Published January 25, 2017
CROP ECONOMICS, PRODUCTION & MANAGEMENT
TheProduction,Consumption,andEnvironmentalImpacts
ofRiceHybridizationintheUnitedStates
L.Nalley,*J.Tack,A.Durand,G.Thoma,F.Tsiboe,A.Shew,andA.Barkley
ABSTRACT
The introduction of hybrid rice in the United States gives
producers an alternative to traditionally cultivated, conventional
lines. The objective of our study is to estimate the economic
(consumer and producer welfare) and environmental impacts of
the commercial adoption of hybrid rice in the Mid-South of the
United States. In our study, the revenue gains associated with
hybrid adoption were estimated at an average of $76.24 million
annually from 2003 to 2013, using existing research findings
in combination with original modeling. Disease packages,
specifically the blast resistance found in all publically released
hybrids, led to both cost and fungicide reductions, which in turn
result in higher profits and increased levels of environmental
sustainability. The cost savings from eliminating fungicide
applications and mitigating yield loss through embedded blast
resistance in hybrids were estimated at $14.35 million annually
from 2003 to 2013. The RiceFlow model results from our
study suggest that the yield premiums through heterosis and
blast resistance associated with hybrid adoption in the MidSouth increased US exports by 383,000 Mg annually and has
fed an additional 5.89 million people annually. Furthermore,
the results from our detailed Life Cycle Assessment show that
hybrid rice has lower environmental (fossil fuel depletion,
ecotoxicity, carcinogenics, eutrophication, acidification, global
warming, and ozone depletion) impacts per megagram of rice
than conventional rice.
Core Ideas
• This study estimated the economic and environmental impact of
hybrid rice adoption.
• The blast resistance in hybrid rice is estimated to be worth
$14.35 million annually.
• Hybrid adoption in the Mid-South feeds an additional 5.89 million people annually.
• The LCA indicated that hybrid rice had lower environmental
impact, mainly due to increase d yields.
Published in Agron. J. 109:193–203 (2017)
doi:10.2134/agronj2016.05.0281
Received 16 May 2016
Accepted 30 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/)
R
ice (Oryza sativa L.) is a global staple food crop that
provides the primary source of calories for more than
50% of the world’s population (World Bank, 2013).
The United Nations estimates that global population will
increase 33% by 2050, from 7.2 billion today to 9.6 billion
persons. Consequently, rice will continue to play an important nutritional role because it is the staple crop in many of
the countries that are experiencing rapid population growth.
In comparison to other rice-producing countries, the United
States is a small producer, generating only 1.3% of the world’s
rice; however, the United States has been among the top five
rice exporters for several decades (Lakkakula et al., 2015).
Because rice provides 21% of global human per capita energy
and 15% of per capita protein (International Rice Research
Institute, 2013), moderate price/supply shocks can have large
impacts on low-income rice consumers. For example, in 2008,
when rice prices tripled due largely to trade restrictions in India
and Egypt, the World Bank estimated that an additional 105
million people were pushed into poverty (World Bank, 2013).
This price turmoil occurred with only an 8% reduction in trade
from 2007 (Childs, 2009).
With this in mind, the international rice market is highly
volatile for a number of reasons. Some of those reasons are that
rice has an inelastic supply and demand (i.e., the percentage
supply response to a change in price is less than the percentage
change in price) throughout much of its primary production
and consumption area (Asia), and it is thinly traded on a
global scale. Global rice exports are highly concentrated, with
the top five exporters (Thailand, India, Vietnam, Pakistan,
and the United States) controlling 87% of global net trade
(Wailes and Chavez, 2012). Given that the global rice market
is so thinly traded and small shocks in price/supply can have
large ripple effects, the adoption of hybrid rice in the United
States (beginning in 2000 with its associated yield increases)
potentially influenced global food security. Since the first
commercially available hybrid rice line was released in the
Mid-South in 2000, the most tangible benefit has been the
increased yields associated with its adoption. Hybrids can yield
15 to 20% more than conventional cultivars on similar land
L. Nalley, A. Durand, F. Tsiboe, and A. Shew, Dep. of Agricultural
Economics and Agribusiness, Univ. of Arkansas, Fayetteville,
AR 72701; J. Tack, Dep. of Agricultural Economics, Mississippi
State Univ. Starkville, MS 39762; A. Barkley, Dep. of Agricultural
Economics, Kansas State Univ., Manhattan, KS 66506; G. Thoma,
Dep. of Chemical Engineering, Univ. of Arkansas, Fayetteville, AR
72701. *Corresponding author ([email protected])
Abbreviations: CL, Clearfield; GHG, greenhouse gas; GM,
genetically modified; LCA, Life Cycle Assessment
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
193
due to yield-improving genetic traits from parent cultivars
(Yuan and Virmani, 1988). Consequently, rice producers in
the Mid-South have been rapidly adopting hybrids since their
commercial release in 2000. In effect, hybrid acreage in the
Mid-South, as a percentage of total harvested acreage, increased
from 15% in 2005 to over 40% in 2013 (Nalley et al., 2016).
Previous research in the Mid-South (Arkansas) found that
hybrid varieties exhibited mean paddy yield premiums of 1.6
to 2.4 Mg ha-1 relative to the best-performing conventional
cultivar (‘Francis’) and were found to be associated with no
increase in yield variability (Lyman and Nalley, 2013). A
broad study by Nalley et al. (2016) that covered Arkansas and
Mississippi observed that hybrid and Clearfield (CL) hybrid
varieties had a paddy yield premium over conventional varieties
of 1.66 and 1.82 Mg ha-1, respectively. The authors found
that, on average, hybrid varieties outperform conventional rice
varieties in terms of absolute profit per hectare and relative
profit margin, defined as profit per cost of production, for both
CL and non-CL varieties.
The most tangible benefit from the adoption of hybrid rice
in the United States is increased paddy yield. Blast resistance is
another imbedded trait that all commercially released hybrid
rice in the United States possesses. Rice blast, caused by the
Magnaporthe oryzae fungus, is one of the most frequent and
costly rice diseases in the Mid-South and in other temperate
rice-growing regions worldwide (Wang and Valent, 2009). The
rice blast fungus is responsible for up to 30% of losses in global
rice production and therefore is a key concern in combating food
insecurity (Skamnioti and Gurr, 2009). It has been estimated
that the worldwide annual loss of rice to blast could feed more
than 60 million people. In addition to yield loss aversion,
embedded blast resistance in hybrid rice also allows producers
to forgo fungicide applications, which can cost rice producers
over $49 ha-1 (Tsiboe et al., 2016). This blast resistance or
“maintenance breeding” is often undervalued by producers and
economists because it is difficult to measure and because it does
not necessarily increase maximum yield potential.
