Published January 25, 2017
CROP ECONOMICS, PRODUCTION & MANAGEMENT
ContinuousCornandCorn–SoybeanProfitsovera45-Year
TillageandFertilizerExperiment
AndrewTrlica,ManinderK.Walia,RonKrausz,SilviaSecchi,andRachelL.Cook*
O
ABSTRACT
Studies comparing profitability of tillage systems often examine
narrow historic windows or exclude annual price fluctuations.
This study uses a continuous corn (Zea mays L.) (CC; 1970–
1990) and corn–soybean [Glycine max (L.) Merr.] (CS; 1991–
2014) Tillage × Fertilizer study in somewhat poorly drained soils
in southern Illinois to reconstruct partial annual budgets with
historical prices for crops, fertilizers, lime, herbicides, fuel, labor,
and machinery. Combinations of tillage (moldboard plow [MP],
chisel tillage [ChT], alternate tillage [AT], and no-till [NT]) and
fertilizer (Control, N-only, N+NPK starter, NPK+NPK starter,
and NPK broadcast) treatments were evaluated. The CC profits were highest in NPK-applied treatments followed by N-only
and Control. The MP treatments were similar to ChT and more
profitable than NT, while AT fell between. In CS, NPK-applied
treatments were similar regardless of tillage. Combined costs for
herbicide, machinery, labor, and diesel were higher in MP and
ChT systems than AT and lowest in NT, but were a small percentage of total costs (26.6, 26.0, 21.5, and 18.2%, respectively).
Nitrogen fertilizer offered a return on investment of 396% in CC
and 133% in CS while P & K returned 78% in CC and 109%
in CS. Sensitivity analysis in CS showed that NT would be less
profitable than MP if herbicide costs increased 850%. A 300%
machinery cost increase would have made MP less profitable than
NT. These findings suggest that since 1991 CS under NT carried
the same potential for profit as other tillage systems under full
fertility management.
Core Ideas
• Cumulative profit (1970–2014) in no-till was similar to other
tillage types with NPK.
• Relative profits were more sensitive to changes in machinery
than herbicide cost.
• Return on fertilizer ranged from 56 to 251% for P and K and 69
to 434% for N.
Published in Agron. J. 109:218–226 (2017)
doi:10.2134/agronj2016.06.0377
Received 28 June 2016
Accepted 18 Sept. 2016
Available freely online through the author-supported open access option
Copyright © 2017 American Society of Agronomy
5585 Guilford Road, Madison, WI 53711 USA
This is an open access article distributed under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
218
ne of the greatest global challenges in coming
decades will be to produce more food from limited
land resources while maintaining environmental
quality. Tillage has been used in crop production for centuries
for weed control, release of nutrients, and seedbed preparation, but can lead to soil degradation and loss through erosion
and requires labor and fuel inputs (Mannering et al., 1987).
Agricultural soil degradation can jeopardize crop production
potential and generate negative environmental impacts that
include eutrophication of surface waters (Carpenter et al.,
1998) and release of stored C (Lal, 2004); excess on- and offfarm costs due to erosion were once estimated at US$44 billion
per year in the United States (Pimentel et al., 1995). Selection
of tillage systems therefore has significant implications for the
farm enterprise, both economically and environmentally (Liu
and Duff y, 1996). Methods for reduced tillage with modern
agricultural production were proposed as early as the 1920s
(Graber, 1928). No-tillage systems have been shown to be effective in the midwestern Corn Belt since the 1960s in corn and
soybean production for conserving soil moisture and reducing
fuel use and soil loss (Dick, 1983; Free et al., 1963; Triplett et
al., 1968). However, the acceptance of NT or other soil conservation practices depends in part on their economic returns
as compared with more intensive tillage systems (Cary and
Wilkinson, 1997). For a cropping practice to be sustainable, it
must also be profitable.
Crop producers invest in farm inputs such as labor, equipment, fuel, seed, fertilizers, and pesticides but face unpredictable yields and market conditions at harvest. No-till
systems could increase net profits by reducing costs associated
with labor, fuel, and machinery, but also typically require
increased investment in herbicides to combat weeds (Siemens
and Oschwald 1978; Brown et al., 1989; Koskinen and
McWhorter, 1986; Chase and Duff y, 1991; Doster et al., 1983).
The other major economic factor in any conservation tillage
system is crop yield, which must compare favorably to that of
a conventional system to compete in profitability. Some longterm studies in the midwestern Corn Belt region have shown
similar corn and soybean yields between NT and other tillage
A. Trlica, Dep. of Earth & Environment, Boston Univ., Boston, MA,
02215; M.K. Walia, Eastern Agricultural Research Center, Montana
State Univ., Sidney, MT 59270; R. Krausz, Dep. of Plant, Soil, and
Agricultural Systems, Southern Illinois Univ., Carbondale, IL 62901;
S. Secchi, Dep. of Geography and Environmental Resources, Southern
Illinois Univ., Carbondale, IL 62901; R.L. Cook, Dep. of Forestry and
Environmental Resources, North Carolina State Univ., Raleigh, NC
27695. *Corresponding author ([email protected]).
Abbreviations:AT, alternate tillage; CC, continuous corn; ChT,
chisel tillage; CS, corn–soybean; MP, moldboard plow; NT, no-till.
