Livelihood sustainability assessment of coffee and cocoa producers in the Amazon region of Ecuador using household types

Viteri Salazar, O., Ramos-Martin, J., Lomas, P.L. (2018): “Livelihood sustainability assessment of coffee and cocoa producers in the Amazon region of Ecuador using household types”, Journal of Rural Studies, Vol. 62: 1-9.

Free download until August 15, 2018:

Screen Shot 2018-07-07 at 09.58.10

Abstract: Supporting small farmer livelihoods in fragile, biodiverse regions, such as tropical forests, is a priority for many development agencies and national governments. These regions tend to be characterized by recent human settlements, increasing populations and infrastructure development, as well as competitive land use activities, which exert pressure on fragile ecosystems. Improvement in livelihood strategies often focuses on increasing yields by improving productivity, but without taking into account alternative methods, such as better agricultural practices and their dependence on agrochemical inputs, changing land use through crop substitution, or improving product commercialization. In this research, we use household types, defined according to different land use patterns, in the Northern Amazon region of Ecuador to explore the limitations of, and identify future options for, improving livelihood strategies based on small-scale coffee and cocoa production. The results of the different types are discussed in order to highlight the methods’ utility and identify benefits in terms of environmental and social objectives versus economic profitability. Lessons are drawn that could be useful in applications of public policy aimed at the betterment of small coffee grower and cocoa farmer livelihood strategies, which involve thousands of families in the Amazon region of Ecuador, without compromising the environment.

Keywords: Household types, Amazon, Ecuador, Livelihoods, Coffee and cocoa, Sustainability

The main objective of this study is to contribute to the evaluation of agricultural production systems. This is done here by using a typology of four predominant production systems and comparing their performance against a set of indicators. A second objective is to inform decision makers about the differentiated outcomes these production systems have, so that tailored policy interventions can be designed based on the evaluation of past initiatives. In particular, this work focuses on: i) identifying the socioeconomic and environmental restrictions implicit in different land use patterns; ii) analysing how different land use patterns improve livelihoods in terms of income; and iii) identifying how certain public policies can lead to the establishment of particular types or lifestyles, thereby generating an impact on the income of small- scale producers.

After collecting the data, a typology of households was established as a method of conceptualization and empirical analysis. Household types have been largely used as societal functional units of analysis within integrated assessments for rural systems (Pastore et al., 1999; Giampietro, 2003; Scheidel et al., 2013; Williams et al., 2015), as well as often employed to strengthen the focus of policies and interventions associated with rural livelihoods (Gomiero and Giampietro, 2001; Niehof and Price, 2001; Senthilkumar et al., 2009; Serrano-Tovar and Giampietro, 2014; Tittonell et al., 2010; Williams et al., 2015).

The characterization of types was made in terms of land use patterns and impacts upon the environment, whether through the use of synthetic inputs, the implementation of monocultures or the expansion of the agricultural frontier and the consequent reduction in the forested area. The classification of typologies was based on the technical data sheets suggested by the National Institute of Agricultural Research (Instituto Nacional de Investigaciones Agropecuarias), in terms of area, level of agrochemical use, crop combination, etc., with consideration given to these recommendations as thresholds.

One characteristic shared by the different types is the need for crops that guarantee a permanent inflow of cash. In all cases, apart from coffee and cocoa, there are “other crops” (plantain, maize, cassava, rice and fruit trees) that help in providing food security to households. The types defined include farmers who share at least one of the cash crops, as follows: Type 1 contains coffee and cocoa plantations (CC); Type 2 contains only cocoa cultivation (C); Type 3 comprises coffee, cocoa and oil palm plantations (CCP); and finally, Type 4 only has coffee farming (Cf).

Screen Shot 2018-07-07 at 10.10.00

Screen Shot 2018-07-07 at 10.10.14

Income from coffee and cocoa production represents about 19% of total household income for Types 1 and 2 despite their low yield, whereas it is only 8% for Types 3 and 4. Type 3 earns more from other crops and oil palm, while Type 4 receives the most from off-farm work, although production at the farm still plays a cultural and environmental role.

Under these circumstances, Type 4 performs very well in environmental terms (LIU and 24 ha of forest) and is very close to Type 2 in economic terms. Hence, there is a need to increase income via improving some agricultural practices (e.g., pruning), selecting plants and particularly changing to commercialization, thus making it possible for smallholders to move up the value chain. Unless measures supporting productive activity of this type are implemented, a shift towards Types 1 and 2 could be expected, or even worse towards Type 3 by means of selling or renting their land.

