Progress 07/01/24 to 06/30/25
Outputs
Target Audience:Other researchers in economics, agricultural economics, and/or in related agricultural disciplines, including scientists in government agencies, whose interest focuses on the implications of pesticide use in agriculture, and the effects of policies and regulation in this contextOther researchers in economics, agricultural economics, and/or in related agricultural disciplines, including scientists in government agencies, whose interest focuses on the implications of pesticide use in agriculture, and the effects of policies and regulation in this context. Changes/Problems:One issue was that the notification of the award was received late (on 7/29/2024) relative to the proposed start date of the project. This was not a major problem, of course, it just delayed the start of the project. What opportunities for training and professional development has the project provided?In this initial year we have focused on recruiting a suitable Ph.D. student to work on this project. We have identified a 3rd-year PhD student with the appropriate training who we plan to employ in the coming academic year. How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?The main activities going forward concern estimation of the structural model/s where farmers choose between pesticides alternatives. The analysis will employ a discrete-choice formulation specified at the market level. In contrast to the seed market, there are fewer reasons to presume spatially segregating markets. Therefore, we plan to consider three different definitions of the relevant markets: a single US market, state-level markets, and CRD-level markets. The first option is mostly as a benchmark, because we believe a more regional basis is required to capture spatial pricing features. Whether it is best to assume the market is at the state or CRD level, however, is somewhat unclear at this stage. There are tradeoffs to defining a market at a more refined spatial level--the available product set will suffer more from measurement error because of sampling error. The other major issue to be studied going forward concerns identification, and the appropriate choice of instrument variables for endogenous pesticide prices.
Impacts
What was accomplished under these goals?
We have undertaken an in-depth exploration of the Kynetec data on farm-level pesticide use in corn and soybean production. We have been able to extend our access to these agrochemical data to include more recent years than outlined in the proposal. Specifically, our current analysis now covers the period 1998-2022. We find that herbicides dominate the agrochemical market for these two crops. For the last six years (2017-2022)--when fungicides, nematicides, and seed treatments are no longer tracked explicitly--herbicides account for 95% of corn and soybean farmers' pesticide expenditures. The year 2022 is somewhat of an outlier, showing a sharp increase in the value of herbicide sales because of a price spike likely due to worldwide shortages in certain compounds. More generally, it seems that total agrochemical sales have increased over time because of increased usage (due to rising weed resistance). Next we have focused more specifically on herbicides. Comparing prices across active ingredients is complicated by the fact that they may be applied at vastly different rates. For example, the mean price per pound of glyphosate is about $7/lb while the mean price per pound of mesotrione is about $85/lb. Application of glyphosate, however, in our sample typically requires 0.75 lb/acre versus about 0.09 lb/acre for mesotrione. To proceed, therefore, as the appropriate price for each herbicide we use the price per standard treatment, where a standard treatment is the standard recommended application rate associated with the label for that product. Our analysis of herbicide use in corn and soybean production reveals the following takeaways: The number of treatments per acre has increased significantly: from ~1.8 (1998) to 4.3 (2022) in soybeans, and from ~2.1 to 3.5 in corn. The increase in treatments has been the primary driver of additional area and sales. The number of products has more than doubled despite the fact that very few new active ingredients (AIs) have entered the market. This is due to both the introduction of additional single-AI products, presumably by a different company and/or as a different formulation (different surfactants and/or concentrations), and the introduction of additional multiple AI products. There is a clear push towards additional product differentiation despite almost no introductions of truly new characteristics/AIs. The market appears to be highly stagnant from an innovation standpoint. Most of the innovation from the companies in this space appears to have come from GE trait innovation. Costs on a per-treatment basis have not changed much and, in real terms, have actually fallen. An exception is the final sample year (2022), as noted, where we see a spike in prices. Farmers' costs on a per-acre basis have risen substantially, due to the need to use more AIs to combat weed resistance. Concerning market structure, we have analyzed the annual revenue shares for companies that had at least 5% market share during the sample period, as well as the HHI index of market concentration. We find that, generally, concentration has gown down in these agrochemical markets, although less so in corn where it has only decreased slightly. The 2018 mergers increased concentration somewhat but not substantially. The market is ultimately characterized by 8-10 moderate size companies. These companies had the entirety of market share early on. Towards the end of the sample, other smaller companies (likely selling generics) accounted for 10-15% of the market. We have also carried out a preliminary analysis of merger effects. There were three major mergers/acquisitions in the agrochemical industry post-2010, all around the year 2018: Dow and Dupont merged in 2017 and spun off to Corteva in 2019. Bayer announced intention to acquire Monsanto in 2016, completing it in 2018. ChemChina acquired Syngenta in 2017. Detailed consideration of the prices for nine of the top-selling products prior to and after the approximate time of the mergers indicates that there is no clear and persistent change in prices for any of these products. These conclusions are supported by a difference-in-difference analysis of herbicide prices at the product-market level.
Publications
|