A g ro n o myJ o u r n a l • Vo l u m e10 9,I s s u e1 • 2 017
systems (Cook and Trlica, 2016; Karlen et al., 2013), while
other studies demonstrated reductions in yield, especially on
poorly drained sites or in sites under CC (Dick et al., 1991;
Griffith et al., 1988).
Studies based on the results of yield trials have estimated the
economic returns of different crop production systems in the
Corn Belt. Al-Kaisi and Yin (2004) evaluated corn yield and
economic return on several different soil types under long-term
NT management in Iowa, with NT providing as much or more
return as other tillage systems under CS rotation. An experiment on a sloping erodible site in southern Illinois showed
increased yield and economic returns in NT vs. MP and ChT
over several seasons (Phillips et al., 1997). Doster et al. (1983)
estimated greater return for NT in sandy well-drained soils,
but poorer performance than other tillage systems in poorly
drained soils. In a survey of Iowa farmers, Liu and Duffy
(1996) reported that most conservation tillage systems (no-till,
reduced-till, ridge-till, or mulch-till) had higher profits than
conventional tillage due to lower production costs, and also
provided environmental benefits.
Adoption of conservation tillage may involve slow shifts in several important factors in the crop production system that affect
yield and profitability, such as operator familiarity with the new
equipment and practices, weed and disease prevalence, and longterm changes in soil properties (Kumar et al., 2012). Overlaid
on changes related to tillage practice are fluctuations on both
short-term (e.g., weather, input prices, and commodity values)
and longer-term scales (climate change, improved plant genetics,
introduction of new technologies) that can also affect both yield
and profits. Any real-world farm in operation for more than a
few seasons will necessarily integrate both these short- and longer-term influences in their overall profits. However, most economic analyses of relative profitability of different tillage systems
have evaluated revenues and costs based on fixed prices and/or
yield histories of 10 yr or less (e.g., McIsaac et al., 1990; Phillips
et al., 1997; Al-Kaisi and Yin, 2004; Karlen et al., 2013). Annual
and longer-term variations in prices and yield have to our knowledge not been accounted for in any detailed study of crop system
economics in the Corn Belt. Moreover, there remains a general
perception that NT is attractive as a production system primarily
in erodible or well-drained soils (Triplett and Dick, 2008).
An experiment begun in 1970 in southern Illinois provides
a unique 45-yr historical record of crop yields and inputs for
several tillage and fertilizer treatments in a flat, somewhat
poorly drained site under CC (1970–1990) and CS systems
(1991–2014). The present study used these records to explore
tillage and fertilizer management influences on historic economic returns. The objectives of this study were to: (i) Evaluate
CC and CS annual profits and cumulative profits in different
tillage and fertility treatments; (ii) Determine return on investment for fertilizer additions; (iii) Examine the sensitivity of
historic profits to machinery and herbicide cost fluctuations in
NT and MP tillage treatments; and (iv) Estimate the relative
soil conservation benefits of NT and MP management. The
annual partial budget approach, combined with reconstructed
historic prices, allows the profitability of the different production systems to be evaluated within the actual economic environment in which crop production occurred.
Materials and Methods
Historic costs and revenues from corn and soybean production
were estimated from operational details and yield data collected
over 45 yr from a long-term Tillage × Fertilizer trial initiated
in 1970 on a somewhat-poorly drained Bethalto silt loam (finesilty, mixed, superactive, mesic Udollic Endoaqualf) at the
Belleville Research Center near Belleville, IL (exact field-plot
coordinates: 38.519179° N, 89.843248° W). Continuous corn
was grown on the site from 1970 to 1990, and CS rotation was
implemented from 1991 to 2014. The experimental design was a
split-plot with tillage as the whole plot treatment and five fertilizer treatments randomized within tillage strips as the subplot.
Tillage treatments for spring seedbed preparation consisted of:
moldboard plow (MP); chisel tillage (ChT); alternate tillage
(AT; with 2 yr no-till and 1 yr MP); and no-till (NT). Fertilizer
treatments consisted of: (i) unfertilized check (Control); (ii)
broadcast nitrogen only (N-only); (iii) broadcast nitrogen plus
NPK starter (N+NPKstarter); (iv) broadcast NPK plus NPK
starter (NPK+NPKstarter); and (v) broadcast NPK only (NPK).
Additional details can be found in Table 1. Each tillage × fertilizer combination was replicated in four blocks. Details of the
experimental treatments and analysis of yield and soil chemical
properties can be found in Cook and Trlica (2016).
Table 1. Fertilizer and tillage treatments 1970 to 2014. Note: No fertilizer was applied before soybean seasons (odd-numbered years after
1990) (from Cook and Trlica, 2016).
Fertilizer treatment
1970–1973†
1974–1999
2000–2014
———————————————————— N–P–K, kg ha–1 ———————————————————
Control
0–0–0
0–0–0
0–0–0
N-only
140–0–0
196–0–0
196–0–0
N+NPKstarter
129–0–0 + 11–25–24 (starter)
179–0–0 + 17–39–112 (starter)
196–24–140
NPK+NPKstarter
129–14–93 + 11–10–24 (starter)
179–25–140 + 17–14–28 (starter)
196–24–140
NPK
140–25–116
196–39–168
196–24–140
Tillage treatment
Moldboard (MP)
Chisel (ChT)
Alternate (AT)
No-till (NT)
Seedbed preparation steps
Moldboard plow, tandem disk, cultivator, culti-mulcher
Tandem disk, chisel plow, tandem disk, cultivator, culti-mulcher
Two years no-till, 1 yr MP treatment
None
† In 1973 total K application rates were 112 kg ha –1 in the N+NPKstarter and 224 kg ha –1 in the NPK and NPK+NPKstarter fertilizer treatments.
