Rewarding the social benefits 

of small-scale farming 

 

December 18, 2025 
 

A framework to inform policies and research to enhance smallholder contributions to public goods.   

 



Rewarding the social benefits of small-scale farming∗

Gashaw T. Abate, Kaleab Baye, Tanguy Bernard, Jordan Chamberlin,

Ruth Hill, Rachid Moussadek and Lamine Senghor

December 18, 2025

Abstract

Small-scale farmers are often encouraged to adopt innovations that generate significant
social benefits, including improved nutrition, food safety, and environmental sustainabil-
ity, yet market mechanisms frequently fail to reward these contributions. Adoption of
such socially desirable practices remains below the social optimum due to unobservable
attributes, high verification costs, limited production scale, and misaligned private incen-
tives. We propose a simple framework that highlights four key dimensions influencing
adoption: returns to farmers (R), adoption costs (C), inspection costs (I), and transac-
tion scale (Q). Interventions can target these dimensions individually or in combination,
such as reducing verification costs, aggregating production, subsidizing adoption, or
enhancing market premiums. We illustrate this framework through three case studies:
aflatoxin mitigation, biofortified crop adoption, and conservation agriculture, demon-
strating how technology, policy, and market-based instruments can be coordinated to
incentivize the adoption of socially beneficial innovations and practices. Our analysis
emphasizes the importance of multidisciplinary approaches, integrated public and private
actions, and dynamic interventions that account for both immediate and long-term
social returns. By explicitly connecting adoption incentives with verification and market
constraints, the framework guides research and policy toward more effective strategies to
reward smallholders for their contributions to public goods and sustainable development.

∗Authors’ affiliations: Abate – International Food Policy Research Institute (IFPRI); Baye – CFSN, Addis
Abbaba University & IPORA; Bernard – BSE, Univ. Bordeaux & IPORA; Chamberlin – CIMMYT; Hill –
International Food Policy Research Institute (IFPRI); Moussadek – International Center for Agricultural
Research in the Dry Areas (ICARDA); Senghor – International Institute for Tropical Agriculture (IITA)

1



1 Introduction

Small-scale farmers are frequently trained or encouraged to adopt new inputs and practices,
hereafter innovations, that generate private benefits for themselves—such as higher yields
or increased resistance to pests and drought— as well as benefits to others. These social
benefits may for instance relate to improved consumer nutrition, enhanced food safety,
better processing quality, or environmental sustainability. In low-income countries, the value
of these social benefits is rarely reflected in market prices, such that small-scale farmers’
adoption of socially beneficial innovations tends to occur socially sub-optimal levels – even
when other barriers to adoption are overcome [1, 2, 3, 4].

Why, then, are farmers in low-income countries not rewarded for the positive externali-
ties they generate when adopting these innovations? A key constraint lies in the combination
of unobservable socially-desirable attributes and limited scale: smallholder production and
marketing typically occur at levels that make it difficult to signal or verify the presence of
desirable qualities in the marketplace. This paper outlines these challenges using a simple
conceptual framework that connects various strands of research along four key dimensions:
the expected returns from adopting an innovation that generates socially desirable outcomes;
the costs associated with its adoption; inspection costs (and the conditions under which
they become relevant); and the scale of production (or the size of transaction) requiring
inspection. We discuss how integrating insights across these dimensions could help improve
our understanding of uptake constraints of innovations by smallholder farmers, and identify
underexplored opportunities where targeted research could contribute to alleviate adoption
barriers.

We illustrate this approach by examining innovations designed to encourage smallholders
to contribute to socially desirable outcomes in the areas of food safety, nutrition, and
environmental sustainability. For each, we examine innovations recognized as effective in
generating social benefits while being accessible to small-scale farmers in Africa, but whose
adoption remains limited compared to the social optimum. Using the proposed framework,
we draw on past and ongoing research (and further discuss possibly new research areas) to
discuss how small-scale farmers could be better encouraged in adopting these innovations in
a way that produces the targeted social benefits.

Our goal in this exercise is fourfold. First, we argue that many aspects traditionally
examined as informational problems within value chains—particularly those related to
quality signaling for attributes that are costly to verify [5, 6]—should also be understood
through the lens of public goods provision [7]. This re-framing highlights the essential role
of public or socially-oriented financing to address this challenge whether that be through the
development of technological solutions, investment in public goods or spending on subsidies
of some form. Second, our framework indicates that there are multiple ways to increase
the incentives farmers face and that cost-effective strategies will require multidisciplinary

2



approaches. Third, we emphasize that the private returns to many current innovations are
scale-dependent. As such, it is crucial to explicitly account for the incentives smallholder
producers face when assessing the viability and potential uptake of these socially desirable
innovations. Finally, we use this framework to explicitly identify where current market
failures could be fixed and where non-market interventions will generally be required to
incentivize smallholders’ transition towards targeted innovations, at least in the short term.

2 A simple framework

Farmers produce goods with multiple attributes, some of which are directly observable,
while others are not. Others in society interested in specific unobservable attributes may
wish to compensate producers for their supply of these attributes. This could be consumers
or a public agency interested in the unobserved quality of the good, or citizens interested
in the environmental impact of how a good is produced (either local citizens interested in
the impact on the local environment or global citizens interested in the impact on global
emissions and resource use). However, when these attributes are unobservable – they cannot
be verified through direct visual inspection of the product – incentivizing their provision
becomes significantly more complex. From the producer’s perspective, the decision to supply
such attributes depends on the corresponding marginal production costs, which must be
offset by an equivalent price premium or transfer. In some cases, the attribute is costless to
provide because it is bundled with other characteristics highly valued by farmers. In these
instances, farmers may supply the unobservable attribute regardless of compensation, though
it may still be possible for the farmer to receive a price premium or a transfer reflecting the
buyer’s utility gain.

In other cases, supplying the unobservable attribute entails positive marginal costs, and
without adequate compensation, farmers have no incentive to do so. The central challenge
lies in verifying the presence of the attribute. When it is strongly correlated with other
observable features of the product, producers can be rewarded through a premium aligned
with the buyer’s valuation. However, as inspection costs increase, for example through
testing an attribute of the good or measuring and reporting on the environmental impact
of production practices, the feasible level of compensation (buyer’s valuation - inspection
costs) declines, potentially to zero. As a general rule, any positive verification cost leads to
an under-provision of the desired unobservable attribute [8].

This issue is particularly problematic in low-income countries agriculture, due to the
typically small scale of production and transactions, combined with the largely fixed costs
of testing for or verifying unobservable attributes in the goods supplied by farmers. Even
when these fixed costs are modest, they can become disproportionately high with respect to
the value associated with their supply.

3



Let I denote the fixed cost of obtaining reliable information on whether a farmer
has adopted an innovation that generates a desired attribute, Q the quantity of product
per transaction, such that the per-unit cost of revealing the presence of the attribute is
(I/Q) ≥ 0. Let R denote the per unit of product premium or transfer that the farmer
receives for the supply of this attribute, and C the marginal cost of producing it by the
farmer. The farmer adopts the innovation if the corresponding per-unit net return weakly
exceeds a given reservation level of returns π̄.1 The per unit of product condition under
which a farmer has an incentive to adopt such innovation is then:

(R − C) − (I/Q) ≥ π̄ (1)

This simple relationship highlights four broad categories of potential interventions.
First, one may seek to lower the inspection cost per unit (I/Q) of product supplied by
farmers.

- Reducing I: This can be pursued through technological and/or institutional solutions.
Technological solutions include the development of testing methods enabling less
costly measuring, reporting and verifying (MRV) of product attributes of production
practices. Institutional strategies may involve public support to the testing or MRV
infrastructure, or the coordination of its support by value chain actors in cases where
they ultimately benefit from the increase supply of the desired attribute by small-scale
farmers.

