2014 Farm Bill Crop Commodity Program Payments in Idaho

There are two primary objectives for this study. The first is to describe and examine the variation of the ARC-CO and PLC program payments for Idaho over 2014–18.¹ The investigation of the payment data over time and across crops will provide insights into whether the programs functioned as expected, especially when also taking into consideration the adjustments in prices. The second is to estimate the extent to which changes in yields and prices explain adjustments in program payments over this period. Notably, the empirical methods employed allow for direct comparison of the estimates of the price and yield effect on program payments.

This study will provide fundamental background information for producers to use when making program decisions under the 2018 FB. One of the key logistical modifications to the ARC-CO and PLC programs between the 2014 and 2018 FB is that producers have more opportunities to change their program elections, which also implies that they more regularly need to reassess their current choices in light of changing market conditions. Specifically, under the 2014 FB, producers chose one program that applied for the entire 2014–18 period. Under the 2018 FB, producers chose by March 2020 for the 2019 and 2020 crop years, and will make elections each year from 2021 to 2023 (USDA-FSA 2019a).

The empirical analysis of this study focuses on barley and wheat. These crops were chosen for emphasis because they are the “covered commodity” crops with the highest cash receipts for Idaho producers.² Specifically, as of 2018, total cash receipts for Idaho barley producers were over $216 million and those for wheat producers were over $523 million (USDA-ERS 2018).

In addition to providing a statewide overview of payment patterns and determinants of payments, an empirical analysis was also conducted on a regional level to account for the wide variation in Idaho producer practices across regions of the state, particularly with respect to dryland and irrigated conditions. Xu et al. (2014) created a map that shows in detail how agricultural cultivation under irrigation is more widespread in eastern and southern Idaho than in western and northern Idaho.³ These differences in production practices across state regions imply that fluctuations in yield caused by changes in weather-related growing conditions are more pronounced in western and northern Idaho than in eastern and southern Idaho. Since yield is a primary variable that influences farmer revenue, such regional differences in production practices plausibly lead to expectations that the degree to which price and yield influence ARC-CO payments also varies across regions.

∧ back to top

Review of ARC-CO and PLC Programs

Under the 2018 FB, eligible producers⁴ have the opportunity to choose between the ARC-CO and PLC programs for each farm and covered crop combination. To make a choice between programs, a producer needs to consider their farm and county production characteristics as well as current and historical prices, all of which are subject to change over time. Payments under both programs are based on 85% of historical base acres rather than actual production (USDA-FSA 2019a). ARC-CO payments are revenue-based, in that a payment is initiated when the actual farm revenue is less than the county revenue guarantee. The guaranteed revenue level is calculated using historical countywide production and national price data. Thus, ARC-CO payments depend on both prices and production. Since historical base acres are fixed for the 2018 FB period, the primary variables that can influence ARC-CO payments are price and yield. For PLC payments, the actual payment amounts are influenced by historical base acres and program payment yields, both of which are fixed values for any program year.⁵ Whether a PLC payment occurs or not depends upon price changes and not yield fluctuations.

The specific ARC-CO and PLC payment formulas are described in detail in CRS (2019). The payment formulas are provided here in a slightly simplified form to emphasize the key variables that influence whether or not a payment occurs. The streamlined payment formulas are

1.   arc = [RG – (p × y)] × 0.85,
2.   plc = (P* – p) × Y × 0.85,

where lowercase terms indicate variables and uppercase terms represent fixed values, such that arc and plc are per base acre payments under ARC-CO and PLC programs, respectively, RG is a countywide revenue guarantee that is based on historical price and yield data, p is the higher of the US marketing year average (MYA) price or the loan rate set in the 2018 FB, y is actual county yields, P* is an effective reference price for which the formula is set within the 2018 FB, and Y are farm PLC program yields that are fixed for a given crop year. Hence, the only variables that can change within and across crop years in equations (1) and (2) are arc, plc, p, and y.

Given the observed effects of prices and yields on program payments, estimation of the the relationship among these variables is a key focus of this study. While the relationships appear direct such that it is expected that there are strong relationships among the variables, there are some statutory program characteristics that can cause the variable relationships to become independent. For example, both the ARC-CO and PLC programs have a payment limit of $125,000/person or legal entity per year (USDA-FSA 2019a). This cap on payments means that in some cases payments would stay capped at $125,000/person or legal entity even if prices continued to decline to a level that would have resulted in payments greater than the limit amount. If that scenario was fairly common, then, in the aggregate, the actual payment amount values would have a weaker relationship with prices than would be the case without the payment cap.

∧ back to top

ARC-CO and PLC Program Payments under 2014 FB in Idaho

The first step in the empirical analysis was to organize the data on ARC-CO and PLC program payments for Idaho by crop to examine how these payments changed over the 2014–18 period. Focus is placed on the state total payment amounts for select crops for both programs for each year over the 2014–18 period and the total payments. The tables in Appendix A provide the payments for a larger set of crops by county.