In other words, productivity enhancement is traditionally
estimated in terms of yield gains per hectare and increased
total supply, whereas productivity maintenance (in our case,
blast resistance) is measured in terms of the yield losses avoided
through embedded seed technology. Marasas et al. (2003)
found that the economic impact of wheat breeding efforts for
pathogen resistance (maintenance breeding) can be as great if
not greater than the impact of the associated yield increases
of a breeding program. As such, the valuation of agricultural
research is not holistic unless it accounts for the losses avoided
by its continual maintenance component (Araji et al., 1978;
Knutson and Tweeton, 1979; Plucknett and Smith, 1986).
To demonstrate the lack of commercially available blastresistant rice varieties in the Mid-South, only 19 of the 59
varieties (32%) planted in the three largest rice-producing states
in the Mid-South (Arkansas, Louisiana, and Mississippi) from
2002 to 2014 were blast resistant; 14 (74%) of those were hybrid
lines. Of the nonhybrid lines that were blast resistant, only
the variety Drew was ever sown to more than 5% of any MidSouth state from 2002 to 2014, indicating that blast-resistant
conventional varieties were not widely adopted by producers
(Rice Technical Working Group, 2001–2013). Although few
194
commercially available varieties are blast resistant, it is important
in the context of producer profitability in that it does not
require the application(s) of fungicide and is beneficial to the
environment through reduced toxicity exposure. Applications
of fungicide lead to increased toxicity as well as increased
greenhouse gas (GHG) emissions. The higher yields and the
lower GHG and toxicity levels associated with hybrids can
result in higher efficiency levels per kilogram of grain produced.
Specifically, Nalley et al. (2014) found that hybrid cultivar
production in Arkansas was estimated to be 23.22% more
efficient in converting GHG inputs into grain output than
conventional cultivars. The authors found that, on average, the
production of 1 kg of hybrid rice results in 0.001 Mg of CO2eq
produced. By comparison, a conventional cultivar is estimated
to release 0.00124 Mg of CO2e, a 23.22% increase over hybrid
rice. However, the aforementioned authors did not account for
differences in GHG emissions from blast resistance or GHG
emissions from yield losses associated with blast-susceptible
cultivars, which would increase the hybrid GHG efficiency level.
Previous studies have analyzed various components of the
economic and environmental impacts of hybrid adoption in
the Mid-South but not the totality of those components. This
study builds off of the existing literature but delves further into
a holistic estimate of the economic and environmental impacts
of hybrid adoption on various stakeholders in the rice supply
chain. Specifically, this study analyzes the additional yield that
hybrids have provided both from genetic enhancements and
maintenance breeding (resistance to rice blast) from 2003 to
2013 in the Mid-South of the United States. This estimated
additional supply from hybrid adoption was then put into the
RiceFlow model to answer the counterfactual question: What
would the implications be if hybrids had not been released in
the United States? The RiceFlow model generates estimates of
changes in rice price given a decreased supply as well as changes
in consumer welfare. Finally, the counterfactual decreased yield
and increased fungicide usage from the absence of hybrids were
analyzed in a Life Cycle Assessment (LCA) model to assess
the environmental impact that would have resulted if hybrid
rice had not been commercially released. These comprehensive
results provide insight into how hybrid rice adoption in the
Mid-South affects producer livelihoods, food security, and
environmental sustainability.
Materials and Methods
The Value of Blast Resistance in Hybrid Rice
This study follows the methodology put forth by Tsiboe et al.
(2016), who estimated the economic cost of cultivating blastsusceptible varieties in several Mid-South states (Arkansas,
Louisiana, and Mississippi). The authors compiled data
pertaining to the cost of mitigation per hectare as a function
of (i) fungicide and application costs, (ii) probability of a blast
outbreak by variety-specific blast susceptibility ratings, and (iii)
variety-specific yield loss by blast susceptibility ratings for all
blast-susceptible varieties planted in 2002 to 2014. In Tsiboe
et al. (2016), the authors disregarded hybrid lines because they
were all resistant to blast, and as such, they experienced neither
yield loss nor additional production costs. This study presents
the counterfactual argument: all hybrid lines are “moderately
resistant” to blast, not “resistant.”
Agronomy Journal • Volume 109, Issue 1 • 2017
Susceptibility ratings put forth by university extensions
range from very susceptible, moderately susceptible, moderately
resistant, to resistant. The more susceptible a variety is, the higher
the probability of its infection being severe, and thus the higher
probability of a yield loss. As such, this study takes a conservative
approach and assumes that all hybrids are “moderately resistant,”
which is associated with the lowest mean yield loss of all the
nonresistant blast ratings. In our counterfactual situation,
rice producers could replace a blast-resistant hybrid line with a
blast-resistant conventional line; however, there has not been
a blast-resistant conventional variety since 2002 (‘Drew’) that
has been sown to more than 5% of Arkansas, Louisiana, or
Mississippi (Rice Technical Working Group, 2001–2013).
Therefore, given the rice varieties currently available to purchase,
it is highly improbable that a producer would choose to cultivate
an available blast-resistant conventional variety. Importantly,
if producers could switch to a blast-resistant conventional
variety then the value of blast resistance in hybrid rice would be
mitigated through substitutability.
Following Tsiboe et al. (2016), actual data were used on
hybrid planting areas from Arkansas, Mississippi, and Louisiana,
simulating two alternative “outbreak” scenarios annually
from 2002 to 2014. The annual hybrid rice areas planted for
each rice-growing county/parish in Arkansas, Louisiana, and
Mississippi were collected from 2002 to 2014 to capture hybrid
adoption rates (Rice Technical Working Group, 2001–2013).
Additionally, annual potential yield (Mg ha-1) data by
commercial hybrid rice and county/parish were obtained from
university-run experiment stations (Arkansas Agricultural
Experiment Station, 2016; Louisiana State University Ag
Center, 2015; Mississippi Agricultural and Forester Experiment
Station, 2015). The data consist of 14 rice hybrids, 32 ricegrowing counties in Arkansas, 27 parishes in Louisiana, and 18
counties in Mississippi totaling to 1478 area/yield observations
from 2000 to 2014 (Table 1). Given that universities do not
conduct yield test plots in each county/parish for each year, the
missing county/parish level variety-specific yields were replaced
with the average yield for that variety in all test plots that it was
produced in that state for that year. The average yields by hybrid
variety and by state are reported in Table 1.
Blast, like other fungi, have favorable conditions for growth,
which can be a function of field conditions, variety type, and
climate. Blast presence one year in a specific field does not
necessarily mean presence next year because a hard overwinter
freeze can kill blast spores. However, if a specific field is located
by a tree line then it is more likely to have consistent problems
with blast because nearby trees tends to extend the dew period
on rice leaves and thus provide more favorable conditions for
fungus growth. Thus, blast presence is partially tied to field
conditions and partially tied to climatic conditions, making
it difficult for a producer to forecast from year to year and
difficult to model outbreak probabilities and subsequent yield
losses. To address these uncertainties, the estimates of Tsiboe
et al. (2016) on (i) the blast-outbreak rate, (ii) the yield-loss
rate for rice cultivars that are moderately resistant to blast, and
(iii) the cost of blast mitigation per hectare were used in this
study. The percentage of the “moderately blast resistant” hybrid
that could have been infected with blast was simulated using
data from the University of Arkansas, which indicate that,
on average, 22.52% of all acreage in Arkansas had a fungicide
application to help mitigate blast from 2000 to 2014. Table 1
indicates that the blast infection rate for all nonresistant
varieties in the Mid-South ranged from 0.00 to 46.95% from
2000 to 2014, with a mean of 21.52. Thus, if hybrid cultivars
were not blast resistant, we assume they would experience the
same infection rate distribution and simulate infection rates
as such. Empirical yield loss data illustrated in Table 1 indicate
that rice cultivars rated as “moderately blast resistant” have
averaged yield loss in the Mid-South of 9.79% of potential yield
(Groth et al., 2015). The mean was 9.79 with an SD of 5.59.
