Authors: Mariah Beverly, Kimberly L. Morgan, and Elizabeth Canales
As artificial intelligence (AI) technology becomes cheaper and more prevalent in everyday life, it is only a matter of time before the banking industry implements some type of AI to streamline workflows, if they have not already. AI technology is hailed as being faster, more efficient, and more accurate than your average human worker, and there are certainly cases where AI is welcomed and necessary. However, the implications of using AI for automatic underwriting and credit scoring should be carefully considered before being fully implemented.
The problem with using AI in underwriting is rooted in the fact that AI models are trained on historical data to predict an operation’s repayment capacity. Because the quality and representativeness of the data used could influence lending decisions, this can create significant problems for portions of the farming population, particularly when operational context is lacking from the data used or when little historical data are available.
Beginning farmers and ranchers, for example, can have little to no historic data for their own operations. Similarly, specialty crop producers, urban producers, or agritourism-reliant farms might have historical data for their operation, but AI models can lack sufficient training on the context of these types of operations due to little publicly available data for these segments of the agriculture industry to make comparisons. This matters, because typical financial metrics can vary greatly across size, scale, and scope of farm operations.
This lack of homogeneous data availability is of particular concern in southern states, which have a significant number of scale-dependent specialty crop operations as well as diverse grower demographics. Because these producers may be underrepresented in traditional credit data, important nuances could be lacking from the underwriting process, potentially reducing access to credit for producers in the region, particularly where inequalities in credit access have existed historically. For example, data from the FINBIN database shows that over the last 20 years, Minnesota dairy farms consistently held the least working capital as a percentage of gross revenue compared to other farm types. This can be expected due to dairy farms receiving monthly milk checks and the fact that most assets held by dairy farms are intermediate or long-term assets. But unless an AI underwriting model is trained using this type of historic data and provided context about the industries, it could unfairly penalize a dairy farm for having lower working capital as a percentage of gross revenue when comparing to other farm types or all farm types averaged together.
AI underwriting models can also unfairly penalize farms when they lack personal context. Consider a farm that had to temporarily scale back the operation because the main operator needed major health treatments, or crop farmers facing their third or fourth consecutive year of losses because of low commodity prices. The farms’ historic data could reflect poorer performance than usual and cutting off capital access for the farms in these temporary downturns could be disastrous. While AI can help identify patterns, it takes a relationship with a lender to understand the full story, not only of the operation but also of the local market conditions surrounding it and evaluate how these conditions realistically impact a credit decision.
Agricultural lending has always been an art as much as it is a science. Numbers gathered from tax returns and income statements only tell part of the story. The introduction of AI in underwriting can make telling that story easier and more efficient for loan underwriters. However, AI cannot replace the relational aspect of lending that can provide context and insight where numbers cannot do justice in giving the whole picture of the farm and its financial situation. It is important that policymakers take these considerations into account and recognize that AI may be most valuable as a decision support tool that assist human judgment rather than as an autonomous decision maker in credit underwriting in the agricultural sector. In conclusion, AI tools can be helpful, but the lending officer should be the one to fill in the context AI lacks to make the final decision on loan approval.
Authors: Bridget Yeboah Bafowaa, Ph.D. Student, Department of Agricultural Economics and Rural Sociology, Auburn University
Adam N. Rabinowitz, Associate Professor and Extension Economist, Department of Agricultural Economics and Rural Sociology, Auburn University
The United States Department of Agriculture (USDA) recently announced the election and enrollment deadlines for the Price Loss Coverage (PLC) and Agriculture Risk Coverage (ARC) programs for the 2026 and 2027 crop years. For the 2026 crop year, elections and enrollment are open now through December 11, 2026, while the 2027 decision period runs from November 2, 2026, through March 15, 2027 (see USDA 2026). As producers prepare to make these decisions, it is important to recognize that program election can have significant financial implications.
PLC is designed to stabilize income from low prices, while ARC covers revenue shortfalls from a combination of low prices and/or low yields. However, decisions are often cited as difficult for producers and can also be related to prior experiences (Fischer, et al. 2022 and Taylor, et al. 2017). Furthermore, when the enrolled program triggers a lower payment than the alternative, the payment received provides less income support than is optimal.
