The complex dynamics of smallholder technology adoption: the case of SRI in Madagascar

This article explores the dynamics of smallholder technology adoption, with particular reference to a high‐yielding, low external input rice production method in Madagascar. We present a simple model of technology adoption by farm households in an environment of incomplete financial and land markets. We then use a probit model and symmetrically censored least squares estimation of a dynamic tobit model to analyze the decisions to adopt, expand, and disadopt the method. We find that seasonal liquidity constraints discourage adoption by poorer farmers. Learning effects—both from extension agents and from other farmers—exert significant influence over adoption decisions.     

Market imperfections,access to information and technology adoption in Uganda: challenges of overcoming multiple constraints

Limited empirical evidence exists on how multiple binding constraints influence the adoption of improved technologies by smallholder farmers. This article uses the case of groundnut variety adoption in Uganda to investigate the role of information, seed supply, and credit constraints in conditioning technology uptake. New data from a household survey in seven groundnut growing districts (n = 945) indicate that 8% of farmers lack information on new varieties, while 18% and 6% of farmers, respectively, cannot adopt mainly due to seed supply and capital constraints. A tobit‐type specification that considers all nonadopters as being uninterested in the technology (i.e., corner solutions) would lead to inconsistent parameter estimates and incorrect conclusions in this context. We therefore estimate a modified multi‐hurdle specification of demand for new varieties, taking into account how information, seed supply, and capital constraints jointly determine adoption probability and intensity. The study reveals new empirical insights on why agricultural technology adoption in Africa has lagged behind: slow uptake is not mainly due to a lack of economic incentives, but rather a reflection of information, seed supply, and credit constraints that prevent farmers from translating their desired demand into adoption of modern varieties. Policy implications are discussed.     

Risk,learning, and technology adoption

This article explores how decision makers learn and use information, with an application to the adoption of biotechnology in agriculture. The empirical analysis relies on experimental and survey data measuring risk preferences, learning processes, and the adoption of genetically modified (GM) seeds among U.S. grain farmers. While controlling for risk aversion, we link individual learning rules with the cognitive abilities of each decision maker and their actual GM adoption decisions. We find evidence that very few individuals are Bayesian learners, and that the population of farmers is quite heterogeneous in terms of learning rules. This suggests that Bayesian learning (as commonly assumed in the analysis of agricultural technology adoption) is not an appropriate characterization. In addition, we do not find a strong relationship between observed learning styles and the timing of GM seed adoption. To the extent that learning is a key part of the process of technology adoption, this suggests the presence of much unobserved heterogeneity in learning among farmers.     

Agricultural diversification and integrated pest management in Bangladesh

We study factors associated with a shift toward diversified, high‐valued vegetable crops and the incentives associated with the use of IPM methods for vegetable producers in Bangladesh. The primary objective is to measure how IPM technologies affect the crop and technology choices of low‐income rice farmers. A three‐season household optimisation model is used to study crop and technology choice under price and yield uncertainty. The model is parameterised using data from vegetable farms and experimental IPM trials conducted in Bangladesh. Simulation results show that access to IPM technology and IPM availability combined with access to credit increase household welfare and lead to higher rates of vegetable adoption. Off‐farm employment opportunities work against vegetable cultivation and IPM use by risk‐averse farmers. Implications for policy and extension efforts are highlighted.     

Risk, uncertainty, and learning in adoption of a crop innovation

Risk and uncertainty have often been suggested as causes of poor adoption of rural innovations, but empirical evidence has been scarce. This study focuses on a new crop‐type, chickpeas, in Western Australia to gather such evidence. The empirical models developed are based on a theoretical framework that conceptualizes adoption as a dynamic decision process involving information acquisition and learning‐by‐doing by growers who vary in their managerial abilities, risk preferences, and their perceptions of the profitability and riskiness of the innovation. Learning encompasses improvements in skill as well as reductions in uncertainty. An annual face‐to‐face survey of over 100 farmers was conducted over 3 years, eliciting the farmers' risk attitudes and their subjective distributions of yields and prices. Two limited dependent variable models, Tobit and Probit, are used to estimate the empirical model. There is a high degree of goodness‐of‐fit for both models. The study provides strong empirical support for the primarily economic character of the adoption decision, and highlight the importance of economic risk in the process. The two risk‐related factors with greatest impact on the adoption decision were risk aversion and relative riskiness of the innovation. Risk aversion tended to reduce adoption, and to do so to a greater extent as relative riskiness and scale increased. Results also reveal the key role that trialing of the innovation plays in adoption.     