This was assumed to be a normal distribution, truncated at 0%
and simulated 1000 times to obtain a range of possible yield
losses. As such, if hybrid rice was no longer blast resistant and
was instead “moderately blast resistant,” yield loss distribution
would occur (Table 1). Thus, this study takes the probability
of infection distribution and multiplies it by the yield loss
distribution, both found on Table 1, to obtain a total yield loss.
The cost of aerial application fungicide was estimated
at $19.77 ha-1, and the cost of fungicide was estimated at
$51.10 ha-1 (Tsiboe et al., 2016). Furthermore, the estimates
of the cost of blast mitigation were based on the two most
commonly used fungicides to treat blast in the Mid-South in
2015, which were Quilt Xcel (Syngenta) (active ingredients:
13.5% azoxystrobin and 11.7% propiconazole) and Quadris
(Syngenta) (active ingredient: 22.9% azoxystrobin) (K. Driggs,
personal communication, 2015). Because there are no accessible
data on what percentage of producers used Quadris and Quilt,
it was assumed that they were used in equal proportions. Other
costs not included in this study include crop consulting costs
associated with blast management, producers time scouting
for blast, and time associated with mitigation efforts. Because
these are minor costs relative to total production, it would
suggest that the total costs estimated in this paper are likely
conservative.
Table 1. Simulated blast infection rate and yield loss rate by blast susceptibility rating.
Mean
Infection rate, %†
21.52
Blast yield loss rate by susceptibility rating, %‡
Resistant
0.00
Moderately resistant
9.79
Moderately susceptible
12.84
Susceptible
15.89
Very susceptible
18.32
SD
12.01
Max
46.95
Min
0.00
0.00
5.59
3.85
5.35
8.06
0.00
21.37
22.88
24.53
34.43
0.00
0.00
0.00
0.00
0.00
† Simulated using estimates of the yearly percentage of rice area that required a fungicide. application, reported by Norman and Moldenhauer (2016).
‡ Groth et al. (2015).
Agronomy Journal • Volume 109, Issue 1 • 2017
195
Blast Outbreak Scenario 1
In Scenario 1, the model simulates the infected area of
hybrid rice yearly, which has had its blast rating changed from
“resistant” to “moderately resistant” (an average of 22.11%)
based on of the estimates put forth by Groth et al. (2015).
The simulated infected area was then assumed as having
been treated with two applications of fungicide to mitigate
this outbreak; as such, Scenario 1 was not associated with a
yield loss. Current Mid-South production practices suggest
two applications of fungicide when blast is observed during
the vegetative stage: one application at the late booting stage
and one application 7 d after the 90% panicle emergence of
the main tiller when blast is spotted in a field (University
of Arkansas Cooperative Extension Service, 2016). Thus,
Scenario 1 was modeled as follows:
parish l, and the season-average farm price for type g rice at time
t (Pgt). Pgt is measured in $ Mg-1 and aggregated at the graintype level (g = 0 for medium; g = 1 for long grain) as reported
by USDA (2015). The variable gi is the simulated average yield
loss percentage (i.e., 9.79%) for a “moderately blast-resistant” rice
variety, which was estimated by Groth et al. (2015).
Scenario 1 provided the total estimated fungicide usage,
which was then incorporated into the LCA (toxicity) model
below, in addition to an associated cost savings value, which
was used in determining the net value of hybrids. Scenario 2
provided both a cost savings value and fungicide usage as well
as the volume (Mg) of rice that would have been lost to blast
if hybrids had not been blast resistant. This volume of lost
production was used in the RiceFlow model below to estimate
the price and supply effects of the counterfactual argument,
which is that hybrid rice was not released in the Mid-South.
TC1t = 2C l ∑ t Ailt [1]
Value of Genetic Gain
where the total economic cost of hybrids moving from “blast
resistant” to “moderately resistant” in year t (TC1t) is the
summation of all actual historic hectares (Ailt) of hybrid rice
varieties (i) sown in each rice-producing county/parish (l) in a
given year (t), multiplied by the simulated infection rate of blast
(l), multiplied by the cost of two applications of fungicide (C).
Equation [1] is a function of time because the distribution of
the county-level hybrid rice varietal and the total area sown
changes yearly. In this scenario, all hybrid varieties have equal
probabilities of infection because they are all assumed to be
moderately resistant to blast.
Blast Outbreak Scenario 2
In Scenario 2, the model simulates the area of hybrid rice
varieties that are infected with blast and simulates a corresponding
yield loss by variety associated with the infection. Yield losses
associated with blast infection for “moderately resistant” varieties
were simulated from distributions derived from Groth et al.
(2015). Similar to Scenario 1, the infected areas were assumed to be
associated with two applications of fungicide, but, unlike Scenario
1, there was a subsequent yield loss associated with the infection.
Although a draw from the simulated distribution provided a
fixed percentage of yield loss for all varieties, each hybrid variety
had a different yield potential for each location and each year. As
such, each variety’s average yield was denoted by county/parish, as
reported by each state’s extension service. Also, a fixed percentage
of that reported yield by location and year was lost to blast
(Arkansas Agricultural Experiment Station, 2016; Louisiana State
University Ag Center, 2015; Mississippi Agricultural and Forester
Experiment Station, 2015). Thus, hybrid varieties with higher yield
potential lose a higher amount of yield due to blast, although it is
the same percentage as all other hybrid varieties. Scenario 2 was
modeled as follows:
TC2
=
TC1t + ∑g iYil Pgt [2]
t
t
where the annual total economic cost of hybrid lines moving
from “blast resistant” to “moderately resistant” (TC2t) is the
summation of the annual total economic cost calculated for
Scenario 1 (TC1t), and the product of the simulated yield loss
due to blast giYil, associated with hybrid variety i in county/
196
Results from Nalley et al. (2016) were used to estimate the
genetic gain associated with hybrid adoption in the Mid-South.
The objective of their study was to quantify the economic
performance of hybrid and conventional rice in the MidSouth. This objective, in partnership with their modeling
approach, complemented the research goals of our study.
Unlike the previously mentioned blast model, Nalley et al.
