Economics of Alternative No-Till Spring Crop Rotations In Washington’s Wheat-Fallow Region

Chapter 10 – New Technology Access, Adaptation and Economics no. 19, December 2003

Authors

Louis Juergens, Douglas Young, William Schillinger, and Herbert Hinman

L.A. Juergens, D.L. Young, and H.R. Hinman, Dept. of Agricultural and Resource Economics, Hulbert Hall 103, Washington State University, Pullman, WA 99164-6210, and W.F. Schillinger, Dept. of Crop and Soil Sciences, Washington State University, Dryland Research Station, Lind, WA 99341.

Abstract

Winter wheat – summer fallow (WW-SF) is the dominant cropping system in the low-precipitation (< 12 inch annual) region of the inland Pacific Northwest (PNW). Intensive tillage during summer fallow often leaves soil vulnerable to wind erosion. While no-till cropping is well known for wind erosion control benefits, previous research in the inland PNW showed that annual no-till hard red spring wheat (HRSW) trailed WW-SF in profitability by $46 /ac/year. Our objective was to assess the agronomic and economic feasibility of alternative no-till spring grain and oilseed rotations in a 5-year experiment near Ritzville, Washington. Spring crops were soft white wheat (SW), barley (SB), yellow mustard (YM), and safflower (SAF) grown in three rotation sequences. Net returns from WW-SF on ten neighboring farms during the 5-year period averaged $8.71 /ac /year. The most profitable no-till spring cropping sequence was continuous SW that averaged net returns of $4.90 /ac /year, statistically equivalent to WW-SF and much more competitive than previous HRSW results. No-till SW-SB and a 4-year rotation of SAF-YM-SW-SW averaged -$4.90 and -$12.73 /ac /year, respectively. Although all no-till spring crop rotations had higher annual income variability than WW-SF, positive net returns for continuous SW is the first economic “good news” for continuous annual cropping using no-till in the low-precipitation region of the inland PNW.

Abbreviations:
HRSW, hard red spring wheat; PNW, Pacific Northwest; SAF, safflower; SB, spring barley; SF, summer fallow; SW, soft white spring wheat; WW, soft white winter wheat; YM, yellow mustard.

Introduction

Potential for economic and environmental benefits is a driving force in the gradual shift by dryland farmers to adopt reduced-till and no-till farming methods. Despite several associated environmental problems, winter wheat-summer fallow (WW-SF) is the dominant cropping system in the low-precipitation zone of the inland PNW because it provides agronomic and economic advantages. Farmers and bankers appreciate time-proven grain yield and income stability of WW-SF and the system’s relatively uniform seasonal demands on farm machinery and labor.

Environmental disadvantages of WW-SF include recurrent wind erosion, especially during drought cycles when straw production is low. Summer-fallowed fields in south-central Washington were reported to have lost 1.6 to 3.9 inches (107 to 270 tons /acre) of topsoil from wind erosion in one season. In addition to degrading soil, blowing dust from summer fallow also inflicts substantial off-site damage on human respiratory health, causes traffic accidents, and incurs cleanup costs. Research in the PNW and elsewhere has shown that no-till cropping controls soil erosion, builds soil quality, and reduces machinery wear and fuel consumption when compared with tillage-based systems. More diverse cropping systems than WW-SF also offer opportunities for weed, disease, and insect control.

Nationwide, the advantages of annual cropping in semiarid regions have led to substantial adoption. Farmers in the USA reduced summer fallow acreage by 43% from 1964 to 1997, with the largest reductions occurring in the Great Plains. In 2000, about 36% of total U.S. cropland was in conservation-till or no-till, whereas in Washington state it was only 23%. In east-central Washington and north-central Oregon, where annual precipitation ranges from 6 to 12 inches and WW-SF cropping is practiced on 3.7 million acres, even minimum tillage is rare. In Adams County, Washington, where this study is located, conservation tillage is practiced on only 17% of the cropland.

Farmers in the WW-SF region are slow to adopt conservation tillage summer fallow despite conclusive research showing environmental benefits with no agronomic or economic disadvantages when compared with intensive tillage summer fallow. Concerns about economic risk and profitability appear to be a barrier to adoption of reduced tillage systems.

Few farmers in the PNW low-precipitation region practice continuous annual cropping. Two recent multi-year experiments in Washington compared profitability of no-till HRSW in 6-inch (Benton County) and 11.4-inch (Adams County) precipitation zones. In Benton County, 1997-2002 net returns over total costs before government farm payments averaged -$44 /ac /year for annual no-till HRSW and -$5.70 /ac /year for WW-SF. In Adams County from 1996-2002, the values were -$49 /ac /year for HRSW compared with $3.60 /ac /year for WW-SF. The average shortfall of -$45.73 /ac /year translates into -$183,000 /year for a typical 4000-acre farm in the region. The WW-SF system was not only more profitable than annual HRSW in both studies, but also demonstrated less annual income risk.