New Maize Variety Adoption in Mozambique: A Spatial Approach

Farmers in developing countries can dramatically improve their productivity by adopting new plant varieties. Yet, informational barriers often mean adoption rates remain low. In this study, we focus on how learning from others represents one means of removing informational barriers. We capture the effect of social learning through an explicitly spatial econometric model applied to farm‐level maize adoption rates in Mozambique. We find that social learning is significant and can help explain the apparent clustering of adoption among farmers. Agencies interested in promoting variety adoption, therefore, would be well served to leverage the strength of existing information networks, rather than imposing solutions that work against interfarmer information flow.     

Heterogeneous information exposure and technology adoption: the case of tissue culture bananas in Kenya

Classical innovation adoption models implicitly assume homogenous information flow across farmers, which is often not realistic. As a result, selection bias in adoption parameters may occur. We focus on tissue culture (TC) banana technology that was introduced in Kenya more than 10 years ago. Up till now, adoption rates have remained relatively low. We employ the average treatment effects approach to account for selection bias and extend it by explicitly differentiating between awareness exposure (having heard of a technology) and knowledge exposure (understanding the attributes of a technology). Using a sample of Kenyan banana farmers, we find that estimated adoption parameters differ little when comparing the classical adoption model with one that corrects for heterogeneous awareness exposure. However, parameters differ considerably when accounting for heterogeneous knowledge exposure. This is plausible: while many farmers have heard about TC technology, its successful use requires notable changes in cultivation practices, and proper understanding is not yet very widespread. These results are also important for other technologies that are knowledge‐intensive and require considerable adjustments in traditional practices.     

Welfare impacts of maize–pigeonpea intensification in Tanzania

This article examines the driving forces behind farmers’ decisions to adopt improved pigeonpea and maize and estimates the causal impact of technology adoption on household welfare using data obtained from a random cross‐section sample of 613 small‐scale farmers in Tanzania. We use seemingly unrelated and recursive bivariate probit regressions to test the endogeneity and joint decision making of pigeonpea–maize production. A double hurdle model is used to analyze the determinants of the intensity of technology adoption conditional on overcoming seed access constraints. To address the impact of adoption on welfare, the article employs both propensity score matching and switching regression techniques. Results from bivariate probit models show that unobservable factors cause both decisions to be correlated but the finding does not support the conjecture that both decisions are made jointly. Overall the analysis of the determinants of adoption identifies inadequate local supply of seed, access to information, human capital, and access to private productive asset as key constraints for pigeonpea technology adoption. The causal impact estimation from both the propensity score matching and switching regression suggests that maize/pigeonpea adoption has a positive and significant impact on income and consumption expenditure among sample households.     

Interaction between crop insurance and technology adoption decisions: The case of wheat farmers in Chile

This paper examines relationships between crop insurance and input technology decisions among Chilean wheat farmers. Using nationwide farm‐level data, a bivariate probit model is estimated. We investigate the extent to which the adoption of production input technologies is associated with farmers’ participation in the insurance program. We find that relationships between insurance and technology decisions are significant only for family farmers. In particular, there is a negative relationship between participation in the insurance program and the adoption of modern irrigation. Interpretations based on the role of input technologies on insurance adoption and adverse selection behaviours are discussed.     

Technology adoption under seed access constraints and the economic impacts of improved pigeonpea varieties in Tanzania

Dry‐land legumes, well adapted to drought‐prone areas, have largely been neglected in the past despite the good opportunities they offer for income growth and food (and nutritional) security for the poor. This study evaluated the adoption and impact of two farmer and market‐preferred and disease‐resistant pigeonpea varieties that were developed and promoted in semi‐arid Tanzania. The new varieties were resistant to fusarium wilt, a fungal disease devastating the crop. However, farmers wanting to adopt new varieties did not adopt due to seed access constraints and under‐developed seed delivery systems. Adoption of new varieties is therefore analyzed using an augmented double hurdle model that allows estimating variety adoption conditional on seed access thresholds accounting for the additional information on sample separation. The study identifies the crucial role of seed access (local supply), extension, education, participatory decision making, capital, and household assets in determining adoption. The social economic benefits of the technology and policies for improved seed access were further analyzed using the extended economic surplus method (DREAM model). Even under restrictive assumptions, overall discounted benefits were found to be quite attractive, indicating the need for additional efforts to scale‐up the success story. Analysis of changes in research benefits from relaxing the seed access constraint showed that net gains would increase by up to 30% if farmer access to improved seeds can be assured. Smallholder farmers are the major beneficiaries along with consumers and rural net‐buyers who gain from productivity‐induced lower market prices.     