(2016) only estimated the amount and dollar value of heterosis
for Arkansas and Mississippi, not Louisiana. This action was
taken primarily because the university test plots administered
by Louisiana State University did not grow hybrid lines
continuously throughout the study’s timeframe (2002–2013);
as such, these data could not be estimated. Consequently, the
results from Nalley et al. (2016) that were used in our study are
conservative in their estimation of the total gains in the MidSouth, given that Louisiana produced 15% of the total US rice
crop in 2015, whereas Arkansas produced 51% and Mississippi
produced 6% (USDA–ERS, 2015).
Nalley et al. (2016) used seven test plot locations in Arkansas
and 16 in Mississippi. Yield data (Mg ha-1) were obtained
from the University of Arkansas Division of Agriculture
Arkansas Rice Performance Trials (2000–2012) and the
Mississippi State Agriculture and Forestry Experiment Station
(2000–2012) rice variety trials from 2003 to 2013. A total
of 65 varieties were analyzed, including 14 pure hybrids, 10
CL hybrids, 10 CL lines, and 31 conventional varieties. Their
multivariate ordinary least squares regression model to elicit
hybrid yield advantage was defined as:
yilt = ai + al + at + wiltb + wilt [3]
where yilt is yield measured in Mg ha-1 for rice type i, at
location l, in trial year t, and ai is a fixed effect for rice type
(conventional, hybrid, CL, and CL hybrid) i. It was concluded
that the use of field-trial data precluded the need to control for
economic variables in the regression model because optimal
management practices were followed regardless of the current
price levels. The vector (wilt) included temperature, vapor
pressure deficit, and solar radiation for each growth-stage
window for each variety–location–year. From this, yield
premiums could be determined for hybrids over conventional
Agronomy Journal • Volume 109, Issue 1 • 2017
Table 2. Environmental impact categories used in the life cycle analysis for hybrid vs. conventional rice.
Category
Units
Description
Ozone depletion
kg CFC-11
accumulated ozone-depleting compounds emissions
accumulated greenhouse gas emissions (IPCC, 2007 characterization factors)
Global warming
kg CO2 eq
small forming potential
Smog
kg O3 eq
terrestrial acidification driven by acid gases
Acidification
kg SO2 eq
Eutrophication
kg N eq
freshwater and marine eutrophication driven by nutrient runoff
Carcinogenics
CTUh
human toxicity units
Noncarcinogenics
CTUh
human toxicity units
Fossil fuel depletion
MJ surplus
nonrenewable energy consumption
Eco-toxicity
CTUe
ecosystems toxicity units
Respiratory effects
kg PM2.5 eq
primary and secondary particulate emissions
varieties and for CL hybrids over CL lines. As such, we use the
hybrid yield coefficients as estimated by Nalley et al. (2016) in
this study for Arkansas and Mississippi as well.
The Nalley et al. (2016) study estimated total revenue
enhancement from hybrids. By dividing their findings by annual
price per megagram, the total volume enhancement (Mg)
premium associated with hybrid adoption could be derived.
These results provide a dollar estimate of enhanced producer
revenue from hybrid adoption and an increased volume of
the rice produced. The volume (Mg) of additional production
resulting from genetic gains and from avoided yield losses
through blast resistance was used in the RiceFlow model below
to estimate the price effects of the counterfactual, which is that
hybrid rice was not released in the Mid-South. Specifically,
the counterfactual case was presented in this way: How much
does rice price increase and consumers lose if hybrid rice is not
introduced into production in the Mid-South?
RiceFlow Model
RiceFlow is a spatial, supply-chain, partial equilibrium
model of the global rice economy in which the behaviors of
producers and consumers are specified according to neoclassical
economic theory (profit and utility maximizers, respectively).
RiceFlow is used extensively to assess different aspects of the
global rice economy. For instance, Thompson et al. (2015)
used RiceFlow to assess the impact of the US Liberty Link rice
contamination. Furthermore, Briones et al. (2012) and Wailes
et al. (2015) used the model to assess alternative rice policy
options in Southeast Asia and Western Africa, respectively. In
our study, the RiceFlow model (Durand-Morat and Wailes,
2010) was used to assess the impact of hybrid rice adoption
on the US rice market due to its improved yield potential
and resistance to blast. That is, the yield gains (both through
genetic and maintenance breeding) were summed and then
run through the RiceFlow model. RiceFlow was calibrated to
market conditions in the calendar year 2013. The global rice
economy was disaggregated into 73 regional markets and nine
rice commodities derived from the combination of rice type
(long, medium, and fragrant rice) and milling degree (paddy,
brown, and milled rice). Whereas the value of genetic gains
and blast resistance were calculated yearly from 2003 to 2013,
the consumer, export, import, and price impacts were only
estimated for the production year of 2013 because this was the
most recently calibrated version of the RiceFlow model.
To achieve the goal of this study, the counterfactual scenario
of removing all gains (from heterosis and maintenance
breeding) from hybrid adoption was entered into the RiceFlow
model. Thus, the results can be interpreted as the market
conditions that would have prevailed in 2013 if all hybrid rice
production in the Mid-South was replaced by conventional rice
varieties that were moderately susceptible to rice blast.
The Environmental Impacts of Hybrid Rice
The LCA was performed to provide quantitative comparison
of the cradle-to-farm gate environmental benefits realized by
hybrid rice production. The goal was to provide a comparison for
the functional unit of 1 kg of rice dried to 12% moisture at the
farm gate ready for transportation to processing. The principal
differences between the two (hybrid vs. conventional) systems
are yield and fungicide application (no Quadris and associated
aerial application emissions). The inputs for each system, in
terms of planting, fertilizer, and pesticide application (except as
noted) as well as harvesting, had the same of crop area. We have
used the TRACI 2.1 lifecycle impact assessment framework,
which was developed by the USEPA for conditions in the
United States (Bare et al., 2006). The impact categories included
in this method are presented in Table 2. To minimize bias in
the comparison between hybrid and conventional writings,
we adopted a paired Monte Carlo simulation approach using
SimaPro 8.1, which selects variates from each unit process in
the supply chain and computes the difference between the two
(hybrid vs. conventional) product systems. This approach ensures
that additional variability from independent simulations of the
supply chains is not introduced. From this methodology we can
ascertain the differences between hybrid and conventional rice
production from a holistic environmental standpoint. Figure 1
illustrates the modeling procedure for estimating total yield gain
from hybrid rice, environmental impacts from hybrid adoption,
and consumer/producer welfare impacts from hybrid adoption.
Results
The Value of Blast Resistance
Embedded in Hybrid Rice
The total (aggregated annual) economic cost results of
Scenarios 1 (mitigation cost with no yield loss) and 2 (mitigation
cost plus yield loss) are presented in Table 3. All monetary
values included in this paper are converted to 2014 USD using
annual consumer price index retrieved from IMF (2016). The
results from scenario 1 (all hybrids go from blast “resistant”
to “moderately resistant” and as such are treated with one
application of aerial fungicide) indicate that the blast tolerance
offered by hybrids has a value of $4.50 million annually and
Agronomy Journal • Volume 109, Issue 1 • 2017
197
Fig. 1. Data sources and flow for estimating total yield gain from hybrid rice, environmental impacts from hybrid adoption and consumer/
producer welfare impacts from hybrid adoption.
