2026 Crop Insurance Optimization Report

2026 Crop Insurance Optimization Report

Introduction

The One Big Beautiful Bill Act (OBBBA) made significant investments in the farm safety net, including changes to premium subsidies for crop insurance and enhancements to the Supplemental Coverage Option (SCO). These changes could improve a farm's net revenue while inducing a change in behavior.

This analysis takes the changes in OBBBA and retroactively applies to the years 2011-2023 to give insight into which policies might optimize farmer's net revenue.

In general, do not take the results of this analysis as recommendations. Every farm is different, with different sets of conditions, cost of production, appetite for risk, and financial situations. Please use this as a point of thought and consult your crop insurance provider.

Background & Assumptions

An increase in premium subsidies is one of the investments OBBBA made in the farm safety net. These changes are outlined in Table 1.

These changes are applied to both Optional and Basic Units. Enterprise Units also received new premium subsidy rates with coverages between 0.50-0.75 receiving 80% subsidy, 0.80 receiving 71% subsidy, and 0.85 receiving 56% subsidy.

This analysis solely looks at Optional and Basic Units.

Similarly, changes to the SCO policy should make it more attractive to farmers. SCO is a county-level crop insurance option that provides additional coverage for a portion of a producer's underlying crop insurance policy deductible. It must be purchased as an endorsement to the underlying crop insurance policy.

The premium subsidy rate for SCO increased from 65 percent to 80 percent, while the maximum coverage level increased from 86 percent to 90 percent. The increased subsidy also applies to the Enhanced Coverage Option (ECO), the Margin Coverage Option (MCO), the Hurricane Insurance Protection Wind Index (HIP-WI), and the Fire Insurance Protection Smoke Index (FIP-SI).

This analysis examines the impacts of changes to SCO.

In Kansas, over 96 percent of corn acres elect Revenue Protection, with over 77 percent of those policies covered at 70 or 75 percent. This is shown in Figure 1.

As such, this analysis assumes that Revenue Protection is the underlying policy.

Data

Counties

This analysis estimated the impacts in the three largest corn-producing counties in each crop reporting region. The counties are shown in Table 2.

Table 2: Yield, Projected Yield, Projected Price, Harvest Price, and Adjusted Rates

All data needed to calculate farmer premiums, indemnities, and guarantees was taken from USDA RMA. The analysis assumes that the increase in subsidy rates will not impact additional demand for coverage in terms of changing actuarial tables.

Actual Production History

A random multiplier was created to simulate a farmer's Actual Production History (APH). The multiplier assumes a random bell curve between 50 and 150 percent of a county's yield, with a disaster exclusion that eliminates any yield less than 60 percent of the 10-year average. The APH is then a 10-year average with the disasters excluded.

Farmer Premiums, Indemnities, and Net Revenue

Farmer Premiums and Indemnities were calculated according to USDA methodology, while Net Revenue is assumed to be yield multiplied by the higher of harvest or projected price, plus a crop insurance indemnity, less the farmer-paid crop insurance premium.

Year

The years examined in this analysis are 2011-2023 for the underlying policy and 2021-2023 for inclusion of SCO. The limited timeframe is due to lack of available data. RMA pricing data is limited to 2011, while SCO was not created until 2021. However, 2000-2010 yields are included for APH purposes.

Methods

This analysis sought to answer the following three questions:

• Which coverage level for the underlying policy provides the highest net revenue over time if that coverage is the same every year?
• Which coverage level, with the option of SCO, provides the highest net revenue each year?
• Which coverage level, between the underlying policies and SCO, provides the highest net revenue over time if that coverage is the same every year?

To answer these questions, this analysis built three separate models.

Model 1, Pick and Hold One Policy (2011-2023), examines only the underlying policy and assumes that the same coverage is chosen every year. It then determines which coverage had the highest net revenue over the time period. This model also shows the difference in per acre net revenue by county, coverage level, and year between the old and new subsidy regime.