The analysis of the surplus generated by each type helps in identifying the reasons why households have chosen different productive patterns. It can be seen how cocoa has gradually replaced the cultivation of coffee. According to the Third National Agricultural Census (2000), 49,389 ha of coffee and 7751 ha of cocoa were present in the area, while, according to our study, there were 44,580 ha of cocoa and 9500 ha of coffee in 2013 (Fig. 2). This is consistent with our results, as Type 2 has a higher surplus than Type 1, which has the lowest surplus of all and depends largely on off-farm work for guaranteeing livelihood. Until 10 years ago, large-scale landowners dominated oil palm plantations. Only recently have smallholders engaged in oil palm production, thus giving rise to Type 3. Despite the fact that plantations are still young (four to eight years), they already show the highest surplus (US $144.63); however, from an environmental point of view, they are the least interesting of all, because they involve a HIU, with forest only covering 4% of land, and the number of families that have benefited from that is very low (4.5%). One can understand why private enterprises and even the central government are interested in this type (profits, but also greater GDP and taxes). However, this type also re- presents more environmental impacts and a greater degree of dependency for the households (a large fraction of their income is ex- pended in buying agrochemicals from the very same intermediaries who are commercializing palm oil). These reasons do not make this type, in our view, an option for future development.

Screen Shot 2018-07-07 at 10.13.31

Creating types for an analytical case study in one of the most important areas of the Ecuadorian Amazon, with highly biodiverse ecosystems and significant forest cover, allows for visualizing different land use patterns linked to economic resource generation as a means of subsistence. This behaviour is strongly linked to public policies being developed in the sector, where it is believed that oil palm monocultures could be a better alternative revenue source for farmers. However, as shown in Types 2 and 4, its utility is very similar to Type 3 (oil palm), but with a much lower dependence on intensive agrochemical use in the case of Type 2 and a practically null value in Type 4, which hardly uses agrochemical inputs. The differentiated performance of types should encourage the design of tailored interventions to address the problems encountered by each of them, promoting those types with less environmental impact, but which still provide the necessary means for livelihoods and incentives for the conversion of the types with more impact.

The intensity of land use, marked in large part by the expansion of the agricultural frontier and the reduction in forest cover on farms, is much lower in Type 4 than in the other types, making this form of production more desirable in environmental terms. However, this type shows a limitation in terms of income generation, as nearly 35% of its income comes from off-farm work performed by dayworkers. This aspect is important to consider in the development of agricultural programmes oriented toward jump-starting production, where time dedicated by farmers is an important variable, since this type would not have any time restrictions when either adopting a more labour-intensive agricultural practice or expanding the agricultural frontier. In contrast, Type 4 reflects the practical impossibility of adding crops to a farm, since agricultural use of the available land surface is found to be at 96% capacity.

Identifying the land use and economic income corresponding to types has also revealed the fact that Type 2 represents the lowest labour demand per hectare, an aspect that could be useful for pushing programmes focused on cocoa production, in which labour availability, on the one hand, and profitability per hectare for cocoa and coffee, on the other, should be analysed.


The concept of caloric unequal exchange and its relevance for food system analysis: The Ecuador case study

Ramos-Martin, J., Falconi, F., Cango, P. (2017): “The concept of caloric unequal exchange and its relevance for food system analysis: The Ecuador case study”, Sustainability, Vol 9(11), 2068. 


Abstract: The impact of food production patterns and food supply upon consumption patterns is usually explained by economies of scale and affordability. Less attention is given to food trade patterns and global insertion of economies affecting dietary changes. This paper contributes to the discussion using the concept of caloric unequal exchange that defines the deterioration of terms of trade in food in units of calories and complements studies on unequal exchange and ecologically unequal exchange. A new perspective to food systems’ analysis is achieved by using this concept. This paper uses the case study of Ecuador to exemplify its potentiality. Exports and imports to and from Ecuador are analyzed for the period 1988 through 2013 in volume, value, and calories, for different groups of products. The conclusion is that Ecuador is increasingly helping to feed the world, at a caloric cost that is decreasing over time. There is a deterioration of the terms of trade of traded food in terms of calories for Ecuador of more than 250% between 1986 and 2013.

Keywords: Latin America; caloric unequal exchange; terms of trade; food; Ecuador.