- Increasing Q: One approach is to enlarge the scale over which MRV is done by
encouraging farmers to aggregate their products. It is however important to note
that when Q is increased by aggregating across farmers it brings its own costs of
information as farmers have to verify the unobserved characteristics of the goods and
practices of others in the group. Thus, I can be lower for farmers based in the same
locality, but these costs are rarely 0. For instance, public support to cooperatives in
charge of commercializing smallholders’ output are often promoted as a means to help
farmers commercialize higher quality products. However, such arrangements often
underperform or become limited to lower-quality products due to coordination failures
and free-riding among cooperative members[9, 10, 11]. Alternatively, farmers can be
brought together earlier in the production cycle and encouraged to produce and market
together (e.g., cluster/block farming in Ethiopia, Ghana, Malawi, Tanzania, and
Zambia). However, sustaining these schemes is often difficult, as individual incentives
for effort may not be aligned, leading to short-lived successes.

1 The reservation level may vary across farmers, reflecting heterogeneity in attitudinal factors (e.g. risk
aversion), knowledge constraints (e.g. education), and other individual-level characteristics that have been
shown to influence the adoption of agricultural innovations among smallholder farmers in low-income
countries.[1]

4



Even if the per-unit inspection cost is lowered, one must ensure that by investing in the
provision of socially desirable attributes of their products, farmers obtain private rewards
that exceed the associated cost: (R − C) .

- Increasing R: Rewards to farmers can be enhanced in several ways. One is by increasing
the market premium for goods that can credibly be linked to socially desirable outcomes.
Another is through direct payments to farmers who adopt socially beneficial innovations.
An example of this would be payments from voluntary carbon markets or public
payments for ecosystem services where farmers are compensated on the basis of land
or water management practices.

- Lowering C : Subsidies can be targeted at lowering the adoption costs of the innovation,
to the point where the corresponding private benefits approximate social gains. For
example, non-linear subsidies for the use of an aflatoxin bio-control product (Aflasafe)
have been designed to more precisely target portions of a farmer’s land dedicated to
local market sales [12].

Figure 1 presents the proposed framework as a two-dimensional graph. The vertical axis
represents the per-unit cost of inspection, while the horizontal axis captures the gap between
the private return of a farmer adopting a socially-desirable innovation and its associated
costs.2

Each innovation can be conceptualized as a specific point within this space, defined
by the prevailing values of I, Q, R and C at a given time and in a particular location.
The (I/Q) = (R − C) line delineates combinations of these values where information costs
offset net social returns. Only technologies or practices that lie below this line (indicating
information costs are not prohibitively high) and to the right of the vertical line at R−C = 0
(where rewards to social returns exceed private costs) fall within the feasible adoption region
(area IV ).

The infeasible region can be further disaggregated. Technologies or practices falling
within area III are characterized by negative net returns to the farmer, irrespective of
information costs. In such cases, the priority is to either enhance the farmers’ rewards for
the social returns (R) or reduce the marginal cost of adoption (C). For technologies located
in area I, adoption would require increased rewards, but may also benefit from reduced
information costs. Those situated within area II may be made adoptable by lowering
either information costs or increasing rewards, or both. This framework thus clarifies
the intervention levers needed to expand the set of viable technologies for smallholder
engagement.

2 As is standard in economics, we refer to the costs of providing the desired attribute as the increase in
production costs compared to the farmer’s best alternative technology or practice.

5



Figure 1: Conceptual framework

(R − C)

(I/Q)

I

IVIII

II

Feasible adoption

(I/Q) = (R − C)

π̄

We note that for initial locations in areas I and II, the shortest path to feasible
adoption (area IV ) will involve a combined reduction in (I/Q) and increase in (R − C),
hinting at the importance of multidisciplinary approaches to addressing adoption incentive
constraints. Accordingly, addressing only one element of the equation - whether I, Q, R, or
C - will yield limited results unless large reduction in I or C, or markedly increase in Q or
R can be obtained. Alternatively, even modest reductions in I or C may be effective when
accompanied by concurrent increases in R or Q.

Table 1 outlines illustrative interventions that can reduce cost of inspection I, increase
transaction quantities Q, enhance returns R, or lower adoption costs C. Efforts to reduce
the cost of inspection (I) include the development of technologies that reduce testing
costs, the provision of subsidies to offset inspection expenses, and the strengthening of
MRV systems. Such interventions, however, require substantial research capacity as well
as institutional mechanisms for the establishment and enforcement of credible standards.
Enhancing transaction size (Q) through collective action and specialization – such as
the organization of farmer groups or the promotion of cluster and block farming – can
reduce inspection costs per unit. The success of these measures depends on effective
coordination among producers, alignment of incentives, and robust risk management systems.
Interventions designed to enhance producer returns (R) include public awareness campaigns
that stimulate consumer demand for products embodying socially desirable attributes or
practices, conditional payment schemes that incentivize adoption, consumer subsidies, and

6



the implementation of regulatory instruments. These approaches rely on mechanisms for
monitoring quality and adoption, adequate public financing, and strong institutional capacity
for oversight and enforcement. Similarly, interventions intended to lower adoption costs (C),
such as agricultural R&D, extension services, and producer subsidies, necessitate advanced
research infrastructure, efficient extension systems, and publicly funded subsidy mechanisms.

Table 1: Examples of interventions to alter IQRC

Intervention Aim Requirements
Reducing I
Testing R&D Develop means to reduce the

cost of testing
Global or national capacity in
research.

Testing subsidies Subsidizing the cost of the test Test is specific to identifying the
characteristic.

Support certification Support the testing and
certification infrastructure (or
coordination by value chain
actors)

Strong public institutions for
defining and enforcing grades
and standards.

Increasing Q
Promote farmers groups Increase the size of transaction

through collective
commercialization

Group capacity to monitor and
enforce farmers’ adoption of
innovation.

Promote block farming Increase the size of transaction
through collective production

Strong coordination and
alignment of individual
incentives between farmers.

Promote specialization Increase the size of transaction
by reducing farmers’ crop
diversification

Well-functioning means to lower
production and market risks at
farmer-level.

Increasing R
Public awareness cam-
paign to consumers

Increase consumer demand for
targeted attribute of good

A way for the unobserved
quality to be observed.

Conditional payments to
producers

Reward producers’ adoption of
innovation

A way for the adoption to be
observed.

Consumer subsidies Subsidize the purchase of goods
with the required characteristic

A way for the unobserved
quality to be observed.

Regulation on consumer
markets

Decrease market return of
non-adopting the innovation

Strong public institutions to
design and enforce regulations.

Reducing C
Agricultural R&D Develop innovations with lower

costs of adopting by small-scale
farmers

Global or national capacity in
research.

Extension Provide farmers with
information on innovation and
cost-effective adoption

Broad reaching and reliable
extension services (physical
and/or digital).

Producer subsidies Lower financial cost of adoption Subsidy schemes targeting
social-returns of innovations.

While extensive research exists on each individual parameter, there is a pressing need

7



to move toward a research agenda that explicitly considers their interactions. Below, we
illustrate this approach by outlining real-world examples where integrating existing strands
of research may generate synergies and yield effects greater than the sum of their parts.

3 Three illustrative case studies

3.1 Food safety – Promoting Aflatoxin-free production

The issue: Foodborne disease-related deaths are estimated to be the highest per capita
in Africa, causing 137,000 deaths and 91 million illnesses annually, underscoring the close
link between agricultural systems and public health [13]. Visually undetectable food safety
problems that contribute to this situation are widespread in the region. Reported examples
include Salmonella and other pathogens in foods at prevalences up to 44% in Burkina Faso
[14], high coliform counts in Kenya’s informal milk chains [15], pesticide residues exceeding
safety limits in tomatoes and vegetables in Burkina Faso and Uganda [16, 17], antibiotic
residues in animal products [18], and heavy metals and pathogens in wastewater-irrigated
vegetables in Accra [19].

