Developing Affordable Seed Corn for On-Farm Production and Sales

The potential for open-pollinated corn varieties to have higher nutrient density than most current hybrids is well known and clear.  It is a potentially important set of traits.  Higher protein and fat can make up for lower yields in corn varieties used for feed because this nutrient density can substitute for some of the more expensive soy that may be purchased or grown to fill out a ration.  However, these traits are quite variable in their expression: every open-pollinated variety is not significantly higher in specific nutrients.  In our data above, based on grow outs on our farms, one can see that most of the open-pollinated varieties and test crosses had higher protein and fat levels than the check hybrids which were fairly low by comparison.  One to two percentage points higher in protein and fat were common features.  Of the current open-pollinated varieties, Stan’s OP was only higher than the hybrids for protein by about a percentage point, and only slightly higher for fat.  This may be due to Minnesota 13 being part of its parentage as Minnesota 13 is not a very nutrient dense variety.  Varieties and test-crosses with Northern Flint or Wapsie Valley parentage often had both much higher protein and fat.  However, the Northern Flint test-crosses were generally not very high yielding.  Of the exotic parents, only the Conico test-cross had somewhat reasonable yields and higher nutrient density.  Although slower to dry, these lines may offer something useful to breeders looking for higher yielding and nutrient dense populations for grain production.  The commercial Iodent lines also were quite impressive for pushing yield while maintaining nutrient density in test-crosses with Dairyland.  Crude protein in our new synthetic variety was 8.7%, which is a very mild improvement over the hybrids reported here.

Using our updated version of CornPro3, a corn protein and yield assessment spreadsheet developed during the SARE-funded Heritage Maize Project in 2004, we were able to identify the dollar value of increasing corn protein for use as feed on farm.  If a farmer were to grow an expensive hybrid that yielded as well as ours did, about 196 bu/A, costs of production followed the 2023 average cost estimates from Iowa State University, and soymeal is $440/ton, we see clear, but modest reductions in feeding costs with increasing corn protein (Table 10).  A hybrid, like our check varieties with 8% protein, could be mixed into a ration for finishing beef cattle that would cost $1.93 per head daily.  With higher protein allowing for the substitution of higher quantities of cheaper corn in the ration, the costs for feed falls modestly by about 2.4-3% for each percentage increase in protein for the expensive hybrid, and by about 2.7-3.4% for the inexpensive hybrid.  Similar trends occur with the other animal diets.

Table 10.  Daily costs to feed livestock using corn hybrids with varying seed costs and levels of protein.  All calculations are based on CornPro3 assumptions for the corn and soymeal in grain-based diets.

Our student economist, Elena Hill, processed a great deal of data from across multiple states to gain a better understanding of the costs of corn production, and the time needed to carry it out.  Data on yield and costs that is published by states in the 90-day corn region show these averages for yields and variable costs related to field work (Table 11).  The general trend is that organic growing provides slightly lower yields, except in North Dakota where the concentration of organic growers in the best corn growing areas of that state likely bias those data.  Organic corn production was about 72-97% of local, conventionally grown corn if we ignore the North Dakota data.  This reduction in yield increased the costs of production on a per bushel basis as well.  However, estimated variable field costs per bushel of grain (or potentially seed) are fairly similar using either system anywhere across this region.  Calculations for the average number of hours spent with field activities concerning growing an acre of corn were about half an hour per acre for conventional growing and about two hours per acre for organic growers.  Part of this is due to cultivation passes, and part of it may be due to scale of production which can be quite large for many conventional operations.  USDA estimates that total organic corn production costs are $83-$98 higher per acre in general, but in December 2023 organic corn prices in the Midwest averaged $8.05/bu, as compared to the $4.75/bu price for other corn.

Table 11.  Average yields (bu/A) and variable costs for field work for conventionally grown and organically grown corn calculated from data published by states in the 90-day corn maturity zone.

