Beef Magazine is part of the divisionName Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

EPDS & Economics Determining the Relative Importance of Traits

With all the Expected Progeny Differences (EPDs) available to commercial producers today, sire selection should be simple. All a producer needs to do is find bulls with EPDs that fit the right "profile" for his operation.That sounds easy enough. The difficulty, of course, comes in determining what that profile should be. Logically, the sire with the perfect set of EPDs is the one that will maximize

With all the Expected Progeny Differences (EPDs) available to commercial producers today, sire selection should be simple. All a producer needs to do is find bulls with EPDs that fit the right "profile" for his operation.

That sounds easy enough. The difficulty, of course, comes in determining what that profile should be. Logically, the sire with the perfect set of EPDs is the one that will maximize net returns over the short and long run without incurring unacceptable risk or compromising the sustainability of the cattle operation.

Note that "maximizing returns" and "incurring risk" are economic terms. Sire selection is not, therefore, just a problem in genetics. It more properly belongs in that area where genetics and economics meet. To select sires correctly means to determine the relative economic importance of traits and weight the EPDs of sires accordingly.

The One-Size-Fits-All Fallacy When you don't know the answer to a question, a sensible tactic is to borrow an answer from someone who does. In the context of sire selection, this means identifying people who are knowledgeable and choosing the same kind of bulls they do.

There are two problems with this approach:

* Your sources may actually have no more insight than you;

* And, even if they do a good job of selection, the sires they choose for their herds may not be appropriate for yours.

In sire selection, one size does not fit all. The right kind of cow (and therefore the right kind of bull to sire that cow) can only be determined in the context of specific environments and management practices.

Two operations can be identical in terms of environment and management, but if the cowherds are genetically different, the kinds of sires best suited to each operation may differ. This happens when optimal performance in one or more important traits is intermediate - in other words, not at either extreme.

Consider milk production, for example. The cows in one herd may have too little milk and should therefore be bred to bulls with higher milk EPDs. The neighbor's cows, on the other hand, may have too much milk and should be bred to bulls with lower milk EPDs.

A similar case can be made for marbling. In today's marketing climate, it seems unlikely a breeder would deliberately look for bulls with lower marbling EPDs. Yet, there's little question that in a herd that already produces well-marbled carcasses, selecting for even higher levels of marbling will not be productive. The relative importance of many traits depends on current levels of performance.

A Complex Question The classic method for selecting optimally for more than one trait is the economic selection index. The starting point for creating an economic index is an equation called a breeding objective. Breeding objectives are typically of the form:


H is a measure of overall genetic merit of the animal, the BV's represent breeding values for different traits of economic importance (i.e., EPDs), and the v's are economic weights associated with each breeding value.

An economic weight is defined as the value (in dollar terms) of an independent, one-unit increase in performance in a trait. Calculating economic weights is then a mathematical method for determining the relative importance of traits.

Finding values for economic weights is rarely straightforward. Suppose, for example, that age at puberty is a trait in our breeding objective. We want to calculate the value of a one-day increase in age at puberty in our herd (independent of changes in performance for all other traits in the breeding objective).

Increases in age at puberty are normally undesirable, so we expect this economic weight to be negative. But how negative? To determine that, we must quantify how a change in age at puberty affects production (output) and costs (input).

Age at puberty influences production through its effect on heifer pregnancy rates. Higher pregnancy rates mean fewer heifers saved for replacement and more designated for sale as young animals - in other words, a change in product mix.

Age at puberty affects costs, too. Fewer replacements to develop means less feed consumed. And, if heifers are genetically more fertile, we can conceivably get away with less feed per heifer without a drop in pregnancy rate.

Logical as this thinking may be, it does not answer the question posed. It does not pinpoint the value of a one-day change in age at puberty. To do that, we must know the mathematical relationship between age at puberty and heifer pregnancy rate.

Unfortunately, this relationship is not constant - it changes with environment and management. When heifers are developed slowly and bred as yearlings, the relationship is likely to be strong, but when heifers are developed rapidly or bred at later ages so that all heifers are cycling well before the breeding season, it may be nonexistent. Differences in feed costs and relative prices of feedlot heifers and culled replacements can similarly complicate the calculation of economic weights.

And, to make matters worse, we do not currently have genetic predictions for age at puberty. Instead, we have predictions for a related trait - scrotal circumference.

With scrotal circumference, we are a step further removed from pregnancy rate. If scrotal circumference were a trait in our breeding objective, we would need to know the mathematical relationshipbetween scrotal circumference and pregnancy rate, and then calculate the economic value of a one-centimeter increase in scrotal circumference - no mean feat.

