Towards a Gendered Agro-Commodity Approach

Commodity or value chains are the dominant means to channel agro-food products from cultivators to consumers. Direct open markets are either non-existent or insignificant . These chains are also the main mechanisms for integrating underprivileged groups into the world economy. Why do global value chains generate sorrow for many and joy for a few, and why are these outcomes heavily gender biased? To look for answers this article critically reviews the post-2000 and earlier gender literature by proponents and opponents of the mainstream value chain approach. The purpose is to provide a methodological contribution on the integration of gender into the commodity chain approach. Most studies have fo cused on the economic effects of chain dynamics on women in agricultural product and labor markets . Some have extended this reasoning with social and cultural effects. Despite these advances, analytical gaps still exist as most existing research has concentrated on the agricultural nodes of modern, high value chains and lacks a gendered conceptual foundation. Scarce attention has been given to traditional staple crops, non-agricultural nodes, and feed back effects of gender relations on the chain. Our results indicate that an appropriate GCC approach should also consider the gendered impacts of the interaction between the governance structure and the institutiona l embeddedness, as well as the consequences of intra-household division of resources and labor in all stages of the chain. These two conceptual complements will be needed to explain the opportunities and constraints to improve gender equity in traditional and modern agro-commo dity chains.

The global commodity chain (GCC) approach is derived from the world system perspective (Wallerstein 2004;Bair 2009) and research in this tradition has largely concentrated on the relationships between the core and periphery.Relatively little attention has been paid to the social dimensions within these areas.The approach "promotes a nuanced analysis of worldeconomic spatial inequalities in terms of differential access to markets and resources" (Gereffi, Korzeniewicz and Korzeniewicz 1994:2).It addresses questions about what products countries import and export in relation to complex institutions like WTO rules, import barriers and others.In a historical study on rubber, Fernandes (2010) further noted that attention had been mainly paid to the macro dimension, falling short on human relations at the micro level.This is why GCC dynamics have been identified by four basic dimensions, which include the input-output, territorial and governance structures, together with the institutional embeddedness (Rammohan and Sundaresan 2003;Unni and Scaria 2009).
Global commodity chains are considered the principal mechanism integrating underprivileged groups into the world economy as stepping stone towards development (see Webber and Labaste 2010) for positive experiences in African agricultural chains).Therefore this article will concentrate on the gender dimension of global chain dynamics and will refer only to broader gender issues-such as those addressed in women and development work by Ester Boserup, Maria Mies and Sylvia Chant--when necessary.Quite a number of studies have researched the consequences of women participation in global commodity chains.However few have included a coherent gendered analysis, for agro-commodity chains (Allen and Sachs 2007;Maertens and Swinnen 2009;Blandon, Hanson and Cranfield 2009).Most are concentrated on non-traditional and traditional export commodities like fruit, vegetables, flowers, coffee and cacao (Webber and Labaste 2010).The studies focus on the agricultural stage only with less attention to other nodes of the chains.
This article offers a gendered agro-commodity approach that we believe provides a more complete and realistic vision on the functioning and dynamics of the chain.Our purpose is to provide a methodological contribution to the GCC approach.
Commodity chains provide more than just commodities and their value generation; this process affects individual and collective identities of actors, linking the material with the cultural (Ramamurthy 2004).In most commodity chain analyses, insufficient attention is paid to daily human and ecological circumstances (Dunaway 2001;2014).According to ILO statistics, 66 million workers were assembling ( agro-)industrial products for global markets in 2006.This global assembly line is often accompanied by the feminization of manufacturing (Bair 2010).This is done mostly under conditions of long working-hours, low wages, health risks, sexual harassment and union under-representation (Pyle 2001;Pyle and Ward 2003;Palpacuer 2008).Lack of gender desegregation makes it difficult to examine how men and women are affected differently so that gendered inequalities can be addressed.In agro-commodity chains as diverse as those of Indian tea, Colombian roses and Guatemalan geraniums, working and living conditions of women are very vulnerable to changes in international business strategies and public policies (Cabezas, Reese and Waller 2007).Focusing only on the nature of the commodity as it goes through the chain does not give an exhaustive view of who benefits and who loses from transactions along the chain.Specific attributes of a chain are shaped by the different societies and communities through which it traverses (Rammohan and Sundaresan 2003).It is further noted that the relations of materials and people in a commodity chain are affected by their social milieu of class, caste, and gender (Ramamurthy 2011 ).The commodity chain concept is a very useful tool for integrating women and households into world-systems analysis (Dunaway 2001;2013).One has to emphasize that gender relations affect and are affected by the ways in which value chains function (Bammann 2007).There is a need to disentangle the gender role s in what Maria Mies (2007) has named the formal and informal underground of the global capitalist accumulation model and their importance for global neoliberal policies (Shepherd and Ferguson 2011).
It is against this background that our study will reflect on what can be learned from post-2000 gender-related studies of agro-commodity chains, and will discuss if the y have been able to focu s on gender conceptually.The advances in this area are undoubtedl y based on research prior to the present centur y.Nevertheless we will not extend very much into the past , to be able to give due attention to the latest developments of studies on gender in agro-commodity chain s and on the evolution of reasoning by authors over time.The credits for earlier work will be there when needed.The other objectives of the article are to draw attention to analytical gaps and contradictory interpretations that still exist and to understand how the global commodit y chain approach should be complemented to give it a gender sensiti ve conceptual foundation.From a Sub-Saharan Africa perspective , the major role of women in both producti ve and care activities should be taken into account.We begin by considering how the selected studies on global agrocommodity chains have included gender in their work.We then discus s the analytical and conceptual gaps in these studies that impede understanding of gender relations in agrocommodity chains.After this the gendered aspects of the interactions betw een the governance structures and the institutional frameworks of agro-commodit y chain s are examined.We consider where there are opportunities and challenges for improving the return s to weak participants.We mak e a case for expanding attention to th e intra-hou sehold division of labor and resources.This is neces sary bec ause man y agriculture-b ased chains in Sub-Sah ara Africa involve a m ajori ty of women as growers and workers.Prescription s for stru cturin g a gendered agrocommodit y appro ach conclude this articl e.