Aflatoxin contamination is among the most serious food safety threats globally, resulting
in significant health developmental, trade and economic problems in the affected regions[20]
and notably in sub-Saharan Africa[21]. Given climate conditions and dependence on
susceptible staple crops like maize and groundnuts, studies across sub-Saharan Africa report
persistently high levels of human and animal exposure [22, 23, 24, 25]. Acute exposure can
be fatal [26, 27], while chronic exposure is linked to child stunting, cancer, and weakened
immunity [28, 29, 30, 31, 32].

Aflatoxin-producing fungi infect crops before, during, and after harvest and can produce
aflatoxins during all these stages if conducive conditions for toxin formation occur[33, 34].
Several practices can reduce crop aflatoxin content at different stages of the crop value
chain[35, 34, 36, 37, 38] but, when used in isolation, aflatoxin contamination often occurs
above safe and permissible levels, particularly in tropical and subtropical regions[39]. Another
approach is biological control, notably Aflasafe, which targets Aspergillus flavus, the fungus
responsible for aflatoxin production. Despite its growing availability across the region since
its introduction in 2014 and its efficacy in reducing aflatoxin levels in maize and groundnut
across diverse farming systems, notably in Nigeria [40], uptake of Aflasafe among smallholders
remains limited in Africa.

Reducing I/Q: Aflatoxins are imperceptible outside of dedicated tests: contaminated
crops are visually, olfactorily, and gustatorily indistinguishable from uncontaminated ones.
Currently available testing for the presence of Aflatoxins is sample-based and largely invariant

8



with the size of the consignment that is tested: the cost is the same whether one tests
for a 50kg or a 5MT consignment. Public and private R&D efforts have recently focused
on developing detection methods better suited to smallholder value chains in low-income
settings, particularly for crops like maize and groundnuts. For effective integration into
these systems, tests must balance accuracy, usability, and affordability. One such example
is ICRISAT’s portable competitive ELISA (cELISA) kit available since 2016, designed to
function without specialized laboratory personnel, offering reliable (+/- 10ppb) results
in under 15 minutes at a cost of less than 2 USD per test. More recently (since 2022),
ICRISAT, in collaboration with PureScan, introduced an AI-based diagnostic tool priced
around 50 USD, capable of delivering rapid and precise (+/- 1ppb) results within seconds at
low marginal costs as it does not require any consumable. If made available (possibly at
subsidized rates) to farmers, cooperatives or rural traders, these equipments can be used
to assess aflatoxin contamination rate at farmers’ level (or their use of Aflasafe) and signal
it so that they are rewarded for it. Public support to the existence of reliable testing and
certification could help ensure that small-scale farmers are indeed rewarded for their efforts
and investments towards the supply of aflatoxin-free food products.[41].

Lowering per-unit inspection costs through product aggregation is another common
strategy in quality assurance programs. In Nigeria, the AgResults program promotes Aflasafe
adoption by aggregating farmers’ outputs for collective testing by corporate buyers [42], a
model also seen in Senegal’s groundnut export cooperatives [43]. But these arrangements,
which impose strict quality oversight, typically serve export markets, leaving local consumer
markets exposed to high aflatoxin risks. Outside such contracts, aggregation has shown
limited effectiveness in promoting non-visible quality traits for domestic value chains.3 A
yet-untested alternative is to introduce more flexibility in the level of Q, by allowing farmers
to choose the scale at which testing occurs instead of systematically promoting it at the
full group level where observability and trust of all others’ engagement with the innovation
is oftentimes limited. Even simple coordination between two farmers could halve testing
costs, offering a more flexible and potentially scalable approach than relying solely on fixed
cooperative or cluster/block farming structures.

Increasing R–C: Enhancing farmers’ returns from Aflasafe use often relies on market-
based mechanisms that leverage consumer willingness to pay for safer food. Evidence on
consumers’ willingness to pay suggests limited but positive demand [45, 46], which tends to
increase following targeted information campaigns [47, 48].4 Even when consumer demand
is established, one must ensure that the food safety premium paid by the final consumer is

3 In Ethiopia, farmers whose products were tested and rated as of higher quality regarding flour extraction
rate sold less through cooperatives than untested peers with similar levels of quality [44].

4 Oftentimes however, the demand for certified and safer products skews toward more affluent consumers,
raising equity concerns, especially if poorer populations are left with the more contaminated products
[46, 49].

9



passed through suppliers at each stage in the market chain all the way to producers. This
may require coordination of a sometimes large number of actors [50]. As few policies directly
subsidize food for its health attributes, an alternative is the enforcement of food safety
standards, as practiced in the EU and US, which shifts the burden of safety from individual
choice to regulation, though it requires accounting for enforcement costs. Importantly, the
enforcement of food-safety standards would likely increase overall food prices by the added
cost of Aflasafe treatment per unit of product, effectively internalizing the health benefit in
the price faced by all consumers.

Regarding the cost of adoption (C), proper application of Aflasafe typically requires 10
kg per hectare. While labor requirements are minimal, the product cost—approximately
USD 20 per hectare—is significant for many small-scale farmers. Even with partial subsidies,
adoption may remain suboptimal. Absent market reward, evidence suggests that farmers who
are aware of aflatoxin risks tend to selectively protect the portion of their harvest intended
for household consumption [49]. There is a role for public research and development on
reducing the cost of Aflasafe production. If this is not possible, public funds could subsidize
Aflasafe to smallholder farmers, thereby safeguarding public health. Where full subsidies
are deemed too costly, geographically targeted approaches could be considered. On-going
research is seeking to leverage time and geolocalized data on aflatoxin contamination to
calibrate AI models able to predict areas most susceptible to aflatoxin contamination given
weather patterns ahead of the agricultural season [51]. This information could be used to
target the subsidies to producers in those locations. Another approach is to design subsidy
schemes that specifically targets the portion of farms which produce food for the market
[12].

Key takeaways: The pre-intervention location of the aflatoxin problem can be mapped
to area II of Figure 1 where the most direct route to feasible adoption will likely involve
both reductions in (I/Q) and increases in (R − C), signaling intervention synergies. In the
absence of affordable and widely accessible diagnostic tools, the most effective approach
may involve combining health information campaigns with fully subsidized (and possibly
targeted) distribution. However, recent advances in aflatoxin detection technologies may
open pathways for market-based solutions, provided there is sufficient demand. If demand
proves robust (potentially in response to other investments in public information campaigns),
Aflasafe could be promoted at cost. Otherwise, adoption will likely continue to rely on some
level of public subsidy to internalize its health-related social benefits.

3.2 Nutrition – Encouraging the adoption of biofortified crop varieties

The issue – While global trends in micronutrient deficiencies have generally declined
between 1990 to 2019, sub-Saharan Africa continues to bear a significant burden, particularly

10



for iodine, vitamin A, folate, zinc and iron deficiencies. These deficiencies are driven by
factors such as inadequate dietary diversity, poor sanitation, frequent infections, and limited
coverage of supplementation programs. The health consequences are severe, including
adverse pregnancy outcomes, congenital malformations, impaired cognitive development,
increased perinatal, infant and child mortality, and reduced resistance to infections, diarrheal
diseases, measles and other illnesses [52].

Supplementation, food fortification, and biofortified crop varieties are three strategies
that can address deficiencies when nutrition requirements cannot be met through dietary
intake alone. Biofortification (increasing the concentration of particular micronutrients in
staple crops through agronomic, transgenic techniques, and plant-breeding approaches) has
proven to be a feasible and cost effective means of delivering micronutrients [53, 54, 55]. Yet,
while more than 284 biofortified (rich in iron, vitamin A, or zinc) varieties of 12 crops have
been released in 22 African countries over the past two decades, smallholders’ adoption of
these varieties remains suboptimal [56, 3, 57, 58].