Table 3. Estimated economic cost of blast mitigation in Arkansas, Louisiana, and Mississippi: 2000–2014. All hectares planted to hybrid
varieties assumed to be moderately resistant to blast were infested with a simulated blast rate of 22.11% and then sprayed twice with fungicide to prevent a yield loss.
Sown hybrid
Blast-infested
Fungicide
Blast mitigation Long-grain rice, Total yield loss Total economic
Year
area†
area‡
usage§
cost, Scenario 1 average price¶
to blast
loss#
-1
—————— ha ——————
liters
Mg
$
$ Mg
2003
6,414
1,418
2,865
167,869
215.62
1,551
502,349
2004
14,805
3,274
6,613
397,396
202.82
3,444
1,095,794
2005
23,172
5,124
10,351
637,548
195.09
5,259
1,665,311
2006
71,196
15,744
31,803
2,030,517
245.17
15,346
5,792,896
2007
123,910
27,401
55,350
3,658,619
312.13
28,196
12,459,401
2008
158,044
34,949
70,597
4,825,564
401.19
36,471
19,457,459
2009
173,353
38,334
77,436
5,293,003
324.12
33,208
16,056,198
2010
254,338
56,243
113,611
7,851,076
280.00
59,758
24,583,408
2011
243,456
53,837
108,750
7,760,223
287.71
53,227
23,074,226
2012
292,491
64,680
130,654
9,519,483
311.95
62,799
29,109,979
2013
224,067
49,549
100,089
7,367,732
337.44
49,471
24,061,072
Avg.
144,113
31,868
64,374
4,500,821
283.02
31,703
14,350,736
Total
1,585,246
350,553
708,119
49,509,030
–
348,730
157,858,093
† Source: Rice Technical Working Group (2011–2013).
‡ Simulated by authors using annual varietal area planted for each rice-growing county in Arkansas, Louisiana, and Mississippi retrieved from the
Proceedings of the Rice Technical Working Group (various years).
§ Fungicide application of 1.01 L ha-1 and at a cost of $75.87 ha-1 (aerial application at $19.77 ha-1 and $51.10 ha-1 for fungicide cost).
¶ Values in 2014 US$; deflated with consumer price index retrieved from IMF (2016).
# In 2005, there was a yield loss of 30.23 Mg for medium-grain hybrid rice valued at the seasonal medium-grain average rice price of $253.62 Mg-1;
this is added into the long-grain loss value.
198
Agronomy Journal • Volume 109, Issue 1 • 2017
$49.51 million over the period 2003 to 2013 (Table 3). If there
is an associated yield loss, which is based on historical trials
of varieties that are moderately resistant to blast, the value of
blast tolerance in hybrids increases to $14.35 million dollars
annually or $157.86 million over the period 2003 to 2013. This
number is partially driven by rice price and adoption rates, both
of which are exogenous, but also by hybrid yield potential and
the protection of that yield potential through resistance to blast,
both of which are endogenous to the hybrid variety.
Agricultural productivity enhancement is traditionally
estimated in terms of yield gains per hectare and increased
total supply; however, in our study productivity maintenance
(i.e., blast resistance) was measured in terms of the yield losses
avoided through embedded seed technology. Thus, the results
for the value of blast resistance can be viewed as the value of
the costs of yield losses and fungicide application that can be
avoided through breeding for blast resistance. In other words,
without blast resistance embedded in hybrid rice varieties in
the Mid-South, producers would incur larger costs at $4.50
million annually and less revenue at $14.35 million annually.
Yield and Total Genetic Gain
from Hybrid Adoption
Using the regression results from Eq. [3], Nalley et al. (2016)
estimated that, in Arkansas and Mississippi, hybrids and CL
hybrids were associated with average yield premiums of 1.66
and 1.82 Mg ha-1, respectively. The average estimated yields
for conventional varieties and CL varieties were 9.05 and
8.79 Mg ha-1, respectively, indicating that on average hybrids
and CL hybrids were associated with an 18.3 and 20.1%
increase in yields, respectively, relative to conventional and
conventional CL varieties. From these regression results, this
study was able to create Table 4, which highlights the estimated
heterosis (genetic) gains associated with hybrid adoption
based on the actual hybrid adoption from 2003 to 2013. On
average, between 2003 and 2013, producers in Arkansas and
Mississippi gained an additional $76.24 million annually from
hybrid adoption. Overall, the total gains in the two states for
the same time period were estimated to be $838.70 million.
Combining the yield enhancement (genetic gains), the
yield loss prevention, and cost savings (maintenance breeding)
provided a holistic benefit of hybrid rice adoption in the
Table 4. Additional yield and revenue (2014 USD) attributed to Clearfield (CL) hybrid and non-CL hybrid adoption in Arkansas and
Mississippi, 2003–2013.
Additional yield
Additional yield
Total additional
CL hybrid
attributed to non- Non-CL hybrid yield attributed
Total revenue
attributed to CL
Year
hybrid adoption†
adjusted price CL hybrid adoption adjusted price
to hybrids
gain‡§
-1
-1
Mg
Mg
Mg
2014 $ Mg
2014 $ Mg
Arkansas
2003
0
$214
14,121
$214
14,121
$3,015,939
2004
41,155
$196
18,343
$194
59,498
$11,608,892
2005
16,106
$173
19,572
$177
35,678
$6,255,727
2006
56,494
$227
39,474
$233
95,967
$21,981,728
2007
96,557
$288
46,760
$280
143,317
$40,870,778
2008
252,197
$343
1,959
$311
254,156
$87,128,409
2009
259,478
$354
29,827
$343
289,305
$102,148,301
2010
415,547
$288
25,060
$288
440,606
$126,762,482
2011
542,587
$313
53,175
$316
595,762
$186,390,914
2012
333,952
$284
29,615
$284
363,567
$103,089,337
2013
136,766
$324
99,288
$326
236,054
$76,605,103
Total
2,528,032
$765,857,610
Mississippi
2003
0
$208
0
$208
0
$0
2004
0
$189
0
$188
0
$0
2005
0
$168
0
$172
0
$0
2006
1,887
$221
2,382
$225
4,269
$953,615
2007
9,203
$282
5,859
$274
15,061
$4,200,707
2008
2,829
$332
61
$299
2,891
$957,809
2009
15,564
$346
1,853
$345
17,418
$6,033,113
2010
53,036
$280
2,836
$280
55,872
$15,649,780
2011
76,263
$303
14,055
$301
90,318
$27,345,757
2012
35,734
$275
305
$275
36,038
$9,921,664
2013
16,013
$315
8,627
$317
24,640
$7,767,804
Total
246,506
$72,830,250
Combined total
2,774,538
$838,687,860
† Source: Nalley et al. (2016).
‡ Values in 2014 US$; deflated with consumer price index retrieved from IMF (2016).
§ In 2005, there was a yield loss of 30.23 Mg for medium-grain hybrid rice valued at the seasonal medium-grain average rice price of $253.62 Mg-1;
this is added into the long-grain loss value.