Changes in consumption patterns are usually referred to as being caused by changes in preferences in consumers or certain constraints, such as those of a budgetary nature. Acknowledging the relevance of demand-side determinants for consumption, we believe the role a country plays in international trade may also be a determinant. A country’s global positioning in trade, concentrated in certain products, may well induce changes in production patterns, which may have an effect, subsequently, on the domestic supply of food products. The traditional way we look at food systems, in terms of supply and demand, seems limited to us. Data in terms of volume and value may well be complemented with data on the relative cost to the calorie of both imports and exports, what we call caloric unequal exchange. In this way, we can have a different look to the concept of domestic availability which links to other concepts such as food sovereignty. The change in domestic availability (and very often affordability) may also induce changes in consumption patterns. With this consideration in mind, the aim of the paper is twofold: (1) to explore the existence of caloric unequal exchange for Ecuador as defined by a recent study; and (2) to explore the links between changes in food consumption patterns, international trade and domestic production of food products.

In order to respond to the two objectives of the paper, we use the recently introduced concept of caloric unequal exchange, defined as the deterioration in the terms of trade of food traded when considering the cost of exported and imported calories, and explained with detail in Section 2. The study originally introducing the concept focused on Latin America and the Caribbean as a block, showing that, even if calories exported were more expensive than those imported, there was a deterioration of the terms of trade since 1961 and the region was increasingly feeding the rest of the world at ever cheaper costs to the calorie. This went along with an increased volume (and value) of exports, making explicit the recent turn, to intensify the international insertion of the region as a provider of commodities to the rest of the world. This international insertion is seen by some as positive from an economic point of view. However, the impacts upon the environment in terms of soil deterioration, export of nutrients, and increased energy consumption and CO2 emissions from those exports, are still not clear as there are only a few studies that analyze the loss of nutrients involved in food exports.

The insertion of Ecuador in the world economy since then has suffered from two facts. On the one hand, the small scale of its economy prevents it from influencing world prices for commodities exported. On the other hand, the loss of monetary policy implied by dollarization prevented the country from using competitive devaluations for gaining market share. The outcome, for all products but particularly for food products, was an increase in volumes exported, as can be seen in Table 1 below.


Not only did Ecuador increase its exports, but its imports also rose after dollarization. The economy became very open to the global market. The openness index, a measure of the share of imports plus exports over GDP, went from 0.25 in 1965 to 0.57 in 2012, having reached a peak in the year 2008 with a value of 0.64. This vulnerability of the economy to the world markets, aggravated by the lack of industry in the country, implied an increase in food trade as well.

Table 1 presents the food trade balance (for the selected product groups) between Ecuador and the rest of the world for the period 1988 through 2013. Data is presented in volume, monetary value, and its conversion into calories. Exports in volume increased by 4.3 times their original size in the period, less than in terms of calories (5.3), whereas its monetary value increased by 7.9 times. In the case of imports, they increased by 2.1 times in terms of volume and calories, while they increased by 10.9 times in monetary terms.

During the 25-year period analyzed, Ecuador went from exporting double the amount of food that was imported in 1988 to exporting more than five times what was imported in 2013, in terms of volume. This is part of the re-primarization experienced by some of the Latin American economies.


Deepening the data analysis shown in Table 1, Figure 6 presents the cost in US dollars of one million kcal exported and imported in real terms (left axis) and the ratio between the cost of the exported calorie and the imported calorie (right axis), that is, an approximation to the terms of trade measured in calories.

The trend observed in the figure is an increase in the cost of exported calories of 47% in the period, but a much higher increase in the cost of imported calories over time (more than 400%), which implied a deterioration of the terms of trade measured in calories, with a decrease of more than 250% in the period analyzed. Thus, Ecuador is feeding the rest of the world at a lower relative cost over time, despite the recent boom in commodity prices experienced worldwide.