Lessons from Previous Studies
Gender refer s to the sociall y constructed roles and behaviour s of men and women , whi ch change s w ith t ime and space.It is a relational concept, whi ch varies w it h socio-political and cultural differentiation (Chant and Pedwell 2008).Some other aspects such as ethni city, age and religion may hav e importance for Sub-S ahara Africa, but will be treat ed only indir ectly as we maintain our focu s on th e gender dimension.
The con cept of agro-commodity chain has been applied in studies of m any indu strial and agricultural product s.Immanuel Wallerstein defined a commodit y chain as "a network of labour and produ ction pro cesses whose end result is a fini shed comm odity " (Hopkin s and Wallerstein 1994: 17).It is interestin g to refer also to Rammoh an and Sund aresan (2003), wh o call it a commodit y web, which is a network of labor and material pro cesses th at results in an end produ ct.The term s comm odity chain s, valu e chain s and comm odity web w ill be applied interchangeably, for the y all have the same meanin g as defin ed above.A sampl e of 11 th eoreti cal and 25 applied post-2000 article s on comm odity chains tha t addre ss gender is used 1 (see appendix for a list of articles and classific ation s) .Below we report on how the se earlier studi es have combin ed gender with th e agro-comm odity approach.
Representative efforts to include gender explicitly in agro-commodity chain research were selected, mostly but not exclusively from experiences in Sub-Sahara Africa.Relevant theoretical and historical literature from other continents was also included.A noteworthy empirical article of York and Ergas (2011) demonstrated the general presence of association between women's status within countries and the structural position of these in the world-system.Another example is Barrientos, Dolan and Tallontire (2003), where the gender sensitive codes of conduct in the African export horticultural sector is explored by combining the global value chain and gendered economy approaches.Nevertheless, many authors still continue to ignore the gender dynamics, overlooking the fact that commodity chains are embedded in households whose survival depends very much on the work of women (Dunaway 2001;2013).As Dunaway (2014:64) observed, there is a "glaring absence of women from commodity chains." The selected post-2000 publications on gender in agro-commodity chains have been categorized according to their main orientation as theoretical (11), historical (3), poverty (10) and codes of conduct ( 6) and commercial (6).Attention has been paid to the methodological focus and the presence of specific gender, household and upgrading aspects.Practically all the studies refer to high value agro-commodities and the outcomes for small farmers.Studies by Rammohan and Sundaresan (2003:903) and Mayoux and Mackie (2007) pay explicit attention to the social embeddedness of the chains, which Rammohan and Sundaresan refer to as "mapping the social linkages of production and movement of commodities." The research on modern high value crops focuses on globalization trends and how poor farmers and workers may benefit.For instance, Dolan and Sutherland (2003) and other commercialization and codes of conduct studies share the view with Swinnen and Vandeplas (2009) that high value chains link many poor workers in developing countries-including increasing numbers of women-with rich consumers in the developed ones.This gives opportunities to spread the gains from globalization to the general population, but negative effects are also possible.
The majority of the studies that explore a gender dimension have considered women as small producers or (family) workers, who may be exploited or excluded in many chains.The feminist commodity chain approach in Ramamurthy (2004) explains this by the power division and its impact on both men and women's involvement.A more explicit and complete methodological framework on how to integrate gender into agro-commodity chains research is needed.
Intra-household dynamics and/or consumption are addressed in less than a third of the studies in our sample (Dunaway 2001 and2012;Maertens and Swinnen 2009;Ramamurthy 2004;Greenberg et al. 2012;Allen and Sachs 2007;Farnworth and Hutchings 2009;Bolwig and Odeke 2007;Dolan and Sutherland 2003).Nevertheless, intra-household issues can explain gender differences in who will participate in the labor market and when and who controls the income generated from production and labor contracts.Little attention is paid to what really goes on in the households that provide the primary labor force and resources for agro-commodity chains (Mies 2007;Dunaway 2001;2013).Consumption patterns of households that produce for the markets in different nodes of the chains need to be known to consider the impact on nutrition levels within families, which may affect their labor supply.This omission is possible for most modern value chains, where only 'free' labor is contracted.Gender also plays a dominant role in ultimate consumption decisions, which have a considerable impact on the total value creation and opportunities for primary producers in agro-commodity chains (Ramamurthy 2004;Pelupessy and Van Kempen 2005;Allen and Sachs 2007).
Only seven of the selected studies consider possible remedies to address gender inequities in agro-commodity chains.Policy interventions to provide higher returns to (women) producers in the value chain, may not effectively reach them.

Two Analytical Models
Two of the theoretical studies in the sample are indicative for different conceptual frameworks on the role of gender in value chains.First, the gender economic approach of Maertens and Swinnen (2009) identifies two basic economic market mechanisms through which the dynamics of the chain may influence rural households and women.The operations in product and labor markets give women two possibilities of benefiting from modern chains.This is the result of either the sale of cash crops as smallholders to trading and processing firms, or the employment as workers by big commercial estates.Examples are the cut flower, horticulture, coffee and cacao chains in the region (Maertens et al. 2012;Bolwig and Odeke 2007;Mayoux and Mackie 2007;Dolan, Opondo and Smith 2002).This is mainly based on field research in Sub-Sahara Africa, which provides a comprehensive view on women's possibilities for getting involved in and taking advantage of export supply chains with positive effects for their households.
These outcomes contradict other cases of women's participation in global agro-commodity chains, which are linked to poverty, exclusion and worsening living and working conditions (Barrientos et al. 2003;Quisumbing and Pandolfelli 2010;Ramohan and Sundaresan 2003).Mixed or conditional results for women's well-being are reported in Cabezas et al. (2007), Dolan and Sutherland (2003), Barrientos and Smith (2007) and Greenberg et al. (2012).