If we consider all costs from crop production budgets prepared by USDA and Extension economists in several states, we see that the costs of production often push corn for grain into a not profitable situation.  Per bushel costs for corn are significantly higher when land costs are included (Table 12).  On-farm prices to grow each bushel are still quite low compared to seed costs, so if we could grow our own seeds, there is space potentially for advantageous cost savings.

Table 12.  Average yields and average total costs of production reported by States and USDA for 2023.

One of the critical calculations worked through by Elena is the costs and returns from growing open-pollinated corn for seed by three organic farmers who are already producing seed in the Midwest.  Data in Table 13 below are based on reported time, expenses, and yields from these farmers.  State averages released by Extension Services were used to fill out the table with our best estimates when our cooperating farmers could not supply all of the numbers.  If we assume all are planting about 30,000 seeds per acre, and about 80,000 seeds per bushel, then farmers can plant about 2.7 acres per bushel.  Using this estimate we can also estimate the cost for seed when these farmers plant their own seeds back as compared to the likely prices if they were to purchase these same varieties.  While the yields of open-pollinated varieties are still considerably lower than commercial hybrids in most cases, it appears that saved seeds could reduce the costs of seed by $21-$40/A each year over purchasing fresh open-pollinated corn seeds each year.  The costs of hybrid seeds range from $85 up to $330 and more per bag, and $31-$122/A for planting at 30,000 seeds to the acre.  Low costs of open-pollinated seed corn seed grown on farm could reduce seed costs compared to some hybrids depending on one’s production and seed cleaning system, and on the variety grown.

Table 13.  Costs of production for three organic open-pollinated corn growers in the Midwest.

Figure 3.  Plot of hand harvest time by number of ears per plot.

Very small-scale producers also would have to dry the corn, and this can be done in a number of ways.  The easiest way to just set the ears on a wire rack, but ears can also be hung on holders, hung by string, braided in the husks, etc.  There would be some as yet unknown time input to drying.  Shelling comes next after drying.  Frank hand shelled three samples of ears from a flour corn population and arrived at an estimate of 0.94 bu/hr.  Three samples of a mixture of floury, flinty, and dent ears were also shelled through an antique John Deere 1b hand cranked sheller.  With this machine 8.7 bu/hr could be shelled, although there would also be some time to clear cobs and remove cobs from the shelled corn.  Winnowing to clean the corn for planting or eating comes next, and this is fairly quick as well, just pouring grain from a bucket into another bucket three times.  Frank’s estimate is winnowing at a rate of about 7.7 bu/hr, although there would be some set up time since one can winnow about 15-25 lbs at a time in each batch.  With labor at $15/hr, cleaning up the grain at this scale adds at least another $3.66-$17.90 per bushel depending on tools, variety, and other factors.  Even with these costs, home seed production could be relatively affordable depending on one’s scale and the commercial availability of the variety one wishes to grow.  Rare varieties, if they are otherwise important to grow, are rarely available commercially at any price.

We can model any potential savings from using seeds with different costs and from using varieties with different yields or nutrient densities.  Once more, we returned to CornPro3, which has been updated with current crop budgets from the USDA and from several states.  Given the current corn price of $4.75/bu and soymeal price of $440/ton, the open-pollinated corn varieties we grew in our trial are not competitive for growing grain for sale.  Even the more expensive hybrids are only marginally profitable or not profitable to grow in this scenario as the cost of production is close to or exceeds the current price (Table 14).  However, we predict that synthetic varieties grown as open-pollinated corn that achieve 176 bu/A could currently meet or beat the profitability of hybrids grown for grain when the hybrids cost $170 or more per bag.  This relationship is still close even if it cost $30/bu to clean the open-pollinated corn seeds.  Currently there are no such open-pollinated varieties with such high yields, but they are theoretically possible.  The search continues.

Table 14.  Feed costs based on corn varieties with different population structure, seed cost, and nutrient density.  Predictions for corn costs and costs of daily animal rations are based on CornPro3 with crop production budgets from Iowa State University from January, 2023.