Our research also shows the degree of concentration of consumption in a few products, measured in kcal, comparing 1961 and 2011 (see Table 4). This high concentration in a few products did worsen in the period. Table 4 presents the cumulative calorie intake per product in year 1961 and 2011. The number represents the ranking of that product in both years, for instance, sugar was the second product in caloric terms in 1961 but the fourth in 2011. The table reads like this: in 1961 rice represented 12% of calorie intake, and rice plus sugar 23.9%, and so on. In the year 2011, rice represented 24%, while rice and wheat 41.4%, and so on. Five products (rice, sugar, maize, banana and wheat) represented 52.7% of calorie intake in 1961, while the share went up to 71.1% in 2011 (with a change in composition: rice, wheat, palm oil, sugar, soybean oil). When extending to 10 products, they represented 77.1% of consumption in 1961 and 88.4% in 2011. Therefore, apart from a change in the diet, with a noticeable increase in vegetable oils (soybean and palm) and a decrease in beans, cassava and potatoes, a large fraction of consumption is concentrated in a very small number of products and this concentration is increasing over time, reducing the variety of the diet in the country.


Supplementary materials: The following are available online at, Table S1: CUEE Data.xlsx which includes all the data and references for both figures and tables in the paper.

Spatial assessment of the potential of renewable energy: the case of Ecuador

Cevallos, J., Ramos-Martin, J. (2018): “Spatial assessment of the potential of renewable energy: the case of Ecuador”. Renewable & Sustainable Energy Reviews, 81:1154-1165. Download at

Screen Shot 2017-09-10 at 18.12.55Abstract: Although renewable energy represents a large share of the electric energy generation sources in Latin America, non-conventional sources such as solar or wind energies have not represented a big share of their electric energy systems. The first step to promote the use of these sources in the region is identifying the potential of each energy source, task that can be estimated with the use of spatial tools such as Geographic Information Systems (GIS). This study has reviewed a large list of GIS publications to select a methodology to identify suitable areas for the development of non-conventional renewable energy projects (REP), in order to estimate the maximum energy these technologies could contribute to a national electric energy system, and its applied to the Republic of Ecuador. By using GIS, it is sought to identify the sites where potential renewable energy plants could be located, and initially recommends geographic locations for the installation of measuring towers of solar and wind resources, in order to obtain more detailed information on their behavior. As a result, the areas with higher potential for the development of REP have been identified, and classified in spatial layers according its technology and location. These results show that solar PV is the technology with most suitable areas in the country and demonstrate particularly large potential in two regions: the Andes cordillera and Insular region, especially in the provinces of Loja, Pichincha and the Galapagos islands.


Keywords: Renewable-energy, GIS, Ecuador, Multi-criteria analysis

This study focuses on the assessment of potential locations for the implementation of non-conventional renewable energy projects, in order to estimate the maximum theoretical amount of energy they could contribute to the Ecuadorean’s energy system. By using Geographic Information Systems (GIS) tools, this study seeks to identify suitable areas where NCRE power plants could potentially be sited and classifies the results by regions. The evaluation can be used as a starting point to select the best locations for the installation of resource measuring towers, in order to obtain detailed behavioral data from these sources.

Screen Shot 2017-09-10 at 18.18.32

Screen Shot 2017-09-10 at 18.18.44

Screen Shot 2017-09-10 at 18.18.52

Results show the provinces with the most suitable areas are located in the Andean and Insular regions. Here it can be observed that appropriate locations vary with each resource, however these are concentrated in only few provinces. The provinces of Loja, Pichincha, and Galapagos are the most benefited from the solar resource. It is in these provinces where optimal areas for implementation of SPV and CSP technologies are found. Provinces benefited with wind resources are distributed in the Andean region, however potential locations are limited to the provinces of Loja, Cañar, Chimborazo, Cotopaxi and Pichincha only.

Screen Shot 2017-09-10 at 18.21.15

Organizational structure and commercialization of coffee and cocoa in the northern Amazon region of Ecuador

Viteri-Salazar, O., Ramos-Martin, J. (2017): “Organizational structure and commercialization of coffee and cocoa in the northern Amazon region of Ecuador”, Revista NERA, Núcleo de Estudos, Pesquisas e Projetos de Reforma Agrária, Nro. 35: 266-287. Download

Screen Shot 2017-09-10 at 18.10.00Abstract: The cultivation of coffee and cocoa is the main source of income for small farmers in the northern Amazon region in Ecuador. As border area, they have been beneficiaries of multiple public and private institutions, principality designed to reactivate the production of coffee and cocoa. The goal was to improve the quality of life of local population threatened by poverty and characterised by high level of immigration from Colombia. The current study was carried out to outline the main obstacles faced by producer associations in order to identify policy measures to address these. This study show data of organisational structure, initiatives for marketing under partnerships, and storage infrastructure with an estimate for the production of coffee and cocoa, based on primary and secondary information. It also implies that the government could play a bigger role supporting peasant organisations in different aspects like: capture of added value by peasants; associative commercialisation with a focus on a popular and solidary economy; and offering flexible credit. All of this would encourage participatory, sustainable rural business ventures. Finally, we present different alternatives for improve the implementation of public agricultural policies, about of organisational structure of the producers, commercialisation processes and environmental concerns.

