GENDER WELFARE EFFECTS OF REGIONAL TRADE INTEGRATION ON HOUSEHOLDS IN GHANA

Emmanuel ORKOHDepartment of Economics, Faculty of Social Sciences, College of Humanities and Legal

StudiesUniversity of Cape Coast

GhanaEmail: aorkoh@gmail.com

Abstract

This study1 assesses the welfare effects of Ghana’s adoption of the common external tariff of the Economic Community of West African States (ECOWAS), with special attention to gender differences. Using data from the 2005 Social Accounting Matrix for Ghana, the World Trade Organization Secretariat, and the 2012-2013 round of the Ghana Living Standard Survey, it combines a computable general equilibrium (CGE) model and a microsimulation model integrated through a top-down approach. The study focuses on one of the channels through which trade integration can affect poverty and welfare, namely the price channel. The CET tariffs are expected to result in changes in domestic prices, which will in turn have an impact on household welfare, depending on whether the households are consumers or producers of the commodities affected by the tariff reform. The results of the study indicate a positive welfare effect of the CET on poor households, and particularly female-headed households, as consumers. However, implementation of the CET is expected to reduce the welfare of both poor and rich households as producers, with poor and male-headed households being the most affected. The net welfare effect will be around zero for the poor male-headed households and negative for the rich. It will be positive for female-headed households at the low and middle income category, but also negative for the rich. While results also depend on the geographic location of the household, among others, the reform is thus expected to have pro-poor and pro-female effects. In order to address the welfare gaps identified, the study recommends that the government could introduce compensatory policies such as income transfers that would target male-headed households that are producers, and streamline the Livelihood Empowerment Against Poverty Programme it is currently implementing to help poor male-headed households. It could also aim to improve infrastructure in the regions which will experience a net welfare loss in order to make them more competitive.

May 2016

1 The author would like to thank David Zavaleta, of Universidad Católica Boliviana, Bolivia, for his very useful advice and excellent mentorship during the development of this study. Sincere appreciation also goes to the Virtual Institute team and Jane Elizabeth Casabianca, of Università Politecnica delle Marche, Italy, whose feedback and guidance have helped shape the study. Financial support from the government of Finland is also gratefully acknowledged.

1 Introduction

Over the past two to three decades, a number of developing countries have opted for regional economic integration as an instrument to harmonize trade policies and increase their weight in global trade. Economic integration, particularly in Africa, has also been seen as a way to diversify the structure of African economies, boost intra-African trade and investment, build supply capacity, and sustainably reduce poverty (Osakwe, 2015). These integration efforts have resulted in the creation of regional blocs such as the Economic Community of West African States (ECOWAS), West African Economic and Monetary Union (WAEMU), Common Market for East and Southern Africa (COMESA), Economic Community of Central African States (ECCAS), Central African Economic and Monetary Community (CEMAC), Southern African Customs Union (SACU), and Arab Maghreb Union (AMU). The trade-related objectives of these blocs include the establishment of custom unions,2 with a common external tariff (CET) as a major trade policy instrument.

Regional trade integration through the creation of a customs union with a CET has been found to have both direct and indirect implications for household poverty and welfare in general. Trade integration affects poverty and welfare through three main channels: (a) changes in employment structures and wages; (b) changes in prices and their impact on consumption and production patterns; and (c) changes in financing for social expenditure by governments (Winters et al., 2004; Harrison and Tang, 2005). Trade reforms can also affect poverty indirectly via economic growth. It has been argued that increased trade openness resulting from trade reforms can improve access to technology and hence foster productivity growth, leading to faster economic growth and reduced poverty (UNCTAD, 2010). Conversely, increased trade restrictions would impede productivity growth and slow economic growth, leading to increased poverty.

Among the various channels of effects, this study analyses the price channel. The focus on this channel is due to the fact that most often trade policies such as a CET affect import tariffs and thus domestic prices of commodities, which in turn affect the consumption and production decisions of households (Marchand, 2012; Nicita, 2009). We can therefore consider the price channel as the mechanism that affects households more directly in the short term.

Between 2007 and 2015, Ghana alternated between its own tariff system and the CET of the regional economic bloc of which it was a member at a given time. From 2007 to 2011, the country implemented its own tariffs, but in 2012 it adopted the WAEMU CET, before switching back to its own tariff system in 2013 and finally adopting the ECOWAS CET starting in 2015. This followed negotiations by ECOWAS members in Dakar, Senegal, on the CET for the region, which concluded in October 2013 (Roquefeuil et al., 2014).3 2 A customs union is a type of trade agreement under which certain countries preferentially grant tariff-free market access to each other’s imports and agree to apply a common set of external tariffs to imports from the rest of the world. That is, they enter into a free trade agreement and apply a common external tariff to imports from non-members (Adams, 1993).3 According to Article 3.2(d) of the revised ECOWAS treaty, one of the aims of ECOWAS is “the establishment of a common market through: (i) the liberalisation of trade by the abolition, among Member States, of customs duties levied on imports and exports, and the abolition, among Member States, of non-tariff barriers in order to establish a free trade area at the Community level; (ii) the adoption of a common external tariff and a common trade policy vis-à-vis third countries; and (iii) the removal, between Member States, of obstacles to the free movement of persons, goods, service and capital, and to the right of residence an establishment.” (see

2

Extant literature suggests that Ghana’s implementation of the ECOWAS CET will result in significant changes in the country’s existing tariff structure for both agricultural and non-agricultural products. Some of the tariff rates will be lower and some will be higher as a result of the CET. For instance, the implementation of the CET will reduce the number of commodities admitted under zero per cent tariff rates from 725 to 85. On the other hand, the range of commodities admitted under the 5 per cent band will be broadened from 375 to 2,146 under the CET. The changes in tariffs related to implementation of the CET will affect prices and, consequently, the welfare of households, depending on their position as either net producers or consumers of these products.