Given the unpromising economic comparison of annual no-till HRSW to WW-SF, a need clearly exists for alternative cropping systems that offer greater economic viability. The objective of this study was to evaluate the economic performance of three annual spring cropping systems involving soft white wheat, barley, yellow mustard, and safflower and compare them with performance of the WW-SF system practiced on neighboring farms.

Materials and Methods

Field Layout and Treatments

A 5-year study of no-till annual spring cropping systems was conducted from 1997 to 2001 at the Ron Jirava farm near Ritzville, in Adams County. Cropping systems were: i) a 4-year SAF-YM-SW-SW rotation; (ii) a 2-year SW-SB rotation and; (iii) continuous SW. The experiment covered 20 acres. The soil is a uniform Ritzville silt loam. Soil depth is more than 6 feet, having no restrictive layers, and slopes are less than 1%. Average annual precipitation at the site is 11.4 inches. The field where the experiment was established had been planted to spring wheat in 1996 following decades of WW-SF.

Experimental design was a randomized complete block with four replications. Each crop in all rotations occurred each year in 20- x 150-yard plots, making a total of 28 plots. During the first 3 years (1997, 1998, and 1999), all plots were planted and fertilized in one pass directly into the undisturbed soil and residue left by the previous crop using the grower’s Flexi-Coil 6000 air-delivery no-till drill equipped with Barton II dual-disk openers on 7.5-inch spacing for simultaneous and precision placement of seed and fertilizer in the same row. In 2000 and 2001, all plots were planted and fertilized in one pass using a custom-built no-till drill equipped with Cross-slot notched-coulter openers on 8-inch spacing for simultaneous and precision placement of seed and fertilizer in the same row. Both openers are low-disturbance and place fertilizer beneath and slightly to one side of the seed. Glyphosate herbicide (Roundup) was applied 2 to 4 weeks before planting at 16 oz/acre to control weeds and disease “green bridge.” Seeding rate averaged across years was 70, 70, 21, and 9 lb/ac for SW, SB, SAF, and YM, respectively. Solution 32 provided the base for liquid fertilizer to supply an average of 36 lb N, 10 lb P, and 15 lb S /acre. The quantity of available soil water and residual N, P, and S were measured in all rotations each spring to determine fertilizer needs based on a yield goal. Between the tillering and jointing phase of growth of SW and SB, in-crop broadleaf weeds were controlled with 8 oz 2,4-D + 0.03 oz/acre Harmony Extra. In-crop herbicides were not used in SAF or YM plots, as no legally labeled broadleaf weed herbicides were available for these crops in Washington.

All plots were harvested with a commercial-size combine, and grain yield was determined on site by auguring grain into a weigh wagon. When Russian thistle and other broadleaf weeds were present at time of harvest in cereals (1999 and 2001 only) and broadleaf crops (all years), 20 oz/acre Surefire (paraquat + diuron) was applied 7 to 10 days after harvest to prevent seed production and halt soil water use by these weeds. A complete list of field operations and timing for each operation throughout the study is shown in Table 1.

Economic Methodology

The machinery complement of farmer cooperator Ron Jirava was used for cost estimation: a 250-hp John Deere 4wd tractor; 30-foot-wide Flexi-Coil 6000 no-till drill with attached air cart; 150-hp John Deere 2wd tractor; 80-foot-wide sprayer with 850-gallon capacity tank; John Deere 8820 combine with 24-foot cutting platform; single-axle 300-bushel capacity grain truck; 850-bushel capacity tractor-trailer semi, one-ton pickup truck; and utility/service vehicle. The age, used or new purchase price, size, use, and service life of machinery were considered typical of the area.

Total cost of production was estimated using standard enterprise budgets that identify fixed and variable costs. For a given land and machinery base, fixed costs do not vary with number of acres planted. Machinery fixed costs are depreciation, interest, taxes, housing, and insurance. Land fixed costs include property taxes and net land rent. Net rent is money paid for rented land or rental income foregone for using owned land. In the study region net rent is based on the prevailing one-third landlord and two-thirds tenant crop share with the landlord also paying land taxes and one-third of fertilizer expense. Other fixed costs include farm-wide insurance, legal and accounting services, and overhead expenses.

Variable costs include any costs that vary proportionately with the area planted. Machinery repair, fuel, labor, custom hire of services, seed, fertilizer, pesticides, and crop insurance are typical variable costs. The actual operations and input rates for the 5-year experiment were used in computing variable costs.

Soft white wheat and feed barley prices used in this analysis are $3.36 /bushel and $84.10 /ton, respectively. These are the regional average 1997-2001 farm gate prices. Safflower and yellow mustard price of $0.12 /pound is the average contract price that regional farmers received during the period.

