PRF Insurance Moves to NCEI Rainfall Data: What It Means for Premiums and Indemnities
By: Geoffrey Childress, James Keeler, and Dylan Turner
Publication date: October 30, 2026 | DOI: 10.22004/ag.econ.415937
For the 2027 crop year, the rainfall index behind Pasture, Rangeland, Forage (PRF) insurance moves from the National Oceanic and Atmospheric Administration’s (NOAA) Climate Prediction Center (CPC) to NOAA’s National Centers for Environmental Information (NCEI). For PRF the change took effect at the August 31, 2026 contract change date and will apply to the 2027 insurance year (USDA Risk Management Agency, 2026). This brief looks at how the move from CPC to NCEI may affect insurance outcomes.
RMA states that “historical data shows that overall loss ratios remain nearly identical under the new data source” (USDA Risk Management Agency, 2026). In aggregate, the results presented here largely confirm this with long-run simulated loss ratios of 0.861 under CPC compared to 0.860 under NCEI. These aggregate results, however, mask substantial variation across individual grids, the unit that matters to producers, who buy coverage for a specific grid and a set of intervals. In the latest full year of observed outcomes, July–August 2025 through June–July 2026, mean indemnity under the two sources differs by more than $1 per acre in 43% of grid-intervals and by more than $5 per acre in 10%. Producer-paid premiums are about 6% lower under NCEI overall, but the change varies substantially across grids and intervals. Even the long-run loss ratio (defined as the simulated loss ratio for the full history behind each source of precipitation data), which changes by less than 1% in most grids, changes by more than 5% in about 9% of individual grid-intervals.
To arrive at the above values, this analysis starts with existing grids used to measure rainfall, which remain the same under both CPC and NCEI sources (USDA Risk Management Agency, 2026). What changes is the precipitation data assigned to each grid and the corresponding index values used to set premiums and trigger indemnities. NCEI’s grid values are aggregated from its nClimGrid product, which is built at a resolution of roughly 5 kilometers (Vose et al., 2014). NCEI values may therefore be smoother, or less variable across space, than CPC values. CPC’s data begin in 1948, while the NCEI series used for PRF begins in 1951, so the historical record behind each grid’s normal rainfall is three years shorter under NCEI.
We focus the analysis on a single hypothetical policy for each grid-interval combination defined by grazing coverage for one interval at 90% coverage, using one insured acre at a productivity factor of 1. RMA’s actuarial data for the 2027 insurance year give each county’s base value and the total set of insurable grids. Every calculation uses these 2027 base values and index values measured against RMA’s crop year 2027 normal, average interval rainfall from 1951 (NCEI) or 1948 (CPC) through 2025. Only the rainfall data and the premium rates differ between the two sources. RMA’s 2027 premium rates are the first ones based on NCEI. Thus, we pair NCEI with the 2027 rates and CPC with the 2026 rates. We use premium in two forms. Total premium, before the subsidy, enters the loss ratio. Producer-paid premium, after the 51% subsidy at 90% coverage, enters net benefit.
We measure indemnity over three windows. The latest year uses each interval’s most recent outcome observed in both sources, July–August 2025 through June–July 2026, since index values for the rest of 2026 are not yet final. These indemnities are hypothetical in the sense that they apply 2025–26 rainfall to 2027 coverage and are not the payments actually made in those years. The long-run indemnity is the average payment over each source’s full record through 2025, 78 years from 1948 for CPC and 75 years from 1951 for NCEI. We also use a final indemnity calculation based on the last ten years, which represents the average payment over 2016–2025 and captures the bulk of what PRF participants would have experienced. To obtain a stable measure of the effects on loss ratios, we focus on the long-run loss ratio defined as the long-run indemnity divided by total premium. Each grid’s value pools the premium and indemnity dollars of its counties and insurable intervals when defining the loss ratio.
Figure 1 maps the percentage change in each grid’s long-run loss ratio from CPC to NCEI. The overall long-run loss ratio is 0.861 under CPC and 0.860 under NCEI. This similarity is driven by the tendency for the expected indemnity and premium to move together in aggregate. The long-run expected indemnity is estimated to be 6.1% lower under NCEI against a total premium that is 6.0% lower. The middle half of grids fall within about ±0.6%, and 99% of grids within ±3.7%. Increases and decreases are scattered across the country, with somewhat larger changes in both directions in the northern Rockies.
Figure 1: Percentage Change in Long-Run PRF Loss Ratio Moving from CPC to NCEI Rainfall, 90% Coverage
Note: Long-run loss ratio is expected indemnity (average payment over 1948–2025 for CPC and 1951–2025 for NCEI) over premium before the subsidy (crop year 2026 rates for CPC and 2027 for NCEI). Colors are capped at ±3.7%, the 99th percentile of the absolute change. Grazing, 90% coverage, one acre in each interval at a productivity factor of 1; each grid pools the dollars of its counties and intervals. Index values are against the crop year 2027 normal. White areas have no grazing actuarial data.