Keywords: Amazon Region; associative commercialization; coffee-cocoa production; public policy.


Esta vez es diferente. La importancia de la energía explicada a expertos de otras áreas

Esta vez es diferente. La importancia de la energía explicada a expertos de otras áreas

Se replica aquí el breve artículo: Ramos-Martin, J. (2017): “Esta vez es diferente. La importancia de la energía explicada a expertos de otras áreas”, in Investigación y Ciencia n°487: 94-95, Abril 2017. Versión alternativa en html aquí, versión pdf aquí.

complexityEl libro Energy, complexity and wealth maximization («Energía, complejidad y maximización de la riqueza»), de Robert Ayres, es la última contribución de este gran científico al análisis energético, tan desaparecido de nuestras facultades. Junto con autores como James Kay o Tim Allen (ecología) o Vaclav Smil (sistemas energéticos), el trabajo de Bob Ayres se ha centrado no solo en analizar y explicar el papel de la energía en el origen de la vida y el funcionamiento de los sistemas naturales y humanos, sino en educar a profesionales de otras áreas en el rol crucial que desempeña la energía. Ese es precisamente el objetivo de este libro: explicar la importancia de la energía a personas formadas en otras ramas del conocimiento que, normalmente, no le prestan demasiada atención, como los economistas.

¿Por qué centrarse en los economistas? Porque, ante la pregunta inocente de un estudiante sobre dónde están los recursos naturales en la función de producción normalmente empleada en economía (la cual solo depende del capital K y del trabajo LY = Af(K,L), donde Y denota el PIB y Aes una variable que, en teoría, recoge el avance tecnológico), la respuesta de muchos profesores de materias como crecimiento económico es, simplemente, añadirlos a la función. Esto tiene el problema de obviar que la producción y la reproducción tanto del capital como del trabajo necesitan recursos naturales: un error repetido por economistas de prestigio como Joseph Stiglitz o Robert Solow. El libro aborda estos problemas desde la perspectiva de los sistemas complejos, las redes ecológicas y la termodinámica de sistemas fuera del equilibrio para mostrar la importancia de la energía en la evolución de los sistemas naturales y sociales, [véase «Economía biofísica», por Jesús Ramos Martín; Investigación y Ciencia, junio de 2012].

La obra se divide en tres partes. La primera se centra en definir conceptos fundamentales, como entropía y «exergía» (o energía útil), así como en analizar el papel de la energía en la creación del universo, el origen de la vida y el funcionamiento del sistema terrestre y los ciclos que lo regulan. La segunda presenta la importancia de la energía en el avance de la ciencia y la técnica, abordando incluso las limitaciones en términos tecnológicos de la actual dependencia del petróleo como recurso agotable. Por último, la tercera recoge desde la visión de la economía ortodoxa hasta las tesis que hoy comparten ecólogos, economistas ecológicos y otros científicos con respecto a la relación entre la disponibilidad y uso de energía y la generación de riqueza material, así como los desafíos que encontramos en una sociedad fuertemente dependiente de los recursos fósiles. El libro termina con una serie de apéndices orientados a profundizar en la modelización de la energía para explicar el crecimiento económico, y en cómo los límites de los recursos no se ciñen a la energía, sino también a la disponibilidad de materiales. Como decía Georgescu-Roegen, la materia también importa (matter matters, too!).

La energía, dice el autor, es la esencia de toda sustancia. La energía útil, o exergía, es la que permite la creación y condensación de la materia y los recursos, pero también la que hace mover el aparato industrial que transforma esos recursos en bienes y servicios que satisfacen las necesidades humanas: el «metabolismo biofísico» del que hablaba Georgescu-Roegen. Sin gradientes de energía disponible no hay evolución de los sistemas.

La evolución prima a los individuos, ecosistemas y sociedades que son capaces de procesar un mayor volumen de recursos, en la línea de lo apuntado por Howard Odum. Son estos gradientes de recursos disponibles los que permiten que las sociedades evolucionen hacia formas cada vez más complejas, con mayor cantidad de estructuras y con una mayor interacción entre individuos. Es decir, una ciudad resulta más compleja que una comunidad de cazadores-recolectores precisamente porque cuenta con una mayor disponibilidad de recursos.