Reducing I/Q: Some biofortification interventions result in visible changes—such as
the orange or yellow colors of Vitamin A-enriched sweet potatoes , cassava and maize.
In contrast, other biofortified traits are not visually detectable and require specialized
testing to assess nutrient levels. As for Aflatoxins, such assessments are typically sample-
based and independent of the size of the product batch to be evaluated. Several analytical
technologies are currently available to accurately measure the concentration of micronutrients
in biofortified crops, including minerals (e.g., via X-ray fluorescence [XRF] spectroscopy),
vitamins (e.g., using high-performance liquid chromatography [HPLC]), and other beneficial
compounds. However, these methods are often costly and largely inaccessible in rural areas
where many smallholder farmers operate. Support to emerging technologies offer promising,
lower-cost alternatives capable of providing rapid assessments at the farm level or at local
aggregation and collection centers. For example, portable near-infrared spectroscopy that
uses light to analyze the chemical composition of materials could yield rapid, non-destructive,
and cost-effective analysis of nutritional content of biofortified grains [59, 60, 61]. The
advancement in sensor technology, such as integrating smartphone-based detection methods
and paper-based sensors can also facilitate real-time and on-site analysis of both nutritional
components in grains [62]. Lastly, remote sensing approaches leveraging hyperspectral
sensors and advanced modeling, also show a strong promise for non-destructive and scalable
assessment of crop nutrient quality across space and time [63].

Aggregation at production (e.g., through cluster/block farming) and marketing stages
(e.g., producer groups) prior to verification would be one way to lower the per-unit costs
for producers (raising Q). However, as with Aflatoxin, challenges include intentional or
unintentional mixing of quality (free-riding or coordination failures in production) prior to
verification, which cannot be traced and are hard to correct for.

11



Increasing R–C: Improving farmers’ returns from producing biofortified crop varieties
generally relies on certification schemes that leverage consumers’ willingness to pay for
nutrient dense food. While the evidence suggests differences across contextual factors and
visibility of nutrient traits, consumers in low-income countries generally show a positive
willingness to pay a premium for biofortified crops that increases with the provision of
nutritional information and the visibility of the trait [64, 65, 66, 67, 68, 69].

Fixing markets for biofortified crops may however require public policy support, as
some studies show a negative willingness to pay for biofortified crops with visible traits,
apparently linked with neophobia around unfamiliar food appearances [67, 70, 71]. Public
policy support may require a financial commitment to increase demand. This could be quite
low cost such as funding marketing campaigns to reduce consumer concerns and increase
consumer demand, or could be more substantial, such as providing time-bound subsidies to
biofortified crops to increase use and familiarity with new food products. This spending can
be justified by the large and lasting economic, social, and medical impacts of (micronutrient)
malnutrition. Even where consumer demand for biofortified crops is positive, enhancing
farmers’ reward to their supply of it may require coordination of key value chain actors to
ensure/facilitate the pass-through of premiums from consumers to farmers, via intermediary
actors. Alternatively, non-market approaches may leverage recent advances in remote sensing
technologies to provide geographically targeted subsidies to farmers who produced nutrient
dense crops [63].

Expected yield is often one of the most critical factors influencing farmers’ production
decisions. Although biofortified crop varieties typically entail production costs comparable
to those of conventional varieties, their expected yields must be at least equivalent to
conventional crops to support widespread adoption [72, 47, 73]. To enhance adoption,
micronutrient traits should be integrated into advanced lines of widely adopted, agronomically
superior (mega) varieties and, when they exist, farmers should be informed about the absence
of tradeoffs between nutrition content and other desired varietal attributes [74]. In addition
to agronomic and economic considerations, sensory and sociocultural attributes of biofortified
crops can impose indirect costs on farmers, particularly since they often consume a portion of
their own production. Although biofortified varieties are generally well accepted from these
perspectives, future breeding efforts can further support large-scale adoption by aligning with
localized preferences related to crop color, preferred food products, cooking ease, storage
characteristics, and other culturally relevant traits (e.g., [58]).

Key takeaways: Non-visible versions of the biofortification problem can be mapped to
area II of Figure 1, where the most direct route to feasible adoption will likely involve both
reductions in (I/Q) and increases in (R − C). The situation may however be different for
visible versions of the problem (e.g., orange-fleshed sweet potato, orange maize) for which
the low verification costs sometimes comes with negative consumer willingness to pay (e.g.,

12



orange maize), which places the pretreatment situation in area III of Figure 1. There,
adoption can only be promoted through increases in R, through direct public subsidies
or targeted information on nutrition and health benefits through individual or collective
behavioral change campaigns aimed at increasing consumer demand.

3.3 Environment – Promoting adoption of conservation agriculture

The issue: Agriculture in low-income countries, particularly in sub-Saharan Africa, is
a significant source of greenhouse gas (GHG) emissions, stemming from both direct farm
activities and land-use changes. Key contributors include methane from livestock enteric
fermentation and irrigated rice production, nitrous oxide from inefficient fertilizer use and
poor manure management, and emissions from widespread residue burning [75]. Indirect
emissions result from land conversion—especially deforestation and grassland clearance for
cropping—which depletes soil organic carbon (SOC) and above-ground biomass. These
impacts are worsened by soil-degrading practices such as intensive tillage, residue removal,
and inadequate nutrient cycling, accelerating carbon loss and limiting long-term carbon
sequestration potential [76].

In response, the 2015 COP21 summit in Paris introduced the voluntary ‘4 per 1000
Initiative: Soils for Food Security and Climate’, promoting regenerative farming practices
to enhance soil health and climate resilience. One recommended practice is conservation
agriculture (CA), which combines crop diversification (e.g., cereal-legume rotations), perma-
nent soil cover (e.g., crop residues or cover crops), and minimal soil disturbance (e.g., no
or minimum tillage). While CA’s effectiveness has been debated, recent studies (e.g., [77])
indicate it can increase SOC, particularly in dry regions like Africa, where yield effects are
neutral or positive. Despite this, adoption by small-scale farmers remains low, covering only
about 1.25% of cultivated land [78].5

Increasing net soil carbon storage offers a promising strategy for climate mitigation and
sustainable food production by limiting CO2 emissions, capturing atmospheric CO2 and
enhancing soil quality. However, incentivizing smallholders to adopt conservation agriculture
through compensation for its positive environmental externalities is challenging, given the
difficulties in monitoring adoption, tracking changes over time, and verifying the integrated
application of its three key practices, which are thought to deliver the most significant
environmental gains.

Increasing R–C: Conservation Agriculture remains sparsely adopted in sub-Saharan
Africa, largely due to the absence of strong economic incentives for smallholder farmers [81].

5 An important literature questions claims about about adoption rates of CA, noting that very different
numbers are obtained whether one considers as "adopter" those who implements at least one of the key
recommended practices (e.g. no-till) or the full set of recommended practices [e.g. 79, 80].

13



A central challenge—especially for resource-constrained farmers—is that many of the private
benefits (e.g., yield improvements) tend to materialize only after several years, whereas
some of the associated costs are incurred immediately. These costs (C) include potential
short-term yield reductions, increased weed pressure requiring additional labor or chemical
control, loss of crop residues for livestock feed, and limited access to appropriate equipment
for direct seeding [82, 78].6In the absence of immediate profitability, the non-adoption of CA
by African smallholders may constitute a rational economic choice and further lead to high
rates of dis-adoption [83]. Several complementary interventions could also help lower the
cost of adoption by easing access to labor-saving technologies such as mechanized planters
or chemical weed management. An additional cost is the cost of learning how to implement
CA practices correctly. A related study in Niger highlights the importance of training and
public awareness campaigns in encouraging take-up of soil conservation practices [84].

