ORIGINAL RESEARCH

published: 30 October 2020

doi: 10.3389/fsufs.2020.564418

Balancing Economic and

Environmental Performance for
Small-Scale Rice Farmers in Peru
Marshal White 1*, Elizabeth Heros 2 , Eduardo Graterol 3 , Ngonidzashe Chirinda 4,5 and
Cameron M. Pittelkow 1,6
1
Department of Crop Sciences, University of Illinois, Urbana, IL, United States, 2 Departamento de Fitotecnia, Univsersidad
Nacional Agraria La Molina, Lima, Peru, 3 Fondo Latinoamericano para Arroz de Riego (FLAR), Palmira, Colombia, 4 Centro
Internacional de Agricultura Tropical (CIAT), Palmira, Colombia, 5 Mohammed VI Polytechnic University (UM6P),
AgroBioSciences (AgBS), Agricultural Innovations and Technology Transfer Centre (AITTC), Benguerir, Morocco,
6
Department of Plant Sciences, University of California-Davis, Davis, CA, United States

Integrated research efforts are needed to provide insights for simultaneously achieving
food production and environmental goals in rice production systems. In this study
we evaluated eight Sustainable Rice Platform performance indicators (net profit, labor
Edited by: productivity, grain yield, water use efficiency, nitrogen use efficiency, phosphorus use
Rebecca Chaplin-Kramer, efficiency, pesticide use efficiency, and greenhouse gas emissions) for 65 rice farmers in
Stanford University, United States
Peru. The aim was to (1) identify the major management factors influencing profitability, (2)
Reviewed by:
evaluate potential tradeoffs between economic and environmental sustainability, and (3)
Julie Gwendolin Zaehringer,
University of Bern, Switzerland understand farmer perceptions and decision-making processes related to profitability.
Kikuko Shoyama, For data analysis, farmers were grouped into profitability categories based on net
National Research Institute for Earth
Science and Disaster Resilience profit: top (top 20%), middle (middle 60%), and bottom (bottom 20%). Our results
(NIED), Japan indicate that rice yields, grain sale price, and total spending drive profitability in this
*Correspondence: region of Peru. An important finding is that the top profitability group achieved higher
Marshal White
yields while using lower inputs of labor, pesticides, elemental P, and irrigation hours,
[email protected]
resulting in improved environmental performance (e.g., higher nutrient and water use
Specialty section: efficiency and lower greenhouse gas emissions). Results from focus group discussions
This article was submitted to
highlighted that rice production has changed in recent decades, and there is a need for
Land, Livelihoods and Food Security,
a section of the journal improved extension support to cope with increasing pests, disease, and water scarcity.
Frontiers in Sustainable Food Systems Moreover, farmers indicated they are vulnerable to exploitation from vendors along the
Received: 21 May 2020 rice supply chain, with limited access to credit facilities and market channels and little
Accepted: 22 September 2020
Published: 30 October 2020
negotiating power over the cost of inputs and rice sale price. Our results suggest that rice
Citation:
productivity and profitability can be improved in this region without negatively influencing
White M, Heros E, Graterol E, environmental footprint per unit production, yet this is unlikely to occur unless key
Chirinda N and Pittelkow CM (2020)
agronomic and socio-economic challenges are addressed, such as improved access
Balancing Economic and
Environmental Performance for to crop management recommendations, alternative financing and credit options, and
Small-Scale Rice Farmers in Peru. technologies that decrease the costs of production.
Front. Sustain. Food Syst. 4:564418.
doi: 10.3389/fsufs.2020.564418 Keywords: rice, environmental performance, Peru, small-scale, economics, sustainability, SRP, perceptions

Frontiers in Sustainable Food Systems | www.frontiersin.org 1 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

INTRODUCTION production using multiple indicators to comprehensively balance

between profitability and efficient management practices leading
Rice (Oryza sativa L.) is a major source of calories for nearly half to a reduced environmental footprint. Benchmarking yield and
the world’s population and a livelihood source for over one billion resource-use efficiency gaps among farmers is a well-documented
people (Van Nguyen and Ferrero, 2006). Rice consumption is strategy to highlight attainable efficiencies and yield-limiting
drastically increasing in developing countries with expanding factors within specific agronomic and socioeconomic contexts
populations (Seck et al., 2012). Global projections show that rice (Van Ittersumvan and Rabbinge, 1997; Papademetriou et al.,
production will need to increase by 26% by 2035 to meet the 2000; Lobell et al., 2009; Foley et al., 2011; Stuart et al., 2016;
growing demand in regions such as Africa and Latin America Devkota et al., 2019). In simple terms, yield and resource
(Seck et al., 2012). While production increases are critical for food use gaps are calculated as the difference between the highest
security, there is also a need for balancing the competing goals attainable values currently achieved in the region and the
of maximizing farmer profits and minimizing negative effects on average value achieved by farmers. While yield gap analysis is
the environment. a crucial first step in setting agricultural intensification targets,
Intensive rice production practices have increased global relatively few studies have holistically examined relationships
cereal output, but higher yields do not always mean increased between yield gaps and sustainability indicators to determine
profitability at the farm-level. Starting with The Green tradeoffs between increasing yields and its adverse impacts on
Revolution, the use of technologies, such as improved seed, environmental footprint.
irrigation and the intensive use of fertilizers and pesticides, has The Sustainable Rice Platform (SRP) is a multi-stakeholder
increased global rice yields by 130% from 1966 to 2000 (Khush, alliance led by the United Nations (UN) Environment
2005). During this period, populations of underdeveloped Programme and IRRI (International Rice Research Institute),
countries grew by 90%, food prices consistently decreased, with the overall goal to create a framework for evaluating
and an estimated 18 million hectares (ha) of agricultural and promoting sustainable rice production (Sustainable Rice
land was spared from cultivation (Khush, 2005; Stevenson Platform, 2015). The SRP contains a set of 12 performance
et al., 2013). However, external inputs including fertilizers and indicators designed to evaluate the sustainability of farming
agrochemicals represent a significant economic cost for farmers, systems while also identifying opportunities for improvement.
many of whom do not have access to financial resources or Utilizing performance indicators for benchmarking yield
the appropriate knowledge to effectively manage their rice and resource use efficiency gaps allows local evaluation
crop and thereby maximize returns to investment (Reardon of performance, as well as regional comparisons within a
et al., 2009; Wiggins et al., 2010). For rice production, labor standardized and internationally accepted framework. The
costs can also be exceptionally high for transplanting as a crop majority of research utilizing indicators from the SRP has
establishment method (Toriyama et al., 2005). Lastly, since focused on Southeast Asia, where it has identified areas of
the 1960’s, global grain prices have continued to decline as improvement for the adoption of sustainable practices (Stuart
input costs have increased (FAOSTAT, 2020). While this is et al., 2018; Watcharapongchai et al., 2018; Devkota et al., 2019;
beneficial for consumers, producers ultimately generate less 2020). To our knowledge, there is no published evidence that the
revenue per unit. Therefore, if costs are too high and yields SRP has been utilized to evaluate rice production practices in
or rice prices are too low, there is a risk that rice production Latin America.
will not be profitable. Given the dependence of millions of In Peru, like many countries in Latin America, rice is a staple
small-scale farmers on rice production as a livelihood strategy, food and cash crop with demand rapidly increasing due to a
understanding the major factors influencing profitability in growing population with changing dietary preferences. Since
specific biophysical, political, and socioeconomic contexts is 1960, Peru’s land area of rice production has expanded by 81%
necessary to inform the development of effective agricultural and yields have increased by 49%, but the country is still a net
research and extension programs. rice importer (FAOSTAT, 2020). Peru has the second-highest
In the face of growing resource scarcity and a changing average rice yields in Latin America and the fifth-highest in the
climate, crop production targets aimed at attaining global food world (FAOSTAT, 2020). However, gains in productivity have
security must also be achieved sustainably without expanding been supported by higher input use which is generally associated
land under crop production (Godfray and Garnett, 2014). High with greater environmental costs (Foley et al., 2011). From the
greenhouse gas emissions (GHG) and water use are two main year 2007 to 2017, Peru increased total agricultural pesticide use
challenges threatening the sustainability of flooded rice systems. by 69%, and since 2002 has increased total nitrogen fertilizer
Methane emissions from rice production account for roughly use by 36% (FAOSTAT, 2020). Thus, there is a need to evaluate
50% of global cropland emissions (Carlson et al., 2016). In how input use affects the sustainability and profitability of farm
addition, it is estimated that rice production uses ∼34–43% systems in this region.
of the world’s irrigation water (Bouman et al., 2007). High Exploring the farmer decision-making process is important
rates of fertilizers and pesticides are also known to pollute air, to adequately understand farm management decisions within
water, and soil resources which has long-term environmental and farm systems (Byerlee and Anderson, 1982; Edwards-Jones,
social ramifications (De Roos et al., 2003; Pimentel et al., 2004; 2006; Wilson et al., 2009; Nakazi and Sserunkuuma, 2013).
Fageria and Baligar, 2005; Rosenbaum et al., 2008). Hence, it Farmers have a variety of perceptions, biases, and personal
is necessary to evaluate the environmental performance of rice beliefs that inform the decisions that they make (Mankad,