Source: ARPC calculations using data from USDA.
Overall, this empirical exercise suggests that holding participation fixed, PRF as a whole performs about the same under NCEI as under CPC, but that a particular grid and interval may see substantial shifts.
Most producers can expect a modest reduction in premium. However, whether a producer is better or worse off depends on which grid and intervals they insure, and in any given year the change in indemnity can be much larger than the change in premium.
To capture the combined effect of changing indemnities and premiums into a single interpretable metric, we also look at expected net benefit, which we define as the indemnity a producer can expect to receive on average, plus the premium subsidy, less the full premium. Here expected indemnity is measured over the last ten years which reflects the conditions producers have recently experienced. Figure 2 maps the change in expected net benefit from CPC to NCEI by grid and interval, in dollars per $100 of protection (defined as the product of the county base value, coverage level, and productivity factor). Measuring the change per $100 of protection removes differences in county base values, so grids in different parts of the country can be directly compared. A value of −$1.00 means a producer would have been about 1% of their protection worse off each year, on average, had NCEI been in place over 2016–2025. These values can be converted to dollars by multiplying the value for each insured interval by that interval’s protection and dividing by 100. For example, a ranch with $18 of protection per acre, about the western average, in an interval showing −$1.00 would have received about $0.18 per acre less each year, or about $900 on 5,000 insured acres.
Figure 2: Change in 10-Year Expected Net Benefit per $100 of Protection Moving from CPC to NCEI Rainfall by Interval, 90% Coverage
Note: Expected net benefit is expected indemnity (average payment over 2016–2025) less producer-paid premium. Change is NCEI minus CPC per $100 of protection. Colors are capped at ±$16.30, the 99th percentile of the absolute change. Grazing, 90% coverage, one acre in each interval at a productivity factor of 1; each grid averages its counties, and intervals are weighted equally, not by participation. Index values are against the crop year 2027 normal. White areas have no grazing actuarial data.
Source: ARPC calculations using data from USDA.
Over 2016–2025, expected net benefit is estimated to be lower under NCEI in 52% of grid-intervals. Per $100 of protection, expected indemnity falls by $1.10 on average while producer-paid premium falls by only $0.61, so expected net benefit falls by $0.50. At face value, this reduction appears to contradict Figure 1, which shows loss ratios will be largely unchanged. This discrepancy arises because the premium subsidy is a fixed share of premium, so when premiums fall, the subsidy falls with them. At 90% coverage, which has a 51% subsidy rate, the producer keeps only 49 cents of each dollar of premium reduction, which means an equal reduction in both premiums and indemnities results in a lower net benefit.
As with the loss ratio, these average effects hide large changes in both directions. The middle half of grid-intervals ranges from −$2.53 to +$1.97 per $100 of protection, and about one in five change by more than $5. The average loss is smaller in the western states, $0.10 per $100 of protection compared with $0.90 elsewhere, but the changes there are larger in both directions: about 29% of western grid-intervals change by more than $5, compared with 11% elsewhere.
Over each source’s full record, the picture is more uniform. Expected net benefit falls in 80% of grid-intervals, by $0.48 per $100 of protection on average, and the largest losses are in the northern Rockies and Great Basin in the winter intervals. The two windows give nearly the same average loss, but over the last ten years the result for a particular grid and interval can differ substantially from its long-run result, in either direction. A lower premium under NCEI is therefore not by itself a better deal, and producers should look at both the recent and the long-run record for the intervals they insure.
Although these results do provide some initial guidance on what to expect from the change from CPC to NCEI as a source of precipitation data, they do not reflect where producers actually buy PRF, which intervals, coverage levels, and productivity factors they choose, or how they may modify those choices in response to the switch to NCEI. Because participation is concentrated in some grids and intervals, and producers may shift toward intervals that look more favorable under NCEI, the loss ratios, premiums, and payments RMA reports for the program can differ from those shown here. Our loss ratios and expected net benefits are averages over the historical record and do not necessarily reflect the future outcomes farmers and ranchers may ultimately experience.
References
USDA Risk Management Agency (2026). USDA Risk Management Agency Upgrades Rainfall Data Source. News release, Washington, DC, April 30. https://www.rma.usda.gov/news-events/news/2026/washington-dc/usda-risk-management-agency-upgrades-rainfall-data-source
Vose, Russell S. et al. (2014). Improved Historical Temperature and Precipitation Time Series for U.S. Climate Divisions. Journal of Applied Meteorology and Climatology 53(5): 1232–1251. https://doi.org/10.1175/JAMC-D-13-0248.1
Recommended Citation: Geoffrey Childress, James Keeler, and Dylan Turner (2026, October 30). PRF Insurance Moves to NCEI Rainfall Data: What It Means for Premiums and Indemnities. ARPC Brief 2026–22. Agricultural Risk Policy Center, North Dakota State University. https://doi.org/10.22004/ag.econ.415937
Comments