Sustainable Development Goals, Gender Equality and the Distribution of Land in Latin America^*

Abstract

Among the advances in the 2030 Sustainable Development agenda is that the goal to achieve gender equality and empower women has nine specific targets. These focus on many of the root causes of gender inequality, including women’s unequal access to economic resources. This article focuses specifically on women’s unequal access to land and the set of proposed indicators to measure progress. Drawing on the available data for Latin America, it demonstrates the current degree of inequality in the gender distribution of landholders and landowners, and why it is important that countries improve their gender statistics, collecting gender disaggregated data on both land ownership and agricultural decision-making.

Gender Equality; Land Ownership; Agricultural Decision-Making; SDG 5

Resumo

Dentre os avanços contidos na Agenda 2030 para o Desenvolvimento Sustentável, está o fato de que o objetivo de alcançar a igualdade de gênero e empoderar mulheres tem nove metas específicas. Essas têm por foco muitas das causas básicas da desigualdade de gênero, incluindo o acesso desigual das mulheres a recursos econômicos. O foco deste artigo é, especificamente, o acesso desigual das mulheres à terra e o conjunto de indicadores propostos para medir progresso em relação a isso. Com base nos dados disponíveis para a América Latina, o artigo demonstra o nível atual de desigualdade na distribuição de gênero de produtores e proprietários agropecuários e por que é importante que países melhorem suas estatísticas sobre gênero, coletando dados desagregados sobre gênero em relação tanto à propriedade da terra quanto ao processo decisório agrícola.

Igualdade de gênero; Propriedade de terra; Processo decisório agrícola; ODS 5

Introduction

The 2030 Agenda for Sustainable Development launched by the United Nations in September 2015 is generally lauded as a major advance for gender equality and women’s rights, including by those who recognize its limitations. Among its seventeen Sustainable Development Goals is SDG 5, to “achieve gender equality and empower all women and girls”. While SDG 5 is similar in language to its predecessor in this process of international goal setting, Millennium Development Goal 3, the sustainable development agenda goes much beyond in at least two ways. First, SDG 5 includes nine specific targets that focus on many of the root causes of gender inequality – such as the burden of women’s unpaid labor and their unequal access to economic resources and political power. It also focuses on one of the main manifestations of women’s subordination, gender-based violence against them.¹ In contrast, MDG 3 had only one target, eliminating gender disparities in primary and secondary education. Second, in addition to the stand-alone SDG 5, gender equality is mainstreamed across many of the other goals.

Another of the notable features of the SDG framework is that it is grounded in the Universal Declaration of Human Rights – the equal and inalienable rights of all – and has integrated these into the global development agenda, even if not as tightly as might be desired (Razavi, 2016). Rather than focusing only on poverty and deprivation in less developed countries, as did the MDGs, the scope of the SDGs is universal and encompasses economic, social, political and environmental dimensions applicable to both developed and less developed countries.² In addition, the need to reduce inequalities both within and among countries is recognized in SDG 10.

At the same time, the SDG framework does not offer a coherent model of socio-economic transformation. As Esquivel (2016) argues, while it encompasses a broad vision of sustainable development, it is built on a conventional view of growth, expecting GDP growth to fuel social progress. While inclusive growth is required to meet many of its goals, such as SDG 8, which seeks full and productive employment and decent work for all, or SDG 2, aimed at ending hunger and achieving food security, there is no roadmap on how to get there. The policy prescriptions do not go much beyond the need for further industrialization, trade liberalization and public-private partnerships, with a nod to environmental concerns. There is little attention to the role of macroeconomic policy or the need for redistributive policies to assure that growth is inclusive.³

Two other general concerns that have been raised about the SDG framework are the lack of financing to implement the 2030 agenda and its relatively weak accountability mechanisms. The means of financing are laid out in SDG 17, which specifically addresses implementation, and include the usual ways of financing development initiatives (domestic resource mobilization, official development assistance, foreign direct investment, and remittances). The problem is that there are no new sources of targeted funding to support the SDG initiatives.

Moreover, the SDG framework does not require mandatory reporting on the many indicators developed to monitor progress towards the 169 targets. Accountability is thus weak, and is dependent on the good will of governments both to implement the agenda and to submit to a voluntary process of follow-up and review. Of course, the United Nations Statistical Agency, the UN regional economic commissions (such as ECLAC) and other UN agencies carry a lot of weight in setting the data collection agenda.⁴ But their power to actually influence policy in support of the sustainable development goals is relatively weak, particularly without new lines of financing. As most feminist analyses conclude, meeting the SDGs, particularly SDG 5, will depend on the lobbying efforts of women’s rights advocates and their allies at all levels-- local, national and international.

This paper focuses specifically on SDG Target 5.a (one of the three “means of implementation” of SDG 5) and Target 1.4, which is related to the goal of ending poverty. Target 5.a encourages governments to

undertake reforms to give women equal rights to economic resources, as well as access to ownership and control over land and other forms of property, financial services, inheritance and natural resources, in accordance with national norms.

Target 1.4 is similar, but much stronger since it has a timeline and implies equality of outcomes rather than equality of opportunity. Its aim is to:

By 2030, ensure that all men and women, in particular the poor and vulnerable, have equal rights to economic resources, as well as access to basic services, ownership and control over land and other forms of property, inheritance, natural resources, appropriate new technology and financial services, including microfinance.

Our primary interest is in women’s ownership and control over land and in the specific indicators developed to measure progress and compliance.

Attention to the importance of women’s land rights in the international arena is, of course, not new. The need for gender equality in land rights has been formally recognized since the 1979 UN Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) and the Program of Action of FAO’s 1979 World Conference on Agrarian Reform and Rural Development (Deere and León, 2001). Since then, there has been growing recognition of the role that women’s land rights play in increasing their welfare, efficiency, and empowerment, in addition to achieving the intrinsic goal of gender equality, a case made initially by Agarwal (1994) and further developed by other researchers.⁵

What is different now is that specific indicators have been adopted by the UN Statistical Office to track progress on women’s land rights, one of which will require significant changes in the content of agricultural censuses and household surveys, at least in Latin America. Until recently, data on the distribution of land ownership by sex has not been collected in the national agricultural censuses, nor even recommended in the FAO guidelines that orient the decennial censuses; moreover, relatively few surveys collect such data. This has limited research on a whole range of issues related to women’s economic empowerment, such as the relation between land ownership and participation in production decisions and the relation that ownership and control over land have to outcomes more favorable to women. In addition, the lack of baseline indicators on the distribution of land ownership by sex have hampered efforts at the micro level to fully evaluate the efficacy of agricultural interventions and their gendered impact.