Global chain dynamics have far-reaching social consequences which influence gender relationships (Palpacuer 2008;Bair 2010).Theoretical reasoning on how these relationships and the status of women and households affect and are affected by global production and trade processes can be found in Dunaway (2001;201 l;2013).She elaborated on earlier conceptual work of Mies (1986;2007), arguing that low wages, flexible and unpaid (family) labor and reproduction tasks of women are in fact the foundations of modern agro-export chains (see also Barrientos 2001).The persistent presence of semi-proletarianized households is another subsidy to the globalized system of surplus generation (Dunaway 2013).The shift of production costs to rural households in the third world and disruption of their structure are other mechanisms that affect women negatively (Dunaway 201 l;2014).
A second model appears in Ramamurthy (2004) who used a feminist commodity chain analysis to understand how global production generates individual and collective identities and why other non-economic aspects are essential in the process of capital accumulation.Ramamurthy (2010) generalized the argument by relating the feminization of (agricultural) labor to the reconfiguration of global and domestic capital, considering a major role for culture and social reproduction.See also Deere (2005) for a good overview of the feminization of agriculture in Latin America under neo-liberal restructuring of the economies.According to Ramamurthy, households and the way they produce and demand commodities become important for the international organization of production.Women's labor should therefore be taken into account, whether under local and subsistence conditions or within international commercialized production and consumption frameworks.Because women are involved in different stages of agro-commodity chains, this feminist analysis will be supportive for development policies (Chant 2008).About a quarter of the selected studies had added cultural, social, intra-household and other aspects to the economic ones (Rammohan and Sundaresan 2003;Allen and Sachs 2007;Quisumbing and Pandolfelli 201 O;Barrientos et al. 2003;Bolwig and Odeke 2007;Ramamurthy 2010;2011 ).Ramamurthy gives examples of the decisive weight of cultural factors above economic ones in South India for the case of hybridization of genetically manipulated cottonseed.She observed the continuation of caste-based smallholding in this global chain, despite the presence of more rewarding wage labor alternatives.The general framework in Allen and Sachs (2007) with the intersection of the material, socio-cultural and corporeal to explain gender relationships in the food system goes in the same direction.
Other studies have turned their focus on participation and empowerment of the poor in agrocommodity chains, with insufficient attention on gender specification in their discussions (Blandon et al. 2009;Ponte 2008a;Laven 2009).Here the question is how poor men and women can strengthen their abilities and capacities to manage parts of a chain or chains in which they are involved.Membership in farmer organizations may have importance, but variables such as improved competences, resources, and technology can also count.
Most efforts to integrate the concept of gender in commodity chain studies resemble that of Maertens and Swinnen (2009).They focus on the economic outcomes of product and labor markets for the directly participating women, following the outstanding empirical research of Boserup (1970).Even the manuals for applications and evaluations of gender mainstreaming in the chain approach prioritize economic outcomes and other market related issues (Riisgaard 2010;Mayoux and Mackie 2007;Nichols, Manfre and Rubin 2009).Several studies address the conditions of marginalized workers in horticulture export sectors from a gender perspective and try to establish their impact in developing countries (Dolan and Sutherland 2003;Barrientos et.al. 2005).The discussion looks at how women are disadvantaged in rural labor markets and identifies cultural, social and religious norms as some of the reasons that prevent them from taking the benefits of off-farm opportunities and work outside the home and family farms.Barrientos (2001) confirms that a complete comprehension of the dynamics of global buyerdriven value chains needs a gender approach.This can be useful for designing value chain interventions, and changes in reporting, monitoring and evaluation (Riisgaard, Escobar and Ponte 2010).Powerful agents tend to benefit from unequal gender relations in and outside (flexible) labor and produce markets.Dunaway (2011) has provided a very useful conceptual framework to explain this.Barrientos et al. (2003) elaborated a method to assess gender sensitivity of codes of conduct in horticulture, where participation in market and household care activities are considered simultaneously.Increasing participation of women in value chains will also influence household gender relations.A general theoretical macro framework in a post Keynesian vein is provided in Todorova (2006).Nevertheless, it is the theoretical work of Ramamurthy (2004;2010;2011) in India that opened up and explored the possibilities that non-economic sociocultural factors may best account for gendered behaviour in global commodity chains.This could be a breakthrough for the analysis of gender in chains emanating from peripheral countries or reg10ns.

Gender in Agro-Commodity Chains
To understand the benefits, contributions, burdens and participation of both men and women in commodity chains, one should consider the changes in the gender division of labor and societal arrangements and how these have particularly impacted the activities streaming from primary production processes through to final consumption (Deere 2005).A gendered analysis will be useful to explain the interaction between modern global processes of production and trade and traditional gendered social relations that varies with geographic location (Carr 2008).
Women have been involved in the production and trade of both traditional and modern crops, and are sustaining many of the export commodity chains at considerable labor disadvantages, risks and insecurities (Cabezas et al. 2007).Trade liberalization has created (underpaid and unpaid) jobs for millions of women especially in the labor-intensive stages of commodity chains.Mainstream analyses have neglected this burden and few have considered its significance for the contribution of women to the generation of surplus.
For Sub-Sahara Africa, gendered studies of value chains have been carried out for modern cash crops.These crops seem to support gender equality mainly because of wage payments in the labor markets and less by the increasing contract-farming earnings of smallholders (Maertens andSwinnen 2009 andMaertens, Minten andSwinnen 2012).A literature review on contract farming in Sub-Sahara countries by Schneider and Gugerty (2010) attributed its disadvantage to women's restricted access to land and other productive means.However, there may be possibilities for empowering women in modern commodity chains.Demands in this sector for both skilled and unskilled labor, as well as wage and non-wage family workers may enhance the empowerment of women in the cases of Sub-Sahara Africa and Mesoamerica (Farnworth and Hutchings 2009;Lyon et al. 2010).Business codes of conduct do not address employment conditions, of women as the majority of the workforce in the Kenyan and Zambian cut flower, South African fruit and Zambian vegetables chains (Barrientos et al. 2003).