Tradeoffs and complementarities in the adoption of improved seeds,fertilizer, and natural resource management technologies in Kenya

There is widespread consensus that agricultural technology has an important role to play for poverty reduction and sustainable development. There is no consensus, however, about the types of technologies that are best suited for smallholder farmers in Africa. While some consider natural resource management (NRM) technologies as most appropriate, others propagate input intensification with a stronger role of the private sector. In the public debate, these two strategies are often perceived as incompatible. Environmental non‐governmental organizations in particular consider low‐external input strategies as the only sustainable form of agriculture, a view that has considerable influence on policymakers and the international donor community. Most existing research studies on smallholder innovation focus on the adoption of individual technologies, so that comparisons between different types of technologies in the same context are not easily possible. We use representative data from maize‐producing households in Kenya and a multivariate probit model to analyze the adoption of different types of technologies simultaneously. Results indicate that NRM technologies and strategies that build on external inputs are not incompatible. Interesting complementarities exist, which are not yet sufficiently exploited because many organizations promote either one type of technology or the other, but rarely a combination of both.     

The impact of the use of new technologies on farmers’ wheat yield in Ethiopia: evidence from a randomized control trial

In 2013, Ethiopia's Agricultural Transformation Agency introduced the Wheat Initiative to increase smallholder productivity. In this article, we measure the impacts of the Wheat Initiative package of technologies, and its marketing assistance component alone, on yields among a promotional group of farmers. The package includes improved techniques, improved inputs, and a guaranteed market for the crop. Relying on crop‐cut measures and farmers’ own assessments, we find that full package led to an average 14% higher yields. Implementation of the Wheat Initiative was successful in making certified seed and fertilizer accessible to farmers and increasing their uptake, though only 61% of the intervention group adopted row planting and few farmers received marketing assistance. The measured yield difference may underestimate the true yield difference associated with the technology because of incomplete adoption of the recommended practices by intervention farmers and adoption of some practices by control farmers.     

Impact of training on the intensification of rice farming: evidence from rainfed areas in Tanzania

Agricultural development is indispensable for poverty reduction and food security in sub‐Saharan Africa (SSA). This study investigates the impact of rice production training in a modified version of the System of Rice Intensification (SRI) on the performance of small‐scale rice farmers in a rainfed area of Tanzania. Utilizing the plot level variation, we employ propensity score matching (PSM) to assess the impact of training on technology adoption, productivity, and profitability. We also estimate a difference‐in‐differences model with plot fixed effects using recall panel data covering the periods before and after training. We found that trainees achieved an average paddy yield of 4.7 tons per hectare and rice profit of 191.5 USD per hectare on the plots where new technologies were adopted, which is higher by about 1.3–1.8 tons and 119–137 USD per hectare than on the other plots. Our study suggests the high potential of transforming favorable rainfed rice growing areas in SSA so as to achieve a rice Green Revolution through training in modern input use and improved agronomic practices.     

Modern variety adoption and risk management in drought prone areas: insights from the sorghum farmers of eastern Ethiopia

Adoption rates of improved or modern varieties (MV) of sorghum in eastern Ethiopia are generally low. Although these MV may represent an effective means of coping with droughts, given their early maturing traits, landraces could prove to be more drought‐tolerant and better adapted to marginal production conditions. Whether MV adoption is a risk reducing technology is very much an empirical question which this article investigates using a unique dataset from eastern Ethiopia in a year of extreme weather conditions. Results show that risk‐factors coupled with access to markets and social capital drive farmers’ decisions to adopt MVs. On the one hand, it appears that farmers use MVs to mitigate moderate risks. On the other hand, farmers who have been most vulnerable to extreme weather events are less likely to use MVs suggesting that MV adoption does not necessarily represent an effective means of coping with drought. Finally, findings show that MV growers are more likely to be affected by sorghum failure once controlling for exogenous production factors.     

The adoption of intensive monocrop horticulture in southern Cameroon

Results from a 1997 survey of 208 households in the humid forest zone of southern Cameroon indicate that African policy makers seeking to intensify agricultural production should focus attention on the horticultural sub‐sector. The survey, which gathered information on horticultural production practices, found that the average expenditure on agro‐chemical inputs by horticultural producers using monocrop production systems was 190US$/ha, which greatly exceeds the FAO reported national average expenditure of 6.50 US$/ha. A logit model of monocrop adoption indicated that the size of land holding per household had a negative effect on adoption, congruent with population‐driven technical change and that increases in unit transportation costs significantly decreased the probability of adoption. These findings suggest that policy makers should target horticultural intensification in areas of higher population density and promote investment in rural roads. The age of the household head had a significant negative and elastic effect on adoption, which in combination with an increase in the cohort of younger farmers in the rural population induced by macro‐economic events contributed to the spread of intensified horticulture. In the study area, roughly two‐thirds of rural households also produce cocoa and the quantity of cocoa produced was positively associated with adoption of intensive horticultural systems suggesting that export crop promotion indirectly facilitated diversification of agriculture. Women's participation in intensive monocrop production was limited and efforts to promote their greater involvement are recommended.     