For the 2025 crop year, producers made decisions under the 2018 Farm Bill parameters prior to April 15, 2025. However, with the passage of the One Big Beautiful Bill Act (OB3) in July 2025, the ARC/PLC program payments for the 2025 crop year were changed to be based on (1) new reference prices, (2) an adjusted ARC formula, and (3) the higher payment rate of ARC or PLC. To illustrate the challenges producers face in making optimal election decisions, this article analyzes the PLC and ARC-County programs for the 2025 crop year. We focus on the outcomes for the five major southern crops: corn, peanuts, soybeans, wheat, and seed cotton for the 13 southern states under the 2018 Farm Bill and under the changes introduced by OB3. For crops within a county that had both irrigated and non-irrigated ARC rates, the overall ARC rate was calculated as the weighted sum of the irrigated and non-irrigated ARC rates, using the respective percentages of irrigated and non-irrigated acreage. For counties and crops designated as “all,” the ARC rate was set equal to the corresponding ARC rate for “all.” There is no consideration of payment limits because our analysis focuses strictly on county-level payment rates.
2018 Farm Bill Elections
The first part of this analysis illustrates the percent of enrolled acres in each program in the southern states that would have received the higher payment rate had the 2018 Farm Bill still been in place for the 2025 crop year payments. As shown in Table 1, the total base acres across the South that were enrolled in ARC was 20,632,981, of which 74.4% are in counties where ARC would have received the higher payment rate under the 2018 Farm Bill. There are, however, vast differences across commodities. Nearly all of the soybean acres that enrolled in ARC would have received the higher payment, but it is vastly less for corn and wheat where only about 2/3 of the ARC enrolled acres would have received the higher payment. Only about 1/3 of ARC enrolled seed cotton and peanut base acres would have had the same outcome. Meanwhile, the outcome for those that enrolled in PLC was less favorable, with only 51.1% of those base acres across the south receiving the higher payment rate. In other words, using 2018 Farm Bill parameters alone (without the OB3 improvements), 48.9% of the base acres enrolled in PLC would have been better off had it been enrolled in ARC for the 2025 crop year. Fortunately for these producers, OB3 created the situation for the 2025 crop year where they will receive the higher of the ARC or PLC payment rate, regardless of their election.
Table 1. Base Acres Enrolled in ARC and PLC for the 2025 Crop Year with Percent Receiving Optimal Payment Rate
Covered commodity
Total base acres enrolled in ARC
Percent of enrolled base in ARC when ARC would have received the higher payment rate under 2018 Farm Bill
Total base acres enrolled in PLC
Percent of enrolled base in PLC when PLC would have received the higher payment rate under 2018 Farm Bill
Corn
5,699,520
62.2%
2,028,292
52.1%
Peanuts
23,417
39.2%
2,253,098
51.3%
Seed Cotton
1,324,425
30.5%
9,567,675
61.3%
Soybeans
6,586,209
99.8%
566,848
10.0%
Wheat
6,999,410
68.9%
7,466,245
40.8%
Total
20,632,981
74.4%
21,882,157
51.1%
Note: The analysis includes Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia. Total base acres enrolled refers to acres enrolled in each program for each crop across the counties within these states. All values are rounded.
Payments under the One Big Beautiful Bill Act
The second part of this analysis focuses on the higher of the ARC or PLC payment rate under the OB3 changes. In this case, the actual enrollment decision does not matter. Figure 1 shows county maps for the selected commodities with shading in green when PLC payment rate is higher and red when the ARC payment rate is higher. For corn and soybeans, the majority of the counties where base acres exist are receiving a higher payment rate under the ARC program; however, there are large areas in Texas, North Carolina, and Virginia where PLC triggered the higher payment rate for corn. Meanwhile, peanuts, seed cotton, and wheat are predominantly being paid under PLC, although areas of Kentucky, Tennessee, and Alabama have many counties where ARC triggered the higher payment rate for wheat. This illustrates the potential under all crops for producers in a given county to benefit more from one program compared to the other. It is also important to note that the analysis in this article is at the county level. While ARC payments for a covered commodity are the same for all producers in a county, PLC varies based on an individual producer’s PLC yield. Figure 1. County Map Illustrating the Higher of the ARC or PLC Payment Rate by Covered Commodity under the One Big Beautiful Bill Act.
Payment Differences between ARC and PLC
Enrolling in a program that does not generate the maximum payment rate results in a smaller safety net. To quantify the additional payment rate that would result from receiving the higher of ARC or PLC, we calculate the payment difference between the higher payment and the alternative. Figure 2 reports these differences using payment rates calculated under the OB3 provisions. Payment differences vary considerably, with additional payment rates over $350 per base acre for a crop in one county. Smaller additional payment rates were more common. A total of 59.2% of the differences were less than $44.30 per base acre, although only 9.5% were under $5 per base acre. This means that if producers had to make a choice in 2025 under the new OB3 provisions AND they made a choice that did not maximize payments, the payment difference would have been over $44 per base acre over 40% of the time.