Agronomy Journal • Volume 109, Issue 1 • 2017
199
Table 5. Value of genetic and maintenance breeding gains from hybrid rice adoption in the Mid-South.
Maintenance breeding for Maintenance breeding for
Genetic +
blast resistance
blast resistance
maintenance
Genetic gain via
(no yield loss)‡
(yield loss)§
Year
hybrid adoption†
(no yield loss)
2003
$3,015,939
$167,869
$502,349
$3,183,808
2004
$11,608,892
$397,396
$1,095,794
$12,006,288
2005
$6,255,727
$637,548
$1,665,311
$6,893,275
2006
$22,935,344
$2,030,517
$5,792,896
$24,965,861
2007
$45,071,485
$3,658,619
$12,459,401
$48,730,104
2008
$88,086,218
$4,825,564
$19,457,459
$92,911,782
2009
$108,181,413
$5,293,003
$16,056,198
$113,474,416
2010
$142,412,262
$7,851,076
$24,583,408
$150,263,338
2011
$213,736,672
$7,760,223
$23,074,226
$221,496,895
2012
$113,011,000
$9,519,483
$29,109,979
$122,530,483
2013
$84,372,907
$7,367,732
$24,061,072
$91,740,639
Avg.
$76,244,351
$4,500,821
$14,350,736
$80,745,172
Total
$838,687,859
$49,509,030
$157,858,093
$888,196,889
Genetic +
maintenance
(yield loss)
$3,518,288
$12,704,686
$7,921,038
$28,728,240
$57,530,886
$107,543,677
$124,237,611
$166,995,670
$236,810,898
$142,120,979
$108,433,979
$90,595,087
$996,545,952
† As derived by Nalley et al. (2016) for Arkansas and Mississippi and illustrated in Table 4.
‡ As derived from Eq. [1], using the mean of the simulated values from Table 3 for Arkansas, Louisiana, and Mississippi.
§ As derived from Eq. [2], using the mean of the simulated values from Table 3 for Arkansas, Louisiana, and Mississippi.
Mid-South. Total gains assuming genetic gains plus mitigation
costs (Scenario 1 in Table 3) were estimated to be $80.75
million annually or $888.20 million in total from 2003 to 2103
(Table 4). Using the assumption that evolving from “resistant” to
“moderately resistant” to rice blast results in yield loss (Scenario 2
in Table 3), the value increases to $90.60 million annually or
$996.55 million over the entire study period (Table 5). These
values should be viewed as conservative estimates because
they only analyze two (genetic gains) and three (maintenance
breeding) of the six rice-growing states in the United States. That
is, the benefits estimated here only account for 60% (genetic
gains) and 65% (maintenance breeding) of the total rice acreage
in the United States in 2013 (USDA–ERS, 2015). Other states,
notably Texas and Missouri, also produce hybrid rice, but those
benefits were not captured here.
Impact on the US Rice Market
Table 6 illustrates the effects on the US rice market if hybrid
rice had not been adopted in 2013, as estimated by the RiceFlow
model. The benchmark scenario accounts for hybrid adoption
and the associated genetic and maintenance breeding gains as
calculated in Tables 5 and 6 for 2013. Scenario 1, as presented on
Table 6, removes these gains and estimates the changes in imports,
exports, prices, and production. As a result of the adoption of
hybrid rice in the United States, the total rice production in
2013 increased 440,000 Mg, or 5.1%, due to an 8.1% increase in
the production of long-grain rice (Table 6). Medium-grain rice
production decreased slightly as a result of hybrid rice adoption,
which was assumed to be an exclusively long-grain rice technology
in this study. The net impact of hybrid rice adoption on rice
acreage showed an increase of 13,000 ha or 1.2%. Total US rice
exports were 383 Mg or 9.0% higher as a result of hybrid rice
adoption. By type, exports of long-grain rice were 411 Mg or 13.9%
larger, and imports were 30 Mg or 17.7% lower, whereas exports of
medium-grain rice decreased by 28 Mg or 2.1%. Additionally, US
milling activity, represented by the domestic sales of brown rice
that enter the milling process, increased by 302 Mg or 4.3% due to
the adoption of hybrid rice.
200
Our findings show that the higher efficiency of hybrid
rice results in greater competitiveness and lower prices across
the US long-grain supply chain. Farm gate prices for longgrain rice decreased by $16 Mg-1 or 4.6%, whereas wholesale
prices decreased by $62 Mg-1 or 4% (Table 6). Farm gate
and wholesale prices for medium-grain rice increased slightly
due to higher competition with long-grain rice for factors
of production. As a result of the changes in the volume of
production and the farm gate prices, the total rice output value
increased by 63 million or 2.1%, attributed to the adoption of
hybrid rice. All of the economic gains estimated in the RiceFlow
model are in the long-grain markets. On the other hand, the
increased competitiveness brought on by the adoption of hybrid
rice generated lower consumer prices and therefore savings of
approximately $165 million. Consequently, all savings accrued to
long-grain rice because the consumer price and value of mediumgrain consumption in the United States increased slightly as a
result of the technology adoption.
Environmental Impact of Hybrid Rice
The life cycle analysis results are shown in Fig. 2. The range
of yield premiums of hybrid varieties significantly overlaps
conventional varieties, and this uncertainty was fully accounted
for in the assessment. It is primarily because of this overlap that
there is a 70 to 75% probability in most impact categories of an
improvement to the environmental impact associated with hybrid
varieties. If in practice this yield variability is not random and is
simulated in the Monte Carlo simulations but is correlated with
annual weather conditions, then this analysis underestimates the
likelihood that hybrid varieties outperform conventional strains.
Figure 2 indicates that over 75% of the time (in this case drawing
from the Monte Carlo simulation) hybrids had less environmental
impact than conventional varieties in terms of fossil fuel depletion,
ecotoxicity, respiratory effects, carcinogenics, eutrophication,
acidification, smog, global warming, and ozone depletion. The
only category for which hybrid rice had a negative environmental
impact was with noncarcinogenics. This is due almost exclusively
to zinc uptake by the hybrid plants because the removal of a metal
Agronomy Journal • Volume 109, Issue 1 • 2017
Agronomy Journal • Volume 109, Issue 1 • 2017
201
7507
Value domestic demand, $ million
-165
-38
63
-11
-13
-30
-27
27
383
247
-3
302
0
-2.2
-2.5
2.1
-3.0
-17.5%
1.2
-17.8
0.6
9.0
9.8
-15.4
4.3
0.0
0.0
9.8
4.1
%
5.1
4602
2073
1500
332
748
138
124
3129
3355
1770
4775
14
65
-101
1520
4826
4788
2008
1562
348
757
168
151
3101
2944
1503
4453
16
58
-101
1384
4494
-186
-62
65
-16
-9
-30
-27
28
411
267
-3
322
8
0
136
332
BENCH
SCEN
D
——1000 Mg, paddy basis——
6245
5777
468
Long grain
-3.9
-4.0
3.3
-4.6
-1.2
-17.7
-17.9
0.9
13.9
17.8
-15.4
7.2
13.0
0.0
9.8
7.4
%
8.1
2906
1049
1500
374
336
1
1
1573
1281
1004
2576
0
276
-46
0
2852
2885
1052
1490
371
340
1
1
1573
1308
1024
2596
0
284
-46
0
2880
-2
21
10
-0.2
0.7
0.7
-1.2
0.8
-2.1
0.0%
-28
0
-4
3
-2.0
0.0
0.0
-0.8
-2.6
–
-1.0
-1.0
0.0
–
%
-20
0
0
-21
-8
0
-28
0
0
BENCH
SCEN
D
——1000 Mg, paddy basis——
2806
2834
-28
Medium grain
† BENCH, benchmark; SCEN, scenario; D, BENCH - SCEN. The benchmark is with actual adoption of hybrid rice for 2013 in Arkansas and Mississippi.
