Show me the money – the case for genetic selection
John Young
Farming Systems Analysis Service, Kojonup
Email: john@farmingsystems.com.au
This article discusses the profitability of using genetic selection to improve the productivity of animals. Selecting animals that have superior production is relatively cheap and can result in permanent improvements in production and profitability. This will be illustrated through the improvements achieved in the South Australian Selection Demonstration Flocks, which demonstrate the profitability of alternative selection strategies.
These results are then compared with the observation from on-farm benchmarking which suggested there is no consistent observable trend that genetic selection is associated with increased profitability.
The assumptions made by economists and geneticists when calculating the increase in profit from genetic selection are examined, whilst the differences between production per head and production per hectare are discussed as possible explanations for this disconnect. Recent research into ‘new’ traits would appear to help to bridge this gap and is discussed in this article.
The South Australian Selection Demonstration Flocks compared the genetic gain achieved using one of three selection methods: measured performance recording utilising quantitative genetics; professional classer appraisal using visual and tactile appraisal; and an elite wool flock.
Each selection method achieved genetic gain relative to the randomly mated control (Table 1). These results demonstrate that selection of sheep can lead to improvements in the productivity of the animals.
Table 1 Production of the South Australian Selection Demonstration Flocks 2004 drop hoggets. Control values are expressed as the actual performance and the three selection lines are expressed as their values relative to the control
|
| Control | Measured preformance | Classer appraisal | Elite wool |
| Clean fleece weight (kg) | 3.55 | +0.17 | +0.16 | +0.14 |
| Fibre diameter (µm) | 20.8 | -2.6 | -1.7 | -2.0 |
| Staple strength (N/kTex) | 34.7 | -1.4 | -1.6 | +2.4 |
| Liveweight (kg) | 47.1 | +1.3 | +2.5 | +2.7 |
Table 2 Gross margins of each flock, expressed in $/DSE and $/ha, based on 10 year average prices
| Flock | GM/DSE ($) | GM/ha ($) |
| Control | 18.6 | 144 |
| Measured preformance | 24.2 | 196 |
| Classer appraisal | 21.4 | 166 |
| Elite wool | 23.1 | 181 |
The improvement was greatest for measured performance flock and resulted in a 36% increase in gross margin per hectare compared with the randomly mated control flock. This is a substantial improvement in estimated profitability and was achieved from identifying outside sires for two matings and then a further five years of selection within the flock.
In contrast to the above results, consultants who carry out on-farm benchmarking report that they do not observe consistent differences between profitability between genotypes, whereas differences of 30-40% between genotypes should be easily observed. This raises the question as to why the differences are not being detected. Is the effect of genotype being overridden by the effect of management, or is there a problem with the practical implementation of breeding programs associated with which traits increase profitability in extensive animal production systems? The remainder of the article examines the second issue.
Current genetic selection is based on improving production per head whereas profitability is more closely associated with production per hectare. The difference between the two is associated with the number of animals that can be carried per hectare.
In carrying out economic analysis of different genotypes a range of assumptions have to be made regarding carrying capacity achievable for each genotype. The assumptions made are similar for most genotype evaluations are:
-
the feed requirement of animals is proportional to their metabolic liveweight and the foraging ability of animals is similar. This means fewer large animals can be carried than smaller animals and that the level of pasture utilisation is similar regardless of genotype.
-
increases in clean fleece weight are achieved without increasing the energy requirement of the animal.
The assumptions may not always be correct, however, the underlying biology is difficult to quantify for more accurate economic analysis.
Sensitivity analysis carried out to determine the importance of these assumptions has shown that varying them within a realistic range has a major effect on the profitability of using different genotypes (Table 3). Further analyses showed that developing a selection index from the values calculated within this range of assumptions would lead to very different breeding directions.
Table 3 Whole farm profit ($) for different pasture and animal production systems based on a standard genotype and changes to profit for more resilient genotypes achieved through increased capacity to consume low quality feed or lower energy requirements for maintenance
| Profit ($) | Wool enterprise - Poor pasture | Wool enterprise - Good pasture | Prime lamb enterprise - Poor pasture | Prime lamb enterprise - Good pasture |
| Standard genotype ($) | 38000 | 92000 | -22000 | 165000 |
| Higher intake of low quality feed ($) | +8800 | +700 | +77000 | +17000 |
| Reduced maintenance requirements ($) | +20500 | +11000 | +39500 | +23000 |
It is commonly observed that the number of animals that farmers carry is associated with the number of animals that can be carried in a poor season.
Recent work has shown that there are marked differences between genotypes in their resilience when faced with feed restrictions. Much of the work has revolved around differences in genetic fat and muscle. Animals with higher genetic fat and muscle are more resilient and are able to maintain production when feed supply is short.
While studies suggest that resilience may be a very important trait, so far it has not be widely used for selection.
A recent analysis compared the estimated profitability of a genotype selected for high fleece weight with a genotype selected for improved resilience under two different nutritional scenarios.
-
Maintaining ewes at condition score three from joining, day 90 and through to lambing (CS 3-3-3).
-
Allowing half a condition score loss during pregnancy from joining in condition score three to lambing in condition score 2.5 (CS 3-2.7-2.5).
The analysis showed the more resilient genotype was $100 000 per annum more profitable when animals were fed to maintain condition and this increased to $250 000 per annum if there was a nutritional challenge and animals were losing 0.5 of a condition score during pregnancy.
This analysis varied the production assumptions outlined above based on some experimental and anecdotal evidence and therefore is not proof that a more resilient genotype is more profitable. However, it suggests that this area needs further work to decide if it is part of the reason for the divergence between theoretical and experimental gains from genetic selection and the on-farm benchmarking observations.
In conclusion, when evaluating the benefits of genetic selection we need to be clear on what is determining the profitability of an enterprise. It may not be production parameters such as clean fleece weight and growth rate as these higher production genotypes require more management to look after those sheep. If stocking rate and the scale of the sheep enterprise is limited by labour requirements and feed costs, then a genotype that reduces these limitations may be more profitable.