Technology To The Rescue? Determining the relative importance of the traits for which we have EPDs would be less difficult if the traits had direct economic impact. An EPD for heifer pregnancy rate, for example, would be easier to work with than an EPD for scrotal circumference.

Similarly, EPDs for death loss at calving and labor required to deal with dystocia would be better than EPDs for birth weight and calving difficulty. An EPD for cow maintenance requirement would be better than an EPD for mature weight.

With EPDs for economically relevant traits, we could more easily put traits in perspective. Researchers are developing EPDs for these traits, but even if they succeed in producing EPDs for every trait with direct economic significance, weighting the traits will be no trivial matter. We must still account for the complicating effects of management, environment and genetics of the existing cowherd.

The technology with the most promise for weighting traits is bioeconomic simulation. Simulation models have the potential to combine genetic data with operation-specific information on environment, economics and management; determine interactions among these factors; and predict animal performance and firm-level profit. Results can be used to produce customized selection indexes, optimal EPD profiles and even buyer-specific rankings of bulls offered for sale.

Because these models simulate underlying biological relationships, they require genetic predictions not for economically relevant traits but rather predictions for biologically relevant traits. It will be some time before breeding technology using bioeconomic simulation is in place, so, in the near term, we will likely see increased emphasis on EPDs for economically relevant traits.

In the longer term, we will have genetic predictions for inputs to simulation models. Whether these predictions will be routinely published remains to be seen.

Rules Of Thumb For Today Bioeconomic simulation/selection technology is years away, currently available sets of EPDs for economically relevant traits are incomplete, and tools for translating those EPDs into profit measures are still in the development stage. Given, therefore, that objective technologies for weighing the relative importance of traits do not exist today, how should producers proceed?

The answer, unsatisfying as it may be, is that they should use the same tools they have used in the past: intuition and common sense. Fortunately, these are not fixed qualities - they can be improved - and one way to improve them is to develop more logical systems for thinking about problems. A good reasoning system improves perspective.

For the problem of determining the relative importance of traits, it's helpful to first understand how individual traits influence profitability - how they affect income and costs. For example, heavier mature weight often results in more income per cow due to heavier sale weights. It also increases maintenance costs for cows and days on feed for slaughter offspring.

Once we understand how the various traits affect profitability, we can place them in the following categories: survival, fertility/longevity, feed consumption, product and non-feed costs.

Currently measured survival traits include all traits related to calving difficulty. Fertility/longevity traits include scrotal circumference and stayability. Feed consumption and product categories contain all the weight and carcass traits. Non-feed costs comprise traits related to calving difficulty, soundness, temperament and time on feed. (Note that many traits fit in more than one category.)

As a rule, traits affecting survival are the most economically important. That's because dead cows and calves represent units of production that are totally lost. The greater the death loss, the fewer the units of product sold.

Traits related to fertility/longevity are important, but not as much as survival traits. They rarely affect the number of product units. Instead, they influence product mix - the proportion of units in different product categories. Increased cow fertility, for example, results in the sale of more young stock and fewer cull cows.

Traits in the feed consumption and product categories are next in importance. Both categories contain similar sets of traits, so it would be speculative to rank one higher in importance than the other.

Traits affecting non-feed costs are typically the least important economically (ignoring any effects they might have in other categories.) They can be important for non-economic reasons, however. Differences in temperament, for example, may have small effects on profits but large influences on the enjoyment producers derive from being around their animals.

This ranking of trait categories is meant only as a guide. Differences in environment, economics, management practices and genetic merit of the current cowherd can change the relative importance of categories and traits within categories.

Until we have better, more objective selection technology, the best that producers can do is try to understand how traits affect profitability. They must then adjust that understanding to fit the particular conditions of their operations, and select sires accordingly.

All production and management changes in a commercial cow/calf operation can only be evaluated over time. And, it must be done in an equation where long-term profitability of the entire cow/calf enterprise equals the value of outputs minus the cost of inputs.

As cow-calf producers seek increased carcass marbling while simultaneously attempting to reduce input costs, concern has grown over what effect selection for increased marbling will have on maternal function and efficiency.

The answer depends on what degree of emphasis the producer continues to place on those convenience and functional traits that EPDs have yet to measure. Selection that includes positive marbling ability need not have a negative effect on those convenience traits. In fact, some producers have found that as they select for easier fleshing cows, marbling ability of progeny is enhanced.

