We study the dynamics of smallholder participation in export value chains focusing on the example of small-scale broccoli producers in the highlands of Ecuador. Combining cross-sectional data from a household survey with 11-year longitudinal data on export market transactions, we explain the hazards of dropping out of a high-value export chain. We apply a multispell cox duration model that allows us to consider multiple entries and exits from the supply chain. We also provide evidence on the welfare impacts associated with participation. The results suggest that small-scale farmers' exit from the export sector is accelerated by high transaction risks experienced in the past. While we find no particular evidence for the exclusion of small-scale farmers from the export sector, we do find that poorer households and female-headed households tend to drop out faster, especially as long as the sector is still prospering. Finally, when considering welfare effects, we do not find evidence that participation translates into tangible benefits for broccoli farmers. We discuss some measures that could help improve the long-term sustainability of smallholder integration in high-value chains.JEL classifications: D23, D81, Q12

Linking smallholder farmers to high‐value markets through contract farming has become an important strategy to increase farm incomes in developing countries. However, various forms of opportunistic behaviour have been reported, threatening the sustainability of inclusive supply chains. When formal institutions are weak, informal codes of conduct like trust can be important complements to facilitate market transactions. We explore the effect of opportunistic behaviour on farmers’ trust, using a field‐framed trust experiment with prior signaling with small‐scale farmers in Ecuador. These farmers are linked to different types of markets, including high‐value export chains, and have been exposed to varying levels of opportunistic behaviour in the past. Our unique dataset allows us to control for self‐selection of game participants. We find that a positive signal triggers a positive response increasing trust, while a negative signal has no effect – possibly reflecting very low overall trust levels. In settings where market linkages have failed previously, ex‐ante credible positive signals could help to build trust and encourage small farmers’ participation in sustainable value chains.

Fermentation of agave sap, also known as exudate, has become an ancestral practice throughout Ecuadorian Andean. In Cayambe, located in this region, grows Agave americana L., which is recollected, and its sap is fermented. The agave-based fermented beverage, locally named "tzawar mishki", exhibits variable features, mainly ethanol concentration. In this work, fermentation conditions of agave sap were studied to enhance ethanol yield. Two thermal treatments for raw exudate were evaluated, pasteurization at boiling point for 30 minutes and sterilization at 121°C for 15 minutes; fermentation temperature, 30°C and room (around 18°C); and two yeast strains. Thermal pretreatments have a positive impact on reducing sugars and sucrose concentration. In the first case, an increase of 76 % and 30 % has been reported, while sucrose concentration quadrupled and doubled in pasteurized and sterilized samples, respectively. The highest ethanol concentration (63,31 g/L) and the best yield (66,21 %) were accomplished through agave sap pasteurized and fermented for 96 hours at 30°C. Negligible differences have been evidenced in ethanol and other volatile compounds content between the two yeast strains evaluated. Keywords: agave sap, ethanol, fermentation, fermentable sugars, yeast