In the next section, we discuss the relevant SDG indicators and how these relate to the information typically collected in the agricultural censuses. Subsequently, we consider the extent to which the legal framework in Latin America guarantees women’s land rights, and then what the available information on the region tells us about women’s ownership and control over land. Finally, we discuss why it is important to have information on both women’s ownership of land and their role in agricultural decision-making, and then conclude.

The SDG gender and land rights indicators

There are three indicators which specifically focus on women’s land rights, two of which require improved quantitative data collection. Indicator 5.a.1 recommends that countries collect and publish data on a) the proportion of the total agricultural population with ownership or secure rights over agricultural land, by sex; and b) the share of women among owners or rights-bearers of agricultural land, by type of tenure. Indicator 1.4.2 focuses on the proportion of the total adult population with secure tenure rights to land, with legally recognized documentation and who perceive their rights to land as secure, by sex and by type of tenure. The main difference between the two indicators is that one focuses on the share of the agricultural population and the other, of the national adult population, information usually derived from different sources, the agricultural census in the former case and national household surveys in the latter.⁶ A third indicator, 5.a.2, requires UN agencies to report on the proportion of countries where the legal framework (including customary law) guarantees women’s equal rights to land ownership and/or control.

The language of these indicators highlights the importance of security of tenure, probably in recognition that, depending on context, security of tenure may be conveyed not just through individual private property, but also through collective forms of tenure.⁷ Nonetheless, the inclusion of “control over land” in indicator 5.a.2, as well as in Targets 5a and 1.4, point to the importance of not only security of tenure, but also to that of enhancing women’s role in agricultural decision-making.

One of the criticisms of the SDGs is that insufficient attention is given to the arena of the household even though inequalities are often most palpable at the household level (Esquivel, 2016). For example, Target 5.5 focuses on women’s participation in decision-making only in the public arena, which leads to the traditional indicators of increasing women’s representation in national legislatures, local government and in managerial positions. No mention is made of the importance of enhancing women’s role in household decision-making, a likely precondition for changing the gender division of labor in unpaid care and domestic work (Target 5.4).

Achieving some of the other targets for rural women may also depend upon increasing women’s role in household and farm decision-making. Take Target 2.3, for example, which is to double the agricultural productivity and incomes of small-scale food producers by 2030, particularly those of women, indigenous peoples, family farmers, pastoralists and fishers. This is to be accomplished “through secure and equal access to land, other productive resources and inputs, knowledge, financial services, markets and opportunities for value addition and non-farm employment”. The indicators proposed here are both outcome measures: the volume of production per labor unit by enterprise size (2.3.1) and the average income of small-scale food producers, by sex and indigenous status (2.3.2).

Whether rural women are able to take advantage of and benefit directly from any potential increase in the resources channeled to the small holding agricultural sector will likely depend on a series of intervening variables. These include women’s participation in decision making on farms and their control over any increase in output and income. The indicators of women’s security of land tenure and specifically, land ownership, are thus incomplete, lacking complementary information on farm decision-making by gender.

Among the reasons it is so important to consider the gendered distribution of land is related to the potential bargaining power that land ownership conveys, both within the household and community (Agarwal, 1994; Deere and León, 2001). Women’s land ownership may be a precondition for their participation in agricultural decision-making or in the allocation of household income. Yet, we know relatively little about the relationship between women’s land ownership (and other secure forms of tenure) and farm decision-making. This is partly because the national agricultural censuses have focused solely on agricultural holders, without gathering information about who owns the land or who is actually participating in the different agricultural decisions.

Following FAO guidelines, the information on “agricultural production units” or farms is provided by the agricultural holder, until recently defined as “the civil/juridical person who makes the major decisions regarding resource use and executes management control over the agricultural holding” (FAO, 2005). A long-standing problem is that the concept of agricultural holder is too often conflated with that of the head of household, so that if an adult male is present, he is considered the head and principal farmer, irrespective of whether he manages the farm alone or with someone else, such as his wife. The guidelines for the World Census of Agriculture (WCA) 2010 recognized the traditional gender bias in this concept and proposed that its definition should be expanded so that joint holders, such as a husband and wife, be enumerated if both of them co-manage the family farm. While an important step forward, no Latin American country, with the exception of Colombia, adopted this recommendation in the 2010 census round. Moreover, this recommendation did not go far enough.

The WCA 2010 guidelines were silent on the critically important issue of who owns the land. The guidelines retained the traditional question on land tenure, asking whether the holding is characterized by legal ownership or owner-like possession (i.e., statutory security of tenure) and other forms of tenure. But by not taking this question one step further and eliciting information on who the owners are, all one can deduce from such census information is whether the agricultural holder manages an owner-operated farm, not whether the agricultural holder him or herself is the owner or joint owner of the farm.

Since the 1980s researchers have pointed out this deficiency in the agricultural censuses for purposes of gender analysis (Deere and León, 2001; Doss, 2014). First, ignoring who specifically in the household owns the land means that the key question regarding a gender equitable distribution of productive resources – land ownership – cannot be addressed. Second, this deficiency also hinders evaluations of the impact of policies designed to promote gender equality since a rigorous baseline cannot be established. Third, an important research question related to the goal of food security is whether land owners in the smallholding sector manage their own farms, and if not, why not? It is important to know, for example, whether female landowners are less likely than male owners to manage their own parcels because of discrimination in credit markets or in access to technical assistance or product markets. Such an analysis cannot be carried out solely with the information on the sex of the agricultural holder; it is essential to know who owns the land.

In response to these concerns, the guidelines for the next decennial round of agricultural censuses, WCA 2020, have modified the definition of the agricultural holder, and added a new proposed census theme (#10): the intra-household distribution of managerial decisions and ownership on the holding (FAO, 2017). The agricultural holder may now include a “group of civil persons” from the same (such as husband and wife) or different households.⁸ Moreover, the guidelines recommend that the censuses collect information on the sex of the household member making a series of managerial decisions; the area of crops by sex of the person managing them; and the area of land owned by sex of the owner. With respect to livestock, it also recommends a disaggregated approach so that information is collected on the number of livestock by sex of the person managing them and their respective owners.

The recommendations to allow for joint agricultural holders and to collect sex-disaggregated data on both management and ownership in the WCA 2020 are thus major steps forward in terms of gender analysis. Now much depends on whether the national statistical agencies and agricultural ministries are up to the task. The poor results regarding the adoption of the WCA 2010 recommendations in Latin America suggest that a major lobbying effort will be required in each country. The SDG indicators requiring disaggregated data on land ownership thus constitute an important new arrow in the arsenal of arguments.

Before turning to what we know about women’s ownership and control over land in Latin America from existing censuses and surveys, it is relevant to first consider the legal framework governing women’s land rights and what might be required to implement SDG indicator 5.a.2.

