by Zhushan Du, J. Arbuckle, and Hongli Feng
Conservation practices such as cover crops and no-till are widely promoted as solutions to the environmental externalities of row crop agriculture, including soil erosion, nutrient loss, and degraded water quality. Federal and state programs invest billions of dollars annually to encourage farmers to adopt these practices. Yet despite decades of effort, adoption rates remain well below the levels needed to meet conservation goals such as those outlined in the Iowa Nutrient Reduction Strategy. A growing body of research suggests that a critical but overlooked factor is suppressing net adoption rates—disadoption, the discontinuation of conservation practices by farmers who had previously adopted them, including intermittent patterns of discontinuation followed by resumption.
Previous research documents changes in conservation acreage across counties (Plastina, Sawadgo, and Okonkwo 2024). This article summarizes findings primarily from two recent studies by the authors that use longitudinal farm-level survey data to examine temporal patterns in cover crop adoption among Iowa farmers. The results reveal substantial flux in adoption behavior over time, with many farmers cycling between adoption and non-adoption. This “churning” effect means that net gains in adoption are far smaller than the number of farmers who try the practice in any given period, because concurrent disadoption offsets part of the new uptake. These findings have major implications for how researchers measure adoption, how policymakers design conservation programs, and how the conservation community sets and assesses progress toward adoption goals.
What the data show: Flux beneath the trend
Most conservation adoption research relies on cross-sectional surveys that measure practice use at a single point in time and also measure practice use as a dichotomy phenomenon of adoption or non-adoption. This approach implicitly assumes that once a farmer adopts a practice, adoption is permanent. However, practices like cover crops and no-till involve annual management decisions with relatively low switching costs. A farmer who plants cover crops one year can choose not to plant them the next and may or may not use them again, based on changing weather conditions, commodity prices, labor availability, or other operational factors.
Our research analyzes data from the Iowa Farm and Rural Life Poll (IFRLP), a longitudinal panel survey of Iowa farmers conducted annually since 1982. Each survey year, the IFRLP covers a set of topics that are relevant to agriculture and farmers in Iowa—though the topics are often also applicable to regions surrounding Iowa. Because some farmers participate in the IFRLP in multiple waves, while new respondents are added to offset attrition, we are able to track cover crop adoption over time for the subset of farmers observed in multiple survey years.
The latest Census of Agriculture confirms continued growth in cover crop use, although the pace has slowed, as shown by figure 1 below. Figure 1 shows that growth after 2017 was slower nationally and across the three-I-states in the central Corn Belt region than during 2012–2017. Cover crop acreage increased substantially from 2012 to 2022, especially in Iowa. US cover crop acreage rose from 10.3 million acres in 2012 to 15.4 million in 2017 and 18.0 million in 2022 (Wallander et al. 2021). Iowa increased from 0.38 million to 0.97 million and 1.28 million acres over the same period, reaching 5.0% of cropland in 2022. Across Iowa, Illinois, and Indiana, cover crops occupied 3.15 million acres, or 5.1% of cropland in 2022 (Plastina, Sawadgo, and Okonkwo 2024). However, growth slowed after 2017. Aggregate acreage figures alone do not reveal whether the slowdown resulted from fewer farmers trying cover crops for the first time, more previous users discontinuing the practice, or both. Longitudinal farm-level data provide evidence on these underlying movements.
Notes: Acreage is indexed to 2012 = 100 to facilitate comparison across geographies of different sizes. Point labels report actual acreage in millions
The first study we reference analyzes cover crop use over three IFRLP survey waves (2014, 2016, and 2018, corresponding to use in 2013, 2015, and 2017). Our analytical sample consists of 519 farmers who participated in all three waves (Du, Feng, and Arbuckle 2025a). The panel data reveals a pattern of significant flux beneath the aggregate trend. Overall, the share of farmers using cover crops rose from 17.7% in 2013 to 22.9% in 2017—a net increase of 5.2 percentage points over four years. However, this modest net gain masks much larger underlying shifts. Figure 2 illustrates that between 2013 and 2015, 10.6% of the sample newly adopted cover crops, but 8.5% simultaneously discontinued use, yielding a net gain of only 2.1%. Between 2015 and 2017, 9.1% adopted while 6.0% disadopted, producing a net gain of 3.1%. Thus, in both periods, the rate of new adoption was substantially offset by concurrent disadoption.
Note: We call this group “new adopters,” but it is possible that they had used cover crops in the past (before our study period) and their use in the second or third year of the study period could be intermittent use.