Frontiers in Sustainable Food Systems | www.frontiersin.org 2 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

2016). Researching extrinsic and intrinsic factors such as fertilizer, irrigation fees, application of water, perceptions of
profitability, praise, and personal motivation can lead to a water use, pesticide, herbicide, weeding, harvest, yield, and sale
better understanding of the adoption or non-adoption of certain price (Supplementary Figure 1). The survey was piloted with
agricultural practices (Dono et al., 2013; Meijer et al., 2015). six farmers in the district to verify that the questions were
An important outcome of researching farmer perceptions is to appropriate and applicable to the local context. Farmer surveys
determine if farmers perceptions of profitability and successful were conducted during July and August 2019. Respondents were
management are different from those of various stakeholders limited to those that cultivated rice in the 2018-2019 growing
tasked with implementing solutions. season and confirmed that they were the sole decision-makers
Balancing the economic and environmental sustainability on management and input purchases for the farm. Farmers
of rice production systems is a complex challenge. Integrated were identified within the Ferreñafe irrigation district through
research efforts accounting for both biophysical and social snowball sampling. This convenience technique relies on initial
dimensions are needed to provide insights for achieving food subjects referring additional subjects and is useful for identifying
production and environmental sustainability goals in the future. hard to reach populations where a sampling database does not
The objectives of this study were to (1) investigate the economic exist (Biernacki and Waldorf, 1981). The total sample size for the
performance of smallholder rice farming systems of Peru to survey was 65 farmers, representing 1.6% of the farmers in the
identify the major factors influencing profitability, (2) evaluate region. Interviews lasted ∼45 min, where farmers were asked to
environmental sustainability using SRP performance indicators answer in-depth questions about the previous growing season.
and assess relationships with economic outcomes, and (3)
understand perceptions and barriers to increasing productivity
and profitability based on discussions of research findings with Sustainable Rice Platform Performance
local farmers. Indicators
The SRP is an international framework for measuring the
METHODS economic and environmental sustainability of rice systems across
12 performance indicators. Information from surveys allowed
Characterization of the Survey Region eight indicators to be calculated in this study: net profit, labor
This study was conducted within the Ferreñafe irrigation district productivity, grain yield, water use efficiency, nitrogen use
of the Ferreñafe province (6◦ 38’01.8 “S, 79◦ 47’14.3” W) in the efficiency (NUE), phosphorus use efficiency (PUE), pesticide use
department of Lambayeque, Peru (Figure 1). The department efficiency, and GHG emission. The indicators for food safety,
of Lambayeque is one of the largest rice-producing areas in worker health and safety, child labor, and women empowerment
the country and is known for high yields and exceptional grain were omitted to shorten the survey to an appropriate length
quality due to favorable climate and availability of irrigation for conducting the interview. Several modifications to indicator
infrastructure. This region is ∼20 km inland of the north-west calculations were made based on local management practices
coast of Peru, and is classified as coastal desert with rainfall and available data obtained from the farmers. Net profit ($
<100 mm per year (Nguyen et al., 2019). The Ferreñafe irrigation ha−1 ) was calculated by subtracting total variable costs from
district is the largest district in Lambayeque with ∼14,000 ha gross returns (grain revenue). The total variable cost is the sum
planted in rice and 4,000 rice farmers. The conventional method of production costs for seedbed and field production (labor,
of rice production consists of manual transplanting and high fertilizer, harvest, transportation, land preparation, irrigation and
rates of fertilizer and pesticide inputs, when compared to small- fees, pesticides, herbicides, and seed), and the cost of milling and
scale irrigated farming regions in Asia (Devkota et al., 2019). rent (if applicable) per hectare. An opportunity cost for land
Current production challenges in the region include labor and ownership could be included in total variable costs, but this is
water shortages, disease pressure, and reduced soil quality due to only valid in situations where land-owning farmers could easily
salinization (Maclean et al., 2013). Farmers in this region typically find a renter. Based on our conversations with farmers, rice
grow one rice crop per year, without crop rotation, but with production is becoming less attractive as a livelihood strategy
small plots of non-rice crops between seasons. The availability in this region and the potential for renting farmland is limited.
of irrigation is dependent on rainfall that occurs in the Andes Therefore, the opportunity cost for land ownership was excluded
Mountains, which is harvested from a reservoir situated 45 km because conditions did not meet the assumption of perfect
away and delivered to the region using canal systems. competition. Gross return was calculated from paddy yield and
the market price the farmer received or expected to receive.
Farmer Survey Labor productivity (kg grain $ labor−1 ha−1 ) was calculated by
A paper-based survey was created in collaboration with dividing grain yield by the total cost of labor for production. Total
CIAT (Centro Internacional de Agricultura Tropical) and the labor hours and costs were calculated from farmer responses
Universidad Nacional Agraria La Molina to collect detailed regarding the amount of time for a field activity and the rate of
data on farm management practices and production costs. pay, including own and family labor. Farmers reported grain yield
In turn, this information was used to calculate eight SRP (kg grain ha−1 ) standardized to 14% moisture content. Water
performance indicators (Sustainable Rice Platform, 2015). The use efficiency (kg grain water-hour −1 ha−1 ) was calculated by
survey consisted of 115 questions in 14 sections: demographics, dividing grain yield by the number of water-hours the farmer
previous crop and residue management, land preparation, seed, purchased. Farmers purchase water-hours from the irrigation

Frontiers in Sustainable Food Systems | www.frontiersin.org 3 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

FIGURE 1 | Map showing the Ferreñafe irrigation sector in the department of Lambayeque, Peru. Surveys were completed in the Ferreñafe irrigation sector. Sources:
Esri, DeLorme, HERE, TomTom, Intermap, increment P Corp., GEBCO, USGS, FAO, NPS, NRCAN, GeoBase, IGN, Kadaster NL, Ordnance Survey, Esri Japan, METI,
Esri China (Hong Kong), swisstopo, MapmyIndia, and the GIS User Community.

commission, as a means to account and limit water usage in the and were calculated from farmer reported fertilizer use. Pesticide
district. The irrigation commission estimates that one water-hour use efficiency was calculated using the IRRI field calculator,
is the equivalent to 575 m3 of water, but field variation made which computes a score based on the number and timing of
the calculation of volume usage imprecise. Thus, water-hour was herbicide, insecticide, fungicide, rodenticide, and molluscicide
used as the input value in this study. Indicators for NUE and applications. Rodenticide and molluscicide were not included, as
PUE are expressed as kg grain kg−1 elemental N or P applied, use of these pesticides is not common. Pesticide use efficiency

Frontiers in Sustainable Food Systems | www.frontiersin.org 4 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

