Integrated Crop Management

Early seed discounts don't always pay

Harvest is in full swing and as we wrap up the 2006 season, many begin to plan for 2007. The 2006 growing season, though, is extremely useful when planning for next year by looking at what did or did not work. One key component is hybrid selection.

Seed companies are encouraging producers to purchase 2007 hybrids early by providing cash discounts. Most companies offer the greatest discount (8 to 10%) if hybrids are selected by approximately mid-November. While this discount is a substantial amount of money, especially with increasing seed costs, you do not want to move too hastily when selecting hybrids.

In the March 13, 2006, ICM article, Choosing corn hybrids [1], we discussed the huge variation in yield levels of today's hybrids (see Figure 2 [2]). Last year, hybrids entered in the Iowa Crop Performance Test--Corn that were on the same field, and therefore had the same management techniques, varied among one another by 30 to 54 bushels per acre (based on location). In other words, the top hybrid in the research trial out-yielded the lowest yielding hybrid by 30 to 54 bushels. At $2.00/bushel corn prices, this equates to a $60 to $108 per acre difference.

Producers and agronomists must spend time studying hybrid research trial data to see which hybrids perform consistently well across numerous locations. Just because a hybrid performs well on a producer's field this year does not mean it will produce high yields next year. That may seem illogical. Consider, though, that no matter what we do, we can never reproduce the dynamics of the 2006 growing season. The environment that the seed is placed into next year will certainly be different than this year. This is why we must always look at multiple-location data when looking at hybrid performance. If a hybrid performs consistently among the top hybrids at every location it is tested, then it will likely do well next year. But if it is not consistent across numerous locations, then we would not necessarily expect it to do well next year--even if it did well in one field, in one county, etc.

It is critically important to wait in purchasing seed until research results are posted and studied.

Iowa State University posts yield results at www.croptesting.iastate.edu [3] within a few days of the harvest date. There are more than 380 hybrids in this year's research trials. Selecting hybrids using these data will give you a much greater probability of having good hybrids next year.

Let's look at an example of how this pencils out:If we are purchasing a bag of $200 seed and receive an early seed discount of 10 percent, this bag of seed now becomes $180 (savings of $20 per bag). If that bag of seed has 80,000 seeds it will plant approximately 2.3 acres (at a 35,000 seeding rate).

Therefore, we are saving $8.70 per acre by purchasing our seed early. If we figure that we will receive $2.00 per bushel for our corn, then this equates to a final yield difference of only 4.3 bushels. One way of looking at this is if your yield goal is 200 bushels per acre, then you have paid for seed for 196 bushels of that corn crop, while the company paid for 4 bushels.

As mentioned above though, 2005 research data showed that selecting the wrong hybrid could cost you at least 30, and possibly up to 54, bushels per acre. So even if we select the low end of this yield loss estimate, we would have to receive an early cash discount of nearly 70 percent to account for the loss in overall yield that a poor hybrid would cause.

The monetary disadvantage from purchasing poor-performing hybrids is significant. It is far better to use current yield trial data to select consistently high performing hybrids than to select one based on hearsay or company promotions. In the next ICM newsletter, we will describe how to select proper hybrids based on examples from 2006 harvest data.

This article originally appeared on pages 241-242 of the IC-496(25) -- October 9, 2006 issue.


Source URL:
http://www.ipm.iastate.edu/ipm/icm//ipm/icm/2006/10-9/discount.html