Regarding the statewide program payments data in Table 1, Idaho producers received over $81 million under ARC-CO and over $202 million under PLC in total over the full program period. ARC-CO payment rates per acre have a cap set at 10% of the county revenue benchmark, while PLC payment rates do not have a per-acre cap. Hence, it is more likely for relatively large payment values to occur under PLC than under ARC-CO. In addition to these overall program payment amounts, ARC-CO payments occurred for more crops than was the case for PLC, which is reflected in both Table 1 and in the county-based tables in Appendix A. Regarding payments over time, ARC-CO payments exceeded those for PLC in 2014, but the reverse was the case for every year from 2015 to 2018. Additionally, payments under ARC-CO peaked in 2015 and declined afterward, while those for PLC peaked in 2016 and were consecutively smaller for the 2017 and 2018 years. Among crops, payments were largest for both ARC-CO and PLC for barley and wheat, although the shares of payments for these crops was more even for ARC-CO than PLC. For PLC, wheat accounted for nearly 80% of all payments. PLC payments for barley were a distant second, comprising about 13% of total payments.

Source: USDA-FSA.
Note: ARC-CO payments for barley, oats, and wheat were aggregated across irrigated, nonirrigated, and for all formatting purposes. Payments demarcated by these categories are in Appendix A. Several crops were excluded from the table due to their relatively low values in most years for both ARC-CO and PLC. These omitted crops include canola, flaxseed, mustard, safflower, sorghum, soybeans, and sunflower. These omitted payments are available in the USDA-FSA ARC/PLC program datasets.

Figure 1 is a plot of US MYA prices for barley and wheat for 2014–18. The observed movements in MYA prices help explain the changes in magnitudes of payments under PLC over this period. For barley, the MYA price was highest for 2015, and no PLC payment occurred that year. However, the prices declined fairly precipitously between 2015 and 2017, and were slightly above their 2017 level in 2018. This decline in prices corresponds with the changes in PLC payments for barley with a large increase between 2016 and 2017, and then a small decrease between 2017 and 2018. For wheat, the MYA price was also highest in 2014, substantially lower in 2016, and then was higher but not up to the 2014 level in 2018. The large decline in the MYA wheat price between 2014 and 2016 corresponds with the dramatic increase in PLC payments over these years. The patterns observed in the data show that payments under the PLC program were highest under low price conditions, which is consistent with the design of the programs.

Econometric estimation can help to determine the extent to which variation in payments in Idaho under the 2014 FB is explained by variation in prices and yields. Although PLC payments are solely determined by price changes, estimation of the payment and price relationship was implemented to determine whether or not there were differences between barley and wheat that could be explained by differences in their market structures.

Data

Payments and yield data for the 2014–18 period for the ARC-CO and PLC programs used in the analysis were obtained from the USDA FSA (USDA-FSA 2019b). These datasets also included information on base acres, which were used to convert countywide total payments to payments per base acre. US MYA price data were gathered from the USDA Office of the Chief Economist World Agricultural Supply and Demand Estimates reports (USDA-OCE 2018).

Some counties were excluded from the dataset used in estimation due to the presence of both irrigated and nonirrigated production within the county. This was done because the variable of interest is payment per base acre. However, while payment and yield data are demarcated by irrigated or nonirrigated, data are not available for the number of base acres in each category. Hence, only the counties for which there is no demarcation between irrigated and nonirrigated production were included to obtain clearer values of payments per base acre. These exclusions, in addition to those omitted due to absence of data, limited the utilized dataset to 27 out of 44 Idaho counties.⁶

Summary Statistics

The standard deviation (SD), mean, and coefficient of variation (CV), which is the standard deviation divided by the mean, for the payment, price, and yield variables across all regions (counties for payments and yields, and nation for prices) and years are displayed in Table 2. Since the CV is a general measure of the degree of variation in a variable relative to the mean, comparing the CV across variables implies that payments are substantially more variable than prices and yield. This is explained in part by the presence of many cases of zero payments when the payment criteria are not met and some substantial payments in some counties and years. Additionally, yields are substantially more variable than prices. This is a result that follows from the nature of the yield data coming from counties with diverse production conditions and the prices being national averages across many markets such that much of the cross-market variation was already averaged out. Comparing across crops, barley payments and yields had more variability than wheat, while wheat prices had slightly greater variation than barley prices.