Figure 2: Distribution of 2025 Crop Year Payment Rate Differences (Higher Payment Rate minus Alternative Payment Rate) Under One Big Beautiful Bill Act.
Conclusion
This analysis indicates the potential payment inefficiencies producers may face when needing to choose between ARC and PLC. The additional payments can have significant financial implications in covering farm income during times of financial struggles due to low prices or revenue shortfalls. Without changes in legislation, it is important for producers to take the time to make an informed decision regarding their ARC and PLC elections. Several decision guides are available to assist in making decisions, such as the Texas A&M Agricultural and Food Policy Center (AFPC) Decision Tool (Fischer et al., 2026). Even though no decision tool can guarantee the maximum payment, these resources can provide useful information for evaluating the relative advantages of both programs.
Producers should also understand the fundamental differences between ARC and PLC. PLC provides price support and triggers payments when the marketing year average price (MYA) falls below the effective reference price. Conversely, ARC provides revenue support and triggers when revenue falls below a guaranteed revenue level, with revenue determined by county yield (and not individual yield for ARC-County) and MYA prices. Consequently, if producers expect prices to decline while yield remains stable, PLC might be the optimal choice if prices are expected to be low enough to trigger a payment. If a decline in county yields is expected, particularly when accompanied by lower prices, ARC may provide a higher payment. In addition to using a decision guide, producers should consider both their price and yield expectations and the information that is available to them for the crop year when evaluating which programs may provide a greater farm safety net.
References
Fischer, Bart L., Henry L. Bryant, James M. Raulston, and Joe L. Outlaw. 2022. “Simplifying the farm safety net.” Applied Economic Perspectives and Policy 44(3): 1195–1203. https://doi.org/10.1002/aepp.13268
One Big Beautiful Bill Act. (2025). H.R. 1, 119th Cong., 1st sess.
Taylor, Mykel R., Glynn T. Tonsor, Nicholas D. Paulson, Brenna Ellison, Jonathan Coppess, and Gary D. Schnitkey. 2017. “Is it Good to Have Options? The 2014 Farm Bill Program Decisions.” Applied Economic Perspectives and Policy 39(4): 533–546. https://doi.org/10.1093/aepp/ppx044
It is not news that Brazil is a powerhouse in agricultural production, especially in soybeans and corn. What may be news for some is that these two crops are strongly linked in Brazil through double cropping. About 80% of Brazil’s corn production happens as a second season corn crop, known as safrinha, immediately after the first crop (usually soybeans) is harvested. However, corn being planted as a second crop is more exposed to weather risk in the most sensitive growth stages of the plant (flowering and grain filling), so it is crucial that it is planted within the low-risk window, and that the first crop is harvested on time. With a record El Niño forming, the question for 2027 is whether the safrinha will be planted on time.
In Mato Grosso (Brazil’s largest corn-producing state), soybeans are planted mostly from mid-September through October and harvested mostly in January and February. However, the low-risk window to plant safrinha, defined by Brazil’s official agricultural climate risk zoning (ZARC), ends between early February and mid-March, depending on the region and the soil.
According to the National Oceanic and Atmospheric Administration (NOAA), there is a 75% chance that the current El Niño will be the strongest recorded since 1950 for the October through December period, which coincides with Brazil’s soybean planting. If the consistency of rainfall is affected by this phenomenon, it can delay soybean planting, especially because most of the soybean area in Brazil is rainfed. If soybean planting is delayed, it may push safrinha corn planting to later dates, which could be after the low-risk window closes. Figure 1 shows the share of safrinha area planted after the low-risk window and the yields compared with the trend for each season in Mato Grosso.
In a typical season, about 3 in 10 hectares of safrinha in Mato Grosso are planted after the low-risk window (Figure 1). In the 2015/16 season, the last very strong El Niño, almost half of the safrinha area was planted after the window. Planting was also heavily delayed in 2020/21, a La Niña year: almost 55% of the area was planted after the low-risk window. In both years, we had a significant decrease in yields compared with the trend: a 29% reduction in 2015/16, and almost 13% in 2020/21. In the other nine seasons, yields stayed near the trend. In 2015/16, El Niño also affected the growing season, which may have compounded the effect of delayed planting: rains that usually last until late May stopped in April, when most of the corn in Mato Grosso was flowering and filling grain (Agrolink, 2016a, 2016b).