Mobile money,agricultural marketing,and off‐farm income in Uganda

Mobile money (MM) services can contribute to welfare gains in smallholder farm households. Previous research showed that one important pathway for these MM‐related welfare gains is through higher remittances received from relatives and friends. Here, the role of other impact pathways is examined, especially focusing on agricultural marketing and off‐farm economic activities. The analysis builds on panel data from smallholder coffee farmers in Uganda. Regression models show that the adoption of MM technology has contributed to higher household incomes and consumption levels. Off‐farm income gains are identified to be an important pathway, also beyond remittances. Typical off‐farm income sources are small businesses in trade, transport, and handicrafts, which benefit from novel savings and money transfer opportunities through MM. In terms of agricultural marketing, MM users sell a larger proportion of their coffee as shelled beans to buyers in high‐value markets, instead of selling to local traders immediately after harvest. MM services help to reduce cash constraints and facilitate transactions with buyers from outside local regions. In conclusion, MM can contribute to rural development through various important pathways. Analysis of adoption patterns suggests that MM services are socially inclusive.     

Cotton farmers’ willingness to pay for pest management services in northern Benin

This study was carried out to assess cotton farmers’ willingness to pay (WTP) for pest management services in northern Benin. Targeted staggered control (TSC) has been introduced to reduce pesticide use in cotton cropping and generate an estimated benefit of FCFA48,800 (€74.40) per cotton hectare accruing from increased productivity and reduced pesticide cost. However, TSC application requires extra time for pest identification and scouting, and its adoption remains low due to the lack of funding to boost farmers’ awareness and cover training costs. An interval regression model was used to analyze responses to a double‐bounded contingent valuation survey with data collected from 300 cotton farmers. The results showed that 87.3% of cotton farmers were willing to pay for TSC services. Annual WTP per cotton hectare was estimated at FCFA16,962 (€25.80), revealing an existing demand for TSC adoption. Respondents' WTP was driven by farm and socio‐economic characteristics. Financial mechanisms managed by farmers could thus potentially foster technology adoption, and in turn, generate economic and environmental benefits.     

Is late really better than never? The farmer welfare effects of pineapple adoption in Ghana

Export agriculture offers potentially high returns to smallholder farmers in developing countries, but also carries substantial market risk. In this article we examine the intertemporal welfare impact of the timing of a farmer's entry into the export pineapple market in southern Ghana. We examine whether farmers who never cultivated pineapple are better or worse off than farmers who decided to adopt pineapple earlier or later relative to their peers and experienced a significant adverse market shock several years prior to our endline survey. We use a two‐stage least squares model to estimate the causal effect of duration of pineapple farming on farmer welfare. Consistent with economic theory, we find that earlier adoption of the new crop brings greater welfare gains than does later uptake. But we find that the gains to later uptake of pineapple—just before the market shock—are small in magnitude, just 0.1 standard deviations of a comprehensive asset index, indicating that the gains to adoption may be precarious and depend on the context, in particular on the severity of prospective market shocks.     

The impact of safety nets on technology adoption: a difference‐in‐differences analysis

This article contributes to a growing body of empirical literature relating credit constraints and incomplete insurance to investment decisions. We use panel data from rural Ethiopia to investigate whether participation in a safety net program enhances fertilizer adoption. Using a difference‐in‐differences estimator and inverse propensity score weighting, we find that participation in Ethiopia's food‐for‐work (FFW) program increased fertilizer adoption in the short run, but not in the long run. Results also indicate that the intensity of fertilizer usage increased with livestock holdings for FFW‐participant households, providing some evidence that the intervention helped asset‐rich farm households more than asset‐poor households. We find no significant effects of free distribution on fertilizer adoption or intensification. Our results are consistent with the hypothesis that safety nets can be viewed as mechanisms that allow households to take on more risk to pursue higher profits. The results highlight the importance of safety net programs, their effectiveness in ensuring farmers that they will be protected against uninsured shocks, and how that assurance can translate into productivity‐enhancing behavior.     

The effect of information sources on technology adoption and modification decisions

This article focuses on adoption and modification decisions on improved maize storage technologies in southern Benin. Modification implies changing a technology to adapt to farmers' circumstances. A sample selection framework is used to account for selectivity bias as some farmers were not aware of the new technologies. Using this framework, the study investigates the effect of alternative information sources on adoption and modification decisions. Farmers are either informed by extension agents or by other farmers. The empirical results show that there are differences in adoption and modification decisions between these two groups.