In areas experiencing increasing seasonal variability in production conditions — par-
ticularly irregular rainfall — the incentives to adopt CA may increase naturally, as its
practices improve water retention and reduce runoff, potentially delivering more benefits
(R) in the short term as compared to conventional approaches. When these short-term
private benefits remain insufficient to sustain farmer adoption of CA, additional incentives
may be necessary to compensate for the positive environmental benefits generated. Such
instruments may mirror performance-based environmental programs, including payments
for forest conservation in Uganda [85] and efforts to curb crop residue burning in India [86].
Publicly funded payments can be justified by the environmental benefits of CA, both those
that are short-term such as improved surface soil protection through residue retention and
long-run benefits such as increased SOC, soil moisture retention and water-use efficiency, and
greenhouse gas (GHG) mitigation. Some of these benefits accrue to the individual farmer,
but they also have public good characteristics — particularly under increasing climate stress
[87]. Market-financed payments utilizing carbon offsetting finance can be explored to support
farmers, in a way that helps reconcile the short term costs born by farmers engaging in CA,
and the longer term social benefits associated with it.

Reducing I/Q: Whether farmers adopt Conservation Agriculture (CA) must be inferred
from observing practices at the field level. Monitoring individual farmers, however, involves
fixed costs that can become prohibitively high in smallholder settings. This challenge has
long been noted in the micro-insurance literature, where recent advances in weather-index
insurance have partially addressed similar constraints. In the context of CA, emerging remote
sensing technologies offer promising solutions for monitoring adoption (I) and providing
the basis for MRV. For example, [88] demonstrate in Belgium that combining time series
from optical satellite imagery (Sentinel-2, Landsat-7 and 8), and radar data (Sentinel-1) can
distinguish conservation from conventional farming with 92% accuracy. Such methods could

6 These may be compensated by reduced costs related to land preparation.

14



be adapted for smallholder systems in Africa, although several operational challenges remain
(e.g., limited land registration systems, dynamic functional field boundaries in smallholder
systems, lower scope for remote detection in inter-cropped systems). Such approaches must
however be complemented by precise measures of the social benefits generated from farmers’
adoption of CA. Evidence shows that the magnitude of SOC gains from CA, as well as their
dynamics through time – whether steadily rising, plateauing, or peaking then declining -—
depend strongly on climate, soil type, and residue management. While a 17-year trial in
Spain find that SOC gains under no-till practices peaked after eight years before gradually
declining,[89] evidence shows sustained SOC accrual in semi-arid Morocco,[90, 91] humid
regions of Brazil,[92] and cool climates of the Canadian Prairies,[93]. These contextualized
patterns highlight the importance of spatially disaggregated evidence in order to guide policy
and design incentive mechanisms favoring farmers’ engagements with CA.[94]

In addition to these technology-driven reductions in the cost of I, market-based systems
could also rely on paying groups of farmers rather than individual farmers, effectively
increasing the size of the operational units to be verified (Q). This could be through groups
of farmers undertaking the same practices through cluster/block farming and similar schemes.
Recent experiences with cluster/block farming in Ethiopia, Ghana, Malawi, Tanzania, and
Zambia suggest that this may be viable, albeit with strong state support [e.g., 95].

Key takeaways: Conservation Agriculture can be mapped to area II of Figure 1, where
the most direct route to feasible adoption will likely involve both reductions in (I/Q) and
increases in (R − C). A further challenge with CA stems from the predominantly long-term
nature of its private and social benefits. The broader issue of reconciling short-term costs
with long-term benefits is common to many economic decisions but is particularly acute for
smallholders, who often lack access to suitable financial instruments. Technological advances
that reduce the cost of verifying compliance with CA practices, combined with payments
for ecosystem services (whether publicly financed or market-based), could help facilitate a
sustained transition. The visibility of adoption plays a critical role in enabling incentive
structures that reward observable compliance (i.e., affecting returns R). As adoption becomes
more widespread and the system approaches a new equilibrium where R−C ≥ 0, the marginal
cost of monitoring (I) may approach zero.

4 Towards an integrated research-policy agenda

This paper presents a framework for promoting small-farmers’ adoption of innovations
associated with important social benefits, illustrated through three case studies. While many
conventional agricultural innovations have negligible inspection costs, generate primarily
private returns, and thus fall outside the framework’s main focus, most contemporary

15



public R&D targets sustainable intensification. These innovations deliver public bene-
fits—environmental services, food safety, improved nutrition—that are often excluded from
R, meaning that focusing solely on reducing C risks leaving significant social gains unrealized.

From this framework, we identify three priorities for improving public R&D systems.

First, integrated, cross-disciplinary research is needed to measure and enhance the
public-good value of smallholder innovations. Examples include quantifying health gains
from reduced aflatoxin contamination, nutritional benefits from biofortified crops, and carbon
emission and sequestration from Conservation Agriculture (CA). In many cases, estimation
remains contested, underscoring the need for robust metrics. Equally important is advancing
tools to cost-effectively verify quality attributes or production practices not visible in final
products, such as remote sensing for CA, portable near-infrared devices for grain quality,
and improved diagnostics for aflatoxin detection.

Second, the framework clarifies when market failures for unobservable quality attributes
can be addressed through institutional or technical innovations and when non-market
instruments are preferable. Market-based approaches work when inspection costs fall to
feasible levels, but public action is often required to coordinate value chain actors and manage
distributional effects. Where fixing markets for unobservable quality remains prohibitively
costly, policies should instead focus on increasing (R−C), through adapted subsidy schedules,
conditional payments or regulations.

Third, the opportunity space—I, Q, C, and R—is dynamic and can shift with early
interventions. Many sustainable intensification practices, including CA, yield modest initial
returns but increase over time, making early-stage investments akin to public goods. Similar
dynamics occur on the consumer side: when private utility diverges from public utility (e.g.,
nutritional or food safety benefits), public investment in R, coupled with education, can shift
demand until markets sustain the desired equilibrium. In this way, short-term non-market
incentives act as transitional “nudges” toward self-sustaining outcomes.

16



References

[1] Tavneet Suri and Christopher Udry. Agricultural technology in africa. Journal of
Economic Perspectives, 36:33–56, 2022.

[2] Antti Autio, Tino Johansson, Lilian Motaroki, Paola Minoia, and Petri Pellikka. Con-
straints for adopting climate-smart agricultural practices among smallholder farmers in
southeast kenya. Agricultural Systems, 194:103284, 2021.

[3] Lidya Samuel, Marcia Dutra de Barcellos, Mulugeta D Watabaji, and Hans De Steur.
Factors affecting farmers’ acceptance and adoption of biofortified crops: A systematic
review. Outlook on Agriculture, 53(1):15–29, 2024.

[4] Matieyedou Konlambigue, Alejandro Ortega-Beltran, Ranajit Bandyopadhyay, Tracy
Shanks, Edward Landreth, and Oscar Jacob. Lessons learned on scaling Aflasafe®
through commercialization in Sub-Saharan Africa. Intl Food Policy Res Inst, 2020.

[5] George A Akerlof. The market for “lemons”: Quality uncertainty and the market
mechanism. The quarterly journal of economics, 84(3):488–500, 1970.

[6] W Kip Viscusi. A note on" lemons" markets with quality certification. The Bell Journal
of Economics, pages 277–279, 1978.

[7] Paul A Samuelson. The pure theory of public expenditure. The review of economics
and statistics, pages 387–389, 1954.

[8] Marcel Fafchamps, Ruth Vargas Hill, and Bart Minten. Quality control in nonstaple
food markets: evidence from india. Agricultural Economics, 38(3):251–266, 2008.

[9] Tanguy Bernard, Alemayehu Seyoum Taffesse, and Eleni Gabre-Madhin. Impact of
cooperatives on smallholders’ commercialization behavior: evidence from ethiopia.
Agricultural Economics, 39(2):147–161, 2008.