Women’s land rights in Latin America: The legal framework

Target 5a requires governmental action, that is, for countries to undertake reforms to give women equal rights to economic resources, among them, ownership and control of land and other forms of property, and inheritance. Progress towards this target is to be measured by indicator 5.a.2, the proportion of countries where the legal framework (including customary law) guarantees women’s equal rights to land ownership and/or control.

FAO is the lead UN agency responsible for producing this indicator, since its mandate includes promoting women’s land rights and increasing their access to and ownership of land.⁹ The organization has already developed a Legal Assessment Tool (LAT) for gender-equitable land tenure to facilitate implementation of the 2012 Voluntary Guidelines on the Responsible Governance of Tenure of Land, Fisheries and Forests (VGGT). It is this Legal Assessment Tool which provides the building-block for current discussions on how progress on SDG indicator 5.a.2. is to be measured. It is useful to consider a few of the features of the LAT and what it tells us about women’s legal land rights in Latin America.

The LAT includes 30 legal indicators clustered around eight key elements, as follows: ratification of human rights instruments; elimination of gender-based discrimination in the constitution; recognition of women’s legal capacity; gender equality of rights with respect to nationality; gender equality in property rights; gender equality in inheritance; gender-equitable implementation, dispute mechanisms and access to justice; and women’s participation in national and local institutions enforcing land legislation (FAO, 2014). The LAT draws upon the FAO’s Gender Land Rights Database Country Profiles which include information (provided by each country) on constitutions, relevant laws and country policies. The LAT is designed to highlight sources of gender differentiation and measure progress by scoring each country according to its “stage” on a continuum, as follows: a zero ranking on an indicator indicates its absence in the legal framework; 1, that a policy is being developed; 1.5, that a policy is in place; 2, that draft legislation is to be submitted for deliberation; 3, that the indicator appears in primary law; 4, that the indicator appears in multiple legal instruments; and NA, when an indicator is not applicable in that particular country.

To date, the LAT has completed assessments for 18 countries, ten of which are from Latin America. These include Brazil, Chile, Colombia, Ecuador, Guatemala, Mexico, Nicaragua, Panama, Peru and Uruguay.¹⁰ This group of countries shows considerable progress, with the majority demonstrating legal achievements (stages 3 or 4) on 20 out of the 27 indicators which are applicable to the region.

One of the main reasons that Latin American countries do so well overall is because of their inherited legal tradition with respect to the marital and inheritance regimes: full or partial community of property in marriage as well as succession laws that treat children of either sex equally. A LAT indicator of gender equality in property rights (#13) is specifically whether the law recognizes full or partial community of property as the default marital regime. This is in keeping with empirical evidence that the distribution of property between men and women is more equal in practice in countries with this default regime compared to those with a separation of property regime, particularly when combined with gender equitable versus unequitable inheritance regimes (Deere et alii, 2013).

Among the main indicators where the majority of these 10 Latin American countries fall short include women’s participation in national and local institutions that enforce land legislation (LAT indicators 29 and 30); constitutional provisions favoring affirmative actions to advance women (LAT indicator 7); and the LAT indicator of most interest to us in this paper, #16, on whether the legal framework includes special measures to guarantee women’s equal rights to land ownership and/or control. Only three of the 10 countries are found to have such provisions (Colombia, Mexico, and Nicaragua).¹¹

Among those scoring zero on indicator #16 is Brazil. Its 1988 constitution established that in the process of agrarian reform, land was to be allocated to men, women or to couples jointly. While specific implementing legislation was never approved, two important directives were adopted by the ministry implementing the agrarian reform, the first, in 2003 making the joint adjudication of land to couples mandatory in agrarian reform, and the second, in 2007, giving priority to female heads of households in the beneficiary selection process (Butto; Hora, 2008). While we agree with the LAT assessment that legislation is far superior to simply announced policy (which could be easily reversed by a new minister), it seems inconsistent with its own methodology not to acknowledge situations where a policy is in place (i.e., a score = 1.5).¹²

Another related problem is that progress in guaranteeing women’s legal land rights has generally happened in Latin America only in the case of specific land redistribution or land titling programs. This means that the gender-progressive aspects of land legislation are not always applicable to all rural women. For example, in the first case, if their households are not landless or otherwise eligible to become beneficiaries of an agrarian reform, or in the second case, if someone in their household already has a land title.

Bolivia, a country not yet included in the LAT analysis, provides a good model of a legal framework relevant to the majority of rural women. Its 2008 Constitution included two articles to guarantee women’s land rights: one that committed the state to eliminate all forms of discrimination against women in access, tenancy, and inheritance of land; and another that guaranteed women’s access to land in titling and redistribution processes, irrespective of their marital status. To my knowledge, this is the first constitution or legislation that specifically draws attention to land inheritance – the main way in which women obtain land in most Latin American countries (Deere; León, 2001).

Bolivia’s constitutional reform was preceded by an amendment to its agrarian reform legislation, the 2006 Law of Communitarian Renewal, which governs its land redistribution and land titling program, and conforms to the new constitutional principles. In this legislation a concrete mechanism of inclusion of women – the mandatory joint titling of land to couples (whether married or in a consensual union) – was given the force of law.¹³ What has been particularly noteworthy is its application of affirmative action principles in the land titling process. Land is being titled in the name of a couple, irrespective of how the parcel was acquired. Bolivia has historically exhibited a strong male bias in inheritance, thus joint titling of such parcels to a couple is a gender-bias compensatory mechanism, one that over-rides the civil code on succession (Deere, 2017).

The Bolivian case also suggests why it might not be sufficient to focus only on general inheritance legislation (civil codes and family law) in measuring progress towards gender equality (LAT indicators 17 to 22). Gender equitable succession legislation in Bolivia going back several centuries did not result in undoing male preference in land inheritance; rather, specific attention to inheritance of land may be necessary to begin to change engrained social norms and customs.

LAT indicator 16 – whether the legal framework includes special measures to guarantee women’s equal rights to land ownership and/or control – might thus be made more precise by specifying such special measures. The Latin American experience suggests that these should include specific mechanisms of inclusion of women such as: i) joint adjudication and titling of land to couples; ii) priority to female heads in state land initiatives; and iii) the use of compensatory measures to overcome male bias in inheritance.

Finally, a gender-progressive legal framework is a necessary but not sufficient condition for generating gender equality in outcomes with respect to land ownership. Where women’s land rights are addressed only in case of specific land-intervention programs, and these programs are of limited scope, the impact of gender progressive land legislation will also be limited (Deere and León, 2001). Moreover, irrespective of scope, feminist researchers have long lamented the gap between law and practice – and have found that compliance with legal frameworks is often lacking (Deere, 2017). In either case, it is essential to have good data on a country’s distribution of land by gender to monitor progress.

What do we know about women’s ownership and control over land in Latin America?