Agronomy Journal • Volume 109, Issue 1 • 2017
219
This study used a partial budget approach which addressed
factors that differed among treatments to permit comparison
of relative profits of varying crop production technologies
(Swinton and Lowenberg-DeBoer, 1998). Factors that were
identical across all treatments (e.g., seed and pre-emergent herbicide costs across treatments) were excluded from this analysis.
Recorded yield and cultivation practices for each plot were used
to represent varying production systems on a typical hectare of
farmland. Annual partial budgets for each Tillage × Fertilizer
combination estimated trends in income, costs, and profits that
varied among treatments over the 45-yr experiment. Treatment
variables that were tracked in the partial budgets were yield,
fertilizer, lime rate, post-emergent herbicide, machinery use,
fuel, and labor.
Yearly revenue for each experimental plot for 1970 to 2014
was estimated using measured crop yield and the national
average price of corn or soybean by marketing year (USDA
NASS, 2016a). National average nominal prices paid per
tonne of ammonium nitrate, urea, potash (potassium chloride,
60–62% K 2O), triple superphosphate (TSP, 44–46% P2O5)
(USDA ERS, 2016a) and agricultural lime (with application
costs included) (USDA NASS, 2016b) were taken for marketing year 2009 to 2014, calendar year 2002 to 2008, and month
of April 1970 to 2001. Costs for paraquat (1,1’-dimethyl-4,4’bipyridinium ion) and glyphosate [N-(phosphonomethyl)
glycine] used for post-emergence control of annual weeds in
NT and AT plots from 1991 to 2014 were estimated based on
records of herbicide application rates and the reported national
average prices for each active ingredient (USDA NASS, 2016a,
2016b). Herbicide prices prior to 1991 were back-projected
from the 1991 national average using the reported Agricultural
Chemicals price index (1977 baseline) (USDA NASS, 2016b).
Machinery use was modeled according to operational history for the different treatments and costs related to machinery
maintenance, depreciation, housing, insurance, and repairs.
Costs for each category were estimated based on 2015 per-area
machinery budgets that assumed the use of a 75 horsepower
diesel tractor, using the middle of the potential range of cost
per area for each tillage, planting, spraying, and fertilizer
implement type (Farmdoc, 2015). The size of the machinery
and implements used were factors in the cost estimate of these
budgets, but this study focused on estimated profits using a
single tractor size and the mid-range cost of each implement,
and as such the sensitivity of profits to machinery sizes was not
evaluated. Machinery use costs were projected for other crop
years using the national agricultural Machinery Totals price
(2011 baseline) (USDA NASS 2016a, 2016b). The machinery
budgets also provided estimates of fuel consumption and labor
time requirements for each machine activity, which were used
to calculate per-activity costs for each. Diesel fuel costs per
gallon were estimated based on the national average January
price (USDA NASS, 2016b). National agricultural hourly wage
labor rates reported by USDA NASS for “wages only” workers
(1970–1973), “machine operators” (1974–1979), “all workers”
(1980–1989), and “labor, hired” (1990–2014) were used to
estimate labor costs for each production system (USDA NASS,
2016a; USDA ERS, 2016b).
All input costs and revenues in each year were adjusted to
2014 U.S. dollars prior to calculating annual profits using the
220
Prices Paid Index for Commodities and Services, Interest,
Taxes, and Farm Wage Rates (PPITW) prepared by USDA
NASS. The 2011-baseline index was used for 1990 to 2014,
and from 1975 to 1989 the 1990-baseline PPITW records
from USDA NASS were recalculated to the 2011 benchmark
(USDA NASS, 2016a). Values of PPITW (1967 baseline)
reported in the Annual USDA NASS Agricultural Price
Summaries for 1970 to 1974 were recalculated to the 2011
baseline using the 1990-baseline index as an intermediary
(USDA NASS, 2016b). Annual net profit for each treatment
was estimated by subtracting 2014-adjusted historic costs from
revenue. A cumulative profit for each treatment subplot was
also calculated using the running sum of the annual profits.
Return on Investment
Fertilizers, particularly N, P, and K, are a critical component
of maintaining high yields and yield is a primary driver of
profit in agronomic systems. A comparison of profits between
Control and N-only treatments allowed for an assessment
of the return on investment for N fertilizer. By a similar
approach, the difference between N-only and NPK-applied
treatments allowed for an assessment of the percent return on
investment for P and K fertilizer. We used fertilizer treatment
profits within each rep and tillage to calculate percent return
on investment as the difference in profit (e.g., N-only minus
Control) divided by the cost of fertilizer.
Sensitivity Analysis
A sensitivity analysis was performed using the NPK broadcast treatments in the CS rotation to assess the robustness of
the profit results to changes in inputs costs. Elasticities were
calculated to judge the relative importance of different input
price fluctuations on profit in MP and NT as the “extreme”
treatments. Elasticities are a measure of the percent change in a
dependent variable (profit) divided by the percent change in an
independent variable (costs) (Pannell, 1997).