Source: 2014, 2016, and 2018 Iowa Farm and Rural Life Poll (N=519).
Three farmer profiles—and what sets them apart
By tracking individual farmers across all three survey periods, we identify three distinct adoption profiles—continuous adopters, continuous non-adopters, and intermittent adopters. Continuous adopters (6.6% of the sample) used cover crops in all three years. Continuous non-adopters (64.5%) never used cover crops during the study period. The largest group of interest was intermittent adopters (28.9%), who showed shifting patterns of adoption and disadoption. This group was more than four times the size of the continuous adopter group.
Another way to look at the impact of disadoption is through a visualized comparison of the cumulative adoption (the total number of farmers who used cover crops at least once) with net adoption (the number using them at any given time). By 2017 (figure 3), cumulative adoption had reached 35.5% of the sample and 184 farmers had tried cover crops at some point during the study period. Yet only 22.9% were using them in 2017. Without any disadoption, the adoption rate would have nearly doubled from its 2013 baseline rather than increasing by just 5 percentage points.
The upper line in figure 3 shows the cumulative share of farmers who had used cover crops at least once by each survey year, while the lower line shows the share using cover crops in that year. The shaded area represents farmers with prior cover crop experience who were not current users at the time of the survey.
Statistical comparisons reveal meaningful differences among the groups. Continuous adopters were significantly more likely to raise livestock (56% versus 29% for intermittent and 22% for continuous non-adopters), to farm highly erodible land, and to have higher gross farm sales. They also planted more cover crop acres on average (87 versus 35 for intermittent adopters) and were more likely to have received cost-share assistance.
Attitudinal differences followed a clear gradient. Continuous adopters perceived the greatest benefits from cover crops and the fewest barriers. Continuous non-adopters reported the opposite pattern. Intermittent adopters fell squarely in between on virtually every measure—they perceived moderate benefits and moderate barriers. Notably, intermittent adopters expressed strong interest in learning more about cover crops at levels similar to continuous adopters and significantly higher than continuous non-adopters. This suggests that the intermittent group is not disengaged; rather, they may need additional support to move from trial use to sustained adoption.
What drives disadoption?
Based on analysis of farm and farmer characteristics in the IFRLP data, the strongest predictor of continued use was the scale of prior adoption—farmers who had planted more cover crop acres were substantially less likely to discontinue. Past adoption behavior was also critical—farmers who had used cover crops in the most recent prior period were much less likely to disadopt than those with gaps in their adoption history. This suggests that continuity itself is an important factor in sustaining practice use, possibly because cumulative experience builds management skills and allows farmers to observe soil health benefits.
Notably, policy-related variables including cost-share participation and crop insurance showed relatively modest impacts on the decision to continue use. Results from machine-learning analysis by Du, Feng, and Arbuckle (2025b) shows that in the logistic model, cost-share participation and crop insurance purchase were associated with 3.6- and 2.4-percentage-point lower probabilities of cover crop use in 2017, respectively, but neither estimate was statistically significant. While these programs may effectively encourage initial adoption, their long-term influence on practice persistence appears to be limited. This finding raises questions about whether short-term incentive programs (typically one-to-three years) are sufficient to promote the sustained adoption necessary for conservation practices to deliver their full environmental benefits.
A related study using a separate, larger data set reinforces these findings (Sun et al. 2026). Sun et al. (2026) surveyed more than 3,200 Iowa farmers across six major watersheds between 2015 and 2019, using a three-category adoption variable—non-adoption, open to adoption, and adoption. The results show that nearly 20% of farmers who used cover crops in one year had stopped using them by the following year, and about 13% of no-till adopters similarly discontinued use after one year (table 1). For cover crops, although 192 farmers moved into the adoption category from one year to the next, 142 shifted out, leaving a net gain of only 50 individuals—a 1.9% increase in overall adoption. For no-till, the net gain was just 26 individuals, or 1.0%. These patterns were consistent across survey years, suggesting that the churning effect is not an artifact of a particular year’s conditions but rather a persistent feature of adoption dynamics. Sun et al. (2026) also find spatial variations, with both no-till adoption and disadoption patterns differing significantly across watersheds—disadoption was notably lower in the Missouri-Nishnabotna watershed, where highly erodible Loess Hills soils appeared to sustain higher persistence.