However, El Niño does not always delay the safrinha. In 2018/19 and 2023/24, weaker El Niño seasons, only 18% and 23% of the area, respectively, were planted after the window, and yields were at or above the trend. What matters is whether the rains start late, which El Niño makes more likely, but does not guarantee.
Figure 1. Share of Safrinha Corn Area Planted after the Low-Risk Window (Upper Panel) and Yield Deviation from the 2005/06–2024/25 Linear Trend (Lower Panel), Mato Grosso, 2015/16–2025/26
Sources: Imea (Mato Grosso’s Agricultural Economics Institute); IBGE-PAM; Conab. Notes: Low-risk window: Brazil’s official agricultural climate risk zoning (ZARC) for early-maturing corn on soils of average water-holding capacity, 20% risk of loss, set for each Imea region. ENSO phase: Based on NOAA’s October–December Relative Oceanic Niño Index (RONI). El Niño and La Niña refer to conditions during October–December, not necessarily the entire crop season. “Other” includes La Niña and neutral conditions. 2025/26 yield is a Conab estimate.
If the 2026/27 El Niño affects the soybean planting and harvesting time, pushing safrinha to be planted after the low-risk window, Brazil may have lower yields again. This can impact global prices and quantities as most of Brazil’s corn exports come from the safrinha, which competes directly with U.S. corn in the second half of the year. If the El Niño effects last for more than just the soybean season, reaching the safrinha growing season, the effect may be even larger. It is still too early to say what the effects of this El Niño will be, but three numbers are worth watching: soybean planting in Mato Grosso by October 31 (in 2015/16 and 2020/21, it was below 60%), the soybean harvest by February 20, and rainfall in April, when the safrinha is flowering.
Instituto Brasileiro de Geografia e Estatística (IBGE). (2026). Produção Agrícola Municipal, Tabela 839: Área plantada de milho 2ª safra [Municipal Agricultural Production, Table 839: Second-crop corn planted area] [Data set]. SIDRA. Retrieved September 28, 2026, from https://sidra.ibge.gov.br/tabela/839
Instituto Mato-Grossense de Economia Agropecuária (Imea). (2026). Boletins semanais e estimativas de safra de milho, safras 2015/16 a 2025/26 [Weekly corn bulletins and crop estimates, 2015/16 to 2025/26 seasons].
While cattle prices have declined sharply in a break that began in July, there has been little action in the boxed beef cutout. The cutout value represents the value of the seven primal beef cuts in the proper proportion of the carcass. It is often thought of as a good, representative wholesale value of the beef carcass.
The Choice cutout declined from $397 per cwt the week before the July 4th holiday to $363 per cwt by the first week of August. Declining cutout values in the weeks following Independence Day are the normal seasonal price pattern. Cutout values tend to increase leading up to Labor Day, and they did this year, climbing back to a weekly average of $388 by late August. The Choice cutout has since remained in a range between $374 and $379. While the cutout remains at historically high levels, it has been below last year’s Labor Day surge. So, comparing to last year’s price levels is a bit misleading in that last September’s $413 per cwt price peak was a record high, except for two weeks in May 2020 when COVID shut packing plants.
Among the primal cut values that make up the cutout, the primal rib is the only one climbing in value and hitting new highs for the year. The Choice primal rib value hit $667 for the week of October 2nd. The rib normally increases seasonally up to the Thanksgiving and Christmas holidays. Most of the rest of the primal cuts, round, brisket, plate, and flank, have been declining in value as they often do seasonally. The decline in the primal flank value has been especially striking, down 27 percent compared to a year ago and 36 percent lower than its peak in June.
Where to From Here?
The Choice beef cutout tends to decline slightly in the Fall. Beef production normally increases and changing consumer purchases often result in lower prices. This Fall, continued tight beef supplies should keep the overall cutout value historically high even if we do see some seasonal declines. Given the normal seasonal pattern, it is difficult to see sharply higher cutout values boosting cattle prices.
Authors: Authors: Adam N. Rabinowitz, Associate Professor and Extension Economist, Agricultural Economics and Rural Sociology, Auburn University
Sthefani Oliveira, Research Associate, Agricultural Economics and Rural Sociology, Auburn University
When planning for the crop year, farmers create budgets and form expectations of costs that will occur during the growing season. With row crop harvesting in full swing, one critical cost is piling up faster than expected with every pass of the field. Diesel fuel is the most significant consumable input during row crop harvesting. It powers the combine, tractor, cotton picker, and peanut diggers. It is also a critical input to transporting grain to the elevator, cotton to the gin, and peanuts to the buying point. Last year, farmers set budget expectations for this growing season that included fall harvest, but diesel prices across the South have since risen about 75%.