Based on elasticity results, we selected machinery and
herbicide costs as inputs to vary within MP and NT systems
and recalculated mean profit each year for each treatment. To
account for differing operational practices in corn and soybean
years, the 24 yr of CS rotation mean profits were segmented
into 12 consecutive CS “year-pairs” (e.g., 1991 soybean with
1992 corn, excluding the year-pair 2007/2008 due to weatherrelated loss of the 2008 corn crop prior to harvest). Each yearpair was considered an independent sample since there was no
significant linear effect of time. Tillage treatments (NT vs.
MP) were compared under different price scenarios using an
independent sample t test.
Soil Conservation Benefit Calculation
We compared estimated soil loss from NT vs. MP using
the universal soil loss equation with rainfall erosivity of R =
220, soil erodibility factor of K = 0.37, topographic factors of
LS = 0.1, and crop and management factor values of C = 0.32
for MP and C = 0.075 for NT (Walker and Pope, 1983). We
used an erosion prevention benefit estimate ($ tonne–1) for
the Corn Belt region (Hansen and Ribaudo, 2008) for the
social and environmental benefit of soil conservation by using
NT instead of MP treatments (e.g., reduced ditch clearance,
Agronomy Journal • Volume 109, Issue 1 • 2017
reduced water treatment burden, fishery productivity, reduced
flood damage; see Table 1 in Hansen and Ribaudo, 2008). We
excluded the value of any direct effects on soil productivity
since changes in yields and profits are already addressed in this
economic analysis.
Statistics
Profit was analyzed with a mixed linear model across each
rotation (CC, CS) including fixed effects of: Tillage, Fertilizer,
and Tillage × Fertilizer. Random effects were included for
the class variables Year (to account for year-to-year variability), Replicate (Rep), Rep×Tillage, and Year×Rep×Tillage (to
account for whole-plot error). The Kenward–Roger method
was used to estimate the denominator degrees of freedom of
the F statistics (Kenward and Roger, 1997). A repeated measure statement was included with a first order auto-regressive
[AR(1)] covariance structure with Rep×Tillage×Fertilizer as
the subject. Tukey’s HSD was applied for separation of means
with α = 0.05. Cumulative Profit was analyzed with a mixed
linear model based on the split-plot design at the end of each
rotation (CC, CS) and after the entire combined experiment
(CC and CS) with fixed effects of Tillage, Fertilizer, and
Tillage × Fertilizer and random effects for the class variables
Rep and Rep×Tillage. Return on Investment for N and for
P+K was analyzed with a mixed linear model with Tillage as a
fixed effect and Rep and Year as random effects for each rotation. PROC GLIMMIX in SAS 9.4 and JMP 12.2 were used
to estimate the linear mixed models (SAS Institute, Cary, NC).
Results and Discussion
Mean Annual Profit
Mean annual profit for the CC rotation showed a significant
influence for tillage (P = 0.0002) and fertilizer (P < 0.0001)
but no interaction (P = 0.7324; Fig. 1). Control fertilizer
showed the lowest relative profit, followed by N-only treatments, and all NPK-applied treatments were greatest and similar. The MP and ChT treatments were similar and had greater
profit than NT, and AT fell in between. In contrast to these
results, Karlen et al. (2013) found in CC in Iowa that NT had
higher profits than other tillage treatments using prices from
the early 2000’s over 9 yr even though NT yield was slightly
lower. Herbicide costs were also higher during the CC rotation
of the present study (1970–1990) than in later years (Fig. 4).
In the CS rotation, there was a significant influence of fertilizer (P < 0.0001) and no significant tillage effect (P = 0.2941),
but the interaction term was significant (P < 0.0001). The NT/
Control had the lowest profit overall, followed by NT/N-only
and AT/Control. The N-only treatments with tillage had significantly more profit than all Control treatments. All NPKapplied treatments were similar in annual profit and higher than
other treatments (Fig. 2). Differences in profitability reflected
differences in yield. Again in contrast, Karlen et al. (2013)
found that in a CS rotation in Iowa that NT had slightly higher
profitability than MP. Other studies have also found no differences in profitability due to tillage, particularly in well-drained
soils (Al-Kaisi and Yin 2004; Doster et al., 1983). In more
highly erodible soils in southern Illinois, NT had higher yield
and profitability than ChT or MP (Phillips et al., 1997).
Cumulative Profits
Cumulative profit during the CC period (1970–1990,
data not shown) showed a significant effect for fertilizer (P <
0.0001) and for tillage (P < 0.0001) but showed no significant
effect for their interaction (P = 0.4094). As in all other situations, Control fertilizer treatments were all similar and lowest
overall, followed by the N-only fertilizer treatment. It has been
previously shown that CC in Wisconsin was not profitable
between 1990 and 2004 with less than 112 kg N ha–1 under an
annually varying price of ammonium nitrate fertilizer (Stanger
et al., 2008). All NPK-applied treatments were highest and
similar. Among tillage treatments, MP, ChT, and AT showed
higher cumulative profits than the NT treatment.
The CS period (1991–2014, data not shown) cumulative
profit showed a significant effect for fertilizer (P < 0.0001),
no significant effect for tillage (P < 0.3588), and a significant
interaction (P < 0.0009). As in CC, all complete NPK-applied
Fig. 1. Least square mean annual profit in 2014 dollars for continuous corn rotation (1970–1990). Significant differences among treatments
are noted by different letters (A) within tillage and (B) within fertilizer. Error bars represent model standard error.