‡ SCEN is the removal of the genetic yield premiums (Mg) associated with hybrids as derived from Table 4, and the increased yield loss associated with going from blast resistant to moderately blast resistant
from Table 3.
7673
3060
3123
356
Wholesale consumer price, $
Value production, $ million
345
1097
1538
Mg-1
1084
169
153
4675
4252
2527
7049
1500
Paddy price at farm gate, $
Acreage, 1000 ha
Mg-1
126
4702
4636
Import milled rice
Domestic demand milled rice
Exports
139
2774
Export milled rice
Imports
7351
14
Domestic sales brown rice
341
341
Export brown rice
Import brown rice
16
-147
1384
7374
0
136
304
BENCH†
SCEN
D
——1000 Mg, paddy basis——
9051
8611
440
-147
1520
7678
Change stock
Export paddy rice
Domestic sales paddy rice
Production paddy rice
Variables
Total
Table 6. Impact of the removal of hybrid rice on the US rice supply chain.
Fig. 2. Results (percentages) of the categories in the life cycle analysis comparison of hybrid vs. conventional rice based on 2500 Monte
Carlo simulations.
from the environment is beneficial in a life cycle analysis. Because
conventionals yield less but uptake the same amount of zinc, the
per unit yield of zinc uptake is higher for conventionals. This
would be mitigated if the LCA was scaled on a per-hectare (not
per-kilogram) basis. Figure 2 shows that using the well-established
categories defined by the TRACI 2.1 life cycle impact assessment
framework that hybrid rice appears to be less detrimental to the
environment. Many of these environmental benefits are driven by
the higher yields that hybrids possesses as well as embedded seed
technology (e.g., blast resistance), which requires less toxic inputs.
Conclusions
We observed several key findings in our analysis. First,
hybrids have increased revenue for producers in the Mid-South
through their higher yields per hectare. These higher yields have
contributed to increased domestic rice supplies, which have in
turn resulted in larger exports. Accordingly, these additional
exports are estimated to be large enough to feed an additional 5.89
million people. Given there are some 795 million people who are
malnourished globally, an increase in the food supply that actually
reaches their plates would be an important benefit to humanity.
Second, the increase of the rice supply from hybrid adoption drives
long-grain rice prices down in the United States, which makes it
more accessible via exports to impoverished countries.
There are two important findings from this study regarding the
value of maintenance breeding: (i) the yield loss avoided through
blast resistance via hybrid adoption is significant both in terms of
cost savings and field yield loss and (ii) by not including the value
of maintenance breeding in the economic valuation of a crop
breeding program, one can vastly underestimate its true return
to stakeholders. Our results indicate that the gains from the cost
savings due to not applying fungicide to combat blast, as well as
yield-loss avoidance, are roughly 20% as large as the gains that were
202
attributed to heterosis. This is a large benefit usually not accounted
for in cost/benefit analysis. This study provides conservative
estimates of the benefits of hybrid rice adoption in the United
States in that it only analyzes three of the six rice-producing states.
Our results indicate that hybrids, since their commercial release,
have increased producer revenue, lowered domestic/global prices,
and increased food supply. Overall, the two latter points are of the
utmost importance given the reality of global population growth
and an increased need for food in many countries.
We also find large environmental benefits from hybrid rice
adoption. Using a LCA, we find that hybrids lead to less fossil
fuel depletion, ecotoxicity, respiratory effects, carcinogenics,
eutrophication, acidification, smog, global warming, and ozone
depletion than their conventional counterparts in the Mid-South
of the United States. This is an important finding because in highincome countries increased consumer demand for food products
with lower environmental impact have prompted row crop
producers to reduce their environmental impact associated with
crop production. More importantly, agricultural producers face
increasing demand and in some cases requirements from private
industry to reduce the impact associated with crop production.
Currently, RiceTec Inc., a private company, is the sole purveyor
of southern hybrid lines. In 2011 a five-state consortium was
developed to create and release hybrid rice through University
(Louisiana State University, University of Arkansas, Texas
A&M, Mississippi State University, and Southeast Missouri
State University) breeding programs. To date there have been
no public releases of hybrid rice because hybrid breeding
is expensive and time consuming. The results of this study
should provide economic and environmental motivation for
these breeding programs to continue funding their hybrid
breeding programs. Furthermore, this study highlights that
when policymakers fund programs such as public rice breeding,
Agronomy Journal • Volume 109, Issue 1 • 2017
they need to look holistically, in terms of the economy and the
environment, when making funding decisions. Globally there
is no commercial production of genetically modified (GM)
rice, but, given the large environmental and economic impacts
between the embedded seed technology in conventional and
hybrid rice, one wonders if there is the same difference between
potential GM and hybrid rice. Given the findings of this study,
future policymakers may want to evaluate both the economic
and the environmental impacts of introducing GM rice and
not simply the traditional economic side of adoption to avoid
underestimating the holistic benefits of adoption.
This study demonstrates that benefits to hybrid rice exceed
simple yield increases. We show that, although yield increases
are the most tangible and often the easiest attribute to derive a
holistic comparison of agricultural substitutes, deeper research is
needed. As hybrid rice adoption grows and new embedded seed
technology emerges, new comparisons will inevitably need to be
made. This study lays the groundwork for a holistic production,
consumption, and environmental impact comparison of
competing agricultural seed types.
References
Araji, A., R. Sim, and R. Gardner. 1978. Returns to agricultural research
and extension: An ex ante approach. Am. J. Agric. Econ. 60:964–968.
doi:10.2307/1240129
Arkansas Agricultural Experiment Station. 2016. Arkansas rice performance
trials. Univ. of Arkansas Division Cooperative Extension Service, Little
Rock. http://arkansasvarietytesting.com/home/rice/ (accessed 8 Apr.
2016).
Bare, J., T. Gloria, and G. Norris. 2006. Development of the method and
U.S. normalization database for life cycle impact assessment and
sustainability metrics. Environ. Sci. Technol. 40(16):5108–5115
doi:10.1021/es052494b
Briones, R., A. Durand-Morat, E. Wailes, and E. Chavez. 2012. Climate
change and price volatility: Can we count on the ASEAN plus three
emergency rice reserve? ADB Sustainable Development Working Paper
Series No. 24. Asian Development Bank, Manila, Philippines.