The United States Department of Agriculture Agricultural Marketing Service publishes regular estimates of #2 diesel fuel for farm use in Alabama and Oklahoma. The U.S. Energy Information Administration also publishes estimates of the retail price of #2 diesel fuel at the Gulf Coast. Figure 1 shows the price in Oklahoma at the beginning of September was $4.72 per gallon, while in Alabama it was $4.86 per gallon. Meanwhile, the retail price at the Gulf Coast was $5.75 per gallon. This was up from the average January-February 2026 price of $2.68 per gallon in Oklahoma, $2.73 per gallon in Alabama, and $3.33 per gallon at the Gulf Coast. Those early 2026 prices were the lowest observed prices in several years, concluding a relatively stable period from the fall of 2024 through February 2026. That changed in March because of the conflict with Iran and continued war between Russia and Ukraine. Prices have continued to rise since.
Focusing on the four major row crops grown throughout the South, Table 1 shows the number of acres that are expected to be harvested this fall. Roughly 32 million acres of corn, cotton, peanuts, and soybeans are expected to be harvested across the 13 southern states. We also calculated estimated diesel use during harvest, in gallons per acre, for each of these crops. Not all southern states included in the acreage numbers publish estimates of diesel use for harvest equipment. To account for this, along with variations in machinery efficiency and farms within a state, we compute an Olympic average of the diesel use from the available estimates to apply across the South.
The estimated diesel price increased by approximately $2.20 per gallon, based on the average of the Alabama, Oklahoma, and Gulf Coast prices and the change from January/February ($2.91/gal) to September ($5.11/gal). This represents roughly a 75% increase relative to the diesel prices producers observed during the budgeting period. As shown in the last two columns of Table 1, the harvest cost increase from the rise in diesel prices is estimated at $214 million for the four row crops discussed. This is a harvest-acreage-weighted average increase of $6.67 per acre across the South, ranging from $5.56 per acre for corn to $18.83 per acre for peanuts. The main reason peanut diesel use and cost per acre are much higher than for other crops is that peanuts require two passes for harvest. The first pass is for digging and inverting the peanuts so they may dry, while the second pass is for picking.
The estimates presented here should be viewed as conservative because they exclude diesel used for transporting harvested commodities, which depends upon the distance to the delivery point. Several states (Alabama, Arkansas, Georgia, Louisiana, North Carolina, Oklahoma, Kentucky, and Texas) have issued directives to temporarily halt enforcement of penalties for using tax-exempt off-road diesel on public roads. This can reduce the price for transporting commodities by the amount of the federal and state taxes. Meanwhile, Georgia has also suspended the state excise tax on regular gasoline and on-road diesel. At the federal level, President Trump issued an executive order on October 5 that temporarily allows the use of off-road diesel on U.S. highways in addition to deferring the federal excise tax of 24.4 cents a gallon. There is still further federal action necessary to provide forgiveness of deferred taxes.
Some might counter that the price for these crops has also increased since planting expectations were formed. While this is true, the reality is that the increase was from an extremely low price point that for many was already below cost of production. Thus, one should not see market price increases as an offset to unexpectedly high diesel prices, but instead as continued pressure on already tight margins. Margins that get squeezed even more when one considers the increases in diesel costs for post-harvest transportation necessary for marketing. For more insight on marketing of row crops, check out the Wednesday Southern Ag Today articles.
Figure 1. Number 2 Diesel Prices: Alabama, Oklahoma, and Gulf Coast: February 2022-September 2026
Source: USDA Agricultural Marketing Service and U.S. Energy Information Administration
Table 1. Estimated Increase in Cost of Diesel for Harvesting Southern Row Crops
Crop
Expected Harvest (1,000 acres)
Estimated Diesel Use for Harvest (gal/acre)
Harvest Cost Increase from Diesel
Cost Increase Per Acre
Corn (grain)
9,424
2.53
$52,438,089
$5.56
Cotton
7,540
4.38
$72,550,508
$9.62
Peanuts
1,334
8.56
$25,124,055
$18.83
Soybeans
13,785
2.11
$63,818,990
$4.63
Unexpected Harvest Cost Increase
$213,931,642
$6.67
Source: Expected harvest from USDA Crop Production September 2026 report. Estimated diesel use based on enterprise budgets for Alabama, Arkansas, Georgia, Mississippi, South Carolina, and Tennessee. Harvest cost increase based on average diesel increase of $2.20/gal from Jan/Feb 2026 to September 2026 across Alabama, Oklahoma, and Gulf Coast.