Agronomy Journal • Volume 109, Issue 1 • 2017
221
Fig. 2. Least square mean annual profit in 2014 dollars for corn–soybean rotation (1991–2014). Significant differences among treatments
are noted by different letters. Error bars represent model standard error.
treatments had higher and similar profits than N-only and
Control treatments. The NT/N-only and NT/Control were
the lowest profit treatments.
Across the entire CC and CS rotations, cumulative profit
from 1970 to 2014 showed a significant effect of fertilizer
(P < 0.0001), tillage (P = 0.0004), and their interaction (P =
0.0238) (Fig. 3). Control fertilizer treatments gave the lowest
cumulative profit overall, with NT/Control less than ChT/
Control treatments. The NT/N-only treatments had lower
cumulative profit than all other N-only tillage treatments.
There were no significant differences among NPK-applied
treatments, which were highest in cumulative profits. Overall,
fertilizer was the dominant determinant of yield as well as
profit. Tillage, when sufficient fertilizer was applied, had no
discernable effect on cumulative profitability, despite any longterm cumulative changes or interactions with soils, climate, or
plant genetics over time.
Price and Income Histories in
Different Production Systems
Profits were affected by the different field operations, yield,
crop prices, and quantity and historic prices of inputs. Crop
revenues were higher in the CC rotation than in the CS rotation (P < 0.0001; data not shown) due to higher corn prices,
particularly from 1974 to 1976, as corn yield was lower in CC
than in CS (Cook and Trlica, 2016). Trends in revenue have
been driven more by crop price fluctuations than by yield, as
yield at this site has remained fairly constant over time within
each rotation. When comparing rotations in the same years, CS
rotations have been shown to be significantly more profitable
than CC rotations by $269 ha–1 in Wisconsin (Stanger et al.,
2008). In our experiment, we could not directly compare CC
to CS profits since they occurred in different periods.
Seedbed preparation costs analyzed in this study included
machinery, labor, diesel, and herbicide costs associated with
field operations. Over the course of the experiment, machinery
Fig. 3. Least square mean cumulative profit in 2014 dollars for each treatment over the entire experiment (1970–2014). Significant
differences among treatments are noted by different letters. Error bars represent model standard error.
222
Agronomy Journal • Volume 109, Issue 1 • 2017
costs have trended upward over time, diesel and labor costs have
remained fairly constant and a small percentage of tillage costs,
and herbicide costs in NT and AT treatments have declined,
possibly due in part to the appearance of generic glyphosate after
patent expiry in 2000 or to the expanded production and use
of glyphosate once crops with engineered resistance were introduced in the mid-1990s (Fig. 4A). From 1970 to 2014, seedbed
preparation costs were lowest (P < 0.0001) in the NT treatments
(70.1 ± 0.68 U.S. dollars [USD] ha–1) followed by AT (86.1 ±
0.90 USD ha–1) due to the lower per-hectare cost of herbicide
compared to costs associated with tillage in MP and ChT. Field
operation costs were higher in ChT (110.3 ± 0.52 USD ha–1) and
MP (114.0 ± 0.53 USD ha–1), which were similar.
Following the per-unit cost of fertilizers (N, P, and K), the
per-hectare cost of fertilizer treatment spiked in the mid-1970s
followed by a general decline until the mid-2000’s, which then
increased to levels similar to the early 1980s (Fig. 4B). Fertilizer
costs in CS were half the cost in CC due to applications only
occurring in corn years.
Return on Investment for Fertilizers
Percent return on investment for N was 396% ± 18 in CC
and 133% ± 13 in CS. The effect of tillage on fertilizer return
was significant for N in CC and for N and P+K in CS rotation
(Table 2). In the CC rotation, the MP treatment had a higher
return than NT or ChT while AT fell in between. However, in
Fig. 4. Trends in costs adjusted to 2014 dollars of (A) seedbed preparation in no-till (NT) and moldboard plow (MP) and (B) fertilizer for
NPK broadcast treatments in corn years only.
Agronomy Journal • Volume 109, Issue 1 • 2017
223
the CS rotation MP, ChT, and AT were all significantly higher
than NT. The lower rate of return on N in NT reflects the poor
corn yield in the NT/N-only treatment relative to other N-only
treatments (Cook and Trlica, 2016). This may be due to intensive tillage practices releasing plant-available P and K allowing
greater response to N in MP, ChT, and AT treatments. Return
on N was lower in CS because the rotated cropping system
eliminates N applications in soybean years. Higher returns on
investment for N fertilizer are expected when other nutrient
limitations such as P and K are alleviated (Schlegel et al., 1996).
Our return on investment for N fertilizer likely underestimates
the potential return since our N-only treatments were likely
limited by P and/or K, especially in NT treatments.
Return on P and K was 78% ± 12 in CC and 109% ± 12 in
CS for all NPK-applied treatments across all tillage treatments.