range from zero to 75 points, with a higher score indicating more data normalized to one from the largest value of each variable.
sustainable practices. Field GHG emissions were expressed as A linear mixed effect model was built using lmer() from R
kg grain kg−1 CO2 equivalent (CO2 e). Estimations of methane package lme4 and analysis of variance (ANOVA) was used to test
emissions were made using Tier 1 emission factors given in for significant factors influencing yield. The assumptions of the
the IPCC’s greenhouse gas inventory methodology (IPCC, 2006) model were examined for residual normality. Analysis of variance
which accounts for flooding duration, irrigation regime, drying (ANOVA) was performed to test differences between profitability
events, the timing of organic inputs, and residue burning. Drying groups for demographics, input use, and performance indicators.
events were reported from farmers as periods of aeration of 3 Assumptions of normality and homogeneity of variance were
days or more (IPCC, 2006). Methane was adjusted to CO2 e by evaluated using the Shapiro-Wilk and Levene tests. All statistical
multiplying by a factor of 28. analyses were completed with R version 3.4 (RStudio Team,
2020).
Focus Group Discussions
In November 2019, two focus groups were conducted within the
Ferreñafe irrigation district to present preliminary results from RESULTS
the study and better understand farmers’ perceptions of profitable
farming systems, barriers to production, and decision-making Production Practices and Input Use
process regarding crop management inputs. Focus groups are The majority of farms practiced transplanting for crop
a qualitative research method that allows insight into the establishment (97%, n = 65) with a small number of farmers
complex nature of personal beliefs within the context of the who practiced manual direct seeding with an average of 100 kg
participants’ lives (Krueger and Casey, 2014). Focus groups allow of seed per hectare. Mean farm size was 7.8 ha with mean
farmers to openly engage with ideas and each other in a semi- planting and harvest dates of January 19 and June 2 (Table 1).
controlled environment, supporting in-depth discussions on a Farms were characterized by high mean production costs (1,793
variety of topics (Onwuegbuzie et al., 2009). Each focus group $ ha−1 ), more than 2-fold than production costs observed in
was semi-structured and lasted ∼1.5 h. Farmers in the region flooded regions in Vietnam, Myanmar, Thailand, and Sri Lanka
were invited from an announcement in the regional irrigation (Devkota et al., 2019). High input use was observed in these same
district. The focus group started with a short questionnaire about rice producing regions (mean fertilizer rates of 290 kg N ha−1 ,
perceptions of spending, and then a guided group discussion was 31 kg P ha−1 , and 35 kg K ha−1 ). On average, farms also had
facilitated by the author and a local translator and agronomy multiple applications of pesticides (six). Rainfall in this region
student. Forty-five farmers attended the focus groups, with was minimal during the growing season, ∼74 mm (Nguyen
22 farmers individually responding to the questionnaire. The et al., 2019), thus farmers relied on multiple irrigations (nine)
moderators presented handouts of figures summarizing the throughout the growing season. A mean of 17.8 water-hours was
results of the survey and had participants provide feedback and purchased, with an estimated volume usage of 10,235 m3 per
assess the validity of the results by discussing their perceptions hectare. At least one soil drying event occurred in 42% of farmers’
of profitable vs. non-profitable systems. Focus groups were
recorded and an assistant took notes to emphasize significant
themes. This stimulus-based methodology allowed participants
to interact with survey results and one another, thereby enabling TABLE 1 | Characterization of farm size and production inputs in Ferreñafe, Peru
from 65 farmer surveys conducted in 2019.
an in-depth understanding of their perceptions and the barriers
preventing the adoption of more profitable or sustainable Area, production inputs, and total variable costs
management practices (Onwuegbuzie et al., 2009). Farm Size (ha−1 ) 7.8 ± 10.4
Sowing/Transplanting date January 19 ± 13
Data Analysis Harvest Date June 2 ± 16
To evaluate the potential for optimizing input use and Seed Rate (kg ha−1 ) 101 ± 22
management factors, we used the concept of attainable Elemental N (kg ha−1 ) 290 ± 86
profitability to compare average- or below-average farmers to
Elemental P (kg ha−1 ) 31 ± 41
those that maximized profitability in the region. Farmers were
Elemental K (kg ha−1 ) 35 ± 54
grouped into three profitability categories based on net profit:
Rainfall (mm) 73 ± 61
top (top 20%), middle (middle 60%), and bottom (bottom 20%).
No. of drying events 0.8 ± 1.1
This method was adjusted for a small sample size but based on
No. of irrigations 9 ± 2.3
similar studies exploring yield and resource use efficiency gaps
No. of fertilizer applications 5 ± 1.0
using the top decile (10%) (Samarasinghe, 2003; Lobell et al.,
No. of Pesticide applications 6 ± 1.0
2009; Stuart et al., 2016). Likewise, to analyze possible yield
Establishment method Transplant = 63 (97%)
gaps and resource use efficiencies for different SRP performance
Direct = 2 (3%)
indicators, relative gaps were calculated among farmers as the
Total variable cost ($ ha−1 ) 1,790 ± 295
percent difference between the mean of the top 20% (target) and
the population mean (baseline). Radar charts show the relative Values are mean ± standard deviation. Establishment values are the number of farmers
difference of SRP indicators between profitability groups, with practicing the method with the percentages shown.

Frontiers in Sustainable Food Systems | www.frontiersin.org 5 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

fields, with some experiencing up to four drying events due to 449 and a target of 772 kg grain water-hour−1 . Yield-scaled GHG
shortages of water and the complexities of irrigation scheduling. emissions had a 42% gap with a baseline of 1.836 and a target of
3.191 kg grain kgCO2 e−1 . All farmers surveyed used N fertilizer,
Yield, Total Variable Cost, Sale Price, and and there was a resource gap of 40% with a baseline NUE of 26.3
Net Profit and a target of 36.3 kg grain kg−1 N. Pesticide use efficiency had
Yield was identified as a major factor driving profitability (R2 a 31% gap with a baseline of 44.5 and a target of 64.5 points.
= 0.65). The top profitability group had the highest yield The smallest resource gap observed was grain yield (22%) with
(9,482 kg ha−1 ) with the average yield 26% higher than the middle a baseline value of 7,091 and a target yield of 9,070 kg ha−1 .
(6,981 kg ha−1 ) and 47% higher than the bottom (5,031 kg ha−1 ) The top profit group generally used lower levels of inputs,
profitability group (Figure 2). Predictor variables explained 30% except for seed and nitrogen, but achieved higher yields
of yield variation across farms (Supplementary Table 1). In the (Figure 5). For example, the top profit group had 22% lower
study regions, planting date was the only variable significantly total variable costs (1,777 $ ha−1 ), 20% lower labor costs (212
associated with yield. The top group had the lowest production $ ha−1 ), 65% lower elemental P (20 kg ha−1 ), and 24% fewer
costs (1,721 $ ha−1 ) with production costs 15% less than the chemicals (7.8 chemicals ha−1 ) than the bottom profitability
bottom (2,019 $ ha−1 ) profitability group. The top group spent group. Seed rate was consistent across the profit groups, but the
20% less on labor (538 $ ha−1 ), 24% less on irrigation and top group used ∼10% more N fertilizer (323 kg ha−1 ) than the
fees (146 $ ha−1 ), and 46% less on pesticides than the bottom bottom group. Tradeoffs between economics and performance
profitability group (Figure 3). When accounting for total variable indicators were not observed, as the top profit group had 154%
costs, the top group (1,790 $ ha−1 ) spent 21% less than the higher profits (1,416 $ ha−1 ), 47% higher yield (9,483 kg ha−1 ),
bottom (2,266 $ ha−1 ), due to fewer farmers renting land in and 61% higher labor productivity (18 kg grain $ labor−1 ). The
the top profitably group. The top group had the highest sale top group also had 44% higher NUE (32 kg grain kg N−1 ),
price (0.352 $ kg−1 ), with a sale price 16% higher than the 59% higher PUE (436 kg grain kg P−1 ), 56% higher water
bottom (0.296 $ kg−1 ) group. The top group had the highest net productivity (615 kg grain water-hour labor−1 ), and 39% higher
profit (1,416 $ ha−1 ), which was 78% higher than the middle yield-scale GHG (2.148 Kg grain kg CO2 equiv−1 ) than the
(305 $ ha−1 ) and 154% higher than the bottom (-768 $ ha−1 ) bottom profitability group.
profitability group.
Farmers’ Perceptions of Production and
SRP Performance Indicators and Resource Profitability
Use Efficiency Gaps Farmers indicated that being profitable was becoming more
The largest resource use efficiency gap was for net profit (78%, difficult, due to increasing economic challenges and production
Figure 4), with a baseline value of 312 $ ha−1 and a target of 1,415 barriers, which limited yields. Frequent water shortages were
$ ha−1 (Table 2). Profit values ranged from $-1,409 to $2,102 per described as one of the most challenging aspects to rice
hectare, with 38% of farmers being unprofitable in this growing production in the region. From the farmers surveyed, 78%
season. Net profit was strongly correlated with yield (r = 0.65, said that there were occasions during the previous growing
p < 0.001). The majority of farmers used fertilizers containing P season when water was not available when needed. In addition,
(75%), and a large P-use efficiency gap (51%) was observed with 58% of farmers said their crop experienced water stress during
a baseline value of 273 kg grain kg P−1 ha−1 and a target value the vegetative period, while 17% of farmers said their crop
of 553 kg grain kg P−1 ha−1 . Labor productivity had a 42% gap experienced water stress during the reproductive phase. Farmers
with a baseline of 12.9 and a target of 22.3 kg grain labor hour−1 . purchase water from the irrigation commission on a per water-
There was a 42% gap in water use efficiency with a baseline of hour basis, and the total hours are limited to each farmer based

FIGURE 2 | Yield, total variable costs, grain sale price, and net profit per hectare delineated by profit group (top, middle, bottom) for 65 farmers in Ferreñafe, Peru in
2019. Farmers are grouped into profitability categories based on net profit: top (top 20%, n = 13), middle (middle 60%, n = 39), and bottom (bottom 20%, n = 13).