Another way to gain value from grain is to feed it to animals on farm.  CornPro3 allows the modelling of feed prices based on cost of production including seed prices, the value of grain based on total cost of production, and the price for soymeal which can be replaced with corn protein to some degree (Table 14).  Across all diets, Stan’s OP variety is currently competitive for making feed when compared to the highest priced hybrids ($330/bag).  Other than that, the other open-pollinated varieties are not competitive given our results.  However, when Tommy Boy reaches 150 bu/A yields as it sometimes does, its total cost per bushel falls to $5.01 if one can produce one’s own seeds at $10/bu, and the cost of finishing beef animals with it in the ration drops to $1.86 per day, greatly improving its competitiveness.  Clearly, anyone who wishes to grow open-pollinated corn in a profitable manner still needs to be mindful of yield too even when nutrient density is high.

Is there any other situation where an open-pollinated variety might be economically competitive in the future?  Our modelling from Table 14 shows that a synthetic variety or greatly improved open-pollinated population that yields 90% of our hybrid average (176 bu/A) would be competitive for grain prices with all but the cheapest hybrids at $4.53-$4.55/bu break even prices.  Varieties that could yield 80% of the hybrid average (157 bu/A) would be fairly competitive with the most expensive hybrids for grain production costs, and any yielding less would not be competitive with the hybrids.  When the corn is fed on farm as feed, the value-added option improves possibilities for these hypothetical varieties.  Any synthetic variety that could yield 90% of the hybrid average would compete with or be slightly cheaper to feed than any of the hybrids given current cost estimates and prices (Table 14).  This is true across all levels of protein and is largely due to the seed cost savings, although increasing protein levels would clearly improve the situation.  Differences in costs for cleaning the seeds at home, whether small or large scale, were small enough that this may not be a big consideration for those interested in growing their own seeds.  They just did not greatly influence predicted feed costs here.

The predicted varieties that could yield 80% of the hybrid average are also competitive for feed prices with the more expensive hybrid seed.  If protein levels could be 10%, two points higher than most hybrids, these that yield 157 bu/A would also be competitive with moderate priced hybrids, and if protein levels could be 11%, on the high end for maize generally, these varieties would also be competitive with any 8% protein hybrids for feed price (Table 14).  Developing synthetic varieties that yield 80% of good hybrids is predicted to be fairly readily attainable, and even those yielding 90% are theoretically possible.  However, research in this area by government agencies and seed companies is limited or non-existent.

To this end, in 2022 we advanced the intermating of promising lines to develop a potential synthetic variety of dent corn for the region.  Parents included the B73/Dairyland cross and a four-way cross of other parents from our experiment (B37s/Iodents//Pioneer Mixed/Mo17s).  We also made a cross between B73s/Dairyland and the Dairyland/Conico Cross for future expansion with a slightly broader mix of lines.  In 2023, Roger Coulter grew out the initial synthetic at his farm in New York State to expand the seed supply and move the population to F2 for use as an OP variety and for further evaluation.  In 2024 NIR showed the variety produced about 8.7% CP in New York last summer, a very modest gain over the hybrids.  Seeds from this grow out were sent to USDA Plant Introduction Station in Ames, IA for public distribution.  Photographs show that some of the plants from this new mixture appear to be healthy and productive.  Only ongoing evaluation can determine whether this population can carry through as a new and higher yielding variety with better standing ability than the existing open-pollinated varieties.  However, given the moisture levels observed in the test crosses, it is probably likely that this new population would still need further selection for early maturity to meet what Minnesota 13 can currently do.  In 2023, Frank Kutka also crossed the B37/Iodent//Pioneer Mixed/Mo17 cross with the B73s/Dairyland//Dairyland/Conico cross to generate the F1 of a second, very closely related synthetic population with slightly broader parentage.  This will also be grown out to the F2 and beyond with hope of finding higher yield maintained in the later generations.

Figure 4.  Roots and stalks of our new synthetic variety.

Figure 5.  A fine ear of our new synthetic variety still on the stalk.