVILLA RATIO

LEHOUX BOLDMAN

SHOREMAR MASON

MALOYA LEO

HALDREY LEADERSHIP

WAUREGAN JOCKEY-ET

ROYCEDALE ACADEMY

GRASSHILL SUPERSTAR

GILLETTE CARLTON

LYNLAND CHIEF MAN

MEADOW BRIDGE FORMULA

DISELLEN DESIGN

STARTMORE LEADMASTER

LA PRESENTATION AERO

MAYERLANE VISTA-ET

MEADOW BRIDGE MILLION

COMESTAR LAST CALL

STARTMORE

2208

1923

2189

2392

2146

1456

2168

2364

1863

1453

1070

1704

963

2287

1861

2156

1752

1104

1285

2224

21

61

76

63

53

49

72

61

37

33

59

59

75

67

55

51

80

49

51

54

57

76

25

65

41

6

49

17

55

33

43

47

84

32

58

11

76

59

69

78

71

64

70

78

64

44

49

66

62

84

46

75

69

58

50

65

68

58

61

54

61

49

47

54

74

62

56

64

46

61

65

15

16

11

8

5

7

5

6

13

16

13

5

6

0

16

5

6

9

12

7

4

8

8

8

9

10

15

7

0

7

7

2

10

3

3

2160

1959

1933

1876

1831

1759

1756

1756

1730

1707

1659

1656

1651

1651

1644

1636

1630

1624

1613

1610

15

835

715

731

710

641

630

585

754

546

515

492

652

623

685

505

776

589

584

572

581

716

501

657

512

554

596

435

606

636

569

521

575

502

544

683

Table 5. Top 50 TEV Holstein bulls in Canada.
www Milk Fat Protein Conf LPI TEV
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

STARTMORE RUDOLPH

LYNLAND CHIEF MAN

LEHOUX BOLDMAN

OLIVEHOLME AEROLINE

MAYERLANE VISTA-ET

MAUGHLIN STORM

MEADOW BRIDGE MEGABUCK

GRASSHILL SUPERSTAR

DELABERGE ALCAN

COMESTAR LAST CALL

WAUREGAN JOCKEY-ET

MADAWASKA AEROSTAR

JAWOOD MADISON-ET

GRANDUC REPUTATION

ROYCEDALE ACADEMY

JEWETT MIDAS

QUIETCOVE-RAR PYRAMID-ET

WANIA ALTA CHARMING

BEAUCOISE BLACK KING

MEADOW BRIDGE FORMULA

KARNVILLA RATIO

DIXELLEN DESIGN

LA PRESENTATION AERO

2208

2156

2364

2189

2198

1923

2392

2287

1928

1878

1704

2146

2561

1456

963

1473

1699

1674

1500

1752

21

61

80

37

63

76

76

53

55

115

65

75

49

33

72

67

55

48

73

49

49

61

51

57

83

84

54

43

38

81

6

82

33

41

58

76

75

78

69

68

59

78

84

65

61

66

71

74

64

62

54

60

65

49

69

70

58

65

55

64

50

62

61

53

61

56

64

50

61

56

15

5

6

11

4

16

8

0

3

8

5

5

0

7

6

7

7

-1

10

6

5

9

7

1

2

12

7

5

6

9

6

13

8

3

6

2160

1636

1756

1933

1575

1959

1876

1651

1450

1521

1656

1831

1477

1759

1651

1480

1391

1370

1501

1630

17

835

776

754

731

716

715

710

685

683

657

652

641

636

630

623

606

599

598

596

589

585

584

581

580

575

572

569

565

560

554

547

546

546

544

541

Looking at the relationships between the two indexes and traits such as protein yield or herd life, either based on both conformation and herd life figures or the direct herd life figures that pertain to older progeny groups, you will see a much higher relationship between total economic value and herd life, a much higher relationship between somatic cell scores and herd life, and a lower relationship with conformation, with the exception of udder depth. So, the TEV index focuses on herd life, resistance to mastitis, and traits such as udder depth, giving more emphasis to that than the LPI. The LPI, on the other hand, puts relatively low emphasis on herd life, somatic cell scores, and udder depth, and relatively high emphasis on final score for conformation capacity, rump, feet, legs, and mammary system.

These are two excellent indexes for use. They are available on every sire, and the new ratings clearly show those sires that have the greatest potential for achieving profitable dairying under the two conditions that I have outlined. Tables 4 and 5 set out the top 50 Holstein sires for each index.

I wish you luck in your sire choices in 1997 and beyond.

References

  1. Burnside, E.B., B.T. McDaniel and J. E. Legates. 1963. The relationships among udder height, age, and milk production. J. Dairy Sci. 46:157-158.
  2. Moore, R.K. 1991. Parameter Estimates Between Production and Management Traits in First Lactation Using Milk Recording Data. Ph.D. Thesis. University of Guelph.
  3. Vander Voort, G.E. 1995. Economic Evaluation of Progeny Testing Programs for AI Firms Operating in a Competitive Market. M.Sc. Thesis. University of Guelph.