Frontiers in Sustainable Food Systems | www.frontiersin.org 6 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

FIGURE 3 | Average production costs delineated by profit groups (top, middle, bottom) for 65 farmers in Ferreñafe, Peru. Farmers are grouped into profitability
categories based on net profit: top (top 20%; n = 13), middle (middle 60%, n = 39), and bottom (bottom 20%, n = 13).

on the number of hectares, non-rice crops grown, and the level greater than the mean production cost of the top profitability
of the reservoir in the current season. In recent years, farmers group (1,721 $ ha−1 ), and the population mean (1,793 $ ha−1 ).
reported that water shortages have become more common due Additionally, when asked if farmers knew the exact amounts
to reduced rainfall and the complexity of irrigation scheduling, they were spending throughout the growing season, 15 farmers
causing extended periods of drought conditions in their fields. responded yes, while seven responded no. In the discussion
The cost of production was also described as a major barrier portion of the focus groups, farmers explained the tension that
to profitability. Concerning access to capital, most farmers lack exists between the pressure to purchase inputs from salespersons,
financial resources and access to cheap credit facilities, so they the lack of knowledge in how to use inputs appropriately, and
rely on the rice milling industry for high interest loans to finance the lack of agricultural extension to provide recommendations.
their production, which ultimately reduces their grain marketing Farmers highlighted the fact that production costs are very high,
options. Historically, farmers explained that rice farming was mainly because the only people they talk to about improving their
profitable, but with increasing production costs and constraints management are sales people. This creates a scenario where there
to crop productivity, farmers are unsure about the future of is an incentive for salespersons to continue a business as usual
rice farming. From the focus groups, 82% of farmers responded approach, which does not support low input use, as this would
that low production costs were important or very important reduce their sales. Farmers feel at odds between improving their
for increasing profitability (Figure 6). When the same group management by purchasing additional inputs, and reducing their
was questioned about their estimates on the amounts of money cost of production. Furthermore, farmers financed by the milling
a farmer should spend to maximize yield and profit, 23 and industry claimed that purchases are often made in the field on
36% of participants responded that farmers should spend more credit from traveling salespersons, which made it difficult to keep
than 1,800 and 2,100 $ ha−1 respectively, both of which are track of expenditures throughout the growing season.

Frontiers in Sustainable Food Systems | www.frontiersin.org 7 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

FIGURE 4 | Percent gap for SRP indicators calculated among farmers as the percent difference between the mean of the top 20% (target) and the population mean
(baseline) for 65 farmers in Ferreñafe, Peru.

TABLE 2 | Baseline (population mean) and target (mean of top 20% performers: for increasing yields and profitability across farms (discussed
each category) values of performance indicators for 65 farms in Ferreñafe, Peru. below). The strength of our study is that factors influencing
Indicator Unit Baseline Target
profitability were analyzed from survey data, and focus groups
revealed additional important insights to the perceptions and
Profit ($ha−1 ) 312 1,415 decision-making process of farmers. Namely, the underlying
Labor productivity (kg grain $labor−1 ) 12.9 22.3 explanation of why some farmers were more profitable than
Grain Yield (kg ha−1 ) 7,091 9,070 others appeared to be primarily related to the decision-
Water productivity (kg grain water-hour−1 ) 449 772 making process regarding financing, spending and management
Nitrogen use efficiency (kg grain Kg N−1 ) 26.3 43.6 decisions. Farmers explained that access to credit for capital
Phosphorus use efficiency (kg grain kg P−1 ) 273.4 553.4 expenses and agriculture education was needed meaning that
Yield-Scaled GHG (kg grain kgCO2 e−1 ) 1.836 3.191 overcoming barriers to higher profitability is as much of a
Pesticide use efficiency (0-75 points) 44.5 64.5 socioeconomic challenge as an agronomic challenge. Below
we discuss the major opportunities for improving profitability
in this region and the need for research and extension
programs that would provide farmers access to knowledge
on improved management and technological options. We also
DISCUSSION discuss the need to not only address yield-limiting factors,
Profitability and Decision-making but navigate the complex socioeconomic challenges related to
Our results suggest that rice farmers in this region of credit facilities and market channels. The small sample size in
Peru can achieve the highest profitability through efforts this study is the primary limitation to the generalization of
aimed at increasing yields, marketing grain at the most these results, and we suggest that more research and extension
competitive price, and decreasing the cost of production. The efforts are needed working across a large number of farms to
importance of these factors is consistent with the concepts of identify crop management practices that can reliably increase
agricultural development for small-scale farmers, which focuses yield in this region without negatively influencing the SRP
on improving productivity while reducing input costs (Norton, performance indicators.
2004). It is noteworthy that agronomic management practices, In this study, we identified a significant gap in profitability,
and specifically the level of inputs, were not directly responsible where the top farmers (top 20%) were 154% more profitable than

Frontiers in Sustainable Food Systems | www.frontiersin.org 8 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

FIGURE 5 | Comparison of SRP indicators (A) and production inputs (B) and among three categories of farmers delineated by profit percentiles in Ferreñafe, Peru for
the 2019 season. Farmers are grouped into profitability categories based on net profit: top (top 20%, n = 13), middle (middle 60%, n = 39), and bottom (bottom 20%,
n = 13). Data is normalized to 1 based on the largest value for each variable. Profit = ($ ha−1 ), Yield = (kg ha−1 ), Labor productivity = (kg grain $labor −1 ), Pesticide
use efficiency = (0-75 points), N use efficiency = (kg grain kg N−1 ), P use efficiency = (kg grain kg P−1 ), Water productivity = (kg grain water-hour−1 ), Yield-Scaled
GHG = (kg grain kg CO2 equiv−1 ), Total variable costs = ($ ha−1 ), Labor cost = ($ ha−1 ), Seed rate = (kg ha−1 ), Total pesticides = (total chemicals), Elemental N =
(kg ha−1 ), Elemental P = (kg ha−1 ), Irrigation = (water-hours purchased).

FIGURE 6 | Farmer’s responses to questions about perceptions of production costs and profitability from 22 farmers in Ferreñafe, Peru. (A) Farmer’s responses to the
importance of low production costs for profitability: 1 = not important, 2 = low importance, 3 = neutral, 4 = important, 5 = very important. (B) Farmer’s responses
(yes or no) to whether they know how much they spend per hectare during the growing season. (C) Farmers response to how much they should spend to maximize
profit and yield.

the bottom group. This gap was associated with considerable decades in Peru has been on increased inputs for higher
variations in yield and management practices, as well as related production. Farmers’ management decisions are influenced by
production costs. The rice yields were identified as a major subjective beliefs pertaining to risk aversion (Babcock, 1992;
factor driving profitability, with the top profitability group Sheriff, 2005; Menapace et al., 2013), hence it is likely that
having yields 46% higher than the bottom group. This gap increasing economic uncertainty has contributed to the high
in productivity is not uncommon in small-scale rice systems seed, nutrient, and agrochemical inputs in this study (Table 1).
(Affholder et al., 2013; Stuart et al., 2016). In focus groups, Combined with a suitable biophysical environment, high input
farmers explained that water scarcity, pests, and diseases often levels may help explain why Peru has the second-highest
cause low yields and there is a lack of research extension average yields in Latin America (FAOSTAT, 2020). Nevertheless,
efforts to help them address these issues. When discussing farmers discussed that production challenges have changed
the rationale behind management decisions leading to high in recent decades and maintaining high yields is becoming
yields, farmers felt strongly that the emphasis during recent more difficult.

