Estimating the Economic Value of Specific Characteristics Associated with Angus Bulls Sold at Auction

Jones et al.: Value of Bull Characteristics

Table 5. Correlation Coefficients for Production EPDs and Actual Measures

Note: p-values are in parentheses. Number of observations 5 4,151.

of bulls selling in the fall would bring a
premium relative to bulls sold in the spring.
Several of the sire variables significantly
impact price, indicating that genetic linkages
to top registered Angus bulls can be impor-
tant. The significance of several individual sale
variables is also of interest, as it suggests that
the reputation of the seller can have an impact
on price and that buyers are likely to pay
premiums/discounts for similar bulls sold at
different sales.^6,7

One of the primary objectives for this
research was to reexamine the relationship
between production EPDs and actual weights,
following up on the research conducted by
Chvosta, Watts, and Rucker. Comparing the
parameter estimates for the EPDs and actual
weights reveals larger estimates for the EPDs
relative to their related actual weights. How-
ever, this comparison does not tell the whole
story because of the varying units involved.

⁶ Data collected from some specific breeders (sales)
were not utilized in this model specification because
they did not report all the information used in this
analysis.

⁷ An alternative model specification replaced the
individual sales variables with individual state dummy
variables. Production, marketing, and genetic factor
results were very consistent across models. Relative to
bulls sold in Kansas (the base), bulls sold in Colorado,
Montana, North Dakota, Nebraska, South Dakota,
and Texas received premiums, while bulls sold in
Missouri and Oregon received discounts (Turner).

Elasticities provide a unitless comparison
between the two genetic measures and offer a
measurement that is readily comparable across
variables. The elasticities for the actual weights
are greater than the elasticities for the EPDs.
The results from the comparison of elasticities
are similar to those reached by Chvosta,
Rucker, and Watts and at first glance would
suggest that actual weights receive a higher
value from buyers relative to EPDs.

However, a problem with the elasticities is
that they show the effect of the variable only
at a certain point, here the mean. This
technique ignores the true behavior of most
variables by assuming that a 1% change in all
variables occurs with equal likelihood. There-
fore, it may be more insightful to examine the
effect a variable has on price across a
standardized range of likely changes. This
provides a means for comparing the realisti-
cally expected relative impact between vari-
ables of differing units.

We were particularly interested in compar-
ing the relative expected impact of actual birth
weight and birth-weight EPD from our first
model. In order to make this comparison,
premiums were calculated in log form by
multiplying the parameter estimates for all the
continuous variables by their mean value.
Sensitivity of price to the variable of interest
(e.g., birthwt or birthepd) was calculated across
a range of two standard deviations above and
below the mean of the variable. The calculated

<<   <   8   9   10   11   12   13   14   >   >>