The lower return for CC reflects the smaller yield increase in corn
between N-only and NPK-applied treatments, while the larger
return for CS reflects the relatively large response of soybean yield
and income to application of P and K. Many studies have shown
positive responses of soybean to added P and K (e.g., Bharati et
al., 1986). There was an indication of differences (P = 0.0716)
due to tillage in the CC rotation. When each NPK-applied fertilizer treatment was analyzed separately, NT treatments tended
to have the highest return for P and K fertilizer. In the NPK
broadcast treatment, NT was higher than MP (P = 0.0230) and
in the NPK+NPKstarter treatment NT was higher than AT
(P = 0.0305) with the other treatments falling in between. The
N+NPKstarter treatment showed all tillage treatments to have
similar return for P and K (P = 0.4684; data not shown). In the
CS rotation, all three NPK-applied treatments showed similar
trends in return on P and K. The NT treatment had a significantly higher return than all other tillage treatments (Table 2),
probably because the NT/N-only benchmark was lower in yield
than other N-only tillage treatments (Cook and Trlica, 2016).
These results show that overall even with large price swings
through time, N, P, and K fertilizer remained an important
investment to maintain profits. The particularly high rates of
return for P and K fertilizer in the NT/NPK-applied treatments
demonstrates the importance of managing P and K fertility for
maintaining competitive profits under long-term NT at this site.
Sensitivity Analysis
Based on elasticities, fertilizer cost (N, P, and K combined)
had the greatest impact on profit in both tillage treatments
(Table 3), as would be expected given that fertilizer makes up
a much greater percentage of total costs than other inputs.
After fertilizer, for MP treatments machinery had the greatest impact on profit while herbicide had the greatest impact in
the NT system. There is often a trade-off between machinery
and herbicide costs in NT and more intensive tillage systems
(Chase and Duffy, 1991), but the more recent relative decline
in herbicide costs for NT did not appear to have meaningfully
reduced the sensitivity of NT profits to these costs over the
whole course of the experiment.
In the CS rotation, annual machinery and herbicide costs
were modified from –100% to the approximate point where
NT and MP profits became significantly different (Fig. 5). A
300% increase in machinery cost was required to significantly
disadvantage MP from NT, but a complete elimination of
machinery cost did not cause profits to be significantly different. In contrast, an 850% increase in herbicide cost was
required to make NT significantly less profitable than MP.
Because MP did not receive post-emergent herbicide, there was
no change in profit regardless of herbicide price. The NT profit
was changed very little with changes in machinery cost because
the only machinery operation included the planter, sprayer, and
fertilizer application.
The results from the sensitivity analysis showed that NT and
MP tillage treatment profits were robust to changes in herbicide or machinery costs. Similarly, Meyer-Aurich et al. (2006)
showed that a 50% increase in energy cost (which affected both
fertilizer cost and tillage-related costs) only decreased profit in
MP by 2 to 4% compared to ChT. Across the CS rotation, average machinery costs in the MP treatment were $98 ha–1 and
trending up while herbicide costs were $49 ha–1 and trending
down (Fig. 4). Based on these past trends we would expect that
a continued rise in machinery costs is more likely in the future
than increased herbicide costs, however neither a 300% rise in
machinery cost nor an 850% increase in herbicide cost in the
near future seems likely.
Table 3. Elasticity of profit across entire CS rotation, that is, the
ratio of percent change in profit to percent change in cost of input
(excluding 2007/2008 rotation due to missing corn yield data).
Profit elasticity
Input
Moldboard plow
No till
Fertilizer (NPK)
–0.174
–0.167
Machinery
–0.075
–0.009
Lime
–0.016
–0.017
Labor
–0.015
–0.004
Diesel
–0.007
–0.002
Herbicide
0.000
–0.036
Table 2. Return on investment for N and P+K fertilizers (standard error of mean in parentheses). P values indicate significance of model
for effect of Tillage on percent return, letters indicate Tukey’s HSD groupings within each return category.
Continuous corn
Corn–soybean
Return
N
P+K
N
P+K
————————————————————— % —————————————————————
Moldboard plow
434 (31)a
56 (23)a
156 (27)a
63 (15)b
Chisel till
363 (36)b
71 (28)a
136 (23)a
58 (15)b
Alternate till
415 (38)ab
62 (22)a
172 (26)a
64 (18)b
No-till
370 (35)b
123 (20)a
69 (22)b
251 (28)a
All tillage types
396 (18)
78 (12)
133 (13)
109 (12)
P value
0.0102
0.0716
<0.0001
<0.0001
Tillage model standard error
65
38
42
34
224
Agronomy Journal • Volume 109, Issue 1 • 2017
Fig. 5. Sensitivity of mean annual profit for no-till (NT) and moldboard plow (MP) in corn–soybean (CS) rotation to changes from baseline
(A) herbicide or (B) machinery costs adjusted to 2014 dollars. Error bars represent standard error of the mean. P values show significance
of independent sample t tests at selected levels of price change (–100%, 0, 100%, 300% for machinery, and 850% for herbicide).
Soil Conservation Benefit
Benefits of NT often include reduction in soil loss on
highly erodible slopes. At this relatively flat site (<2% slope),
we estimated that MP would have lost 0.58 Mg ha–1 yr–1 of
topsoil compared to 0.13 Mg ha–1 yr–1 in NT (Walker and
Pope, 1983). A reduction in soil loss from MP to NT would
be equivalent to $0.98 ha–1 yr–1 in social and environmental benefits, excluding estimated soil productivity changes
(Hansen and Ribaudo, 2008). In contrast, at a more erodible
long-term tillage experiment in southern Illinois on a 6% slope,
the estimated reduction in soil loss from 30 Mg ha–1 yr–1 for
MP to 8 Mg ha–1 yr–1 for NT (Olson et al., 2013) would be
equivalent to approximately $47.74 ha–1 yr–1 (under the same
soil conservation benefit estimates) which is approximately half
of the estimated seedbed preparation costs for MP in the present study. However, since external benefits do not accrue to the
farmer, additional incentives may be necessary to encourage
adoption of conservation tillage (Grandy et al., 2006).