[10] Tanguy Bernard, Marie-Hélène Collion, Alain De Janvry, Pierre Rondot, and Elisabeth
Sadoulet. Do village organizations make a difference in african rural development? a
study for senegal and burkina faso. World development, 36(11):2188–2204, 2008.

[11] Kodjo Aflagah, Tanguy Bernard, and Angelino Viceisza. Cheap talk and coordination in
the lab and in the field: Collective commercialization in senegal. Journal of Development
Economics, 154:102751, 2022.

[12] Tanguy Bernard, Jérémy Do Nascimento Miguel, Yanyan Liu, and Tesfamichael Wossen.
Promoting unobservable food safety in smallholder agriculture, (preliminary) evidence
from nigeria. mimeo, 2025.

17



[13] Steven Jaffe and Delia Grace. Flattening the curve on foodborne illness and its costs in
Africa | IFPRI : International Food Policy Research Institute. IFPRI Blog: Issue Post,
November 2020.

[14] G. Dinede, K. Amenu, S. Alonso, L. Gazu, F. Mutua, K. Roesel, J. F. Lindahl, F. M.
Sousa, P. Ulrich, T. Guadu, M. Dione, G. Ilboudo, T. J. D. Knight-Jones, and D. Grace.
Foodborne hazards in food in burkina faso, 1990–2019: a systematic review and meta-
analysis. Frontiers in Sustainable Food Systems, 7:1232992, 2023.

[15] C. Zavala-Nacul and C. Revoredo-Giha. Microbial risks in kenya’s informal milk supply
chain. Agriculture & Food Security, 11(1):49, 2022.

[16] Hanseni Bognini, Assèta Kagambèga, WP Bertrand Tiendrebeogo, Abdallah Sawadogo,
Nicolas Barro, Théodore Knight-Jones, and Michel Dione. Prevalence of salmonella in
fresh produce tomatoes sold in open market in ouagadougou, burkina faso. Food and
Nutrition Sciences, 14(12):1259–1268, 2023.

[17] Charles Ssemugabo, David Guwatudde, John C Ssempebwa, and Asa Bradman. Pesticide
residue trends in fruits and vegetables from farm to fork in kampala metropolitan area,
uganda—a mixed methods study. International Journal of Environmental Research
and Public Health, 19(3):1350, 2022.

[18] Oluwaseun Mary Oladeji, Liziwe Lizbeth Mugivhisa, and Joshua Oluwole Olowoyo.
Antibiotic residues in animal products from some african countries and their possible
impact on human health. Antibiotics, 14(1):90, 2025.

[19] Shine Francis Gbedemah, Alain Attasse Gbeasor, Gideon Selorm Hosu-Porbley,
Louis Kusi Frimpong, Richard Amfo-Otu, Selase Kofi Adanu, and Eric Kofi Doe.
Analysis of heavy metals and pathogen levels in vegetables cultivated using selected
water bodies in urban areas of the greater accra metropolis of ghana. Heliyon, 10(7),
2024.

[20] Fang Wu. Global impacts of aflatoxin in maize: trade and human health. World
Mycotoxin Journal, 8(2):137–142, 2015.

[21] Nathan Meijer, Gijs Kleter, Monique de Nijs, Marie-Luise Rau, Ria Derkx, and
HJ van der Fels-Klerx. The aflatoxin situation in africa: Systematic literature review.
Comprehensive Reviews in Food Science and Food Safety, 20(3):2286–2304, 2021.

[22] S Watson, PM Diedhiou, J Atehnkeng, A Dem, Ranajit Bandyopadhyay, C Srey,
MN Routledge, and YY Gong. Seasonal and geographical differences in aflatoxin
exposures in senegal. World Mycotoxin Journal, 8(4):525–532, 2015.

18



[23] AJ Sirma, Johanna F Lindahl, Kohei Makita, D Senerwa, Nadhem Mtimet,
EK Kang’ethe, and Delia Grace. The impacts of aflatoxin standards on health and
nutrition in sub-saharan africa: The case of kenya. Global Food Security, 18:57–61,
2018.

[24] GK Blankson, FC Mills-Robertson, and IW Ofosu. Survey of occurrence levels of
aflatoxins in selected locally processed cereal-based foods for human consumption from
ghana. Food Control, 95:170–175, 2019.

[25] Wayne L Bryden. Mycotoxin contamination of the feed supply chain: Implications
for animal productivity and feed security. Animal Feed Science and Technology, 173(1-
2):134–158, 2012.

[26] C Probst, F Schulthess, and PJ Cotty. Impact of aspergillus section flavi community
structure on the development of lethal levels of aflatoxins in kenyan maize (zea mays).
Journal of Applied Microbiology, 108(2):600–610, 2010.

[27] A. Kamala, C. Shirima, B. Jani, M. Bakari, H. Sillo, N. Rusibamayila, S. De Saeger,
M. Kimanya, Y.y. Gong, and A. Simba. Outbreak of an acute aflatoxicosis in Tanzania
during 2016. World Mycotoxin Journal, 11(3):311–320, September 2018.

[28] P Coursaget, N Depril, M Chabaud, R Nandi, Vl Mayelo, P LeCann, and B Yvonnet.
High prevalence of mutations at codon 249 of the p53 gene in hepatocellular carcinomas
from senegal. British journal of cancer, 67(6):1395–1397, 1993.

[29] CP Wild and PC Turner. The toxicology of aflatoxins as a basis for public health
decisions. Mutagenesis, 17(6):471–481, 2002.

[30] Vivian Hoffmann, Kelly Jones, and Jef L Leroy. The impact of reducing dietary aflatoxin
exposure on child linear growth: a cluster randomised controlled trial in kenya. BMJ
global health, 3(6):e000983, 2018.

[31] Lee E Voth-Gaeddert, Matthew Stoker, Olga Torres, and Daniel B Oerther. Asso-
ciation of aflatoxin exposure and height-for-age among young children in guatemala.
International journal of environmental health research, 28(3):280–292, 2018.

[32] Sinead Watson, Sophie E Moore, Momodou K Darboe, Gaoyun Chen, Yu-Kang Tu, Yi-
Ting Huang, Kamilla G Eriksen, Robin M Bernstein, Andrew M Prentice, Christopher P
Wild, et al. Impaired growth in rural gambian infants exposed to aflatoxin: a prospective
cohort study. BMC public health, 18(1):1247, 2018.

[33] Paul W Kachapulula, Juliet Akello, Ranajit Bandyopadhyay, and Peter J Cotty. Afla-
toxin contamination of groundnut and maize in zambia: observed and potential concen-
trations. Journal of applied microbiology, 122(6):1471–1482, 2017.

19



[34] Anitha Seetha, Wills Munthali, Harry W Msere, Elirehema Swai, Yasinta Muzanila,
Ethel Sichone, Takuji W Tsusaka, Abhishek Rathore, and Patrick Okori. Occurrence
of aflatoxins and its management in diverse cropping systems of central tanzania.
Mycotoxin Research, 33(4):323–331, 2017.

[35] Meriem Aoun, William Stafstrom, Paige Priest, John Fuchs, Gary L Windham, W Paul
Williams, and Rebecca J Nelson. Low-cost grain sorting technologies to reduce mycotoxin
contamination in maize and groundnut. Food Control, 118:107363, 2020.

[36] George Mahuku, Henry Sila Nzioki, Charity Mutegi, Fred Kanampiu, Clare Narrod, and
Dan Makumbi. Pre-harvest management is a critical practice for minimizing aflatoxin
contamination of maize. Food Control, 96:219–226, 2019.

[37] Farid Waliyar, M Osiru, BR Ntare, K Vijay Krishna Kumar, H Sudini, A Traore, and
B Diarra. Post-harvest management of aflatoxin contamination in groundnut. World
Mycotoxin Journal, 8(2):245–252, 2015.