| Watershed | Cover Crops: Adoption | Cover Crops: Disadoption | No-till: Adoption | No-till: Disadoption |
|---|---|---|---|---|
| All watersheds | 80.4% | 19.6% | 87.5% | 12.5% |
| Iowa | 81.4% | 18.6% | 82.3% | 17.7% |
| Missouri–Little Sioux | 71.4% | 28.6% | 84.1% | 15.9% |
| Upper Mississippi–Maquoketa–Plum | 84.5% | 15.5% | 86.8% | 13.2% |
| Des Moines | 73.2% | 26.8% | 81.2% | 18.8% |
| Missouri–Nishnabotna | 80.5% | 19.5% | 95.4% | 4.6% |
Policy implications and next steps
Findings in the recent studies summarized in this article have several implications for conservation policy and practice. First, conventional measures of adoption may provide an incomplete picture of progress. Aggregate statistics on conservation practice acreage can conceal substantial movement into and out of practice use at the farm level. Tracking new adoption, continued use, and disadoption separately could provide a more accurate assessment of progress toward conservation goals, including those in the Iowa Nutrient Reduction Strategy.
Second, the results point to the importance of considering continued use alongside initial adoption. Intermittent adopters—farmers who used cover crops in some, but not all, survey periods—could be more than four times as numerous as continuous adopters in the studies. Because many environmental benefits of cover crops accumulate through repeated use, programs focused primarily on initial uptake may generate smaller long-term gains when substantial disadoption occurs.
Several approaches may help farmers move from initial trials toward more sustained use. Longer-term or more flexible incentive arrangements could give farmers additional time to gain experience and manage difficult years. Technical assistance and peer-to-peer learning may be particularly valuable for intermittent adopters, who expressed substantial interest in learning more about cover crops but perceived greater barriers than continuous adopters. Strategies that improve the economic returns to cover crops—especially for farms without livestock or grazing opportunities—may also reduce discontinuation.
At the same time, the available evidence provides limited support for the idea that financial incentives alone lead to continued use. In Du, Feng, and Arbuckle (2025b), cost-share participation was not significantly associated with cover crop use in 2017. This result does not establish that cost-share is ineffective, but it raises the possibility that initial financial assistance may be more effective when accompanied by technical support, follow-up engagement, and programming responsive to farmers’ changing circumstances.
Understanding those circumstances is important because intermittent use does not necessarily indicate a lack of commitment to conservation. It may reflect deliberate management flexibility or factors outside farmers’ control, including weather variability, commodity prices, labor constraints, and changes in seed availability and cost. Different causes may call for different policy responses. When discontinuation is associated primarily with limited information or management challenges, education and technical assistance may be particularly relevant. When it is driven by drought, input-price increases, or unfavorable market conditions, flexible or contingent incentives may help farmers maintain the practice through difficult years.
Overall, focusing solely on initial adoption provides an incomplete picture of conservation practice use. The large group of intermittent adopters represents both a challenge and an opportunity—these farmers have experience with cover crops and remain interested in the practice, but they may face barriers that limit continued use. Future research linking longitudinal adoption records with weather, market, and program-participation data could help distinguish among the causes of disadoption and provide a stronger empirical basis for designing policies that support sustained conservation practice use.
References
Du, Z., H. Feng, and J. Arbuckle. 2025a. “Not Ready for a Long-term Commitment? Analyzing Temporal Variability in Cover Crop Adoption.” Journal of Soil and Water Conservation 80(5):551-565. DOI: 10.1080/00224561.2025.2533102
Du, Z., H. Feng, and J. Arbuckle. 2025b. “Exploring the Complementarity between Traditional Econometric Methods and Machine Learning—An Application to Adoption and Disadoption of Conservation Practices.” Applied Economics. DOI: 10.1080/00036846.2025.2462792
Plastina, A., W. Sawadgo, and E. Okonkwo. 2024. “Pervasive Disadoption Substantially Offsets New Adoption of Cover Crops and No-Till.” Choices 39(2):1-14.
Sun, H., S. Upadhaya, C. Morris, J. Arbuckle, L. Nowatzke, and Z. Zhu. 2026. “The Flux of Agricultural Conservation: Understanding Changes in Iowa Farmers’ Adoption of Cover Crops and No-Till Over Time.” Society & Natural Resources 39(2):200-219.
Wallander, S., D. Smith, M. Bowman, and R. Claassen. 2021. Cover Crop Trends, Programs, and Practices in the United States (Economic Information Bulletin No. 222). US Department of Agriculture, Economic Research Service.
Suggested citation
Du, Z., J. Arbuckle, and H. Feng. 2026. “Beyond Initial Adoption: Farm-Level Evidence on Cover Crop Disadoption and Conservation Policy.” Agricultural Policy Review, Spring 2026. Center for Agricultural and Rural Development, Iowa State University.