Conclusions
Fertilizer use had dominant influence on profitability while
tillage had a relatively minor influence overall. Under the current CS system, NT management has been as profitable on
average as other, more intensive, tillage system in NPK-applied
plots. Fertilizer had a positive return on investment even with
large fluctuations in costs year to year. The NT only had a
significant profit disadvantage compared to other tillage types
under conditions of prolonged absence of P and K fertilization.
Sensitivity analysis for herbicide and machinery costs indicated
that the competitiveness of NT and MP were not likely to
be reversed by small errors in estimated costs or large future
changes in costs. While soil loss is likely minor at this site, the
additional soil conservation measures (NT) did not reduce
profitability. We conclude that with adequate application of
NPK fertilizer, profits under NT can be comparable with other
tillage systems in CS production in non-sloping, somewhat
poorly drained soils of southern Illinois.
Acknowledgments
We owe a great deal of thanks to the late Dr. George Kapusta
(1932–2008), who established this experiment in 1970, and Ronald
Krausz, Farm Manager at the Belleville Research Center, who has
diligently maintained the experiment and records. We also owe a
great deal of thanks to Ms. Joy Smith and Dr. David Dickey in the
Department of Statistics at North Carolina State University for their
guidance in the statistical analysis.
References
Al-Kaisi, M.M., and X. Yin. 2004. Stepwise time response of corn
yield and economic return to no tillage. Soil Tillage Res. 78:91–
101. doi:10.1016/j.still.2004.02.011
Bharati, M., D. Whigham, and R. Voss. 1986. Soybean response to
tillage and nitrogen, phosphorus, and potassium fertilization.
Agron. J. 78:947–950. doi:10.2134/agronj1986.000219620078
00060002x
Brown, H., R. Cruse, and T. Colvin. 1989. Tillage system effects on
crop growth and production costs for a corn-soybean rotation. J.
Prod. Agric. 2:273–279. doi:10.2134/jpa1989.0273
Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W. Howarth, A.N.
Sharpley, and V.H. Smith. 1998. Nonpoint pollution of surface
waters with phosphorus and nitrogen. Ecol. Appl. 8(3):559–568.
doi:10.1890/1051-0761(1998)008[0559:NPOSWW]2.0.CO;2
Cary, J.W. and R.L. Wilkinson. 1997. Perceived profitability and
farmers‘ conservation behaviour. J. Agric. Economics 48:13–21.
doi:10.1111/j.1477-9552.1997.tb01127.x
Chase, C.A., and M. Duffy. 1991. An economic analysis of the Nashua
tillage study: 1978–1987. J. Prod. Agric. 4:91–98. doi:10.2134/
jpa1991.0091
Cook, R.L., and A. Trlica. 2016. Tillage and fertilizer effects on crop
yield and soil properties over 45 years in southern Illinois. Agron.
J. 108:415–426. doi:10.2134/agronj2015.0397
Dick, W. 1983. Organic carbon, nitrogen, and phosphorus concentrations and pH in soil profiles as affected by tillage
intensity. Soil Sci. Soc. Am. J. 47:102–107. doi:10.2136/
sssaj1983.03615995004700010021x
Dick, W., E. McCoy, W. Edwards, and R. Lal. 1991. Continuous
application of no-tillage to Ohio soils. Agron. J. 83:65–73.
doi:10.2134/agronj1991.00021962008300010017x
Doster, D., D. Griffith, J. Mannering, and S. Parsons. 1983. Economic
returns from alternative corn and soybean tillage systems in Indiana. J. Soil Water Conserv. 38:504–508.
Agronomy Journal • Volume 109, Issue 1 • 2017
225
Farmdoc. 2015. Machinery economics. FAST Tools, Univ. of Illinois
Farmdoc.
http://www.farmdoc.illinois.edu/pubs/FASTtool.
asp?section=FAST (accessed 11 May 2016).
Free, G.R., S. Fertig, and C. Bay. 1963. Zero tillage for corn following
sod. Agron. J. 55:207–208. doi:10.2134/agronj1963.000219620
05500020040x
Graber, L. 1928. Evidence and observations on establishing sweet clover in permanent bluegrass pastures. J. Am. Soc. Agron. 20:1197–
1205. doi:10.2134/agronj1928.00021962002000110008x
Grandy, A., G. Robertson, and K. Thelen. 2006. Do productivity
and environmental trade-offs justify periodically cultivating
no-till cropping systems? Agron. J. 98:1377–1383. doi:10.2134/
agronj2006.0137
Griffith, D., E. Kladivko, J.V. Mannering, T. West and S. Parsons. 1988.
Long-term tillage and rotation effects on corn growth and yield
on high and low organic matter, poorly drained soils. Agron. J.
80:599–605. doi:10.2134/agronj1988.00021962008000040011x
Hansen, L., and M. Ribaudo. 2008. Economic measures of soil conservation benefits. Tech. Bull. 1922. USDA-ERS. http://www.ers.
usda.gov/media/196118/tb1922.pdf (accessed 6 June 2016).