[38] YA Xu, Andrew Doel, Sinead Watson, Michael N Routledge, Christopher T Elliott,
Sophie E Moore, and Yun Yun Gong. Study of an educational hand sorting intervention
for reducing aflatoxin b1 in groundnuts in rural gambia. Journal of Food Protection,
80(1):44–49, 2017.

[39] A Ayalew, M Kimanya, L Matumba, Ranajit Bandyopadhyay, A Menkir, and PJ Cotty.
Controlling aflatoxins in maize in africa: strategies, challenges and opportunities for
improvement. Achieving sustainable cultivation of maize, 2:1–24, 2017.

[40] Ranajit Bandyopadhyay, Joseph Atehnkeng, Alejandro Ortega-Beltran, Adebowale
Akande, Titilayo DO Falade, and Peter J Cotty. “ground-truthing” efficacy of biological
control for aflatoxin mitigation in farmers’ fields in nigeria: from field trials to commercial
usage, a 10-year study. Frontiers in Microbiology, 10:2528, 2019.

[41] Gashaw T Abate, Tanguy Bernard, Alain De Janvry, Elisabeth Sadoulet, and Carly
Trachtman. Introducing quality certification in staple food markets in sub-saharan
africa: Four conditions for successful implementation. Food Policy, 105:102173, 2021.

[42] Tulika A Narayan and Judy Geyer. Can results-based prizes to private sector incentivize
technology adoption by farmers? evidence from the agresults nigeria project that uses
prizes to incentivize adoption of aflasafetm. Agriculture & Food Security, 11(1):37,
2022.

[43] Joshua W Deutschmann, Tanguy Bernard, and Ouambi Yameogo. Flexible contracting
and quality upgrading under market volatility: experimental evidence from senegal.
Available at SSRN 5014227, 2024.

20



[44] Gashaw T Abate and Tanguy Bernard. Farmers’ quality assessment of their crops and
its impact on commercialization behavior: A field experiment in Ethiopia, volume 1624.
Intl Food Policy Res Inst, 2017.

[45] Stacy Prieto, Jonathan Bauchet, and Jacob Ricker-Gilbert. Improved drying and
storage practices that reduce aflatoxins in stored maize: Experimental evidence from
smallholders in senegal. 2019.

[46] Vivian Hoffmann, Christine M Moser, and Timothy J Herrman. Demand for aflatoxin-
safe maize in kenya: Dynamic response to price and advertising. American Journal of
Agricultural Economics, 103(1):275–295, 2021.

[47] Hugo De Groote, Clare Narrod, Simon C Kimenju, Charles Bett, Rosemarie PB Scott,
Marites M Tiongco, and Zachary M Gitonga. Measuring rural consumers’ willingness
to pay for quality labels using experimental auctions: the case of aflatoxin-free maize in
kenya. Agricultural Economics, 47(1):33–45, 2016.

[48] Nadhem Mtimet, Maria Walke, Derek Baker, Johanna Lindahl, Monika Hartmann, and
Delia Grace. Kenyan awareness of aflatoxin: An analysis of processed milk consumers.
2015.

[49] Vivian Hoffmann and Ken Mwithirwa Gatobu. Growing their own: Unobservable quality
and the value of self-provisioning. Journal of Development Economics, 106:168–178,
January 2014.

[50] Gashaw Tadesse Abate, Tanguy Bernard, Erwin Bulte, Jérémy Do Nascimento Miguel,
and Elisabeth Sadoulet. Buyers’ response to third-party quality certification: Theory
and evidence from Ethiopian wheat traders. Intl Food Policy Res Inst, 2024.

[51] Tanguy Bernard, Joshua , Deutschmann, Jose Morales, and Tesfamicheal Wossen. A
predictive model of aflatoxin contamination for public policy. mimeo, 2025.

[52] Xu Han, Shuangning Ding, Jinxin Lu, and Yongze Li. Global, regional, and national
burdens of common micronutrient deficiencies from 1990 to 2019: A secondary trend
analysis based on the global burden of disease 2019 study. EClinicalMedicine, 44, 2022.

[53] Howarth E Bouis and Amy Saltzman. Improving nutrition through biofortification: a
review of evidence from harvestplus, 2003 through 2016. Global food security, 12:49–58,
2017.

[54] Howarth E Bouis and Ross M Welch. Biofortification—a sustainable agricultural
strategy for reducing micronutrient malnutrition in the global south. Crop science,
50:S–20, 2010.

21



[55] Jonnalagadda V Meenakshi, Nancy L Johnson, Victor M Manyong, Hugo DeGroote,
Josyline Javelosa, David R Yanggen, Firdousi Naher, Carolina Gonzalez, James Garcia,
and Erika Meng. How cost-effective is biofortification in combating micronutrient
malnutrition? an ex ante assessment. World Development, 38(1):64–75, 2010.

[56] Howarth Bouis, Jennifer Foley, Keith Lividini, Jaya Jumrani, Russell Reinke, Dominique
Van Der Straeten, Ronan Zagado, Erick Boy, Lynn R Brown, Bho Mudyahoto, et al.
Biofortification: future challenges for a newly emerging technology to improve nutrition
security sustainably. Current Developments in Nutrition, 8(12):104478, 2024.

[57] Samantha L Huey, Saiful Islam, Neel H Mehta, Elsa M Konieczynski, Valerie M Friesen,
Jesse T Krisher, Mduduzi NN Mbuya, Eva C Monterrosa, Annette M Nyangaresi, and
Saurabh Mehta. Review of the facilitators and barriers to adoption of biofortified foods
and food products. Nutrition Research Reviews, pages 1–22, 2024.

[58] Elise F Talsma, Alida Melse-Boonstra, and Inge D Brouwer. Acceptance and adoption
of biofortified crops in low-and middle-income countries: a systematic review. Nutrition
reviews, 75(10):798–829, 2017.

[59] Melvin R Huber III and Kari Jänkälä. Portable near-infrared spectroscopy for analysis
of crops. 2020.

[60] Sunaina Yadav, Maharishi Tomar, Tripti Singhal, Nilesh Joshi, HA Bhargavi, Naveen
Aavula, Sapna Langyan, Tanay Joshi, C Tara Satyavathi, Jai Chand Rana, et al.
Near-infrared reflectance spectroscopy (nirs): An innovative, rapid, economical, easy
and non-destructive whole grain analysis method for nutritional profiling of pearl millet
genotypes. Journal of Food Composition and Analysis, 142:107373, 2025.

[61] Xiaobo Jiao, Dongliang Guo, Xinjun Zhang, Yunpeng Su, Rong Ma, Lewen Chen, Kun
Tian, Jingyu Su, Tangnuer Sahati, Xiahenazi Aierkenjiang, et al. The application of
near-infrared spectroscopy combined with chemometrics in the determination of the
nutrient composition in chinese cyperus esculentus l. Foods, 14(3):366, 2025.

[62] Bedanta Rajbongshi, C Nickhil, and Sankar Chandra Deka. Colorimetric sensor
technologies for quality detection in grains: A comprehensive review. Journal of Food
Measurement and Characterization, pages 1–36, 2025.

[63] Mariana Belgiu, Michael Marshall, Mirco Boschetti, Monica Pepe, Alfred Stein, and
Andy Nelson. Prisma and sentinel-2 spectral response to the nutrient composition of
grains. Remote Sensing of Environment, 292:113567, 2023.

[64] Shyamal Chowdhury, JV Meenakshi, Keith I Tomlins, and Constance Owori. Are
consumers in developing countries willing to pay more for micronutrient-dense biofortified

22



foods? evidence from a field experiment in uganda. American Journal of Agricultural
Economics, 93(1):83–97, 2011.