Frontiers in Sustainable Food Systems | www.frontiersin.org 9 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

Although the identification of variables leading to yield influence of information and knowledge to the decision-making
increases can help establish guidelines for closing yield gaps process, which can benefit famer profitability and sustainable
(Lobell, 2013; Zhang et al., 2016), we found that only planting management practices (Feder et al., 1985; Phillips, 1994;
date was significantly associated with yield. This finding is Anderson, 2008; Nakano et al., 2018). Also, while yields have
likely due to the relatively small sample size combined with a increased due to the use of technologies, fewer efforts have
large variation in management and crop production outcomes. been made to teach farmers about technologies that can
In addition, farmers reported that yields were lower than decrease costs such as direct manual or mechanized seeding. For
average due to disease pressure that was not captured in the example, in this region, the overwhelming majority of farmers
survey data. Unfortunately, in this region, planting date is transplant rice, which represents nearly 50% of labor costs and
primarily out of farmers’ hands, as farmers plant as soon 16% of total production costs. Farmers that practiced direct
as the water is available in their irrigation branch. Efforts seeding, rather than transplanting, saved nearly 300 dollars per
to make water available earlier, to those who receive it the hectare. Farmers discussed their reluctance to try direct seeding
latest, could improve yields. Despite the lack of significant due to worries about water scheduling and increased weed
management factors influencing yield, yields were still a driver pressure. Recent research has highlighted the benefits of direct
of profitability. seeding in Peru (Pando et al., 2017), but there is a need to
Further research and extension geared toward closing yield increase extension opportunities to inform farmers about specific
gaps is critical for improving farmer livelihood outcomes production or post-production options that could generate
(Tittonell and Giller, 2013; Dzanku et al., 2015) without higher profits.
negatively influencing the SRP performance indicators below. Another driving factor for profit was farmers selling their
In the pursuit of maintaining high yields, farmers explained a grain for higher prices. This was accomplished by either selling
tension that exists in the decision-making process of whether their grain first after the start of harvest, price shopping among
to manage more intensively, often in a prophylactic manner, or different mills, or milling their rice and then selling to retailers.
to risk yield loss by reducing production costs. Interestingly, we However, these options are not feasible for most farmers who
found that the top profitability group spent significantly less on do not have sufficient capital and rely on external loans to cover
production than the bottom profitability group (Figure 2). This production costs each season. The majority of farmers (74%)
result suggests that the top profitability group was optimizing sold their paddy at the farm gate immediately after harvest. This
their spending, mainly related to labor, fertilizer, water, and practice is common in rice-growing regions and is consistent
pesticides. Conversely, farmers spending large amounts on with research highlighting that small-scale rice farmers with
production do so at economic risk. For labor, the top profitability limited access to market channels are more likely to sell at
group spent 20% less on total labor costs (538 vs. 676 $ ha−1 ) the farm gate for lower prices (Cazzuffi and McKay, 2012;
than the bottom group, with a 20% reduction in transplanting Muhunyu, 2012; Soe et al., 2015). While not as profitable, this
costs (269 vs. 333 $ ha−1 ), mainly from hiring fewer laborers. In arrangement allows farmers to be paid quickly, and it reduces
the case of fertilizer, the top profitability group generally spent the costs associated with drying, storing, or milling. Of the 65
less (397 vs. 436 $ for NPK inputs), but applied higher rates of farmers surveyed, 43 were financed by the milling operations,
N fertilizer in their fields (323 vs. 290 kg N ha−1 ). These farmers which typically has the stipulation of selling the paddy at the
may be optimizing N inputs due to more extensive knowledge farm gate. Furthermore, if farmers enter into debt with a given
of fertilizer management and crop requirements, or possibly milling operation, a cycle of reduced market choices ensues until
better access to higher N fertilizer sources such as urea (46-0-0) the farmer pays off debt or decapitalization occurs. Ultimately,
compared to the more expensive compound fertilizers (20-20- this relationship reinforces a lack of bargaining power, which
20) with lower N content. For pesticides, the top profitability makes farmers vulnerable to exploitation and further reduces the
group spent 43% less than the bottom group (111 vs. 204 $ ha−1 ), farmer’s sense of agency. In our focus groups, farmers expressed
which could be due to increased knowledge of when pesticides a desire to be organized so they could protect themselves from
are required or from reduced disease and pest pressure. For low prices and aggressive financial intervention from the milling
water usage and fees, the top profitability group spent 25% less operations. Farmers in this region have concerns about the future
than the bottom group (146 vs. 193 $ ha−1 ), mainly from using of rice production due to the influence of the milling operations
fewer water-hours (16 vs. 21 hours ha−1 ), although irrigation fees over input use and sale prices. Many studies suggesting that
varied based on the cost of cleaning and maintaining the section small-scale farmers lacking power, negotiating capacity, and
of the canal where the farm is located. Furthermore, farmers with political representation would benefit from a farmer organization
limited access to purchase water-hours in person, opted to pay (Stockbridge et al., 2003; Alene et al., 2008; Aku et al., 2018),
an additional price to send someone to purchase water-hours although the success of farmer organizations is mixed (Key
for them. and Runsten, 1999; Masakure and Henson, 2005; Thorp et al.,
When these findings were discussed in focus groups, 2005). In addition to the sale price, farmers desired additional
farmers suggested that the top profitability group was more options for credit, including access to regulated long-term
knowledgeable about inputs and their appropriate usage, which credit, and research efforts concentrated toward helping
allowed them to achieve high yields with lower production farmers make educated management decisions will likely boost
costs. This observation is consistent with studies indicating the their confidence.

Frontiers in Sustainable Food Systems | www.frontiersin.org 10 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

TABLE 3 | Production inputs and SRP indicators among three categories of chemicals applied multiple times over the season. Ultimately,
farmers delineated by profit percentile in Ferreñafe, Peru for the 2019 season. many farmers felt conflicted between their knowledge gaps in the
Input use Top Middle Bottom
use of pesticides, the rising costs of such products, and the desire
to protect yields. Extension efforts geared toward integrated
Total variable costs ($ ha−1 ) 1,790 1,847 2,266 pest management will help farmers make informed choices, and
Labor cost ($ ha−1 ) 212 231 268 potentially reduce the environmental footprint of these systems.
Seed rate (kg ha−1 ) 101 102 101
For both labor and water use, we found that the top
profitability group used less than the middle and bottom profit
Number of pesticides 8 8 10
−1
groups. This practice translates to lower production costs and
Elemental N (kg ha ) 323 278 290
reduced input use and environmental footprint. We hypothesize
Elemental P (kg ha−1 ) 20 25 58 that this is due to the level of involvement the farmer has in
Water-hours (hours purchased) 16 17 21 day-to-day management activities. Production activities such as
SRP Indicators transplanting and irrigation require in-depth understanding and
Profit ($ ha−1 ) 1,416 305 −768 involvement of the activity, field, and community. An earlier
Yield (kg ha−1 ) 9,483 6,981 5,030 study (Abdulai and Huffman, 1998) found that small-scale rice
Labor productivity (kg grain $labor −1 ) 18 13 7 farmers engaged in non-farm activities tended to exhibit higher
Pesticide use efficiency (0–75 points) 50 45 39
levels of inefficiency due to a reduction of time allocated to
farm work. Similarly, our focus group discussions highlighted
Nitrogen use efficiency (kg grain kg N−1 ) 32 28 18
that many farmers are managing their fields while also focusing
Phosphorus use efficiency (kg grain kg P−1 ) 436 252 179
on other means of employment. If rice production continues
Water productivity (kg grain hour H2 O−1 ) 615 454 270 to be viewed as a poor or risky way to make a living, then
−1
Yield-Scaled GHG (kg grain kg CO2 equiv ) 2.148 1.912 1.300 production efficiencies may decrease as farmers continually
shift their resources to find more profitable means of work. A
reduction in the farmer population is a trend that is not only
happening in rice systems but agriculture worldwide (McNally,
Resource Use Efficiencies and Prospects 2002; Phimister and Roberts, 2006; Rigg, 2006; Nasir and Kotu,
for Sustainable Intensification 2014).
To achieve the dual goals of enhancing food security and Our research shows that the top profitability group achieved
environmental performance, yield increases cannot come at the the highest scores for all SRP performance indicators. In the case
cost of lower resource use efficiencies. In this study, we observed of labor productivity, pesticide use efficiency, PUE, and water
a wide range of input levels and resource use efficiencies for eight productivity, a combination of high yields and low input use was
SRP performance indicators across the three profitability groups a driving factor for these high efficiency rates. For NUE, the top
(Table 3, Figure 4), indicating significant room for improvement profitability group applied higher N rates, but still were 40% more
among farmers. Notably, the top profitability group used lower efficient than the bottom group due to achieving higher yields.
rates of inputs, with the exception of N fertilizer and seed, Likewise, for GHG emissions, the top profitability group also had
suggesting that intensive use of inputs does not translate to higher methane emissions per unit area, but 40% lower emissions
increased yield but rather economic and environmental risk. than the bottom group when expressed per unit production.
As discussed below, this finding goes against the common Consistent with our findings, previous studies have shown that
assumption that high yield farming, which is associated with measuring the efficiencies of farm systems by per unit area of
profitability, is often viewed synonymously with environmental farmland underestimates the fact that lower-yield systems require
degradation (Foley et al., 2011). more land per unit yield (Balmford et al., 2018). A recent study
In the case of P fertilizer and chemical applications, lack of assessing SRP indicators in six countries (Devkota et al., 2019),
recommendations and farmer knowledge appeared to be the observed that high yielding farmers in Indonesia used higher
basis for the highest rates of input use. Farmers that chose rates of inputs, and were more efficient in their use, with higher
to use compound fertilizers, in a bag-for-bag manner, applied indicator scores. Similarly, in Guangdong, China, Devkota et al.
excessively high P rates at a higher cost and environmental risk. (2019) observed a similar synergy between high yields and high
Extension efforts aimed at helping farmers understand nutrient SRP indicators.
requirements and the elemental composition of fertilizer will help All farms in the target group (i.e., top 20%) for GHG
reduce over or under application of fertilizers and potentially efficiency experienced at least one period of drying in their
increase profitability. Similar programs have been effective in field, which can reduce estimated methane emissions by 48%
optimizing fertilizer efficiency for increasing environmental (IPCC, 2006). These drying periods constitute a significant cause
sustainability and profitability in rice production elsewhere of stress for farmers, as they are not planned but occur due
(Pingali et al., 1998; Huang et al., 2008; Emmanuel et al., 2016). to limited control over irrigation scheduling and community
For pesticide use efficiency, the top profitability group used water use. Still, the farmers with the highest GHG efficiency
25% fewer chemicals than the bottom group. Many famers were able to maintain yields while experiencing drying events,
applied prophylactic pesticides or took recommendations from which is consistent with studies showing that alternate wetting
sales representatives that consisted of tank mixes of several and drying (AWD) can significantly reduce methane emissions