Bill Hodge of Hodge Cattle Co., Pine Mountain, GA, draws on 40 years' experience with Angus cows to produce a list of convenience traits related to profitability. These are a few of the sought-after qualities that producers should never sacrifice in the pursuit of marbling or any other narrow focus: moderate size, calm disposition, easy fleshing on available resources, adequate body capacity, longevity, and structurally sound feet, legs and udders.

Notice that maintenance of these traits requires subjective judgment while working with the objective balance of EPD-based selection. Only time measures the maternal efficiency traits.

The good news is that there is no evidence to suggest selection that includes marbling as a criterion in balance with other traits has any deleterious effect on these traits. It is, however, difficult to justify single-focus selection for marbling, or even a combination of positive carcass traits, with less regard for other traits.

There is limited scientific literature that explores the impact of selection for marbling on beef cow production traits. But, it reveals little relationship (positive or negative) with measures of reproduction, milk production, preweaning calf growth, feedlot performance, external fat deposition and mature size (Marston et al., 1999).

The authors concluded that single-trait selection for marbling would be ill advised as would single trait selection for any beef cattle production trait. The solution lies with multiple-trait selection based on EPDs. These traits should include measures of reproduction, growth, maternal and carcass characteristics.

Superior carcass characteristics can be achieved without loss of maternal function, depending on sire choice. However, acceptable maternal function at an industry-relative low cost is not achievable with primary selection for carcass characteristics.

The Bottom Line. From a commercial cow-calf perspective, maximum profitability demands a balance between the factors that affect value of outputs and cost of inputs. There is an optimum for everything we do, including carcass traits.

Beyond that optimum, it costs more to achieve than the economic benefits received in return. The commercial cow base of this country has relied on maternal efficiency for years. As carcass traits become increasingly more important, maternal efficienc well into the new millennium.

T. T., J. F. Gleghorn and L. E. Wankel. 1999. "The Impact of Selecting for Marbling on Beef Cow Herds." Certified Angus Beef Program "White Paper." August, 1999.

Can we have too much marbling? My answer is "No." I strongly disagree with the notion that there may be genetic antagonisms between maximizing price on the rail and successful cow/calf production.

"Genetic antagonisms" give the connotation that it is impossible to breed animals contrary to correlated traits. However, through the use of unusual sires with light birth weights but heavy yearling weights, the Angus breed decreased their average birth weight EPD 11% from 1991 to 1997 while raising their yearling weight EPD 36%.

This is an unbelievable result and a dramatic illustration of the power of genetics when you think outside the box.

Research indicates that marbling and backfat are not highly correlated. Thus, you should be able to change either trait in either direction without "genetic antagonisms." Increasing marbling should be easy to do if you have a large enough database from which to select.

Until now, beef cattle breeders have not had adequate or accurate enough carcass data to change carcass traits. This is about to change. The use of ultrasound data will drastically increase the size and the reliability of the current carcass database.

The U.S. beef industry can produce a product that will have greater consumer acceptance if marbling is increased. Many researchers continually point out the fact that marbling minimally increases tenderness. But, consumer acceptance is not just evaluated on tenderness alone.

Gary Smith of Colorado State University has described marbling as the "taste fat" and outside fat cover as the "waste fat." He's suggested that breeding cattle to increase marbling and decrease outside fat would increase consumer acceptance.

Taste panels have indicated that 25% of Select steaks, 10% of Choice steaks and 0% of Prime steaks yield an undesirable eating experience. It would seem that increased marbling could cause a large change in acceptability.

At the present time, only about 10% of fat cattle processed in the U.S. are in the upper 2/3 Choice or Prime grades. There are several things happening that could dramatically change that mix in the next few years. Grids exist now that pay substantial premiums for higher quality grades.

If you look at the economics of beef production, it becomes quite evident that growth traits are still a major factor in maximizing profitable production. Single-trait selection of genetics will continue to be a serious mistake.

But, now we have the information available to us to produce a relatively light birth weight calf with milk genetics fitting the cow's environment, tremendous growth to a year of age, relatively large scrotal circumference, and relatively low mature weight. This new and improved genetic container can also simultaneously have a little more ribeye, a little less backfat and a lot more marbling.

There will be breeders and researchers who say this can't be done, but other breeders will get it done. In only a few years, we will have bulls that will produce as high a percent of Prime carcasses as we now have bulls that produce a very high percent of Choice carcasses.

When that happens, the demand for beef will increase dramatically. This will produce beef so delicious it will make meat-eating addicts out of vegetarians.