Generally a focus on labor and product markets will lead to positive evaluations of women's participation, which can be observed by the adherents of the pro-poor value chains approach.However, there are also researchers like Dunaway who are sceptical about these outcomes.Those who concentrate on cultural impacts, like Ramamurthy and others mentioned in this article, have mostly a negative assessment.But there is also the critical cultural study of Rammohan and Sundaresan (2003) which considers the labor of women in coir chains with technological innovation in Kerala (South India) as a breakthrough in regard to caste restrictions.This may be valid for the few women with office jobs, but the price may be too high for most women in the chain with all-day heavy manual work.
In what follows we discuss two important drawbacks found in the literature in regard to gender equity.One corresponds to women's participation in labor markets and the other, in upgrading practices in product markets of the chain.

Discrimination in the Labor Market
According to Maertens and Swinnen (2009), feminization and gender discrimination in labor markets have been analysed for urban areas and manufacturing sectors, while knowledge on rural areas is very limited.Yet rural women face many challenges that directly impact their ability to enjoy the benefits of their work in commodity chains.
There are considerable risks for women compared to men in the agro-export commodity chains in developing countries.Women, whether working as waged or non-waged laborers, have experienced disadvantages ranging from more casual work, less pay and worse working conditions than men to physical and sexual exploitation (Greenberg et al. 2012).In global chains sexual harassment is a common abuse of power.Women faced with this kind of exploitation may prefer not to take up these jobs, but are often obliged to accept them as a consequence of neoliberal restructuring strategies (Pyle 2001;Deere 2005).
Different studies have shown that women often work at the bottom of fruit, flower and horticulture chains as pickers, sorters, graders and packers while higher paid personnel such as supervisors and technicians tend to be men (Oxfam 2004;English 2007).The common explanation is that women are perceived as low-skilled with limited training and know-how.Downstream consumers and laborers at core nodes may thus be seen as exploiting upstream households and women (Dunaway 2001 ).Palpacuer (2008) also observed that upstream workers are typically the most deprived social groups, and millions of young women execute manual work under oppressive and precarious conditions.Hence, women tend to be more disadvantaged than men in labor markets of both low and high quality wine chains from South Africa (Greenberg et al. 2012).
The trend of profitable organic production has had mixed effects on women.A lack of capital and other factors of production have limited women's production of organic commodities to small scales, mainly for home consumption rather than for the market.The new global drive to produce commercial organic foods in developing countries has created new opportunities for women workers.Women carry out most of the additional farming and processing work needed to meet the stricter organic certification, quality and farm management requirements of the organic exporter.This has increased the total burden on women and reduced the time available for earning other incomes (Gibbon et al. 2008).Nevertheless, household food security may still be improved by women's integration in organic export value chains, providing additional cash income to buy food from the market.Drawing from the research on Uganda's coffee production, Mayoux (2009) added that in spite of the big share of women's work in organic and fair trade production; men have the control of its income.The corresponding codes of conduct could not neutralize these effects because they are lacking sufficient gender sensitivity (Barrientos et al. 2003).Patriarchy not only gives men control of resources, but it also often allows them to control women's labor.A gendered analysis should systematically identify the challenges and opportunities for women to gain more from their work.

Upgrading of Women as Producers
Even when producing for the market, rural women may not be able to get a larger share of the surplus they produce due to social and economic discrimination.Upgrading has been identified as one of the ways in which poor and marginalized producers-of which often women constitute the majority-can benefit from commodity chains.There are various types of upgrading which include product, process, functional and intersectoral upgrading (Humphrey and Schmitz 2001 ).Value chain interventions aim to improve the contribution to national economic growth, but also to reduce poverty by assuring that workers and underprivileged producers are the beneficiaries.Through this, small farmers in developing countries may get an opportunity to strengthen their business and become competitive in the global market.
However, power relations play an important role in determining what opportunities are available for upgrading and to whom, which may be detrimental for women in developing countries (Farnworth and Hutchings 2009;Mayoux and Mackie 2007).They may have fewer opportunities for upgrading options due to their positions in both the household and business.Even where governments or others outside the chain, or lead firms within, support upgrading, women may have limited possibilities to access this support due to socio-economic barriers like the lack of capital and credit collateral (Quisumbing and Pandolfelli 2010).This also means that the results from the upgrading process will often be unequally distributed.Women are likely to receive little or nothing in conditions where their contribution has minor or no recognition at all.This may result in a search for other income generating activities of small scale and low income yields.Women's limited access to credit and other production means makes their conditions worse since they may not be in position to raise enough resources for even very small-scale businesses.In agro-commodity chains, this means that women may not be able to opt for the modern high-yielding varieties with more commercial opportunities and instead must concentrate on the traditional less profitable crops.
In conclusion, women's participation as farmers or laborers in agro-commodity chains remains a challenge.They are frequently stuck at the bottom of labor markets working in precarious and low-paying positions, often facing poor working conditions and overwork in addition to sexual abuse.Amidst these challenges, women must also negotiate their care work within the households.They may not be in a strong enough position to share the care-giving responsibilities and may end up continuing to do most of the unpaid housework.

The Analytical Gaps
The literature that explicitly documents the impacts of commodity chains on gender or women's empowerment has some analytical gaps.The selected studies could be classified as either gender economic, feminist commodity chain, or a mix of these approaches.Most of the works have attempted to consider the participation of women in the chains, but they frequently fail to address whether this will lead to a durable form of empowerment that allows women to make meaningful decisions and socio-economic choices.A central objective of many studies has been the reduction or elimination of gender inequality and poverty.Nevertheless, effects on decisionmaking and differentiation among women should also be taken into account.For these reasons commodity chain research should not only focus on the material aspects of the product itself, but also on the socio-cultural conditions under which global production and trade are taking place.Both gender economic and feminist commodity chain analyses have useful elements to advance this goal.

