Breaking silos: Rethinking how rice breeding programs work

• Many new rice varieties in India are not being adopted by farmers, with only 26% adopting improved seeds despite decades of breeding progress.
• A new brief suggests that this gap is due to different institutional logics across public, private, and development sectors, not just genetics.
• The authors urge “structural synergy”: integrating Market Intelligence and aligning rewards so biophysical, social, and market disciplines jointly drive demand-driven breeding.

By Glenn Concepcion

Since the 1960s, the narrative of agricultural progress has been shaped by the genetic gains of the Green Revolution, which helped transform India into the world’s second-largest rice producer. Yet in recent decades, a growing misalignment has emerged between in-laboratory breakthroughs and on-the-ground realities. 

Despite continuous advances in rice breeding, many new and improved varieties are not being adopted by farmers. In Kerala, for example, the weighted average age of rice varieties is over 25 years. A large household survey of nearly 20,000 paddy farmers across 27 Indian states found that only 26% had adopted new varieties 

A recent brief published by International Rice Research Institute (IRRI) scientists Drs. Arindam Samaddar, Swati Nayak, and Melanie Connor suggest that this adoption gap reflects not only genetic factors but also institutional and social dynamics. The researchers argue that rice breeding is influenced by “institutional logics”, or the embedded values and reward systems that shape which traits are prioritized and which are overlooked. 

To bridge this gap, the authors advocate for Market Intelligence (MI)—the systematic gathering and analysis of data on consumer preferences, market dynamics, and value chain requirements—to evolve from fragmented primary and secondary datasets into a unified, foundational input for breeding. Without such integration, new seeds risk remaining scientific achievements that do not necessarily translate into widespread farmer adoption. 

The social construction of a rice variety 

The researchers identified three primary actors in the Indian rice ecosystem, each operating with a distinct logic that shapes decision-making 

• Public Sector (NARES): National agricultural research systems follow a policy-driven logic focused on national food security, with yield as a non-negotiable priority. Scientists are often rewarded based on the number of varieties officially released, which can create a “closed-loop” feedback system where success is defined by technical trials rather than field adoption. 

• Private Sector: Seed companies operate with a market-driven logic. A rice variety is a product that must generate a return on investment and be scalable. Private firms typically will not commercialize a new variety unless it shows at least a 2–3% yield gain over previous versions, and they often prioritize “mega-varieties” that work across large areas, sometimes overlooking niche or localized needs. 

• Development Organizations (e.g., CGIAR): International organizations follow a research-and-development logic focused on global impact and inclusivity. They try to balance standardized “global pipelines” with the specific needs of regional environments and respond to an “impact narrative” aimed at poverty reduction and development. However, local requirements may not exactly match with global pipelines and can quickly evolve.  

Each of these logics reflects legitimate priorities within its context. The challenge arises when these priorities are not aligned, leading to varieties that perform well in one domain (e.g. research trials) but fall short in others (e.g. market uptake or farmer preferences). 

Different approaches to Market Intelligence 

One of the paper’s key insights is how information is validated and used across sectors. 

• In the public sector, MI is largely “experiential,” often gathered through informal field visits where breeders interact with farmers. While intuitive, this information is often unquantified and can be prone to professional bias. 

• The private sector relies on rigorous internal validation, often collecting proprietary data over months to ensure a variety’s business viability. 

• Development organizations use structured tools like Target Product Profiles (TPPs) and consumer surveys, but the integration of these socioeconomic data into actual breeding decisions can be inconsistent. 

The researchers argue that for breeding to be truly demand-driven, MI cannot simply be a report handed over at the end of a breeding cycle. It must be a structural function that guides a variety from initial design to deployment in farmers’ fields. This requires coordinated input from biophysical scientists, social scientists, economists, and market analysts, each contributing their expertise at the right stage. 

Quantifying “subjective” traits 

The lack of systematic MI has contributed to the marginalization of traits related to Gender and Social Inclusion (GESI). While agronomic traits like disease resistance have strict quantitative benchmarks, traits often preferred by women (such as cooking quality, ease of manual threshing, or labor reduction) are frequently dismissed as “subjective” or “nice to have”. 

Because women are heavily involved in post-harvest processing and cooking, their preferences are vital for a variety’s success. However, when yield is the sole non-negotiable metric, GESI-responsive traits are often sacrificed. For example, Direct-Seeded Rice (DSR) suitability significantly reduces labor for women, but product developers may view it as an indirect benefit rather than a core objective. If a high-yielding variety is difficult to thresh or has poor culinary quality, farmers (particularly women) may simply not adopt it. 

MI is critical because it allows these subjective traits to be quantified into strict benchmarks, making them scientifically actionable for breeders. This does not diminish the importance of genetic or agronomic traits; rather, it highlights the need for a more holistic set of criteria that reflects the full range of user needs. 

A path to “structural synergy” 

The paper concludes that the slow turnover of rice varieties is not necessarily a failure of genetics or extension services, but also a result of institutional alignment, goal-setting issues. To bridge these gaps, the researchers call for stronger “structural synergy,” where different sectors coordinate to connect scientific outputs with actual delivery systems. 

The adoption gap is influenced by who holds the final decision-making authority. In the public sector, breeders often hold the power, leading to a “technology push” where a variety is bred first and a market is sought later. In the private sector, commercial managers chair the committees, ensuring financial viability is the ultimate gatekeeper. 

The researchers propose a realignment of incentives across all sectors: 

• Public sector scientists could be rewarded for actual adoption and commercialization rates, not just the number of varieties released. 

• The private sector could explore new incentive structures to make socially beneficial traits financially viable within ROI models. 

• CGIAR and similar organizations could break down structural silos, empowering social scientists, market researchers to contribute in defining product profiles alongside biophysical scientists. 

Ultimately, improving genetics alone will not fix the adoption gap. A shift from a “technology push” to a “market pull” approach is needed, where Market Intelligence serves as a foundation for every breeding decision. This requires every discipline—genetics, agronomy, economics, social science, and market analysis—to play its part, with the greatest value emerging from how these disciplines work together and complement each other.