The gender effects of Economic Partnership Agreements: The case of Mozambique
This article summarises a research project undertaken by One World Action (OWA) and the Commonwealth Secretariat which develops a gender-aware economy-wide framework to assess the distributional effects of the Economic Partnership Agreements (EPAs). For each country analysed, the approach involves reviewing both the list of goods to be liberalised and the list of exemptions, and examining the gender characteristics of production and consumption of selected products. The potential revenue loss from tariff reduction is also estimated. This article focuses on the findings for Mozambique. The full study also assesses gender effects in Tanzania and Jamaica.
Gender relations interact with trade relations. Gender inequality may limit the gains from trade, through its negative impact on the process of innovation in export-oriented manufacturing, or by undermining supply responses in agriculture, among others. Trade policies, in turn, have gender differentiated effects because of women's and men's different economic roles and control over productive resources. For instance, many women have fewer skills than men, and tend to concentrate in fewer occupations: this makes it more difficult for them to find new jobs if the sectors they work in contracts because of greater import competition.
Trade liberalisation alters the distribution of income between different social groups by changing the relative prices of goods (1). These may induce reallocation of factors of production among sectors that use them with different intensities, and therefore entail changes in their employment and/or remuneration. The same variations in relative prices bring about changes in real incomes that affect groups differently, due to differences in their consumption patterns. Trade liberalisation is also likely to reduce tariff revenues, with group-specific effects on the size and composition of government expenditure. The intensity of these effects, as well as the direction of change, is not easy to predict and will depend on the country structure and the institutional context.
Trade liberalisation can thus affect gender inequalities at the macro, meso and micro levels. For example, gender gaps in market participation might narrow if the sectors that expand are more female-intensive than the sectors that contract (macro); public provision of social services that support women might be undermined, if loss of government revenue from reduced tariffs leads to cuts in such services (meso); female control over household spending is reduced or extended, depending on whether trade liberalisation destroys or creates sources of independent income for women (micro). As these examples illustrate, some changes can be positive whilst others can be negative and there may be tensions between different dimensions.
EPAs are based on the idea of reciprocal market access. Under EPAs, ACP countries are expected to offer duty-free access for ‘substantially all' EU exports to them. By joining an EPA, those ACP countries which are not least developed countries (LDCs) can avoid tariffs being increased on their exports to the EU however further improvements in market access for their exports are likely to be small. The LDCs do not face the negative shock if they remain outside an EPA because they already enjoy duty-free access under the Everything But Arms (EBA) trade. (2)
The gender distribution of the EPA effects will depend on a number of factors. Whether an increase in EU imports will affect female jobs more than male jobs in a country will depend on the gender composition of the labour force in those sectors that are especially sensitive to import competition. Women as either wage workers or small producers are likely to be negatively affected if they are disproportionately employed by the sectors that decline, and, importantly, if their opportunities to find employment in other sectors are limited due to discrimination.
Whether the liberalisation of imports from the EU will benefit poor consumers, and in particular women in their role as household managers, will depend on tariff cuts translating into cheaper consumer goods, and on whether the cheaper imported goods constitute a significant share in low income households' consumption baskets.
As for the revenue effect, loss of government revenue from reduced tariffs might undermine gender equality if it leads to reduced provision of social services such as health, education, water, sanitation and other infrastructure to meet household needs. Even if the government manages to replace tariffs with alternative indirect taxes such as VAT or excise duties, these may have a gender differentiated impact that needs to be assessed.
There is considerable variation across countries and regions in the range of goods to be liberalised under the EPAs, the timeframe set for liberalising them and the exemptions (suggesting also differences in political agendas, as well as skills and power of different negotiating teams). Within the Southern Africa Development Community (SADC), for example, Mozambique agreed to liberalisation schedules considerably different from those jointly decided by the rest of the group (just one fifth of the items are being excluded by both Mozambique and the Botswana, Lesotho, Namibia and Swaziland group). This may risk undermining efforts towards further regional integration. Within the East African Community (EAC) group, in contrast, the liberalisation commitments taken are identical among all the countries included.
The project by OWA and the Commonwealth Secretariat applied a gender-aware economy-wide framework to three countries, each belonging to a different regional grouping: Jamaica (Caribbean Forum), Mozambique (Southern Africa Development Community) and Tanzania (East African Community). A common lesson learnt is that, unless the capacity of women to participate in trade is strengthened through improved infrastructure (including communication technologies), skill development, and support for women's inclusion in farmers organisations and trade unions, the positive outcomes from trade liberalisation are likely to be small.