Table 1 presents the most recent agricultural census information on the distribution of landholders by sex for 17 of the 19 Latin American countries.¹⁴ That census data on the principal agriculturalists is now available for so many countries is certainly an advance, since until the 1980s most census questionnaires did not even ask their sex (Deere; León, 2001). This lacuna reflects cultural norms that assume that only men are the farmers. Once information on the sex of the farmer began to be collected, it was not always processed and published in the official census reports. Moreover, few statistical agencies make use of this information to carry out and publish a gender analysis across all of the pertinent variables.

Table 1
: Distribution of agricultural holders by sex, most recent agricultural censuses for Latin America (%)

Country	Year	Women	Men	Total
Argentina	2002	16.2	83.8	100
Bolivia	2013	25.2	74.7	100
Brazil	2006	12.7	87.3	100
Chile	2007	29.9	70.1	100
Colombia	2014	33.5	66.5	100
Costa Rica	2014	15.6	84.4	100
Dominican Rep.	1998	10.2	89.8	100
Ecuador	2000	25.4	74.6	100
El Salvador	2007	11.5	88.5	100
Guatemala	2003	7.8	92.2	100
Mexico	2007	15.7	84.3	100
Nicaragua	2011	23.3	76.7	100
Paraguay	2008	21.7	78.3	100
Panama	2001	29.3	70.7	100
Peru	2012	30.8	69.2	100
Uruguay	2011	19.7	80.3	100
Venezuela	2008	19.7	80.3	100

• Sources and Notes: FAO Gender and Land Rights Database [www.fao.org/gender-landrights-database/en/ – accessed 22 May 2017]; Bolivia, INE (2015); Costa Rica, INEC (2015, 40); Paraguay, Guereña (2017, Table A3); and for Colombia, Gender Statistics Office, DANE, communication to the author, 24 May 2017. Colombia’s information includes joint holders; the estimate above excludes the 13.8% of landholders whose sex is unknown.

There is considerable heterogeneity across Latin American countries in the share of female landholders, ranging from a low of 8% in Guatemala (in 2003) to almost one-third in the recent censuses in Colombia, Peru, Chile, and Panama. The median for the region is around 20%. Countries with a relatively large share of indigenous population are found on both the high (Peru) and low (Guatemala) end of this range.

Gender disaggregated data over two census periods is only available for seven countries. These show different trends: a significant increase in the share of female agriculturalists in Chile, Nicaragua, Paraguay and Peru, a very modest increase in Guatemala and Uruguay, and a slight decrease in the Dominican Republic.¹⁵ Given the traditional bias in recognizing women as agriculturalists, one might expect the significant increase in the share of female farmers in the former countries to be related to a growth in the share of female-headed households in rural areas.

Take the case of Chile, for example. According to national population censuses, the share of rural female heads increased from 17.2% in 1992 to 21.9% in 2002 (Rueda et alii, 2008:71). But the share of female landholders (30% in 2007) is probably higher than what the share of female rural heads might have been in that year. This can be further investigated, since the agricultural census collected information on the marital status of the landholder: some 54% of the female landholders reported that they were single, widowed, separated or divorced, compared to only 22% of the men; men were thus more likely to be married or cohabitating (88%) than women (46%) (Rueda et alii, 2008, Table 1). Thus, given the relatively large share of female farmers who are partnered, the increase in the share of female landholders in Chile cannot be solely explained by the growth of de jure female headship.

Another gender analysis of the 2007 Chilean agricultural census investigates further the situation of partnered women who declare themselves as the principal farmer and reside on their farms (Namdar-Irani, 2007:147). This study shows that some 82.8% of the partnered women report that their husband currently lives with them, thus only a minority are de facto female heads because their husbands are temporarily away from the homestead. Of the resident husbands, 24.6% do not participate in farm activities at all, 44.9% participate only temporarily, while 30.3% participate on a permanent basis. The first two situations suggest a gender division of labor where the wife is taking primary responsibility for the farm, perhaps because her husband is engaged in off-farm activities. The third situation is the most surprising, since it suggests a gender role reversal, with women declaring themselves to be the landholder even though their partner also dedicates himself permanently to the farm.

Namdar-Irani (2007) raises the pertinent question of whether this latter situation arises because the woman is the landowner, a question that cannot be answered since, in the case of owner-operated farms, the census did not inquire as to who specifically in the household owns the farm. This example also suggests why it is important to provide for the possibility of joint landholders, where both husband and wife are involved in farm decision-making. It may be that both are involved in decision-making but that the husband was not available the day of the census to complete it. Unfortunately, comparable information, on the contribution of the wives of male landholders to farm labor, is not presented in this study.

Analysis of the 2012 Peruvian agricultural census also revealed that a relatively large share of female landholders have a partner (whether married or in a consensual union), 42%, although a lower share than male landholders, 75%. Asked whether their partner participated in agricultural and/or livestock raising activities, the gender difference was minimal: 92.8% of the women and 90.7% of the men landholders reported in the affirmative (INEI, 2014, Table 14.1). Similar results are apparent in the earlier 2000 Ecuadorian agricultural census. In this case, 36% of the women and 82% of the men landholders living on the farm were married and they reported that 81% and 80%, respectively, of their spouses participated in farm decision-making.¹⁶ Thus, it is highly likely that joint management is very common in both of these Andean countries.

Only one agricultural census in the WCA 2010 round – that of Colombia – considered the possibility that two or more people are joint managers of the agricultural and/or livestock activities on their farms. Its 2014 census revealed that 61.4% of the farms are managed by a man, 26% by a woman and 12.6% jointly, by both men and women (DANE, 2016:600). Summing the latter two figures gives us a better idea of women’s participation in farm management: on 38.6% of the farms in Colombia, women are landholders, either alone or jointly managing the farm with others.

Colombia’s main census publication also disaggregates this gender data by ethnic group. It shows that joint management is particularly important in households located in either indigenous territories or Afro-Colombian communities, constituting 26% and 22.1%, respectively, compared to the national figure of 12.6% (DANE, 2016:667).¹⁷ Thus, ignoring joint management may particularly underestimate women’s participation in agriculture, and specifically, that of indigenous and Afro women.

Turning to farm size, women tend to manage smaller farms than men. In Colombia (DANE, 2016), Costa Rica (INEC, 2015), Ecuador (INEC, 2000) and Nicaragua (INIDE, 2012), women are over-represented, compared to men, among those in the smallest farm size category, with access to less than five (or in the case of Nicaragua, seven) hectares. Table 2 presents the average size of holdings in Brazil, Chile, Paraguay and Peru, and confirms this same trend: the landholdings of female managers are always smaller than those of men,¹⁸ and in the case of Brazil, irrespective of the form of land tenure.