Gender Inequality
A variety of value chain (mini-) manuals with a specific gender focus exists.Some have developed action-oriented methods such as the Gender Action Learning System (GALS), the Integrating Gender Issues in Agricultural Value Chains (INGIA-VC) and the Gender Equitable Value Chain Action Learning (GEVCAL) (see respectively Nichols et al. 2009;Mayoux 2009 andMayoux andMackie 2007).These manuals aim to create opportunities for equity and empowerment of women in agro-commodity chains.It is assumed that these goals may be obtained as a win-win outcome of value chains for both men and women (Riisgaard et al. 2010).The application of e.g.GALS should catalyze a process of community-led change in which both men and women benefit from participating in planning aimed at improving living conditions.This method was applied on 1500 growers in the Uganda coffee value chain by the elaboration of specific business and marketing plans from which women could gain more support from men in the community (Mayoux 2009).However, this approach is more an operational device than a coherent analytical or conceptual methodology.More empirical studies are needed to substantiate their practical effectiveness in reducing gender inequality and providing long-run positive outcomes.A majority of the existing guides for promoting gender equity are oriented towards the generation of economic benefits for women, while scarce attention is given to sociocultural factors.
Some researchers have made distinctions between men's and women's production or cropping systems in agricultural chains following the seminal work of Boserup (1970) on male and female farming systems.For example, Golan and Lay (2008) noted that female-owned and managed plots of coffee in Uganda exhibit substantially lower yields because they are less intensively farmed.For cash crops as cacao in Ghana, lack of access to modern agricultural inputs, equipment and credits are often the underlying reasons for lower yields of women's plots (Vargas and Vigneri 2011).Women's farms are hoe-and hand-oriented and their small size prevents the use of tractors even when there is such a possibility.The small size of the farm may be attributed to cultural and legal frameworks in Sub-Sahara Africa, which have for so long restricted women's rights to own large extensions of land (Joireman 2008;Daley 2010).Hence women use the small plots of land around the homestead to grow crops that men would otherwise not produce, due to the lack of profitability or their status as women's crops.Because of low productivity of their farms, U gandese women usually produce for domestic consumption and men produce for the market in Uganda.Moreover, where women's subsistence production has become part of commercial chains as for example exportable fruits and vegetables, their traditional control mechanisms on the land may be lost by the formalization of (Western style) property rights (Howard and Nabanoga 2007;Meinzen and Mwangi 2008).
Studies on gender in agro-commodity chains have often fallen short of explicit consideration of the different adversities for the poor.The participation of poor women may not only affect their income opportunities, but also expose them to risks.Poor farmers sometimes referred to as small farmers in the literature, are often not taken into account in the high value "supermarket driven horticulture marketing and trade" (Minten, Randrianarison and Swinnen 2005).Frequently they are not able to comply with the requirements of contract agriculture or the terms of preferred supplier of big retailers in these chains.
Expenditure and income distribution in agro-commodity chains are other areas where relatively little attention has been put in gender research focused on decision-making capacities in value chains.This aspect would involve asking questions like who earns what from the sales of agro-commodities at the household level and why?Who spends this income and on what?And does the income distribution and spending bring any indication of equality or family wellbeing improvement?Together with the access to land and other productive resources, income and wealth differences are important constraints for equitable gendered decision-making (Nichols et al. 2009;Vargas and Vigneri 2011) All these issues need to be understood in order to make meaningful conclusions about whether participation in agro-commodity chains has positive or negative impacts on gender relations and family welfare.The small number of publications available on these issues indicates that women exert little or no control on the revenues from export crops, due to the ruling customs and practices in the household and the community.Men make independent decisions regarding the expenditure of the money raised from the sale of the agro-commodities.A study carried out in the Kapchorwa district (Uganda) indicated, "the husband buys what he thinks is needed in the household and spends the rest as he wants" (Bolwig and Odeke 2007:37).Spending for food is not a typical priority, and women often have to quarrel for money for food.The explanation for this is also rooted in the traditional norms and habits that only women have the responsibility of care giving in families where it's her duty to make sure that all the needs, including food, are met.Social and cultural barriers are frequently underlying the economic ones that curtail the decision-making capacities of rural women in Sub-Saharan Africa.

Differentiation of Women
Most post-2000 literature still focuses on women as one category and hence makes conclusions about them as a single homogeneous group.Women are viewed as raw material producers who don't benefit fully from commercialized agro-production.This also means that women are generally engaged in food production at or near home and when they move out to produce for the market, the benefits frequently do not accrue to them.However, women as a group are not homogeneous and hence treating them as a single category has overshadowed their different contributions, abilities, struggles and challenges (McCune 2006with comments on Boserup 1970).Female farmers may have problems with access to land, capital, credit and knowledge, but women farm laborers may be even more disadvantaged.This is quite different from the conclusions reached by Maertens and Swinnen (2009), who have concentrated on monetary outcomes of the markets for either smallholders or farm workers.Therefore, there is a need for further research to look at the different categories of women all through the various nodes in the chain so that correct conclusions can be made for each category (Carr 2008).This will improve the identification of women's diverse contributions in agro-commodity chains.
Research has not paid sufficient attention to the impact of increased women's participation in agro-commodity chains on household food production and nutrition.According to the study done by Lastarria (2006), the findings from South Africa indicated that female farmers' increased participation in commercial activities reduced their ability to provide enough food for their households.As more women get involved in agro-commodity chains this may impact the nutritional levels of household members negatively, since women are the traditional food providers of the family.However, as mentioned before, more money income may also give them the means to buy food from the market.
The existing literature on gender in agro-commodity chains lacks analytical treatment of inequalities in traditional low-value chains, decision-making on household expenditures, differences among women, and the impact of women's engagement in commercial agriculture on household nutrition.Bhargava (1997) modelled nutritional status and time allocation at subsistence level of households in Rwanda.Traditional chains are more frequent in Africa and cultural differentiations are playing a bigger role in these cases.More research needs to be done to show how women could increase their empowerment in agro-commodity chains, especially in developing countries where the traditional and current conditions differ considerably from the developed world.