Model 2, SCO vs. Underlying Policy by Year (2021-2023), allows for a choice between having SCO and not having SCO. It compares all 16 possible choices (0.50-0.85 coverage with no SCO and 0.50-0.85 with SCO) for each examined year and determines which policy and coverage had the highest net revenue. This model also compares the winner to it's alternate policy. For example, if the winner for 2023 in Gray County is a 0.50 coverage level without SCO, a comparison is made between the per acre net revenue at 0.50 coverage without SCO and 0.50 coverage level with SCO.

Model 3, Pick and Hold One Policy, Including SCO (2021-2023), allows for a choice between SCO and not having SCO and assumes that the same coverage is chosen every year. It then determines which policy and coverage had the highest net revenue.

Each model was ran 5,000 times to ensure an adequate sample size.

Results: Pick and Hold One Policy (2011-2023)

We first see a small increase in per acre net revenue across every coverage level with a subsidy increase, ranging from an average of $1.59 in Northeast Kansas to $5.22 per acre in Northwest Kansas. Overall, across all counties and coverage levels, the new subsidy regime created by OBBBA should increase by $3.37 per acre.

Across all years and counties, the 0.75 coverage level sees the largest increase in per acre net revenue, while the 0.55 coverage level sees the smallest. This is intuitive given percent increase in subsidy compared to the coverage level. The new subsidy means that 0.75 will be cheaper relative to other coverage levels.

Eight counties see at least one year and coverage level where the difference in per acre net revenue was larger than $10.00. Sherman County had the most such instances, with nine different occasions in which per acre net revenue was larger than $10.00. These instances largely occurred in Western Kansas. Only one instance occurred outside of the years 2019-2023. On the flip side, there were 129 instances where the change in net revenue was less than $1.00. These instances largely occurred in Eastern Kansas.

The year that experienced the largest per acre increase in net revenue on average across all counties and coverage levels was 2022. The smallest increase in net revenue was 2016.

The coverage level with the highest number of wins was either 0.50 or 0.85 in every county except Sherman.

Four counties had one coverage level win 100 percent of the time, with another 12 having one coverage win over 90 percent of the time.

Only the North Central and Northeast districts saw every county had a single coverage level winning less than 90 percent of the time.

Cheyenne, Cloud, and Sherman counties had the largest variance, each with four coverage levels winning more than 10 percent of the time.

The three northern districts had significantly more variance amongst their highest winning percentages compared to the rest of the state, with average highest winning percentages of 46.73 percent in Northwest, 51.28 percent in North Central, and 62.27 percent in Northeast. This means that those three districts see more coverage levels winning more often compared to the other six districts.

The change in the subsidy regime led to the variance of winning coverages increased in seven districts and decreased in the West Central and East Central districts. This means that in seven of the districts, the likelihood of a coverage level that is not the most likely winner, has a better chance of winning under the new subsidy regime.

Tables that show each district can be found in the Appendix.

Results: SCO vs. Underlying Policy by Year (2021-2023)

Looking at which coverage level and which policy has the highest net revenue for an individual year can be helpful for comparison purposes. In general, farmers tend to relate the current crop year to a past year. As such, it can be helpful to show which coverage was the best policy for a specific year.

The table to the right shows each county's best coverage and policy in each year.

Similar to the model that held one coverage over a long period of time, either the 0.50 or 0.85 coverage had the most wins the majority of the time. Only Sherman County in 2022 and Brown County in 2023 had a coverage level with the highest winning percentage other than 0.50 or 0.85.

Across most counties, a noSCO policy is the most common winner. SCO wins most often in 2023 with 10 of the 24 counties have SCO as the most common winner. SCO is never the most common winner in the Northwest, Central, Southwest, and South Central districts.

Of the 81 County-Year combinations, there are 12 combinations that have one coverage that is the winner 100 percent of the time. Of these, SCO is the most winning coverage seven times. The 0.50 coverage is the most winning coverage only twice.

Another 36 combinations have winning coverages with winning percentages of over 90 percent. Of these, only four times is SCO the most winning coverage, with three of the four coverages being 0.85.

The Southwest district has the least variance, with every County-Year combination winning at least 93.32 percent of the time.

The East Central District had the most variance, with 15 total County-Year combinations.

The comparison between per acre net revenue of the SCO and noSCO policies brought interesting results.