Frontiers in Sustainable Food Systems | www.frontiersin.org 11 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

without yield loss (Linquist et al., 2015; Xu et al., 2015; Liang resource efficiency and profitability. Management practices such
et al., 2016; Setyanto et al., 2018). Recent research (Pando et al., as AWD and direct seeding have the potential to change farmers’
2017) has demonstrated AWD and direct seeding as successful perceptions of water use and reduce production costs, to increase
options in northern Peru, but changes in water distribution farm profitability while reducing GHG emissions. Therefore,
and suitable cultivars that tolerate drying periods are needed. our results are critical for exploring pathways for achieving
Extension efforts demonstrating effective AWD and efficient sustainable development goals, and to help policy-makers
weed management could change the perception that drying prioritize investments in agriculture and food security.
events are harmful for the crop. If farmers view this as a viable
technology rather than an insurmountable problem, farmers DATA AVAILABILITY STATEMENT
could likely regain a sense of confidence, which is a critical aspect
of the farmer decision-making process. The raw data supporting the conclusions of this article will be
Our results are the first to benchmark resource efficiency made available by the authors, without undue reservation.
gaps for rice production in Latin America using SRP indicators,
which is an essential step for identifying options for sustainable ETHICS STATEMENT
intensification. While previous work in other countries used
similar sample sizes to assess SRP indicators (Devkota et al., The studies involving human participants were reviewed and
2019), we recognize there are limitations to interviewing 65 approved by The Urbana-Champaign Office for the Protection of
farmers and we encourage more extensive assessments in Peru Research Subjects (OPRS) and Institutional Review Board (IRB).
and other countries. Overall, synergies were found between crop Written informed consent for participation was not required for
production and sustainability goals because the top profitability this study in accordance with the national legislation and the
group used fewer inputs, achieved the highest yields, and institutional requirements.
scored the highest across all SRP performance indicators. While
this provides necessary evidence that sustainable intensification AUTHOR CONTRIBUTIONS
can be achieved at the farm-level, there is still considerable
room for improvement for the majority of farmers across MW and CP conceived the research idea. MW carried out the
SRP indicators (resource use efficiency gaps ranged 22-77%). experiment, data analysis with supervision from CP, EH, EG,
Indeed, the largest gap was for profitability, underscoring and NC, and led the writing of the manuscript, with all authors
the need for efforts aimed at helping farmers manage inputs providing edits and input to the analysis and final version. All
wisely while simultaneously increasing yields. Our results authors participated in the design of study objectives, the survey
suggest that farmers are capable of achieving high levels of tool, helped interpret results, and develop conclusions.
profitability and productivity at low environmental costs per
unit production. FUNDING
CONCLUSIONS This research was funded by the Crop Sciences Department of the
University of Illinois Urbana-Champaign. Additional funding
Globally, agriculture faces the challenge of balancing the need to was awarded through The Global Food Security Graduate
feed a growing population, improving the livelihoods of farmers, Fellowship from the ACES Office of International Programs and
and protecting the environment for long-term sustainability. In International Food Security at Illinois.
this study, we provide a holistic assessment of these challenges
for rice production in Peru by analyzing potential synergies ACKNOWLEDGMENTS
and tradeoffs across multiple performance indicators, while
also exploring the farmer decision-making process. Our results The authors are thankful to the Ferreñafe irrigation district in
indicate that the factors driving profitability are yield, market Peru and the many farmers who shared data and helpful insights.
price, and production costs. However, socio-ecological factors Robert Andrade (CIAT), Sergio Urioste (CIAT), Diego Marin
such as farmer access to knowledge and financing also impact (CIAT), and Moises Bustamante provided valuable input on the
how rice production systems are managed and specifically creation and implementation of the survey tool for this research.
the level of inputs and costs of production, influencing both Thank you to Sandra Loaiza and Rossana Porras Jorge for input
economic and environmental footprint. These findings are on the creation of the map. The authors would also like to thank
essential to inform future research and extension in this region, Andrew Margenot and Hope Michelson at the University of
which should concentrate efforts toward identifying improved Illinois for critical feedback on this work.
agronomic practices to increase yield under on-farm conditions,
while also recognizing the role of economic and socio-ecological SUPPLEMENTARY MATERIAL
factors that constrain farm management practices. This study
indicated the absence of tradeoffs between profit, yield, and The Supplementary Material for this article can be found
environmental performance per unit of production, which online at: https://www.frontiersin.org/articles/10.3389/fsufs.
suggests a synergy can exist between simultaneous gains in 2020.564418/full#supplementary-material

Frontiers in Sustainable Food Systems | www.frontiersin.org 12 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