Variable Standardization

While it is of interest to determine whether and the extent to which yields and prices are associated with program payment variation, producers may want to know which of the variables is relatively more important in explaining payments. To help make this determination, the payment, yield, and price variables were standardized so that each has a mean of 0 and standard deviation of 1. Standardization essentially involved calculation of z-score values, which are the difference between observation value and sample mean divided by the sample standard deviation, for each observation. These standardized variables were then included in the regression models that estimate the relationships among program payments, yields, and prices. Since all variables are “unit free” after standardization, preestimation standardization can allow for direct comparison of the relative size of effects of price and yield on payments.

∧ back to top

Estimation Results

Since the data comprise a panel such that the variables are county-based for each year of the 2014–18 period, a fixed effects panel regression model was used. Essentially, fixed effects regressions account for factors that are different across counties but remain constant over time to measure more directly the relationships between payments, prices, and yields. More details of the estimated econometric models are provided in Appendix B.

The econometric models were estimated for barley and wheat for all included counties in Idaho, and then subsequently reestimated by each region, East, North, South, and West.⁷ Estimation by region was done because it was expected that the effect of yield variation on payments would be larger in the regions for which irrigated production is less widespread, the North and West, than in the more heavily irrigated East and South. The estimation results were summarized so that readers could more easily discern what the authors believe are the most important results. Complete estimation results are provided in Appendix C.

A summary of the regression results for the ARC-CO payments models are displayed in Table 3. The values in the table are absolute values of the estimated coefficients associated with the explanatory variables (price or yield), and are only presented if they were statistically significantly different from zero at the 5% significance level. The absolute values are presented because all estimated coefficients were negative, as was expected since payments decline (increase) with price/yield increases (decreases). The main takeaways from the results include whether or not the coefficient is statistically significant, which implies that the variable is important for explaining variation in payment amounts, and the magnitude of the coefficient. Even though standardization is helpful to view the relative magnitude differences between prices and yield, the interpretation is somewhat difficult to translate in practical terms. For example, the statewide results for barley imply that a 1 standard deviation (SD) change in the price translates into a 0.37 SD change in payments, while a 1 SD change in yield implies a 0.56 SD change in payments. While this may be interpreted as yield having a somewhat larger effect on payments than prices, one must also consider that yield has a higher SD than prices, which means that small changes in prices can have pronounced effects on ARC-CO barley payments. Regardless, more definitive results are found across regions such that ARC-CO payments for barley were determined almost exclusively by yield variation for the North and West regions, while both price and yield variation mattered statewide and in the East and South. The wheat results differed from the barley results in that both price and yield were found as important for explaining ARC-CO wheat payments statewide and in all regions. While acknowledging the differences in SD between price and yield, the larger magnitudes for the coefficients associated with wheat prices statewide and in all regions but the North imply that prices have been relatively more important for explaining ARC-CO payments than yield for wheat in all regions but the North. These results are consistent with expectations given differences in cultivation practices such that dryland wheat cultivation is relatively more common in the North.

The relationship between PLC payments and prices was also estimated in similar fixed effects regression models. Since price is the sole variable determinant of PLC payments, it may be expected that there would be a 1:1 correspondence between PLC payments and prices. However, statutory features of the PLC program such as the aforementioned payment limits per legal entity can cause payments and prices to stop moving together. The absolute values of the regression coefficients were 0.88 and 0.96 for barley and wheat, respectively. These variables were not far from the expected value of 1⁸, but since the barley coefficient was lower, it could be that there were more cases in which the payment limit was reached for barley than was the case for wheat.

∧ back to top

Conclusions

The goals of this study were to analyze how payments for the crop commodity support programs ARC-CO and PLC changed over the course of the 2014 FB period of 2014–18 among crops grown in Idaho, and to determine whether price or yield was relatively more important for influencing program payments for the key crops of barley and wheat. The relationship of ARC-CO payments with yield and prices for barley and wheat was estimated within a fixed effects regression model that accounted for countywide differences that are constant over time for both a statewide sample and by region. The regional analysis was done because the extent of production under irrigation and its impact on yield fluctuations due to weather changes varies across regions. Production under irrigation is more prevalent in the East and South, but relatively less so in the West and North. Standardization of the payment, yield, and price variables allowed for direct comparison of the relative effects of the variables.

Determinants of ARC-CO payments varied between barley and wheat in that yield effects were found to exceed price effects in explaining the variation in ARC-CO payments for barley at the statewide level and in nearly all regions. For barley, only yield was found to have a statistically significant relationship with ARC-CO payments for the North and West regions, while both price and yield effects were important in the East and South. Regarding wheat, both yield and price effects were statistically significant at the statewide level and for all regions. Price-coefficient magnitudes for wheat were roughly twice those of yield in several regional cases, suggesting that program payments were more determined by price than yield changes.