Net returns include only market returns, excluding government payments or crop insurance indemnities. Although government payments have been and are a very important source of farm income, our study compared rank in market profitability of different rotations, not total farm income. Adding recent predetermined government payments will not change the economic ranking of different treatments. Inclusion of government payments requires assumptions on historic grain yields and base acreage of individual “representative farms.” These histories vary from farm-to-farm, and government programs vary substantially annually and from farm bill to farm bill. Readers may add government payments to base market returns reported here consistent with their particular assumptions if desired.

Net return per rotational acre is used to correctly measure profitability of different crop rotations. Net returns for each crop year in the rotation are summed and divided by the number of years in the rotation, thereby standardizing all rotations to a 1.0-acre basis. For example, a rotational acre of WW-SF includes 0.5 acres of winter wheat and 0.5 acres of fallow. This approach also correctly portrays annual income of farmers who commonly allocate 1/n of their land to each crop in an n-year rotation. This annual diversification also reduces annual income risk by growing a “portfolio” of crops and permits more efficient use of machinery and labor over time.

Safflower was discontinued from the 4-year rotation in 2001, but the remaining crops of the original 4-year rotation were planted in the original sequence. To permit estimating profitability of the 4-year rotation for 2001, the profit for SAF was estimated based on its historic yield relationship with YM.

Although WW-SF was not included in the replicated experiment, economic comparison of this traditional system to the experiment’s no-till annual spring crop rotations was accomplished by conducting a multi-year grain yield survey of 10 WW-SF farmers within a 5-mile radius of the experiment site. A one-page mailed questionnaire was used with telephone follow-up as necessary. The sample size of 10 farmers represents 53% of the original mailing to 19 farmers. The 10 neighboring farmers had climate and soils similar to those on the experiment site. Of the 10 participating farmers, one reported on three different fields, with varying yields. This farmer’s data were added independently, increasing the sample size to 12.

The survey approach permitted observing variation of winter wheat yields over time and across farmers, as well as deriving average yields. Reported grain yields from the survey were divided into top, middle, and lower thirds to permit comparisons of each group with spring crop rotations from the experiment. Typical fixed and variable costs for WW-SF were computed from standard enterprise budgets developed for WW-SF for Adams County.

Results and Discussion

Variation in annual precipitation had considerable effect on spring crop yields (Table 2). Crop-year (1 Sept. to 31 Aug.) precipitation was 20.3, 11.1, 7.9, 9.1, and 8.0 inches for 1997, 1998, 1999, 2000, and 2001, respectively. The 5-year average annual precipitation of 11.3 inches was near the long-term average of 11.4 inches. Substantial variation in year-to-year precipitation is common in the region and underscores the importance of using several years’ data to accurately compare cropping systems. Yellow mustard exhibited relatively high yield variation, having a standard deviation (S.D.) of 518 versus a mean of 544 lb /acre. Grain yield of YM ranged from 1430 lb /acre in 1997 to 110 lb /acre in 1999 (Table 2). Safflower displayed relatively low yield variation with a S.D. of 339 and a mean of 896 lb /acre. We theorize that SAF experienced lower yield variation because it extracted soil water below the 6-foot depth (data not shown) with its long taproot. Although deep overwinter soil water recharge is rare, recharge occurred to a depth of 6-foot or greater during the wet 1996-1997 winter (data not shown). Spring wheat does not extract soil water below 5 feet, thus residual water below 5 feet was available to SAF during the first 3 years of the study.

Exceptionally high precipitation in the 1997 crop year resulted in high yields for all crops, but yields of all crops except SAF fell sharply during the low precipitation years. For example, SB yield fell 85%, from 2.3 ton/acre in 1997 to 0.35 ton/acre in 2001. Wheat following oilseeds in the 4-year rotation fared the worst of all crops in 2001, yielding only 8 and 6 bu/acre.

Results of a 1997-2001 WW-SF farmer survey are shown in Table 3. The WW-SF yields were obtained from farms within a 5-mile radius of the study site for the spring crop yields reported in Table 2. Soils of the surveyed farms are similar in texture and depth to those of the study site and are all classified as Ritzville silt loam. The weather station at the experiment site was located at the center of the 5-mile radius and is considered representative of the surveyed farms. Like the spring crop yields in Table 2, WW yield varied with annual precipitation. Highest yields were recorded in 1997 when precipitation was almost double the long-run average, and lowest yields occurred during the 2001 drought year. Over all farms and years, reported winter wheat yields averaged 56.8 bu /acre with a S.D. of 14.5 bu /acre. Average 5-year yields ranged from 50 (farmer 5) to 67.6 (farmer 4) bu /acre. Yield variation among farms is likely due to management and possibly to minor differences in microclimate. Annual average winter wheat yields ranged from 35.5 to 72.3 bu /acre. Comparison of Tables 2 and 3 reveals somewhat less annual yield variation over 1997-2001 in surveyed WW-SF yields than in the experiment’s spring crop yields. Dividing the WW-SF farmers into upper, middle, and lower thirds gives average yields of 65.6, 56.3, and 48.6 bu /acre. Standard deviations ranged from 12.9 to 14.7 bu /acre and are positively correlated with average yields.