To fill the existing gaps of the two main gender economic and feminist commodity chain approaches and improve their explanatory power, two additional methodological components will be proposed.It is not enough to give only a description of the material and cultural conditions and consequences of the operations of women in the different nodes of a commodity chain.Further integration of gender and commodity chain approaches is required to reach a more complete view on their development.This will enhance the gender sensitiveness of the tools to study the dynamics of global agro-commodity chains in Sub-Sahara Africa.It improves the possibility to include the mutual interaction of gender and commodity chain aspects in the analysis.Frequently the impacts of chain dynamics on gender relations are considered in a unilateral way only, while feedback effects may even be more important.A first addition is the consideration of gender bias in the interaction between chain governance and institutional embeddedness; and second is the integration of intra-household dynamics in chain analytics.The interaction may affect gender inequalities and differences among women in the chain, while household relations have considerable effects on the decisions of both producers and ultimate consumers in the different nodes of the chains.Both the referred interaction and household dynamics are also hardly discussed in mainstream commodity chain studies.

Chain Governance and Institutional Framework
In every chain-whether industrial or agrarian-the question of who holds the power and decision making authority derives from the nature of its governance structure.Governance is a process by which lead firms or lead forces (frequently big corporations) control and organize the division of labor, resources and allocation of surpluses and profits along a value chain (Gereffi et al. 1994;Bolwig et al. 2008;Economy & Society 2008).To ensure particular product specifications and the exchange of information, lead firms will frequently apply coordination mechanisms, which are often employed between suppliers in developing countries and buyers in developed countries (Muradian and Pelupessy 2005).
Many general chain studies have been focused on governance because it delineates the prospects of firms and farms in developing countries to engage in global production, trade and the generation of value along the chain.The partition of costs and benefits should be studied from a gendered point of view.Because this refers to socially constructed roles and behaviours, it will be necessary to consider the social-political and institutional embeddedness of the chain, which frequently will be gender related.The interaction among governance structure, power relations and institutional context has been rarely discussed.An exception is Quark (2014), who analyzes this at a macro level.However, for gender bias, the micro outcomes are relevant.Even the critical gender economic and feminist commodity chain studies have given scarce attention to the role and impact of the chain governance and the interaction with the institutional context.The presence of female CEOs in lead firms is as unrelated to gender equity chain outcomes, as the elections of the prime ministers Sheikh Hasina in the national context of Bangladesh and Margareth Thatcher in the UK.Lead firms might continue to utilize society and household traditions of women subordination to benefit from casual and cheap labor.
Most of the commercial agro-commodity chains of today are or tend to be buyer-driven, while the traditional ones may sometimes be producer-driven.In the first mentioned, downstream's agents or lead firms concentrate the authority and power relationships that can affect the gender position of participants.
The institutional framework includes the policies, arrangements and regulations of governments and other agents in the society outside the chain that may affect its outcomes.The state and non-state institutions aim at regulating or deregulating production, distribution and consumption at local, national and supra-national levels.Examples are the European Union and globally the World Trade Organization, international environmental NGOs and the International Union of Food Workers (Muradian and Pelupessy 2005;Pelupessy 2007).These formal rules can all have intentional or unintentional gender effects, which could be positive or negative for equality.The framework also includes informal traditional, religious or customary laws and rules, which may be gender-biased.Traditions, customary law and (world) religions may hamper the access of women to property rights, capital ownership, and employment and stimulate gender inequitable attitudes in general (Joireman 2008;Seguino 2010).Nevertheless there are also experiences where traditions are useful to reduce the disparities of formal regulations, like the cases of witchcraft in Kenyan contract horticulture (Dolan 2002), the customary gender rights on trees and crops of the Buganda in Uganda (Howard and Nabanoga 2007) and the historical women dominance of shea nut chains in Ghana (Wardell and Fold 2013).
Therefore commodity chains should not be seen as closed systems, but as embedded in larger socio-political, institutional and household frameworks, as also brought forward by Ramamurthy.Chain participants receive external inputs as knowledge and information from ministries and research institutes, and influences by pressure groups like trade unions and NGOs with environmental and social concerns.Formal and informal social-cultural rules of the game as established by family, community and class relationships, should be given due consideration (Carr 2008).These institutional aspects are important because they either provide or block access to upgrading opportunities.The impact of lead firms, which determines the final generation and distribution of income throughout the chain, will only become effective through the interaction with the institutional context.This may bring about socially progressive results for the developing country concerned, but could also have a regressive character.In agro-commodity chains the introduction of environmentally friendly regulations belongs to the first category, while that of discriminatory labor rules is of the second type.
Gender inequity is one of the social concerns that chain intervention must address.Many studies on commodity chains have not been able to identify how women and men have been affected or impacted differently by its governance.The socio-political surrounding provides opportunities to participate successfully in commodity chains, through policies that are gender sensitive.
In a study on agrarian reform in the Philippine banana chain, it became clear that the governance by a few multinational trading-companies obstructed the upgrading opportunities of (female) smallholders (De los Reyes and Pelupessy 2009).This outcome gives policy makers the understanding of the intervention needed to improve the situation.
As mentioned, there are many formal and informal institutional factors that deprive women of a decent share in the possible beneficial effects of the actions of lead firms and outside institutions of in agro-commodity chains.
Standardization and certification are part of the commodity chain coordination mechanisms used to establish technical characteristics of a product, specific processing and producing methods, quality traits and safety needs.These may be the results of coordination with actors within or from outside the chain, which could be mandatory, voluntary or private.Whichever form it takes, many public and private regulatory bodies do not explicitly consider gender discrimination (Barrientos et al. 2003;Luetchford 2008).This has been shown by research on codes of conduct where women have been left out of organic export schemes in Africa (Bolwig et al. 2008).Standardization favours more men than women in developing countries, despite the ILO Convention 111, which prohibits gender discrimination in generic standards (Lyon et al. 2010).Men are the ones who benefit directly from training and price premiums given to those meeting the new conditions.