REFERENCES Feder, G., Just, R. E., and Zilberman, D. (1985). Adoption of agricultural
innovations in developing countries: a survey. Econ. Dev. Cult. Change 33,
Abdulai, A., and Huffman, W. (1998). An Examination of Profit Inefficiency of Rice 255–298. doi: 10.1086/451461
Farmers in Northern Ghana. Staff General Research Papers. Econ. Staff Pap. Foley, J. A., Ramankutty, N., Brauman, K. A., Cassidy, E. S., Gerber, J. S., Johnston,
Ser. 298, Iowa State University, Department of Economics. Available online at: M., et al. (2011). Solutions for a cultivated planet. Nature 478, 337–342.
https://lib.dr.iastate.edu/econ_las_staffpapers/298 doi: 10.1038/nature10452
Affholder, F., Poeydebat, C., Corbeels, M., Scopel, E., and Tittonell, P. (2013). The Godfray, H. C. J., and Garnett, T. (2014). Food security and sustainable
yield gap of major food crops in family agriculture in the tropics: assessment intensification. Philos. Trans. R Soc. B Biol. Sci. 369:20120273.
and analysis through field surveys and modelling. Field Crops Res. 143, 106–118. doi: 10.1098/rstb.2012.0273
doi: 10.1016/j.fcr.2012.10.021 Huang, J., Hu, R., Cao, J., and Rozelle, S. (2008). Training programs
Aku, A., Mshenga, P., Afari-Sefa, V., and Ochieng, J. (2018). Effect of market access and in-the-field guidance to reduce China’s overuse of fertilizer
provided by farmer organizations on smallholder vegetable farmer’s income without hurting profitability. J. Soil Water Conserv. 63, 165A−167A.
in tanzania. Cogent Food Agric. 4:1560596. doi: 10.1080/23311932.2018.1 doi: 10.2489/jswc.63.5.165A
560596 IPCC (2006). “Guidelines for National greenhouse gas inventories. Volume
Alene, A. D., Manyong, V. M., Omanya, G., Mignouna, H. D., Bokanga, M., and 4: agriculture, forestry and other land use, Ch 5: Cropland,” in National
Odhiambo, G. (2008). Smallholder market participation under transactions Greenhouse Gas Inventories Programme, eds H. S. Eggleston, L. Buendia, K.
costs: maize supply and fertilizer demand in Kenya. Food Policy 33, 318–328. Miwa, T. Ngara, and K. Tanabe (Hayama: Institute for Global Environmental
doi: 10.1016/j.foodpol.2007.12.001 Strategies), 66.
Anderson, J. R. (2008). Agricultural Advisory Services. Washington, DC: Key, N., and Runsten, D. (1999). Contract farming, smallholders, and
World Bank. Available online at: https://openknowledge.worldbank.org/ rural development in Latin America: the organization of agroprocessing
handle/10986/9041 firms and the scale of outgrower production. World Dev. 27, 381–401.
Babcock, B. (1992). The effects of uncertainty on optimal nitrogen applications. doi: 10.1016/S0305-750X(98)00144-2
Rev. Agric. Econ. 14, 271–280. doi: 10.2307/1349506 Khush, G. S. (2005). What it will take to feed 5.0 Billion rice consumers in 2030.
Balmford, A., Amano, T., Bartlett, H., Chadwick, D., Collins, A. L., Edwards, D., Plant Mol. Biol. 59, 1–6. doi: 10.1007/s11103-005-2159-5
et al. (2018). The environmental costs and benefits of high-yield farming. Nat. Krueger, R. A., and Casey, M. A. (2014). Focus Groups: A Practical Guide for
Sustain. 1, 477–485 doi: 10.1038/s41893-018-0138-5 Applied Research. SAGE Publications.
Biernacki, P., and Waldorf, D. (1981). Snowball sampling: problems and Liang, K., Zhong, X., Huang, N., Lampayan, R. M., Pan, J., Tian, K., et al. (2016).
techniques of chain referral sampling. Sociol. Methods Res. 10, 141–163. Grain yield, water productivity and ch4 emission of irrigated rice in response
doi: 10.1177/004912418101000205 to water management in South China. Agric. Water Manag. 163, 319–331.
Bouman, B., Lampayan, R. M., and Tuong, T. M. (2007). Water Management in doi: 10.1016/j.agwat.2015.10.015
Irrigated Rice. Coping with Water Scarcity. (Manila, Philippines: International Linquist, B. A., Anders, M. M., Adviento-Borbe, M. A. A., Chaney, R. L., Nalley,
Rice Research Institute). L. L., da Rosa, E. F. F., et al. (2015). Reducing greenhouse gas emissions, water
Byerlee, D. R., and Anderson, J. R. (1982). Risk, utility and the value of information use, and grain arsenic levels in rice systems. Glob. Change Biol. 21, 407–417.
in farmer decision making. Rev. Market. Agric. Econ. 50:12524. doi: 10.1111/gcb.12701
Carlson, K. M., Gerber, J. S., Mueller, N. D., Herrero, M., MacDonald, G. Lobell, D. B. (2013). The use of satellite data for crop yield gap analysis. Field Crops
K., Brauman, K. A., et al. (2016). Greenhouse gas emissions intensity Res. 143, 56–64. doi: 10.1016/j.fcr.2012.08.008
of global croplands. Nat. Clim. Change 7, 63–68. doi: 10.1038/nclimate Lobell, D. B., Cassman, K. G., and Field, C. B. (2009). Crop yield gaps: their
3158 importance, magnitudes, and causes. Ann. Rev. Environ. Resour. 34, 179–204.
Cazzuffi, C., and McKay, A. (2012). Rice Market Participation and Channels doi: 10.1146/annurev.environ.041008.093740
of Sale in Rural Vietnam (No. 1007-2016-79533). AgEcon Search. Maclean, J., Hardy, B., and Hettel, G. (2013). Rice Almanac, 4th Edition: Source
doi: 10.22004/ag.econ.126566 Book for One of the Most Important Economic Activities on Earth. IRRI
De Roos, A. J., Zahm, S., Cantor, K., Weisenburger, D., Holmes, F., Burmeister, Mankad, A. (2016). psychological influences on biosecurity control
L., et al. (2003). Integrative assessment of multiple pesticides as risk factors and farmer decision-making. a review. Agron. Sustain. Dev. 36:40.
for non-hodgkin’s lymphoma among men. Occup. Environ. Med. 60:e11. doi: 10.1007/s13593-016-0375-9
doi: 10.1136/oem.60.9.e11 Masakure, O., and Henson, S. (2005). Why do small-scale producers choose to
Devkota, K. P., Pasuquin, E., Elmido-Mabilangan, A., Dikitanan, R., Singleton, produce under contract? lessons from nontraditional vegetable exports from
G. R., Stuart, A. M., et al. (2019). Economic and environmental indicators Zimbabwe. World Dev. 33, 1721–1733. doi: 10.1016/j.worlddev.2005.04.016
of sustainable rice cultivation: a comparison across intensive irrigated McNally, S. (2002). Are ‘Other Gainful Activities’ on farms good for the
rice cropping systems in Six Asian countries. Ecol. Indic. 105, 199–214. environment? J. Environ. Manag. 66, 57–65. doi: 10.1006/jema.2002.0576
doi: 10.1016/j.ecolind.2019.05.029 Meijer, S. S., Catacutan, D., Ajayi, O. C., Sileshi, G. W., and Nieuwenhuis,
Dono, G., Cortignani, R., Doro, L., Giraldo, L., Ledda, L., Pasqui, M., et al. M. (2015). The role of knowledge, attitudes and perceptions in the
(2013). Adapting to uncertainty associated with short-term climate variability uptake of agricultural and agroforestry innovations among smallholder
changes in irrigated mediterranean farming systems. Agric. Syst. 117, 1–12. farmers in Sub-Saharan Africa. Int. J. Agric. Sustain. 13, 40–54.
doi: 10.1016/j.agsy.2013.01.005 doi: 10.1080/14735903.2014.912493
Dzanku, F. M., Jirström, M., and Marstorp, H. (2015). Yield gap-based Menapace, L., Colson, G., and Raffaelli, R. (2013). Risk aversion, subjective beliefs,
poverty gaps in rural Sub-Saharan Africa. World Dev. 67, 336–362. and farmer risk management strategies. Am. J. Agric. Econ. 95, 384–389.
doi: 10.1016/j.worlddev.2014.10.030 doi: 10.1093/ajae/aas107
Edwards-Jones, G. (2006). Modelling farmer decision-making: concepts, Muhunyu, J. G. (2012). Is doubling rice production in Kenya by 2018 achievable?
progress and challenges. Anim. Sci. 82, 783–790. doi: 10.1017/ASC20 J. Dev. Sustain. Agric. 7, 46–54. doi: 10.11178/jdsa.7.46
06112 Nakano, Y., Tsusaka, T. W., Aida, T., and Pede, V. O. (2018). Is farmer-to-
Emmanuel, D., Owusu-Sekyere, E., Owusu, V., and Jordaan, H. (2016). Impact of farmer extension effective? the impact of training on technology adoption
agricultural extension service on adoption of chemical fertilizer: implications and rice farming productivity in Tanzania. World Dev. 105, 336–351.
for rice productivity and development in Ghana. NJAS-Wagen. J. Life Sci. 79, doi: 10.1016/j.worlddev.2017.12.013
41–49. doi: 10.1016/j.njas.2016.10.002 Nakazi, F., and Sserunkuuma, D. (2013). Factors affecting the decision and extent
Fageria, N. K., and Baligar, V. C. (2005). Enhancing nitrogen use efficiency in crop of rice-milling before sale among ugandan farmers. Asian J. Agric. Rural Dev. 3,
plants. Adv. Agron. 88, 97–185. doi: 10.1016/S0065-2113(05)88004-6 1–8. doi: 10.22004/ag.econ.198235
FAOSTAT (2020). Retrieved from: http://www.fao.org/faostat/en/#data (accessed Nasir, M., and Kotu, B. H. (2014). The effect of off farm employment on agricultural
January 12, 2020). production and productivity: evidence from gurage zone of southern Ethiopia.