Table 2
: Average farm size of landholders (hectares)

Country	Year	Category	Women landholders	Men landholders
Paraguay	2008	Total holding	15.5	24.5
Peru	2012	Total holding	1.8	3.0
Chile	2007	Total holding	38.8	52.8
		Agricultural land	15.9	23.6
Brazil	2006	Total, owner operated	33.2	84.2
		Agrarian reform beneficiaries	24.3	31.3
		Renters	13.9	41.0
		Sharecroppers	8.0	14.4
		Squatters	8.0	16.5

• Sources: Paraguay, derived from Guereña (2017, Table A1); Peru, INEI (2014:18); Chile, Namdar-Irani (2012:147); and Brazil, Nobre (2012:58).

Women landowners and landholders

Owner-operated farms are often assumed to be the property of the landholder or household head. But this is an unsatisfactory assumption since a “family farm” may be owned by the husband, the wife, jointly by both of them, or by one of them with a parent or other relative who may not reside in the household. Moreover, each of the parcels which make up a household’s landholdings may be owned by different persons and may have been acquired in different ways. Thus, it is important for agricultural censuses and surveys to inquire, in the case of owner-operated farms, who specifically in the household (or family) owns each parcel of land.

Table 3 presents estimates for a few Latin American countries where there is data on both the share of female landowners (from surveys) and of female landholders (from the census). Although drawn from different sources and for different years, this comparison is instructive in thinking about what these different sources of information reveal. Note, however, that the survey data on landownership is not strictly comparable across countries since the estimates for El Salvador and Nicaragua refer to only titled parcels, whereas those for the other countries refer to reported plot ownership. Whether the estimates for these two countries are biased upward or downward depends on whether there are gender differences in the likelihood of having a titled parcel.¹⁹ To comply with SDG indicator 1.4.2, surveys will need to gather information on both reported and documented ownership.

Table 3
: Female share of landowners, selected household surveys, and female share of landholders, agricultural censuses

Country	Survey Year	% Women landowners	Census Year	% Women landholders
Ecuador	2010**	53.6	2000	25.4
El Salvador	2005*	14.1	2007	11.5
Mexico	2002**	32.2	2007	15.7
	2015**	25.0
Nicaragua	2005*	19.9	2011	23.3
Paraguay	2000**	29.7	2008	21.7

• Notes: *Refers only to titled land parcels. **Refers to all owned parcels whether these are owned individually or jointly with another person, and whether or not they are titled.
• Source: For landholders, see Table 1. For landowners, see Deere (2011), except for Ecuador 2010 survey, which is drawn from Deere and Contreras (2011) and Mexico 2015, from Gutiérrez (2016).

With the exception of Nicaragua, survey estimates of the share of women landowners tend to exceed census estimates of the share of women landholders. There are a number of reasons why we might expect this to be the case. First, with the exception of El Salvador, these survey estimates include women landowners whether they are individual or joint landowners. Since the census estimates allow for only one person in the household to be designated as the holder, the censuses would tend to undercount women’s participation.

Second, the census data on landholders include farms irrespective of the form of tenure (i.e., titled and untitled owner operated farms, as well as land that is rented, sharecropped, owned by a community, etc.). If women are more likely to be landholders of owner-operated farms than farms in other forms of usufruct (because they are less likely to rent or sharecrop), then one might expect the share of women landholders on owner occupied farms to be higher than what is shown in this table, and the gap between survey and census data to be smaller. This is the case in Paraguay, where women constitute 21.7% of the agricultural holders overall, but makeup 23.2% of the holders on owner occupied farms (Guereña, 2017, Table A2). In Ecuador, however, they make up similar shares, thus the direction of this bias appears to be country specific.²⁰

Third, household surveys and agricultural censuses have different sampling frames. Nationally representative household surveys (which include all those shown in Table 3) generally draw their sample from the census units defined by the prior national population census, often stratified by the socio-economic indicators included in the census. In contrast, the agricultural census is an inventory of the agricultural production units (farms) existing in rural areas, however “rural” is defined in the national context. We would expect the census to give us a fairly accurate picture of the distribution of farms by size. Household surveys, in contrast, will not necessarily do so. Since they are designed to be representative of socio-economic characteristics, they do a better job of capturing poor rural households (the landless and near-landless), and moreover, urban households who own land, whether in peri-urban or “officially” rural areas, however nationally defined. Given the tendency for women managers to be associated with smaller farms than those managed by men, to the extent that management is associated with ownership, we would expect household surveys to reveal a higher share of female owners than might appear in the agricultural censuses (if such information were to be asked).

Fourth, the unit of analysis may matter greatly – whether ownership and management is measured at the parcel or at the farm level. The agricultural censuses focus on agricultural production units, defined as areas under one management. Such units may be composed of multiple parcels, each acquired in different ways by different owners. This point can be illustrated with the results from the 2005 Nicaragua LSMS survey (Deere, Alvarado and Twyman, 2012). It considered ownership of titled land at the parcel level and found that women individually owned 16.9% of the parcels, 4.1% were jointly owned by a man and a woman, and 79% were owned by men individually. The decision-making question, however, was only asked at the farm level and only one person could be designated as the manager. This method resulted in women being reported as only 8.8% of the farm managers, much below the share of women landowners, 19.9%, and the share of women landholders in the subsequent agricultural census, 23.3%.

Finally, a reason that we might expect survey estimates of the share of female landowners to exceed census estimates of the share of female managers is if female landowners are less likely than male owners to work their land directly. This is what Deere and Twyman (2014, Table 3) found in Ecuador, a country with a very high share of female landowners. The most frequent form of land ownership in Ecuador is joint ownership by a couple, constituting 36.6% of the parcels, thus women represent over half of the landowners when both individual and joint ownership are taken into account. Yet the parcels owned by individual female landowners are less likely to be worked directly by the owner or someone in the household than parcels owned jointly or by individual male landowners, with over a quarter of these having been rented, lent or sharecropped out. Mardon (2005) found a similar result with respect to commercial farms in Brazil (those larger than 50 hectares). Women were 10.5% of the owners, but only 7.5% of the owner-administrators, suggesting they were more likely than men to either rent their farms to others or to contract a full-time administrator.

One of the reasons that women landowners may be more likely to not work their land directly could be because of insufficient family labor, for example, if they are lone female heads. Or it could be related to the discrimination women face in credit markets or with respect to access to training or technical assistance.²¹

This analysis suggests the importance of collecting information on both ownership and management at the level of the parcel. This is the only way to ascertain the gendered relationship between owning land, the probability that it will be worked directly, and the outcome of the decisions made on that parcel, such as gender productivity differences. A growing body of research has shown that outcomes may differ depending on who actually owns the land (Deere; León, 2001; Doss, 2012; Peterman, 2012). A central question for gender analysis is precisely the relationship between women’s ownership of land and their participation in agricultural decision-making, the question to which we now turn.