Karlen, D.L., J.L. Kovar, C.A. Cambardella, and T.S. Colvin. 2013.
Thirty-year tillage effects on crop yield and soil fertility indicators. Soil Tillage Res. 130:24–41. doi:10.1016/j.still.2013.02.003
Kenward, M.G., and J.H. Roger. 1997. Small sample inference for fixed
effects from restricted maximum likelihood. Biometrics 53:983–
997. doi:10.2307/2533558
Koskinen, W.C., and C.G. McWhorter. 1986. Weed control in conservation tillage. J. Soil Water Conserv. 41:365–370.
Kumar, S., A. Kadono, R. Lal, and W. Dick. 2012. Long-term tillage
and crop rotations for 47–49 years influences hydrological properties of two soils in Ohio. Soil Sci. Soc. Am. J. 76:2195–2207.
doi:10.2136/sssaj2012.0098
Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science (Washington, DC)
304(5677):1623–1627. doi:10.1126/science.1097396
Liu, S., and M.D. Duffy. 1996. Tillage systems and profitability: An
economic analysis of the Iowa MAX program. J. Prod. Agric.
9:522–527. doi:10.2134/jpa1996.0522
Mannering, J.V., D.L. Schertz, and B.A. Julian. 1987. Overview of
conservation tillage. In: T.J. Logan et al., editors, Effects of conservation tillage on ground water quality. Nitrates and pesticides.
Lewis Publ., Chelsea, MI. p. 3–17.
McIsaac, G.F., J.C. Siemens, J.W. Hummel, and A.T. Tyrrell. 1990.
Economic comparisons of six corn and soybean tillage systems,
two soybean row spacings on three farm sizes. Appl. Eng. Agric.
6:557–564. doi:10.13031/2013.26428
Meyer-Aurich, A., K. Janovicek, W. Deen, and A. Weersink. 2006.
Impact of tillage and rotation on yield and economic performance in corn-based cropping systems. Agron. J. 98:1204–1212.
doi:10.2134/agronj2005.0262
226
Olson, K.R., S.A. Ebelhar and J.M. Lang. 2013. Effects of 24 years of
conservation tillage systems on soil organic carbon and soil productivity. Applied and Environmental Soil Sci. 2013:ID 617504.
doi:10.1155/2013/617504
Pannell, D.J. 1997. Sensitivity analysis of normative economic models: Theoretical framework and practical strategies. Agric. Econ.
16:139–152. doi:10.1016/S0169-5150(96)01217-0
Phillips, S., K. Olson, J. Siemens, and S. Ebelhar. 1997. Economics of
conservation tillage systems for CRP land in southern Illinois. J.
Prod. Agric. 10:483–489. doi:10.2134/jpa1997.0483
Pimentel, D., C. Harvey, P. Resosudarmo, K. Sinclair, D. Kurz, M.
McNair et al. 1995. Environmental and economic costs of soil
erosion and conservation benefits. Science (Washington, DC)
267(5201):1117–1123. doi:10.1126/science.267.5201.1117
Schlegel, A., K. Dhuyvetter, and J.L. Havlin. 1996. Economic and environmental impacts of long-term nitrogen and phosphorus fertilization. J. Prod. Agric. 9:114–118. doi:10.2134/jpa1996.0114
Siemens, J.C., and W.R. Oschwald. 1978. Corn-soybean tillage systems: Erosion control, effects on crop production, costs. Trans.
ASAE 21:293–302. doi:10.13031/2013.35292
Stanger, T.F., J.G. Lauer, and J. Chavas. 2008. The profitability and
risk of long-term cropping systems featuring different rotations and nitrogen rates. Agron. J. 100:105–113. doi:10.2134/
agrojnl2006.0322
Swinton, S., and J. Lowenberg-DeBoer. 1998. Evaluating the profitability of site-specific farming. J. Prod. Agric. 11:439–446.
doi:10.2134/jpa1998.0439
Triplett, G., and W.A. Dick. 2008. No-tillage crop production: A revolution in agriculture! Agron. J. 100:S-153–S-165. doi:10.2134/
agronj2007.0005c
Triplett, G., D. Van Doren, and B. Schmidt. 1968. Effect of corn (Zea
mays L.) stover mulch on no-tillage corn yield and water infiltration. Agron. J. 60:236–239. doi:10.2134/agronj1968.00021962
006000020028x
USDA ERS. 2016a. Fertilizer use and price. USDA ERS. http://
www.ers.usda.gov/data-products/fertilizer-use-and-price.aspx
(accessed 11 May 2016).
USDA ERS. 2016b. Farm employment and wage rates (91005). USDA
ERS. http://usda.mannlib.cornell.edu/usda/nass/91005/index.
html (accessed 11 May 2016).
USDA NASS. 2016a. Quick Stats 2.0. USDA NASS. https://quickstats.nass.usda.gov/ (accessed 11 May 2016).
USDA NASS. 2016b. Agricultural prices summary. USDA Economics, Statistics and Market Information System. http://
usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.
do?documentID=1003 (accessed 11 May 2016).
Walker, R.D., and R.A. Pope. 1983. Estimating your soil losses with
the Universal Soil Loss Equation (USLE). College of Agriculture, Coop. Ext. Serv, Univ. of Illinois, Urbana-Champaign.
Agronomy Journal • Volume 109, Issue 1 • 2017