Table 4 shows annual net returns per rotational acre for all no-till spring crop rotations and for the different yield groups of surveyed WW-SF farmers. Five-year averages and S.D.’s of net returns also are reported for each rotation. The WW-SF survey results were excluded from formal statistical comparisons of mean profitability because the survey results were not part of the replicated randomized complete block design of the experiment. Since the surveyed farmers represented more than 50% of the population of all WW-SF farmers within a 5-mile radius of the experiment, average returns WW-SF are treated informally as point estimates of the population means for this group of neighbors. Statistical comparisons of S.D.’s of profitability between the spring crops and WW-SF were not possible, but these S.D.’s permit useful informal comparisons of the economic riskiness of the different cropping systems.

Among the spring crop rotations, continuous no-till SW had the highest average net return at $4.90 /ac /year followed by SW-SB and SAF-YM-SW-SW at -$4.90 and -$12.73 /ac /year, respectively (Table 4). An LSD0.05 of $12.62 /ac /year indicates that the 4-year rotation was significantly less profitable than continuous SW. In addition, continuous SW exceeds SW-SB in mean profitability at the LSD0.10 level. Using this LSD to compare SW to WW-SF, with (assumed population) mean of $8.71 /acre, indicates equivalent net returns (Table 4). However, the top one-third of WW-SF farmers exceeded the average profit for continuous annual SW. Equivalent average profitability between no-till annual SW and WW-SF is a welcome result given the -$45.75 /ac /year shortfall in profitability of previous research comparisons of no-till HRSW with WW-SF.

The WW-SF system also was the least risky rotation over 1997-2001 with a S.D. of $14.96 /ac /year compared to $36.70 for SAF-YM-SW-SW, $40.86 for continuous SW, and $41.69 for SW-SB (Table 4). Farmers and lenders generally prefer cropping systems that sustain profitability and reduce economic risk. Results show that during 1997-2001 WW-SF had this advantage.

Low relative variance of SAF-YM-SW-SW is attributable to consistently negative, but slightly more uniform, net returns throughout the study period. In contrast, the other two spring crop rotations enjoyed positive net returns in 1997, 1998, and 2000 (Table 4). Yields in Table 2 drive the annual profit variation in Table 4. Drought depressed yields in 1999 and 2001 decreased average profitability and increased the economic riskiness of the three spring crop rotations (Table 4). While net returns for WW-SF were not immune from the 1999 and 2001 drought years, this rotation was able to withstand yield reductions to a greater extent when compared with annual spring cropping, especially in 1999.

The upper, middle, and lower thirds of the WW-SF survey showed average net returns over total costs per rotational acre/year of $18.51, $8.09, and -$0.49, respectively. Under the possibly untenable assumption that continuous SW yields could hold at average levels on the lower third of WW-SF farms, then continuous SW would exceed the estimated average profitability of WW-SF by $5.39 /ac /yr ($4.90 – [-$0.49]) on these farms.

Summary and Conclusions

The most promising result of this study was that continuous annual no-till soft white spring wheat was economically competitive with traditional winter wheat-summer fallow. Results may be somewhat robust considering that the 5-year study contained 4 years of below average precipitation and 2 major drought years. Two previous multi-yr studies in east-central Washington showed that no-till HRSW lagged WW-SF by $45.73 /ac /year.

Continuous no-till SW showed considerably more economic risk when compared with WW-SF. Future production and breeding research should focus on improving the yield stability of spring wheat under variable precipitation. Targeted agricultural policies such as “green payments” for no-till farming in areas vulnerable to wind erosion could also help tip the scale toward adoption of these soil conserving cropping systems. Subsidized crop insurance for farmers adopting no-till could also reduce their economic risk. A negative $47.19 /acre net return for SW, as evidenced in 2001, is an unacceptable risk for farmers, even if long-run average prospects are positive.

Given the potential for continuous annual no-till SW to markedly reduce dust emissions when compared with WW-SF, the equivalent profitability of these two systems provides the first reported potential “win-win” solution for no-till farmers and the environment in the low-precipitation zone of the inland PNW.

Acknowledgments

The authors gratefully acknowledge the cooperation of Ron Jirava on whose farm the research was conducted. Appreciation is extended to Washington State University agricultural research technicians Harry Schafer and Steve Schofstoll for their excellent support. Funding for the research was provided by the Columbia Plateau Wind Erosion/Air Quality Project and the Solutions to Economic and Environmental Problems (STEEP) Project.