Female landowners and agricultural decision-making

No rigorous quantitative study has been done to date in Latin America, of which we are aware, on whether women landowners are more likely to be engaged in agricultural decision-making than women who are not landowners themselves on owner-operated farms, although qualitative research supports this relationship (Deere, 1990; Stephen, 1997; Hamilton, 1998). The 2010 EAFF survey in Ecuador only collected information on women landowner’s participation in agricultural decision-making in the case their land parcels were worked directly by members of the household. Survey results show that the great majority of women landowners participate in the decisions over their own parcels, but that their participation varies according to their marital status, the particular decision under consideration, and whether they own land individually or jointly with their partners (Deere; Twyman, 2014).

Most women landowners in Ecuador are partnered (married or in a consensual union), 78%, as opposed to lone female household heads, 22%. While the great majority of partnered women participate in the four major decisions (what to plant, the inputs to use, how much to sell, and use of the income generated), they are less likely to participate in these than unpartnered women landowners, who generally make all the decisions regarding their plots themselves. The authors also find a relationship between the form of decision-making – whether decisions are taken alone or with someone else – and the form of land ownership – whether the woman owns the parcel individually or jointly with someone else. As expected, unpartnered women (who generally are individual owners) are more likely than partnered women to make all of the decisions on their plot alone. The majority of partnered women (who are usually joint owners), when they participate in decision-making, do so jointly with someone else, usually their partners.

There are also differences among partnered women landowners in the form of their participation, depending on whether they own their parcel individually, or jointly with their partners. For example, consider the decision about the control of income from agricultural sales. While 54.4% of partnered women who are individual owners make this decision on their own, only 7.7% of those who are joint landowners do so, with the great majority (77.2%) making this decision jointly. Overall, in Ecuador, where the majority of land parcels are jointly owned by a couple, the majority of decisions are made by the couple together.

The EAFF 2010 asked not only women landowners about their participation in agricultural decision-making, but also asked the spouse about their partners’ participation. Twyman, Useche and Deere (2015) show how among couples, men tend to underestimate their wife’s participation in agricultural decision-making compared to reporting from wives. To test this proposition they created an index of the decisions applicable to each parcel, where 1 is equal to participation in all four decisions when relevant. Whereas slightly over 60% of the wives reported that they participated in all decisions taken, only 52% of the husbands reported similarly. At the other end of the scale (index = 0), while only 10% of the wives reported that they did not participate at all in agricultural decision-making, 12% of the husbands considered that their wives did not participate at all. The gender differences in the distribution of this index was statistically significant, confirming that it matters who you interview in a survey.

A recent survey field experiment in a highland province of Ecuador focusing on responsibilities and decisions related to pesticide use came to a similar conclusion. Alwang, Larochelle and Barrera (2017) found that perceptions vary by gender and the type of interview. This study found that men interviewed alone are more likely to claim that decision-making is a male-only arena, while women interviewed alone either consider themselves to make the decisions or that decisions are undertaken jointly with their spouse. Interviewing both members of a couple separately, knowing that the other will be interviewed, led to less drastic claims of exclusive responsibility and higher reporting of joint decision-making. This study adds to those which argue that women’s role in agriculture will be underestimated if surveys (or the agriculture census) are directed primarily to male household heads (Deere; León, 2001; Doss, 2012).

In order to establish the relationship between women’s land ownership and their role in decision-making it will be important to undertake such studies in more Latin American countries and analyze a broader set of agricultural and livestock decisions. FAO (2017) recommends that data be collected on the following decisions: the area of land cultivated versus the area left fallow; the types of crops grown; the types of livestock raised; applying for agricultural credit; investing in capital assets; marketing of agricultural products and/or livestock; and the types of inputs used (fertilizers, pesticides, irrigation, hired labor, etc.). To this list it would be important to add decisions about whether the parcel is to be worked directly or rented, sharecropped or lent out, and who decides on the use of income generated by the different agricultural and livestock activities.

Conclusion

Irrespective of its limitations, the 2030 SDG process of setting goals, targets, and precise indicators has undoubtedly moved the gender equality agenda a step forward. For those of us concerned with the distribution of land and other resources by gender, the adoption of SDG targets 5.a and 1.4, along with their respective indicators, represent significant accomplishments, but much depends on whether appropriate policies are implemented to reach these targets and the relevant data is collected to measure progress.

As a minimum, the collection of data on the legal land rights indicators and on women’s ownership of land in surveys and the agricultural censuses will allow research on a number of pressing issues. For example, consistent information on legal land rights and women’s actual ownership of land will allow testing the extent to which specific legal frameworks (such as marital or inheritance regimes) and their reforms make a difference. The insistence in the SDG process and indicators on security of tenure should at least encourage government statistical offices to disaggregate the information they currently collect on landholders by gender and form of tenure. This would allow questions to be answered such as whether women are more likely to declare themselves the landholders on owner-operated farms as compared to those which are rented or sharecropped, or on communally owned land. The addition of one question – to specify who is the landowner on owner-operated farms – would indicate whether women are more likely to declare themselves the landholder when they are in fact the owners. In addition, information on whether the declared owner holds a formal land title in their own name will allow more detailed investigation into the barriers that women may confront in enforcing their claims.

I have argued that gender disaggregated data on land ownership by itself is insufficient; disaggregated information on agricultural decision-making is also needed. This would allow more rigorous analysis of who in fact in the household are the managers of the farm, to overcome the existing ideological biases in the concept of landholder (i.e., its conflation with the household head). More detailed information on both ownership and decision-making should allow analysis of such important questions as to why women landowners do not always farm their land, or whether women are more likely to be involved in agricultural decision-making if they are owners as compared to non-owners, or individual versus joint land owners.

It is heartening that the importance of these questions has been recognized in FAO’s WCA 2020 guidelines, which recommends that future agricultural censuses collect information on joint holders, at the very least. FAO’s recommended survey module about the intra-household distribution of managerial decisions and ownership of the holding goes far beyond the SDG land ownership/security of tenure indicators, and would represent a crucial step forward if, indeed, it is adopted by a significant number of countries.

