Certification schemes in agriculture set voluntary standards and monitor and support compliance, with the objective of making production socially sustainable and terms of trade fairer for smallholder farmers and workers. There has been a proliferation of these schemes since the early 1990s, linked to promoting either product quality from the consumer perspective or ethical and social standards in production to help address the plight of small producers and workers. Certified products such as bananas, cocoa, coffee, cotton, sugar and tea are popular among ethical consumers and are sold in supermarkets worldwide. Certification schemes tailor the standards they develop and the inputs they provide to the needs of a variety of stakeholders, including processors, producers and workers. Their socio-economic objectives are usually to improve farm yields and prices received by farmers, possibly leading to higher farm incomes, total household incomes, assets and social development indicators. By committing to better labour standards, they also aim to improve wages and other working conditions.
Primary review question: What are the effects of certification schemes for sustainable agricultural production, and their associated interventions, on socio-economic outcomes for farmers, wage-labourers and households?
Secondary review question: Under what circumstances, and why, do certification schemes for agricultural commodities have the intended and/or unintended effects? What are the barriers to and enablers of certification's intended and/or unintended effects?and/or any one statement does not always contain all of the objectives (this information might be at several places in the review).
Includable studies had to examine farm-level interventions in the production of agricultural commodities under certification schemes that have clearly defined socio-economic goals and third party auditing. Populations of interest were agricultural producers (farmers) and wage workers in low-and middle-income countries. To address review question one (RQ1) on effectiveness, the authors included rigorous impact evaluation studies using experimental and quasi-experimental designs, including controlled before and after (CBA) studies with contemporaneous data collection and with two or more control and intervention sites, as well as ex post observational studies with non-treated comparison groups and adequate control for confounding. These studies had to report on at least one of the following outcomes: household income or consumption or other measure of socio-economic status; health and education of adults and children; gender equity; producers' and workers' empowerment; gross or net returns to certified production; productivity of commodities; price levels, wages, non-wage labour conditions, organisational empowerment of producers' and workers' organisations; or investments in services and infrastructure.
To address review question two (RQ2) on barriers and enablers, they included any study reporting on implementation dynamics, distributional dynamics or a range of other barriers and enablers using a range of qualitative studies, or through qualitative research which was either part of quantitative impact evaluations or mixed-methods evaluations.
The authors searched relevant websites and databases including AgEcon, JOLIS, IDEAS, Social Sciences Citation Index (SSCI) and Econlit with all searches completed by December 2015. They employed meta-analysis and thematic synthesis to address their research questions.
Headline Findings: a summary statement
The review finds some evidence that certification schemes improve'intermediate outcomes, such as prices, but very little evidence for an effect on endpoint outcomes, such as household income.
The impact on yields is mixed, however only some certification schemes target increasing yields and some, like adopting organic standards, may actually decrease it. When schemes did focus on increasing yields, a lack of credit and the inadequate implementation of trainings were important constraints. Prices for certified products did increase; the possible mechanisms behind this improvement are price guarantees, such as from Fairtrade, and/or access to more remunerative markets, such as can be facilitated by GlobalGAP.'Some evidence suggests that comparing the higher certified prices with certification costs leads to a more realistic measure of the incentives associated with certification. The income from certified products was slightly higher overall, though there is substantial variation in the results across the 10 studies, On average, schemes such as GlobalGAP and UTZ do better by combing effective capacity building with access to remunerative markets, than Fairtrade does as the higher prices may not compensate for the lower yields. Some evidence suggest that limited demand for certified products is a constraint.
Wages for workers at smallholder farms were either unaffected by the schemes or slightly lower than outside them; limited monitoring mechanisms, as well as a country's labour laws and enforcement mechanisms, may impede the adoption of the labour standards mandated in the certification schemes.Average household income and asset ownership did not increase; some evidence suggests that households may derive only a small part of their incomef rom the sale of certified products. There was a marginal improvement school attendance in studies conducted in Sub-Sahran Africa, but no effect was seen in Asia or Latin America. The studies that reported on illness also found no effect.
The review includes included 43 impact evaluations measuring effects of certification schemes and 136 high quality qualitative studies examining barriers and enablers. Eighty-three included studies were in Latin America and the Caribbean, 65 in Africa, 15 in South Asia and 5 in East Asia and the Pacific. Studies evaluated the 12 types of certification schemes including Fairtrade, Organic, UTZ, GlobalGAP, Rainforest Alliance, and 4C.
Implications for policy and practice
Evidence suggests that although certification schemes improve prices and income from agriculture, they do not automatically lead to an increase in household income and wages, nor improve education and health outcomes. Future certification scheme designs should focus on translating higher agricultural income into greater overall household income. Implementation bottlenecks around uneven targeting, certification costs, producer organisation buy-in and efficient operation, and effective monitoring and auditing systems need addressing for certification schemes to enhance farmer and wage worker well-being. Producers need more assistance to secure selling contracts, expand access to certified markets and switch to new and more favourable trading relations. Certification coverage should be improved to include wage workers in all forms of agricultural production and establish enforceable labour standards. This may need to be complemented by changes that contribute to strengthening national labour institutions and labour market conditions. Certification complements strong national and international systems regulating trade and labour conditions in global supply chains.
Implications for further research
The number of certification schemes for which evidence is available is limited, reflecting a bias towards a handful of schemes, especially Fairtrade, which accounts for more than half of the included studies. More research is needed on the impacts of schemes that have not yet been evaluated, and on the impacts, in new contexts, of schemes that have been evaluated elsewhere. Most schemes have bundled interventions, making it hard to identify the causes of given impacts. Future research should adopt a theory-based approach, ensuring that outcomes are measured consistently along the causal chain, from prices, yields and output quality to incomes and human development outcomes.
This is a strong review with a comprehensive search, and thorough screening, coding and synthesis procedures. One minor limitation is that the authors were not inclusive of studies in all languages, though they did include studies in multiple languages. A further minor limitation is that while the authors summarise the findings of all RQ2 studies on barriers and enablers, they do not provide a study-by-study list of the results from RQ1 effectiveness studies.