In this section we have shown how both chain governance actions and institutional embeddedness might be gender-biased at macro and micro levels.This implies that special attention must be given to the outcome of this interaction for women.In a gendered commodity chain approach it is necessary to examine if and why control and coordination mechanisms of lead firms enhance the gender equality or inequality characteristics of communities rules in which they are embedded.

Intra-Household Dynamics
The definition of a household as income-pooling unit for reproduction is not adequate for many developing countries (Dunaway 2001).Households are defined here beyond income and reproduction, while other social and cultural aspects also count.It is better to consider households simply as resources pooling units for all purposes, which include all contributions that are made by both men and women.These cannot be measured in monetary terms alone and serve for both their material and immaterial well-being.Men and women are not just producers and consumers in commodity chains.They are units of labor supply, production, decisionmaking, resource allocation, and act sometimes as small businesses that produce commodities for the market.Dolan and Sutherland (2003) note that local dynamics of household and family arrange the participation in the horticulture value chain of Kenya.Semi-proletarian households cover a multi-layered invisible economy that warrants the emergence and development of global commodity chains (Mies 2007;Dunaway 20 II).The norms and rules of households determine women's and men's time availability and capacity to negotiate in both paid and unpaid work, to control income from this work and decide if the employment benefits them.All these different roles are key in the agro-commodity chains dynamics in African countries where production, resource allocation and decision making for markets are made at the household level.However, there has been little research on the interactions between agro-commodity chains and household dynamics, specifically on the pivotal role of females.Sometimes households appear in chain studies, but basically to indicate the economic effects of chain dynamics, without explicit consideration of their feedback to the different nodes (Bolwig and Odeke 2007;Maertens and Swinnen 2009;Vargas and Vigneri 20 II).This is an important drawback since all nodes of agrofood chains are embedded in household dynamics.
In this section we consider the role of the gender power distribution, the influence of household dynamics on chains and the effects of the allocation of chores and care duties.Women are not only often excluded by decision makers, but gender inequalities may in fact increase with policy interventions, further exacerbating gender disparities in power relations (Mayoux 2009).To understand how the impacts of chains are transmitted into the lives of each household member, a first important question is whether there are gender biases in farm-firm contract agriculture or rural labor markets (see above).Frequently the impact of households as a key institution in the global organization of production has not been recognized (Dunaway 200 I;201 l;2014).Intra-household processes should also be taken into account to assess the gendered impact of trade reforms in global commodity chains.
A unitary household model has often been used to study its dynamics, which assumes that households behave homogenously, maximizing a single uniform welfare function.Within the household all resources (land, labor, and capital) and consequently, all production and incomes from factor applications are pooled (Golan and Lay 2008).Nevertheless, in most developing countries, women may pool labor, but are often lacking access to land and capital, which restricts their participation in commodity chains.Therefore, most decisions regarding the production and the distribution of yields are left to the male head of the household, hence making it difficult for women to share the benefits proportionally.
N juki et al. (2006) found that in Uganda and Malawi women are in charge of the purchase, slaughter and sale of small livestock and its revenues.However, they are mostly excluded from decisions about the family's land, larger livestock, commercial crops (except for home grown vegetables sold in local markets), and extractive products such as timber and bricks.For commercial products, men as assumed head of the household might take over many production and expenditure decisions (Mayoux 2009).In many developing countries this kind of household relationship literally means that women are left out of the commodity chains of modern commercial crops.Men frequently allocate land and other means of production to high incomegenerating commodities sold in official markets.However, the experience in Sub-Sahara Africa is in some cases quite different, because women do almost all the cultivation work and sometimes even make decisions about what to produce and how to trade, as in the cases of groundnuts in Cameroon and shea nuts in Ghana (Elad and Houston 2002;Wardell and Fold 2013).All productions and trade nodes of subsistence and commercial shea chains of Ghana and other SSA countries have been operated and governed by women since pre-colonial times.However, modernisation policies and new technologies may change this, with negative effects on women's livelihoods and environments.
The historically female and ecologically friendly managed shea chains in 19 SSA countries are operated by more than two million women.Their gender equity potential has not drawn much attention in the feminist literature, which can be attributed to a generalized development stages vision as e.g.outlined in Boserup (1996).The bargaining power in the household depends more on cultural, relational and opportunistic factors rather than only the economically rational ones, as manifested by research in Eastern Uganda (Iversen et al. 2011 ).The importance of cultural factors to understand intra-household processes had also been shown for other continents (see Ramamurthy 2010 for India; for Chile by Cuesta 2006 andContreras andPlaza 2010).
Decision-making at the household level determines the level and conditions of participation of members of the household in commodity chains as shown for Rwanda (Vargas and Vigneri 2011).In most cases, men make the decisions regarding these questions.In Mesoamerica there is evidence that despite the fact that women devote a large share of their labor to coffee production, they often have little say in agricultural decision-making (Lyon et al. 2010).This leads to unequal sharing of resources at household level.Women and girls typically contribute more labor to household survival than men, but receive an unequal part of the total resources (Dunaway 2001;Vargas and Vigneri 2011).In this way polygamy in Sub-Sahara Africa is considered as only economically profitable to men by the prevailing opinion following Boserup (1970).However, some African anthropologists argue that the system provides greater flexibility and autonomy in decision making to the wives, living in separate houses, which give them advantages (McCune 2006).
It will be clear that household dynamics should be put into consideration when designing intervention programs to include men and women in commodity chains.When the impacts of power relations and balances in the household are not taken into account, the results of interventions may differ from the desired goals.Ignoring these dynamics will lead to negative outcomes for women.An example is an aquaculture program in Bangladesh where intrahousehold dynamics were disregarded.As a result, women were minimally involved, and the project designed to benefit them was never accomplished (Riisgaard et al. 2010).Understanding intra-household dynamics in commodity chains would prevent interventions from creating or enhancing unequal opportunities in male-dominated societies.