Referências bibliográficas

• Agarwal, Bina. A Field of One’s Own: Gender and Land Rights in South Asia. Cambridge, UK, Cambridge University Press, 1994.
• Alwang, Jeffrey; Larochelle, Catherine; Barrera, Victor. Farm Decision Making and Gender: Results from a Randomized Experiment in Ecuador. World Development (92), Elsevier, 2017, pp.117-129.
• Butto, Andrea; Hora, Karla. Mulheres e reforma agrária no Brasil. In: Lopes, Adriana; Butto, Andrea (ed.). Mulheres na reforma agrária: A experiência recente no Brasil. Brasília, MDA, 2008, pp.19-37.
• Dane (Departamento Administrativo Nacional de Estadística). Tercer Censo Nacional Agropecuario. Tomo 2. Resultados Bogotá, DANE y Ministerio de Agricultura, 2016. [http://www.dane.gov.co/files/images/foros/foro-de-entrega-de-resultados-y-cierre-3-censo-nacional-agropecuaria/CNATomo2_Resultados.pdf – acesso em: 30 maio 2017].
  » http://www.dane.gov.co/files/images/foros/foro-de-entrega-de-resultados-y-cierre-3-censo-nacional-agropecuaria/CNATomo2_Resultados.pdf
• Deere, Carmen Diana. Household and Class Relations: Peasants and Landlords in Northern Peru. Berkeley, University of California Press, 1990.
• Deere, Carmen Diana. Tierra y autonomía económica de la mujer rural: avances y desafíos para la investigación. In: Costas, Patricia (comp.). Tierra de mujeres. Reflexiones sobre el acceso de las mujeres rurales a la tierra en América Latina. La Paz, Fundación Tierra and International Land Rights Coalition, 2011, pp.41-69.
• Deere, Carmen Diana. Women’s Land Rights, the State, and Rural Social Movements in the Twentieth-first Century Latin American Agrarian Reforms. Journal of Agrarian Change (17), Wiley-Blackwell, 2017, pp.258-278.
• Deere, Carmen Diana; Alvarado, Gina; Twyman, Jennifer. Gender Inequality in Asset Ownership in Latin America: Female Owners vs. Household Heads. Development and Change 43 (2), The Hague, Institute for Social Studies and Wiley, 2012, pp.505-530.
• Deere, Carmen Diana; Contreras D. Jackeline. Acumulación de activos. Una apuesta por la equidad. Quito, FLACSO-Ecuador, 2011.
• Deere, Carmen Diana; León, Magdalena. Empowering Women: Land and Property Rights in Latin America. Pittsburgh, University of Pittsburgh Press, 2001 [O Empoderamento da Mulher. Direitos a terra e direitos de propriedade na América Latina. Porto Alegre, UFRGS Editora, 2002].
• Deere, Carmen Diana; Oduro, Abena D.; Swaminathan, Hema; Doss, Cheryl. Property Rights and the Gender Distribution of Wealth in Ecuador, Ghana and India. Journal of Economic Inequality 11(2), Springer US, 2013, pp.249-265.
• Deere, Carmen Diana; Twyman, Jennifer. ¿Quién toma las decisiones agrícolas? Mujeres propietarias en Ecuador. Agricultura, Sociedad y Desarollo (11), Texcoco, Colegio de Postgraduados, 2014, pp.425-440.
• Doss, Cheryl R. Intrahousehold Bargaining and Resource Allocation in Developing Countries. World Bank Research Observer 28(1), Washington, D.C., The World Bank and Oxford Academic, 2013, pp.52-78.
• Doss, Cheryl R. Data Needs for Gender Analysis in Agriculture. In: Quisumbing, Agnes et alii (ed.). Gender in Agriculture: Closing the Knowledge Gap New York, Springer and FAO, 2014, pp.55-74.
• Doss, Cheryl R. et alii. Gender Inequalities in Ownership and Control of Land in Africa: Myth and Reality. Agricultural Economics (46), International Association of Agricultural Economics and Wiley, 2015, pp.403-434.
• Esquivel, Valeria. Power and the Sustainable Development Goals: A Feminist Analysis. Gender & Development 24(1), Oxfam and Taylor & Francis, 2016, pp.9-23.
• Fao (Food and Agriculture Organization). World Programme for the Census of Agriculture 2010. FAO Statistical Development Series #11, Rome, FAO, 2005.
• Fao. FAO Policy on Gender Equality: Attaining Food Security Goals in Agriculture and Rural Development Rome, FAO, 2012. [http://typo3.fao.org/fileadminis/templates/gender/docs/FAO_FinalGender_Policy_2012.pc – acceso em: 10 jul. 2012].
  » http://typo3.fao.org/fileadminis/templates/gender/docs/FAO_FinalGender_Policy_2012.pc
• Fao. What is the Legal Assessment Tool (LAT) for gender-equitable land tenure? Rome, FAO, 2014. [http://www.fao.org/3/a-i4006e.pdf – acesso em: 5 jun. 2017].
  » http://www.fao.org/3/a-i4006e.pdf
• Fao. World Programme for the Census of Agriculture 2020 vol. 1. FAO Statistical Development Series #15. Rome, FAO, 2017. [http://www.fao.org/3/a-i4913e.pdf – acesso em: 25 maio 2017].
  » http://www.fao.org/3/a-i4913e.pdf
• Fukuda-Parr, Sakiko. From the Millennium Development Goals to the Sustainable Development Goals: Shifts in Purpose, Concept, and Politics of Global Goal Setting for Development. Gender & Development 24(1), Oxfam and Taylor & Francis, 2016, pp.43-52.
• Guaraňa, Arantza. Kuña Hayvy. Desigualdades de género en el acceso a la tierra en Paraguay. Asunción, Oxfam and ONU Mujeres, 2017.
• Gutiérrez, M.A. Principales resultados de la encuesta piloto sobre la medición de la propiedad de los activos del hogar y el emprendimiento con perspectiva de género, Proyecto EDGE-México. Power point presentation to the International Seminar on Gender Statistics, INEGI, Aguascalientes, Mexico, September 2016.
• Hamilton, Sarah. The Two-Headed Household: Gender and Rural Development in the Ecuadorean Andes. Pittsburgh, University of Pittsburgh Press, 1998.
• Hora, Karla; Butto, Andrea. Políticas públicas para mulheres rurais no contexto dos territórios da cidadania. In: Butto et alii (ed.). Mulheres rurais e autonomia: Formação e articulação para efetivar políticas nos territórios da cidadania. Brasília, MDA, 2014, pp.14-45.
• INE (Instituto Nacional de Estadística). Censo Agropecuario 2013 Bolivia. La Paz, INE, 2015. [http://www.sudamericarural.org/images/en_papel/archivos/CENSO-AGROPECUARIO-BOLIVIA_final.pdf – acesso em: 30 May 2017].
  » http://www.sudamericarural.org/images/en_papel/archivos/CENSO-AGROPECUARIO-BOLIVIA_final.pdf
• INEC (Instituto Nacional de Estadística y Censos). Censo Nacional Agropecuario 2000. Quito, Ecuador, INEC, 2000. [http://www.ecuadorencifras.gob.ec/censo-nacional-agropecuario – acesso em: 11 jul. 2018].
  » http://www.ecuadorencifras.gob.ec/censo-nacional-agropecuario
• INEC (Instituto Nacional de Estadística y Censos). VI Censo Nacional Agropecuario 2014. Resultados Generales San José, Costa Rica, INEC, 2015. [http://www.mag.go.cr/bibliotecavirtual/a00338.pdf – acesso em: 30 maio 2017].
  » http://www.mag.go.cr/bibliotecavirtual/a00338.pdf
• INEI (Instituto Nacional de Estadística e Informática). Inequidades de Género en la Actividad Agropecuaria. IV Censo Nacional Agropecuario 2012. Lima, INEI, 2014.
• INIDE (Instituto Nacional de Información de Desarrollo). Informe Final. IV Censo Nacional Agropecuario (CENAGRO) 2011. Managua, INIDE y Ministerio Agropecuario y Forestal, 2012. [http://www.inide.gob.ni/Cenagro/INFIVCENAGRO/IVCENAGROINFORME/assets/basic_html – acesso em: 30 maio 2017].
  » http://www.inide.gob.ni/Cenagro/INFIVCENAGRO/IVCENAGROINFORME/assets/basic_html
• Mardon, Merillee. Three Essays on Gender, Land Rights, and Collective Action in Brazil’s Political Economy. Ph.D. dissertation. Amherst, MA, University of Massachusetts, Department of Economics, 2005.
• Namdar-Irani, Mina. Mujeres y agricultura en Chile: una mirada de género al censo agropecuario 2007. In: Butto, Andrea; Dantas, Isolde; Hora, Karla (ed.). As mulheres nas estatísticas agropecuárias: Experiências em países do Sul. Brasília, Ministério de Desenvolvimento Agrário (MDA), 2012, pp.119-216.
• Nobre, Miriam. Censo Agropecuário 2006 – Brasil: uma análise de gênero. In: Butto, Andrea; Dantas, Isolde; Hora, Karla (ed.). As mulheres nas estatísticas agropecuárias: Experiências em países do Sul. Brasília, MDA, 2012, pp.41-118.
• Peterman, Amber; Quisumbing, Agnes; Behrman, Julie. A Review of Empirical Evidence on Gender Differences in Non-land Agricultural Inputs, Technology, and Services in Developing Countries. In: Quisumbing, Agnes et alii. (ed.). Gender in Agriculture: Closing the Knowledge Gap. New York, Springer & FAO, 2014, pp.145-186.
• Razavi, Shahra. The 2030 Agenda: Challenges of Implementation to Attain Gender Equality and Women’s Rights. Gender & Development 21(1), Oxfam and Taylor & Francis, 2016, pp.25-41.
• Rueda R., Alexandra et alii. La mujer en la agricultura Chilena. Resultados Censo Agropecuario 2007. Santiago, Instituto Nacional de Estadísticas, 2008.
• Stephen, Lynn. Women and Social Movements in Latin America. Austin, University of Texas Press, 1997.
• Twyman, Jennifer; Useche, Pilar; Deere, Carmen Diana. Gendered Perceptions of Land Ownership and Agricultural Decision-making in Ecuador. Land Economics 91(3), 2015, pp.479-500.