Frontiers in Sustainable Food Systems | www.frontiersin.org 13 October 2020 | Volume 4 | Article 564418
White et al. Balancing Economic and Environmental Performance

J. Econ. Sustain. Dev. 5, 85–98. Available online at: https://cgspace.cgiar.org/ Stevenson, J. R., Villoria, N., Byerlee, D., Kelley, T., and Maredia, M. (2013).
handle/10568/76047 Green revolution research saved an estimated 18 to 27 million hectares from
Nguyen, P., Shearer, E. J., Tran, H., Ombadi, M., Hayatbini, N., Palacios, being brought into agricultural production. Proc. Natl. Acad. Sci. U. S. A. 110,
T., et al. (2019). The CHRS data portal, an easily accessible public 8363–8368. doi: 10.1073/pnas.1208065110
repository for persiann global satellite precipitation data. Sci. Data 6:180296. Stockbridge, M., Dorward, A., and Kydd, J. (2003). Farmer Organizations for
doi: 10.1038/sdata.2018.296 Market Access: A Briefing Paper.
Norton, R. D. (2004). Agricultural development policy: Concepts and experiences. Stuart, A. M., Devkota, K. P., Sato, T., Pame, A. R. P., Balingbing, C., Phung, N.
Agricultural Development Policy: Concepts and Experiences. Available online at: T. M., et al. (2018). On-farm assessment of different rice crop management
https://www.cabdirect.org/cabdirect/abstract/20056701982 practices in the Mekong Delta, Vietnam, using sustainability performance
Onwuegbuzie, A. J., Dickinson, W. B., Leech, N. L., and Zoran, A. G. (2009). A indicators. Field Crops Res. 229, 103–114. doi: 10.1016/j.fcr.2018.10.001
qualitative framework for collecting and analyzing data in focus group research. Stuart, A. M., Pame, A. R. P., Silva, J. V., Dikitanan, R. C., Rutsaert, P.,
Int. J. Qual. Methods 8, 1–21. doi: 10.1177/160940690900800301 Malabayabas, A. J. B., et al. (2016). Yield gaps in rice-based farming systems:
Pando, L. R. G., Villacorta, H. S., Peralta, G. S., and Aguilar, E. C. H. (2017). insights from local studies and prospects for future analysis. Field Crops Res.
Siembra directa: una alternativa para mejorar la sustentabilidad del cultivo 194, 43–56. doi: 10.1016/j.fcr.2016.04.039
de arroz (Oryza sativa L.) en el Perú. Producción Agropecuaria y Desarrollo Sustainable Rice Platform (2015). Available online at: http://www.
Sostenible 6, 13–26. doi: 10.5377/payds.v6i0.5716 sustainablerice.org/assets/docs/SRP%20Standard%20for%20Sustainable
Papademetriou, M. K., Dent, F. J., and Herath, E. M. (2000). Bridging %20Rice%20Cultivation%20v%201.0.pdf. (accessed March 26, 2020).
the Rice Yield Gap in the Asia-Pacific Region: This Publication Brings Thorp, R., Stewart, F., and Heyer, A. (2005). When and how far is group
Together Edited Manuscripts of Papers Presented at the Expert Consultation formation a route out of chronic poverty? World Dev. 33, 907–920.
on “Bridging the Rice Yield Gap in Asia and the Pacific.” Bangkok: doi: 10.1016/j.worlddev.2004.09.016
FAO. Retrieved from: http://www.fao.org/publications/card/es/c/6f856aa2- Tittonell, P., and Giller, K. E. (2013). When yield gaps are poverty traps: the
aa27-58e5-93d6-fa4ff0a076f8/ (accessed October 8, 2020). paradigm of ecological intensification in african smallholder agriculture. Field
Phillips, J. M. (1994). Farmer education and farmer efficiency: a meta-analysis. Crops Res. 143, 76–90. doi: 10.1016/j.fcr.2012.10.007
Econ. Dev. Cult. Change 43, 149–165. doi: 10.1086/452139 Toriyama, K., Heong, K. L., and Hardy, B., (eds.). (2005). “Rice is life: scientific
Phimister, E., and Roberts, D. (2006). The effect of off-farm work on the perspectives for the 21st century,” in Proceedings of the World Rice Research
intensity of agricultural production. Environ. Resour. Econ. 34, 493–515. Conference held in Tokyo and Tsukuba, Japan (Los Baños: International
doi: 10.1007/s10640-006-0012-1 Rice Research Institute; Tsukuba: Japan International Research Center for
Pimentel, D., Berger, B., Filiberto, D., Newton, M., Wolfe, B., Karabinakis, E., et al. Agricultural Sciences), 590.
(2004). Water resources: agricultural and environmental issues. BioScience 54, Van Ittersumvan, M. K., and Rabbinge, R. (1997). Concepts in production ecology
909–918. doi: 10.1641/0006-3568(2004)0540909:WRAAEI2.0.CO;2 for analysis and quantification of agricultural input-output combinations. Field
Pingali, P. L., Hossain, M., Pandey, S., and Price, L. L. (1998). Economics of Crops Res. 52, 197–208. doi: 10.1016/S0378-4290(97)00037-3
nutrient management in asian rice systems: towards increasing knowledge Van Nguyen, N., and Ferrero, A. (2006). Meeting the challenges of global rice
intensity. Field Crops Res. 56, 157–176. doi: 10.1016/S0378-4290(97)0 production. Paddy Water Environ. 4, 1–9. doi: 10.1007/s10333-005-0031-5
0126-3 Watcharapongchai, A., Yooprasert, B., and Keowan, B. (2018). Erd area based
Reardon, T., Barrett, C. B., Berdegué, J. A., and Swinnen, J. F. M. (2009). Agrifood management extension for sustainable rice production: lessons learned from
industry transformation and small farmers in developing countries. World Dev. sustainable rice farmers groups in Det Udom District, Ubon Ratchathani
37, 1717–1727. doi: 10.1016/j.worlddev.2008.08.023 Province. Int. J. Environ. Rural Dev. 9, 1–6.
Rigg, J. (2006). Land, farming, livelihoods, and poverty: rethinking the links in the Wiggins, S., Kirsten, J., and Llambí, L. (2010). The future of small farms. World
Rural South. World Dev. 34, 180–202. doi: 10.1016/j.worlddev.2005.07.015 Dev. 38, 1341–1348. doi: 10.1016/j.worlddev.2009.06.013
Rosenbaum, R. K., Bachmann, T. M., Gold, L. S., Huijbregts, M. A. J., Jolliet, Wilson, R. S., Hooker, N., Tucker, M., LeJeune, J., and Doohan, D. (2009).
O., Juraske, R., et al. (2008). USEtox—the UNEP-SETAC toxicity model: Targeting the farmer decision making process: a pathway to increased
recommended characterisation factors for human toxicity and freshwater adoption of integrated weed management. Crop Protect. 28, 756–764.
ecotoxicity in life cycle impact assessment. Int. J. Life Cycle Assess. 13, 532–546. doi: 10.1016/j.cropro.2009.05.013
doi: 10.1007/s11367-008-0038-4 Xu, Y., Ge, J., Tian, S., Li, S., Nguy-Robertson, A. L., Zhan, M., et al. (2015).
RStudio Team (2020). RStudio: Integrated Development Environment for R. Effects of water-saving irrigation practices and drought resistant rice variety
(Boston, MA: RStudio). on greenhouse gas emissions from a no-till paddy in the Central Lowlands of
Samarasinghe, G. B. (2003). Growth and yields of Sri Lanka’s major crops China. Sci. Total Environ. 505, 1043–1052. doi: 10.1016/j.scitotenv.2014.10.073
interpreted from public domain satellites. Agric. Water Manag. 58, 145–157. Zhang, W., Cao, G., Li, X., Zhang, H., Wang, C., Liu, Q., et al. (2016). Closing
doi: 10.1016/S0378-3774(02)00130-0 yield gaps in China by empowering smallholder farmers. Nature 537:19368.
Seck, P. A., Diagne, A., Mohanty, S., and Wopereis, M. C. S. (2012). Crops that feed doi: 10.1038/nature19368
the world 7: rice. Food Secur. 4, 7–24. doi: 10.1007/s12571-012-0168-1
Setyanto, P., Pramono, A., Adriany, T. A., Susilawati, H. L., Tokida, Conflict of Interest: The authors declare that the research was conducted in the
T., Padre, A. T., et al. (2018). Alternate wetting and drying reduces absence of any commercial or financial relationships that could be construed as a
methane emission from a rice paddy in central java, indonesia without potential conflict of interest.
yield loss. Soil Sci. Plant Nutr. 64, 23–30. doi: 10.1080/00380768.2017.14
09600 Copyright © 2020 White, Heros, Graterol, Chirinda and Pittelkow. This is an open-
Sheriff, G. (2005). Efficient waste? why farmers over-apply nutrients and the access article distributed under the terms of the Creative Commons Attribution
implications for policy design. Appl. Econ. Perspect. Policy 27, 542–557. License (CC BY). The use, distribution or reproduction in other forums is permitted,
doi: 10.1111/j.1467-9353.2005.00263.x provided the original author(s) and the copyright owner(s) are credited and that the
Soe, W. P. P., Moritaka, M., and Fukuda, S. (2015). An analysis of the factors original publication in this journal is cited, in accordance with accepted academic
influencing marketing channel choice by paddy rice farmers in Myanmar. J. practice. No use, distribution or reproduction is permitted which does not comply
Faculty Agric. 60, 535–542. with these terms.

Frontiers in Sustainable Food Systems | www.frontiersin.org 14 October 2020 | Volume 4 | Article 564418