[65] Claudia Meier, Nadja El Benni, Srinivasaiah Sakamma, Simon Moakes, Christian
Grovermann, Sylvain Quiédeville, Hanna Stolz, Matthias Stolze, and K Basegowda
Umesh. Are non-farming consumers willing to pay “a good market price” for iron-
biofortified finger millet? evidence from experimental auctions in karnataka, india.
Journal of Agribusiness in Developing and Emerging Economies, 10(5):751–779, 2020.

[66] Adewale Oparinde, Abhijit Banerji, Ekin Birol, and Paul Ilona. Information and
consumer willingness to pay for biofortified yellow cassava: evidence from experimental
auctions in nigeria. Agricultural Economics, 47(2):215–233, 2016.

[67] Adewale Oparinde, Ekin Birol, Abdoul Murekezi, Lister Katsvairo, Michael T Diressie,
Jean d’Amour Nkundimana, and Louis Butare. Radio messaging frequency, information
framing, and consumer willingness to pay for biofortified iron beans: evidence from
revealed preference elicitation in rural rwanda. Canadian Journal of Agricultural
Economics/Revue canadienne d’agroeconomie, 64(4):613–652, 2016.

[68] Syed Imran Ali Meerza, Khondoker Mottaleb, and Alwin Dsouza. Consumer valuations
of zinc rice: A tale of positive, negative, and weight of memory. 2025.

[69] Hans De Steur, Joshua Wesana, Dieter Blancquaert, Dominique Van Der Straeten,
and Xavier Gellynck. The socioeconomics of genetically modified biofortified crops: a
systematic review and meta-analysis. Annals of the New York Academy of Sciences,
1390(1):14–33, 2017.

[70] Hugo De Groote and Simon Chege Kimenju. Consumer preferences for maize products
in urban kenya. Food and Nutrition Bulletin, 33(2):99–110, 2012.

[71] Robyn Stevens and Alex Winter-Nelson. Consumer acceptance of provitamin a-
biofortified maize in maputo, mozambique. Food Policy, 33(4):341–351, 2008.

[72] Kate Vaiknoras, Catherine Larochelle, Ekin Birol, Dorene Asare-Marfo, and Caitlin
Herrington. Promoting rapid and sustained adoption of biofortified crops: What
we learned from iron-biofortified bean delivery approaches in rwanda. Food Policy,
83:271–284, 2019.

[73] Matin Qaim, Alexander J Stein, and Jonnalagadda V Meenakshi. Economics of
biofortification. Agricultural Economics, 37:119–133, 2007.

[74] Paola Mallia. You reap what (you think) you sow? evidence on farmers’ behavioral
adjustments in the case of correct crop varietal identification. 2022.

23



[75] Keith O Fuglie, Thomas W Hertel, David B Lobell, and Nelson B Villoria. Agricultural
productivity and climate mitigation. Annual Review of Resource Economics, 16(1):21–
40, 2024.

[76] Keith Paustian, Johannes Lehmann, Stephen Ogle, Dave Reay, G. Philip Robertson,
and Pete Smith. Climate-smart soils. Nature, 532:49–57, 2016.

[77] Zhanli Sun, Benjamin Henderson, Zhihao Xu, Eddison N. Masvaya, and Leslie Lipper.
Carbon sequestration potential through conservation agriculture: A global meta-analysis.
Science Reports, 10(1):5473, 2020.

[78] Tesfaye Araya, Christian Thierfelder, and et al. Adoption of conservation agriculture in
sub-saharan africa: Evidence, trends and challenges. Agricultural Systems, 223:103500,
2024.

[79] Jens A Andersson and Shereen D’Souza. From adoption claims to understanding
farmers and contexts: A literature review of conservation agriculture (ca) adoption
among smallholder farmers in southern africa. Agriculture, ecosystems & environment,
187:116–132, 2014.

[80] Dominic Glover, James Sumberg, and Jens A Andersson. The adoption problem; or why
we still understand so little about technological change in african agriculture. Outlook
on AGRICULTURE, 45(1):3–6, 2016.

[81] James R. Stevenson, Rachid Serraj, and Kenneth G. Cassman. Evaluating conservation
agriculture for small-scale farmers in sub-saharan africa and south asia. Agriculture,
Ecosystems Environment, 187:1–10, 2015.

[82] Sylvie M Brouder and Helena Gomez-Macpherson. The impact of conservation agricul-
ture on smallholder agricultural yields: A scoping review of the evidence. Agriculture,
ecosystems & environment, 187:11–32, 2014.

[83] Olaf Erenstein, Ken D. Sayre, Patrick C. Wall, John Dixon, and Jon Hellin. Conservation
agriculture in maize- and wheat-based systems in the (sub)tropics: Lessons from
adaptation initiatives in south asia, mexico, and southern africa. Journal of Sustainable
Agriculture, 36(2):180–206, 2012.

[84] Jenny C Aker and B Kelsey Jack. Harvesting the rain: The adoption of environmental
technologies in the sahel. Review of Economics and Statistics, 107(5):1197–1214, 2025.

[85] Seema Jayachandran, Joost de Laat, Eric F. Lambin, Christopher Y. Stanton, Robin M.
Audy, and Nancy E. Thomas. Cash for carbon: A randomized trial of payments for
ecosystem services to reduce deforestation. Science, 357(6348):267–273, 2017.

24



[86] B. Kelsey Jack, Vikram Bhamoriya, and Sumedha Hegde. Incentivizing environmentally
sustainable agriculture: Evidence from crop residue burning in india. Forthcoming,
American Economic Review, 2025. Forthcoming.

[87] Duncan Knowler and Ben Bradshaw. Farmers’ adoption of conservation agriculture: A
review and synthesis of recent research. Food Policy, 32(1):25–48, 2007.

[88] Le Zhou, Jochem Verrelst, Roberto Della Peruta, Guido Lemoine, and François Kay-
itakire. Monitoring conservation tillage with multi-source satellite time series using
machine learning in belgium. Remote Sensing of Environment, 314:113543, 2025.

[89] Cristina López-Fando and María Teresa Pardo. Soil carbon storage and stratification
with conservation tillage in a spanish cereal–legume rotation. Soil and Tillage Research,
111(2):186–194, 2011.

[90] F Bessam and R Mrabet. Long-term changes in soil organic matter under conventional
tillage and no-tillage systems in semiarid morocco. Soil Use and Management, 19(2):139–
143, 2003.

[91] Ibtissame Lembaid, Rachid Moussadek, Rachid Mrabet, Ahmed Douaik, and Ahmed
Bouhaouss. Modeling the effects of farming management practices on soil organic
carbon stock under two tillage practices in a semi-arid region, morocco. Heliyon, 7(1),
2021.

[92] JLN Carvalho, Carlos Eduardo Pellegrino Cerri, Brigitte Josefine Feigl, Marisa de Cassia
Piccolo, VP Godinho, and Carlos Clemente Cerri. Carbon sequestration in agricultural
soils in the cerrado region of the brazilian amazon. Soil and Tillage Research, 103(2):342–
349, 2009.

[93] HH Janzen, AM Johnston, JM Carefoot, and CW Lindwall. Soil organic matter
dynamics in long-term experiments in southern alberta. In Soil Organic Matter in
Temperate AgroecosystemsLong Term Experiments in North America, pages 283–296.
CRC Press, 2019.

[94] Stephen M Ogle, Cody Alsaker, Jeff Baldock, Martial Bernoux, F Jay Breidt, Brian
McConkey, Kristiina Regina, and Gabriel G Vazquez-Amabile. Climate and soil
characteristics determine where no-till management can store carbon in soils and
mitigate greenhouse gas emissions. Scientific reports, 9(1):11665, 2019.

[95] Guyo Godana Dureti, Martin Paul Jr Tabe-Ojong, and Enoch Owusu-Sekyere. The new
normal? cluster farming and smallholder commercialization in ethiopia. Agricultural
Economics, 54(6):900–920, 2023.

25