Women's empowerment within households is one of the means to ensure that they can fairly benefit from agro-commodity chains.The empowerment in an organic food-growing project in Uganda increased the agricultural food output and improved the rural levels of nutrition (Bolwig and Odeke 2007).This had a direct effect on decision making within households, which includes the allocation of time and food, as shown for Rwanda (Bhargave 1997).The added decisionmaking power in turn increases the self-esteem and enhances the status of women within the family and community.This means that women's empowerment in agro-commodity chains would lead to improved status within the family and vice versa.This is also assumed to provide women with the necessary abilities to make meaningful choices about whether to produce for the market or the household.
Understanding care responsibilities is another important feature in the analysis of household dynamics related to agro-commodity chains.These responsibilities are in most cases attached to women within households; hence obligations and performances of chores and care work are experienced differently by men and women (Lewis and Giullari 2005).In most households, and especially in poor ones, women do all the care activities, and this has an impact on their available time, potential participation, nutrition, productivity and ability to upgrade in chains.According to Tarkowska (2002), women in Poland traditionally do most of the domestic and parenting work, even when they are more involved in out of home paid work.This is attributed to social and cultural expectations and traditions.However, these care activities sometimes act as a blockade to participation in those chain nodes that would demand more of their time and energy.The significance of the care economy should be considered in order to understand the ways in which women could get more involved in agro-commodity chains.Especially in semi-proletarian households, women's work is not remunerated or valued otherwise.Little or no recognition makes it difficult for them to improve their position and enjoy the monetary and other benefits that come with it.This unrecognized work effectively subsidizes other non-agricultural nodes within many agro-commodity chains (Dunaway 2014).Numerous unpaid or underpaid family workers in the harvest of fresh fruit and vegetables as well as women's household work, make low export prices feasible.In sum, a gender perspective is necessary to analyze the position of households in commodity chains in the attempt to alleviate poverty.This is because women tend to be over-represented in poor groups, have greater domestic responsibilities than men, are often underpaid and tend to direct their earnings towards improving household welfare (Carr 2004;Simister 2009).The outcomes of intra-household bargaining will also influence the participation potential of women and men in agro-commodity chains.

Conclusion
The identification of winners and losers of globalization processes is one of the greatest virtues of the global commodity chain approach.Therefore it can be a good instrument to assess the impacts of these processes on gender equity.The fact that agro-commodity chains give women (potential) access to labor and produce markets is for many authors reason to qualify them as winners.But others may emphasize its adverse conditions and the meagre results for empowerment.The consideration of socio-cultural aspects may lead to worse assessments of commodity chain impacts.Underprivileged groups themselves could provide a more complete specification of the assessment criteria.
However, winners may also be found in the non-agricultural and other non labor-intensive stages of the chains.The lack of an appropriate gender approach has been a major shortcoming in many analyses of agricultural based chains.This is important to explain both social outcomes and gender dependence of chain dynamics.Our study provides insights into steps to be taken towards an integrated gendered agro-commodity chain approach.Our analysis of a selection of representative post-2000 studies on gender in value chains and earlier references indicate that despite significant advances, there are still important analytical gaps to reach this goal.Most agro-commodity chain studies that included gender analytics explicitly have referred to high value commercial crops concentrated on horticulture and some traditional export commodities.Few have explored the chains of traditional less commercial staples like maize and millet, which are of great importance to the nutritional conditions of rural households, especially in Sub-Sahara Africa.In the tradition of Boserup and others, about three-fourths of our sample has focused exclusively on economic effects of chain dynamics on the position of women in product and labor markets.The latter seems to provide more benefits to women, but other empirical studies question this outcome.Only one-fourth of the publications extended the economic reasoning with socio-cultural aspects looking at the well-being of women in a broader sense.In the post-2000 period the market-oriented gender economic and socio-cultural feminist commodity chain approaches were developed based on earlier works on the role of women in global production.A necessary first step to reduce the analytical gaps would be the combination of the main aspects of both approaches to assess its explanatory potential.However, only a limited number of publications has included the feedback effects of gender on chain dynamics, mostly as consequences of feminization of the agriculture node, labor force and poverty evolution.
Governance in commodity chains has been viewed by many as a concept of power and authority, concentrated on lead firms and subordinated producers without explicit consideration of the gender dimension.The discussion of institutions has centred on the contextual impact on firms in global chains and regulation of the movement of commodities.Gender relations in the socio-political and institutional realm of commodity chains remain often an unknown feature.There is a need for a more explicit gender-oriented analysis in this field, more specifically in developing countries where gender disparities are still posing many development challenges.This may be a result of continued beliefs in and strong attachment to traditional values in communities and within households.Nevertheless, the modernization of political and economic structures will not always bring improvements to this situation.The insertion in value chains may reduce women's traditional rights of access to resources.It can also diminish the opportunities to make meaningful decisions and take advantage of the benefits from markets or other trading channels.Our review of the research showed that the development of modern agrocommodity chains have mixed effects on gender equity.
The interaction of governance structures and institutional embeddedness may have important consequences for the gender division of labor, leading not only to the feminization of agriculture but also that of other labor intensive nodes in the value chain.The consideration of gender biases in the interaction is a second step to integrate gender and agro-commodity chain approaches.The integration of intra-household dynamics in commodity chain analysis is the third step to be taken in the achievement of a final outcome.This is essential to explain the consequences of feedback effects on the dynamics of agro-commodity chains.These effects and the mentioned interaction make equity outcomes uncertain.The addition of the interaction between governance and institutional context, as well as intra-household dynamics can be seen as our main contribution to the design of a gender sensitive agro-commodity approach.
The three steps together will help us better assess the flows of resources, the distribution of costs and benefits, and the impacts of socio-cultural processes along commodity chains.Both men and women have unique knowledge of production, processing, trade and consumption of different agro-commodities.To validate this three-step methodology, comparative empirical research is needed of both traditional and modern chains.A gendered approach is required not only from an equal rights perspective, but also to improve the effectiveness of interventions, taking advantage of the considerable knowledge of both men and women and their respective cultural backgrounds to assure a positive impact on their well-being.