With reference to Mozambique, the study finds that job losses from import displacement are likely to be small and not disproportionately female. This is because most of the liberalised imports from the EU are not goods which are either produced domestically or in which the EU is a major exporter. These results are of course dependent on the specific socio-economic structure of Mozambique and its liberalisation schedule, and should not be generalised.
In Mozambique, 97 percent of the goods to be liberalised are manufactures and only 3 percent are agricultural goods. The only agricultural product of some significant value (and with a high tariff) is almonds, which do not grow in the country and are consumed mostly by wealthy households. Most of the manufactured imports that may increase are intermediate goods such as irrigation pumps, agro-processing machinery and electrical devices, which are not produced locally. A higher use of these inputs could in principle contribute to enhancing agricultural productivity, but it is very unlikely that this would benefit small female farmers who would have neither the capital nor the knowledge to invest in the adoption of new technologies to cultivate their plots.
Some of the traditionally ‘female sectors', such as processed agriculture, remain protected but this is in no way a sufficient step towards gender equitable trade outcomes. An important finding relative to Mozambique is that, because of restricted access to economic resources and limited bargaining power in most markets, the majority of vulnerable women are unlikely to take advantage of any new economic opportunity resulting from better trade incentives. For example, the few female farmers who own land in Mozambique have smaller plots than men even though official land laws grant them equal access as men. This in turn limits their access to capital and credit. Women farmers use less mechanised production technologies and only 23 percent of them participate in extension services compared with about 45 percent of men. Women also appear to be less informed about prices of agricultural goods and their rights as workers. This undermines their ability to negotiate fair prices and wages with their employers, traders and other intermediaries.
A common argument in favour of trade liberalisation as a tool for poverty reduction is that cheaper imports will enable the poor to increase their consumption. The study finding is that it is improbable that cheaper imports likely to result from current EPAs would benefit vulnerable low-income women, as these imports include final consumption items such as washing machines and gas cookers, which can only be afforded by households with high incomes and easy access to energy sources. Measures need to be taken to avoid a regressive impact from tariff cuts, in the sense of improving consumption of well-off households whilst making goods and services consumed by vulnerable groups less affordable.
The fiscal impact of the EPA liberalisation and its gender effects depends on how much revenue is actually collected from the tariffs, the relative importance of tariff revenue in government financing, the alternative taxes that the government may introduce to compensate for the loss, and the extent to which public expenditure addressing gender disadvantage is a priority for the government. The study calculates the potential revenue loss by applying the base tariff to the value of imports in the reference year in order to generate the ‘hypothetical' revenue being collected and finds that this is about two percent of total tax revenue for Mozambique. (3) This is a modest amount but still likely to have significant impact since a high proportion of the losses (85 percent) is in the first phase.
To conclude, the main lesson for policy is that trade-related technical assistance frameworks such as the Aid for Trade initiative (4), need to give priority to comprehensive and well-designed interventions to facilitate women's economic mobility across sectors and occupations, and to widen their options. These interventions should include, among others: supporting the full enforcement of core labour standards and anti-discrimination legislation; promoting institutional mechanisms that foster small female producers and traders' participation in groups; designing agricultural vocational training and extension services to meet the specific needs of female farmers; promoting gender audits of trade-related administrative procedures; financing physical infrastructural projects that reduce women's time and energy burdens; protecting women's rights over their own financial assets and assisting them in claiming fair remuneration for contributing to their family business.
Several gaps remain in sex-disaggregated statistics relevant to trade analysis. More comprehensive and regular collection of gender aware data and in-depth sectoral studies of gender-differentiated trade impacts need to inform Diagnostic Trade Integration Strategies and must be strongly promoted.
Author: Marzia Fontana is a Research Fellow at the Institute of Development Studies, University of Sussex, UK.
This article is based on a more detailed study by One World Action and the Commonwealth Secretariat entitled Gender Justice in Trade Policy: The gender effects of Economic Partnership Agreements. This study draws on the examples of three specific case studies (Jamaica, Tanzania and Mozambique).
1 In practice, some exemptions are granted, with their extent varying by country.
2 Even though based on strong assumptions, this method can provide some broad useful indication.
3 See related article :Karin Ulmer; Equity in Trade Negotiations: a Gender Review of the EPA; TNI Vol.6, No. 2, March-April 2007, http://acp-eu-trade.org/library/files/TNI_EN_6-2.pdf
4 See related article: Soraya Hassanali, Gender mainstreaming and Aid for Trade: The case of the EIF; TNI Vol.1, No.1, February -March 2011 http://www.ictsd.org/sites/default/files/news/tni/tni_en_10-1.pdf