There are several implications from these results that can help Idaho producers understand 2014–18 ARC-CO payments, and form expectations for ARC-CO payments under the 2018 FB period of 2019–23. First, farm location greatly influenced the degree of yield or price effects on ARC-CO payments. Second, the relative influence of yield and price on ARC-CO payments varied across crops due to differences in their individual markets. These findings can help producers determine the likelihood of an ARC-CO payment occurring in advance of the release of the official FSA program payment, yield, and price data. While it is useful to form expectations, it is recommended that producers also use a decision aid such as the “Texas A&M University 2018 Farm Bill Decision Aid” or the “University of Illinois FarmDoc Gardner Program Payment Calculator” when electing either ARC-CO or PLC for the 2021–23 crop years. These decision aids encompass historical individual farm characteristics, countywide yields, national prices, and the program parameter information into an expected payment estimator. Each provides specific payment and probability estimates that are helpful supplements to general expectations.

∧ back to top

Appendices

Appendix A. ARC-CO and PLC payments by Idaho county for 2014–18. (PDF)

Appendix B. Description of estimation method. (PDF)

Appendix C. Full fixed effects regression estimation results. (PDF)

∧ back to top

Endnotes

¹The focus is on the ARC-CO and PLC programs and not the ARC-IC program because the USDA-FSA enrollment data utilized for the study show that only 2% of farms and 3% of base acres in Idaho were enrolled under that option for the 2014–18 period. Due to its unpopularity among producers, the other programs are analyzed instead. (return to text)

²Note that potatoes, hay, and sugar beets, which are a few of the other crops with the highest cash receipts, are not covered commodities. (return to text)

³See especially Figure 1 on p. 9686. (return to text)

⁴Program eligibility criteria are included in the 2018 FB, such as farms with more than 10 base acres, unless exceptions apply (USDA-FSA 2019a). (return to text)

⁵Owners have an opportunity to update PLC payment yields during the 2020 crop year. If these are not updated, then those under the 2014 FB still apply. If they are updated, then the updated yields apply for the 2020–23 period (USDA-FSA 2019a). (return to text)

⁶Those excluded for each region were, for East: Bannock, Bear Lake, Bonneville, Caribou, Franklin, Fremont, Oneida, Power, and Teton; for North: Bonner, Kootenai, and Shoshone; for South: Blaine and Custer; and, for West: Adams, Boise, and Washington. (return to text)

⁷The included counties in each region are, for East: Bingham, Clark, Jefferson, Lemhi, and Madison; for North: Benewah, Boundary, Clearwater, Idaho, Latah, Lewis, and Nez Perce; for South: Butte, Camas, Cassia, Gooding, Jerome, Lincoln, Minidoka, and Twin Falls; and, for West: Ada, Canyon, Elmore, Gem, Owyhee, Payette, and Valley. (return to text)

⁸Note that hypothesis tests were conducted regarding whether the estimated parameters are equal to -1. The t-statistic values for these tests were 2.60 and 1.46 for barley and wheat, respectively. Thus, with a t-table value at the 5% significance level of 1.96, the estimate for barley was statistically significantly different from -1, while that for wheat was not. (return to text)

∧ back to top

References

CRS. 2019. 2018 Farm Bill Primer: ARC and PLC Support Programs. Research Brief IF11161. Washington, DC: Congressional Research Service.

USDA-ERS. 2018. Annual Cash Receipts by Commodity. Data files. https://data.ers.usda.gov/reports.aspx?ID=17832.

USDA-FSA. 2019a. The 2018 Farm Bill—What Is New and What Has Changed. Issue brief. Washington, DC: USDA-FSA.

USDA-FSA. 2019b. ARC/PLC Program Data. Data files. https://www.fsa.usda.gov/programs-and-services/arcplc_program/arcplc-program-data/index.

USDA-OCE. 2018. World Agricultural Supply and Demand Estimates. Various reports. https://www.usda.gov/oce/commodity/wasde/.

Xu, W., S.E. Lowe, and R.M. Adams. 2014. “Climate Change, Water Rights, and Water Supply: The Case of Irrigated Agriculture in Idaho.” Water Resources Research 50(12): 9675–95.

∧ back to top

About the Authors

Patrick L. Hatzenbuehler—Extension Specialist, Crops Economics, University of Idaho (UI) Twin Falls Research and Extension Center

Benjamin Eborn—Extension Agricultural Economist, UI Extension, Bear Lake County

Ashlee Westerhold—Area Extension Economist, UI Twin Falls Research and Extension Center

BUL 969 | Published June 2020 | © 2022 by the University of Idaho

Issued in furtherance of cooperative extension work in agriculture and home economics, Acts of May 8 and June 30, 1914, in cooperation with the U.S. Department of Agriculture, Barbara Petty, Director of University of Idaho Extension, University of Idaho, Moscow, Idaho 83844. The University of Idaho has a policy of nondiscrimination on the basis of race, color, religion, national origin, sex, sexual orientation, gender identity/expression, age, disability or status as a Vietnam-era veteran.