Para numerosos economistas, como Stiglitz y Solow, el conocimiento constituye una especie de nuevo recurso que explica la mejora de la productividad y la creación de riqueza. Sin embargo, no tienen en cuenta la base material de ese conocimiento. La información y el conocimiento resultan baratos de reproducir pero costosos de producir. La generación de conocimiento requiere estructuras (universidades y centros de investigación) y mano de obra dedicada, cuyo mantenimiento consume importantes cantidades de recursos. Para Ayres, es la energía útil, o exergía (o, mejor dicho, el trabajo útil realizado con su consumo) lo que provoca los aumentos de productividad y, por ende, la creación de riqueza. Por tanto, ese cambio tecnológico (la A que aparece en la función de producción de Solow) depende de la disponibilidad de recursos.

Es aquí donde entra a colación el título de esta reseña. Como dice el autor, «esta vez es diferente», pues la sociedad se enfrenta por primera vez en la historia a la escasez de energía y de materiales, los cuales ponen en juego la posibilidad de crear riqueza material. Esto sigue siendo obviado por la mayoría de los economistas, que, en un exceso de optimismo tecnológico, todavía ven el conocimiento y el progreso técnico como disociado del consumo de recursos. Es decir, se sigue sin tener en cuenta la energía como fuente de riqueza y motor del crecimiento económico. Esto es así porque la mayoría de los modelos económicos que todavía se usan en la toma de decisiones asumen, erróneamente, que no hay límites a la oferta de energía. La misma miopía, nos indica el autor, parece aplicarse también a los recursos materiales, pues se sigue obviando la escasez geológica de los mismos.

En conclusión, no se puede permitir que la toma de decisiones se base en teorías ni en disciplinas que resultan extremadamente simplificadoras, como sucede con la economía. La riqueza material y el aumento del nivel de vida asociado van íntimamente ligados al consumo de recursos. Y la energía constituye el recurso fundamental, al ser necesario en cualquier proceso de transformación de unos recursos en otros.

Garantizar la continua provisión de estos recursos debería ser el objeto de la ciencia económica. Sin embargo, dado que hoy una parte considerable de la riqueza no es de carácter material, sino financiero, la disociación con nuestro entorno parece cada vez mayor. Esto hace que contribuciones como la de Bob Ayres con este libro sean cada vez más importantes, especialmente —aunque no solo— en las facultades de economía. Para que todos, incluidos quienes toman las decisiones, entendamos la base material de la vida, del proceso económico y, por ende, de la satisfacción de las necesidades humanas. Energy, complexity and wealth maximization debería ser de lectura obligada tanto para ecólogos como para economistas.

Impact of two policy interventions on dietary diversity in Ecuador

This is a summary of the paper: Ponce, J., Ramos-Martin, J. (2017): “Impact of two policy interventions on dietary diversity in Ecuador”, Public Health Nutrition, Vol. 20 (8):1473-1480,

PHNObjective: To differentiate the effects of food vouchers and training in health and nutrition on consumption and dietary diversity in Ecuador by using an experimental design.

Design: Interventions involved enrolling three groups of approximately 200 randomly selected households per group in three provinces in Ecuador. Power estimates and sample size were computed using the Optimal Design software, with a power of 80 %, at 5 % of significance and with a minimum detectable effect of 0·25 (sd). The first group was assigned to receive a monthly food voucher of $US 40. The second group was assigned to receive the same $US 40 voucher, plus training on health and nutrition issues. The third group served as the control. Weekly household values of food consumption were converted into energy intake per person per day. A simple proxy indicator was constructed for dietary diversity, based on the Food Consumption Score. Finally, an econometric model with three specifications was used for analysing the differential effect of the interventions.

Setting: Three provinces in Ecuador, two from the Sierra region (Carchi and Chimborazo) and one from the Coastal region (Santa Elena).

Subjects: Members of 773 households randomly selected (n 4343).

Results: No significant impact on consumption for any of the interventions was found. However, there was evidence that voucher systems had a positive impact on dietary diversity. No differentiated effects were found for the training intervention.

Conclusion: The most cost-effective intervention to improve dietary diversity in Ecuador is the use of vouchers to support family choice in food options.