• 1
  The full list of the 169 SDG targets and the many indicators may be found on the UN’s Sustainable Development Knowledge Platform: https://sustainabledevelopment.un.org.
• 2
  See Fukuda-Parr (2016) for a detailed analysis of the differences in the SDG and MDG processes, and their respective goals and targets.
• 3
  Target 10.1 on reducing income inequality within countries, for example, relies on income growth among the bottom 40% to reduce income poverty, without mentioning the potential role of wealth or inheritance taxes, or structural reforms such as redistributive land reform or limits on land ownership (Razavi, 2016).
• 4
  For example, the Gender Division of ECLAC, UN Women, and the Mexican statistical institute (INEGI), hold an annual workshop on gender statistics, attended by representatives of the national statistical institutes and women’s offices and ministries and academics to discuss data needs and best practices.
• 5
  See Doss (2013) and Doss et alii (2015) on the empirical evidence on how women’s ownership of land is related to improved outcomes for women.
• 6
  In a comparative context, the latter indicator will generally reflect the relative weight of the rural population in the total population. National household surveys also capture landownership by those who reside in urban areas.
• 7
  See Doss et alii (2015) for a discussion of this issue in the African context.
• 8
  The precise language is as follows: “Joint holder is defined as a person making the major decisions regarding resource use and exercising management control over the agricultural holding operations, in conjunction with another person” (FAO, 2017:46).
• 9
  See https://unstats.un.org/sdgs/files/metadata-compilation/Metada-Goal-5.pdf and FAO (2012).
• 10
  See http://www.fao.org/gender-landrights-database/legislation-assessment-tool/en. Accessed 29 May 2017. The compilation of scores for the Latin American countries is available from the author.
• 11
  For what has been accomplished in these countries, see Deere and León (2001).
• 12
  Moreover, these policy directives were implemented, resulting in a relatively high share of female agrarian reform beneficiaries, over one-third by 2007 (Butto; Hora, 2008), and a growing share of female household heads among them (Hora; Butto, 2014). On the other hand, the Ministry of Agrarian Development was closed in 2016 by a new government, and agrarian reform is currently at a standstill.
• 13
  For added emphasis, the Bolivian legislation stipulates that in joint titling to couples, the woman’s name should come first.
• 14
  Missing are Cuba and Honduras following the conventional definition of Latin America consisting of 19 Portuguese and Spanish-speaking countries.
• 15
  This comparison is based on Table 1 and the earlier census data presented in Deere (2011, Table 1).
• 16
  Derived by the author from INEC (2000).
• 17
  One ethnic group shows a different trend, the “raizales”, the ancestral inhabitants of the San Andres archipelago. Of their farms, only 1.2% are managed jointly, 26.6% by women and 72.2% by men.
• 18
  See Deere (2001, Table 9.11) on this same trend based on 1990s data.
• 19
  The evidence on this is mixed. The Mexican 2015 INEGI asset survey revealed that women were 25% of the total landowners, but only 21% of the documented landowners (Gutiérrez, 2016). In contrast, in Ecuador, women represent a slightly larger share of the documented landowners (55%) than of the share of reported landowners (54%) (author’s calculations from EAFF 2010 database).
• 20
  Similar data is not available for the other countries reported in Table 3. In Peru’s 2012 agricultural census; women constituted 30.8% of the landholders on all farms, irrespective of form of tenancy, but 32.1% of the agricultural holders on owner occupied farms (INEI, 2014), similar to Paraguay.
• 21
  See Peterman, Quisumbing and Behrman (2012) for a summary of gender differences in access to non-land inputs and services internationally.

Publication Dates

• Publication in this collection
  30 Nov 2018
• Date of issue
  2018

History

• Received
  22 Aug 2017
• Accepted
  18 July 2018