Childs, N. 2009. Rice situation and outlook yearbook. RCS 2008. USDA,
Washington, DC.
Durand-Morat, A., and E. Wailes. 2010. RICEFLOW: A multi-region,
multi-product, spatial partial equilibrium model of the world rice
economy. Department of Agricultural Economics and Agribusiness,
Univ. of Arkansas, Fayetteville. http://ageconsearch.umn.edu/
bitstream/92010/2/RICEFLOW%20model%20documentation%20
SP%2003%202010.pdf (accessed 3 Mar. 2016).
Groth, D., C. Dischler, L. Monte, and M. Frey. 2015. Rice disease control
studies, 2014. Rice Research Station, Crowley, LA. http://www.
lsuagcenter.com/portals/our_offices/research_stations/rice/features/
annual%20report/2013-rice-research-station-annual-report (accessed
31 Dec. 2015).
International Monetary Fund (IMF). 2016. International financial
statistics. http://data.imf.org/?sk=5DABAFF2-C5AD-4D27-A1751253419C02D1 (accessed 17 Mar. 2016).
IPCC. 2007. Climate change 2007: The physical science basis. Contribution
of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge Univ. Press,
Cambridge, UK and New York.
International Rice Research Institute. 2013. Rice knowledge bank.
International Rice Research Institute, Los Baños, Philippines. http://
knowledgebank.irri.org/step-by-step-production/growth/pests-anddiseases/diseases (accessed 8 Feb. 2016).
Knutson, M., and L. Tweeton. 1979. Towards an optimal rate of growth in
agricultural production research and extension. Am. J. Agric. Econ.
61:70–76. doi:10.2307/1239502
Lakkakula, P., B. Dixon, M. Thomsen, E. Wailes, and D. Danforth. 2015.
Global rice trade competitiveness: A shift-share analysis. Agric. Econ.
46(5):667–676. doi:10.1111/agec.12162
Louisiana State University Ag Center. 2015. Rice research station
annual reports, commercial-advanced trials, multiple years.
ht t p://w w w. l s u a g c ent er. com /~/me d i a /s y s t e m /a /0/d /9/
a0d9ea59a7a70b1b388a435f6b2bfc55/2016%20rice%20research%20
report_finalpdf.pdf (accessed 10 June 2015).
Lyman, N., and L. Nalley. 2013. Economic analysis of hybrid rice performance
in Arkansas. Agron. J. 105:977–988. doi:10.2134/agronj2012.0461
Marasas, C., M. Smale, and R. Singh. 2003. The economic impact of productivity
maintenance research: Breeding for leaf rust resistance in modern wheat.
Agric. Econ. 29:253–263. doi:10.1111/j.1574-0862.2003.tb00162.x
Mississippi Agricultural and Forester Experiment Station. 2015. Mississippi
rice variety trials. Mississippi State Univ. Extension, Missippippi State.
http://msucares.com/pubs/infobulletins/ib0491.pdf (accessed 19 May
2015).
Nalley, L., B. Dixon, K. Brye, C. Rogers, H. Myrteza, and R. Norman. 2014.
Estimating cultivar effects on water usage and greenhouse gas emissions
in Arkansas rice production: A methodology using standard cultivar
trial data. Agron. J. 106(6):1981–1992. doi:10.2134/agronj14.0120
Nalley, L., J. Tack, A. Barkley, K. Jagadish, and K. Brye. 2016. Quantifying the
agronomic and economic performance of hybrid and conventional rice
varieties. Agron. J. 108:1514–1523. doi:10.2134/agronj2015.0526
Norman, R., and K. Moldenhauer. 2016. B.R. Wells Arkansas rice research
studies. Arkansas Agricultural Experiment Station, Little Rock. http://
arkansasagnews.uark.edu/1356.htm (accessed 4 July 2016).
Plucknett, D., and N. Smith. 1986. Sustaining agricultural yields. Bioscience
36:40–45. doi:10.2307/1309796
Rice Technical Working Group. 2001–2013. Proceedings of the RTWG. Rice
acreage summaries. CDROM.
Skamnioti, P., and S. Gurr. 2009. Against the grain, safeguarding rice from
rice blast disease. Trends Biotechnol. 27:141–150. doi:10.1016/j.
tibtech.2008.12.002
Thompson, J., E. Wailes, A. Durand-Morat, and A. Leister. 2015. Welfare
effects of U.S. liberty link rice contamination. J. Agric. Applied Econ.
47(2):243–259.
Tsiboe, F., A. Shew, and L. Nalley. 2016. The economic impact of blast
alleviation in the Mid-South of the United States. Presented at the
Southern Agricultural Economics Association Annual Meeting,
San Antonio, TX. 6–9 Feb. 2016. http://ageconsearch.umn.edu/
bitstream/229706/2/Final%20Blast.pdf (accessed 19 Feb. 2016).
University of Arkansas Cooperative Extension Service. 2016. Arkansas Rice
Performance Trials (ARPT). Variety testing programs. http://www.
arkansasvarietytesting.com/home/rice/ (accessed 2 Feb. 2016).
USDA. 2015. National agricultural statistics service, quick stats and world
agricultural supply and demand estimates report. USDA, Washington,
DC. http://www.usda.gov/oce/commodity/wasde/latest.pdf (accessed
18 Feb. 2016).
USDA–ERS. 2015. Rice yearbook 2015. USDA, Washington, DC. https://
www.ers.usda.gov/data-products/rice-yearbook/ (accessed 3 Mar. 2016).
Wailes, E., and E. Chavez. 2012. ASEAN and global rice situation and outlook.
Asian Development Bank Sustainable Development Working Paper
Series. http://www.adb.org/sites/default/files/publication/29969/adbwp-22-asean-global-rice-situation.pdf (accessed 24 Mar. 2016).
Wailes, E., A. Durand-Morat, and M. Diagne. 2015. Regional and national
rice development strategies for food security in West Africa. In:
A. Schmitz, P.L. Kennedy, and T.G. Schmitz, editors, Frontiers of
economics and globalization, volume 15: Food security in an uncertain
world. Emerald Group Publishing, Bradford, UK. doi:10.1108/
S1574-871520150000015025
Wang, X., and B. Valent, editors. 2009. Advances in genetics, genomics, and
the control of rice blast disease. Springer Science and Business Media,
New York. doi:10.1007/978-1-4020-9500-9
World Bank. 2013. Global food crisis response program. World Bank,
Washington, DC. http://www.worldbank.org/en/results/2013/04/11/
global-food-crisis-response-program-results-profile (accessed Feb. 2016).
Yuan, L., and S. Virmani. 1988. Status of hybrid rice research and development.
In: Hybrid rice: Proceedings of the International Symposium on Hybrid
Rice. Changsha, Hunan, China. 6–10 Oct. 1986. IRRI, Manila,
Philippines. p. 7–25.
Agronomy Journal • Volume 109, Issue 1 • 2017
203