Research Brief: Soybean Network Diffusion
Once again, I'm going to discuss a recent publication of mine (potentially the last one for a while). This time is different. Rather than try to give a high-level overview that abstracts away from all the math and statistics and coding - though some would argue that is the interesting part - I'm going to put the research almost into story form.
It all started with my paper with Tanner McCarty about hemp pollen drifting from one farm to another and messing things up. We had no data but had to act quickly, so I took part in a paper that was precisely what I thought I swore off of: economic theory. After the KY Department of Agriculture provided us with some key farm-level data on hemp production, we wrote and published another paper in the AJAE.
In the midst of that publishing process, it was brought to our attention by a fellow ag economist that this sort of problem was going on in the soybean sector. Needless to say, Tanner and I were eager to branch out of an area as niche as hemp and into something more mainstream (and with a larger economic footprint) like soybeans.
The context: farmer A plants soybeans with Enlist genetics (resistant to 2,4-D, glyphosate, etc.) while farmer B plants soybeans with Xtend technology (resistant too all of the above plus one more: dicamba). This is fine, as long as the non-resistant soybeans don't get a whiff of any dicamba. But that's not a problem unless dicamba drifts, right? Well, that's exactly what has earned dicamba its notoriety these last few years. Bear in mind, planting decisions are annual and fixed. If a farmer plants Enlist soybeans and their neighbor plants Xtend and sprays dicamba, then they're opened up to risk of significant yield loss --- no backsies.
What's to be done? Well, in the short run, sometimes the best offense is a good defense. In other words, maybe the farmer who got their soybeans damaged by dicamba this year decides to plant Xtend soybeans next year as a defensive maneuver. May as well go all out and spray dicamba now that your new choice of seed is resistant to it, right? But now that farmer risks harming another non-adopting neighbor's soybeans, and if they do then that neighbor will switch the next year, and so on until there will be large pockets of acreage where it's no longer safe to be a holdout and grow the non-resistant variety of soybeans. Note that all of this excludes anything about the effects on social capital between neighbors who farm together and may have known each other for most of their lives.
So Tanner and I teamed up with some coauthors who study more of the scientific side of dicamba damage to non-resistant crops, and we went to work. We used a modified version of our method we developed in the AJAE paper, since soybeans are more conventionally produced than hemp. At long last, all our work paid off and we got the paper across the finish line at the JAAE right at the beginning of summer.
We mentioned in the discussion that at the time of the paper, Minnesota was weighing two pilot solutions that intrigued us. The first is a simple higher taxation on gross sales of dicamba, and the second is a Coase-type bargaining framework where Xtend and Enlist growers reach an agreement together concerning potential damages.
I can't help but think that overly stringent taxation would lead to (1) black market buying & selling of dicamba between states as it has in the past, and (2) a problematic resurgence of weeds that are resistant to all other herbicides (e.g. palmer amaranth or "pigweed"). The second solution seems, for lack of a better word, quite clever to me. In part that's because Coasian bargaining fascinates me, as does game theory in general. Indeed, "The Problem of Social Cost" by the late Ronald Coase in 1960 analyzes a hypothetical problem between a crop farmer and a cattle rancher that is not only easy to understand, but still very interesting and applicable today. Give it a read when you get 5-10 minutes.
Whatever the case may be, we're beyond excited to see what's next for this track of research. I'm hopeful our findings now and those still to come will benefit all the players in this intricate ecosystem of production agriculture, but especially --- speaking as one of them --- the farmers.