The model creates 150 different winners for each unique county-year combination. Of those combinations noSCO was the optimized net revenue combination in 128 of them. In these, the average difference between the optimal noSCO policy and its SCO counterpart with the same coverage level was $15.10. In the 22 instances in which the SCO policy was optimal, the average difference between the optimal SCO policy and its noSCO counterpart with the same coverage level was $121.75, nearly ten times larger.

The results are more impressive when looking deeper. Of the 22 instances in which an SCO policy was optimal, 12 of them had differences in per acre net revenue larger than $100 per acre, six had had differences in per acre net revenue larger than $200 per acre, and two had had differences in per acre net revenue larger than $300 per acre. Then, 20 of the 22 instances had per acre net revenue differences larger than the average noSCO differences ($15.10).

A cross-model analysis between Model 2 and 3 showed that the counties in which SCO was the optimal policy at any point in Model 2, were also the ones in in which SCO was the optimal policy when holding one policy in Model 3. This seems to indicate that if SCO as a policy were to trigger at all, that automatically makes it the best long-term policy to hold.

This leads to a big question regarding long-term data. If additional historic data were available for years prior to 2021, would this same trend apply to all counties?

Tables that show each district can be found in the Appendix.

Results: Pick and Hold One Policy, Including SCO (2021-2023)

Results largely mirrored Model 1 with the coverage-policy combination with the most wins was a 0.50 coverage with noSCO. The 0.85 coverage level saw an equal split between SCO and noSCO policies.

All three counties in both the West Central and Southeast districts had one coverage-policy combination win 100 percent of the time, while all three counties in the Northeast and Southwest districts had one coverage-policy combination winning over 90 percent of the time.

Geary County was the only county whose coverage-policy combination with the most wins didn't reach a plurality. This led to the East Central district having the most variance in winning percentages.

Implications

The new subsidy regime created by OBBBA is uneven but real. Some areas will see higher overall benefits, though every county and every district do see increases in net revenue because of the increased subsidy.

If the goal is to optimize net revenue, the highest and lowest coverages tend to win more often than the middle coverages. This would also be true if the model replaced 0.50 as the lowest coverage with a different lowest coverage, such as one required by a bank.

In many counties and years, the noSCO policy wins most often. When it does win, it tends to win at a high coverage. It also varies significantly across districts.

District differences matter a lot. Northern districts tend to show more variance, while the southern districts showed significantly less variance. This means that varying coverage in the northern districts makes more sense than it does in the southern districts, regardless of whether SCO is chosen or not. There are also large differences between overall per acre net revenue increases between the eastern and western portions of the state.

Limitations

The small sample size for SCO means that takeaways should be limited. Though it's clear that 2023 would have been a good year for SCO, a longer timeframe might change the outcomes in Model 3. As such, disregarding SCO as an option based on this analysis is not encouraged.

There are also numerous assumptions being made in this analysis that may not translate into real world applications. Yield assumptions, unit election, and minimum coverage requirements by bankers could all impact the way these results translate to the real world.

This analysis also does not take into account cost of production. In some instances, the higher premium prices associated with higher coverage levels could be significant enough to push farmers into negative margins. As such, the optimal coverage for individual operations could vary significantly. It also does not take into account individual farm finances. While 0.50 coverage has the highest net revenue over time, there could be situations in which a farmer is put out of business due to not having enough coverage.

Appendix

Each table in this section shows the number of wins for a coverage level, defined as having the highest net revenue in one of the 5,000 iterations, and its winning percentage, under both the old and new subsidy regimes. This gives a comparison of how the new subsidy regime has affected the optimal coverage level if held over the examined timeframe.

Coverage levels that win in less than ten percent of iterations were not included in the tables.

Each table in this section shows the number of wins for a coverage level, defined as having the highest net revenue in one of the 5,000 iterations, and its winning percentage, as well as the policy. This shows the most winning coverages and policies for each county in each year.

Coverage levels that win in less than ten percent of iterations were not included in the tables.

Each table in this section shows the number of wins for a coverage level, defined as having the highest net revenue in one of the 5,000 iterations, and its winning percentage, as well as the policy. This shows the most winning coverages and policies for each county in each year.

Coverage levels that win in less than ten percent of iterations were not included in the tables.