As explained above, the different channels through which regional trade integration influences household poverty and welfare have been found to result in both opportunities for gains and risks of losses for households (Martinez and Poole, 2004). Literature suggests that these gains and losses are distributed differently among different segments of the population, including men and women (UNCTAD, 2012). According to UNCTAD (2011), given the differences in roles played by men and women in a society and an economy, trade policies such as those enshrined in the ECOWAS CET have different implications for male- and female-headed households. This assertion has been widely supported by findings of several trade-gender specific studies. For instance, Bird (2004) emphasizes that changes associated with trade integration may be positive or negative for women and men depending on their individual characteristics, including education and skills, marital status, family size, social group characteristics such as whether the household is a net producer or a net consumer, urban/rural location, and economic and social status. In the Ghanaian context, the extent to which this assertion holds remains an important policy question, especially in the wake of the country’s recent adoption and implementation of the ECOWAS CET.

The aim of this study is to empirically assess the effects that implementation of the CET will have on household poverty, income, and consumption through the price channel, with a special focus on gender. Specifically, the objectives are to (a) examine the effects of Ghana’s adoption of the ECOWAS CET on household consumption and poverty; and (b) assess the poverty and welfare distribution by gender. A review of the extant literature reveals that even though some ex-ante studies of this nature on Ghana have looked at the poverty and income effects of trade liberalization (Bhasin and Annim, 2005; Bhasin and Obeng, 2007; Bhasin, 2012), no study has focused on the impact of the different ECOWAS CET bands on household welfare and the gender dimensions of this impact. The aim of this study is to fill these gaps and contribute to the existing literature on the links between international trade policy and household welfare.

To assess the implications of the CET, the study applies a top-down approach by combining a macro computable general equilibrium (CGE) (top) model and a micro (bottom) household model, with the latter using data at the household level (Bourguignon and Savard, 2008). Section 2 provides an overview of the evolution of Ghana’s tariffs and poverty over the years. Section 3 briefly reviews the empirical literature on the trade-gender link. Section 4 explains the

http://www.courtecowas.org/site2012/pdf_files/revised_treaty.pdf#page=4&zoom=auto,-82,12). The member countries of ECOWAS are Benin, Burkina Faso, Cape Verde, Côte d’Ivoire, the Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Niger, Nigeria, Sierra Leone, Senegal, and Togo.

3

methodology and data used in the study, and Section 5 describes the simulations and presents the results. The final section provides conclusions.

It is hoped that the findings of this study may serve as input to Ghana’s Ministry of Trade and Industry in formulating gender-aware policies to ensure that all population segments and household categories share appropriately in the gains and losses associated with the country’s adoption of the CET. More broadly, it is hoped that the study may help the ministry in formulating policies to enhance gender equality and promote human development. The gender analysis on the impact of the CET may be useful to the government of Ghana in designing compensatory policies to target segments of the population that may come out as losers due to implementation of the CET.

2 Overview of Ghana’s trade reforms and poverty trends

Ghana’s trade policy evolved from a fairly liberal one in the 1950s to a significantly controlled regime in the 1970s, after which the country embarked on major trade and economic reforms in the 1980s accompanied by trade liberalization. This policy has been largely influenced by developments in international trade under the General Agreement on Trade and Tariffs (GATT) and the World Trade Organization (WTO). It has also been shaped by trade agreements between Ghana and its major trading partners, the country’s economic development policy, and the structural adjustment programmes of the World Bank and the International Monetary Fund, particularly in the 1980s and 1990s (Ackah and Aryeetey, 2012). Significant trade liberalization in Ghana began with the downward adjustment of tariff rates in 1983 from 35 per cent, 60 per cent, and 100 per cent to 10 per cent, 20 per cent, 25 per cent, and 30 per cent. The tariffs were further simplified and lowered to 0 per cent, 25 per cent, and 30 per cent the following year to create a uniform pattern of protection, although some import controls remained in place. Further reductions were made in 1986, when the higher rates were lowered to 20–25 per cent (Ackah and Aryeetey, 2012).

Major trade policy reforms took place between 2007 and 2015, when the most-favoured-nation (MFN) tariff applied by the country was frequently modified. In 2012, the country adopted the WAEMU CET, which was based on four tariff bands comprised of a zero duty on social goods such as medicine and publications, 5 per cent duty on imported raw materials, 10 per cent duty on intermediate goods, and 20 per cent duty on finished goods (Office of the United States Trade Representative, 2014). In 2013, Ghana switched from the WAEMU CET back to its own national tariffs. In this context, it abolished the non-ad-valorem tariffs applied to petroleum products, and replaced them with ad valorem rates in January 2014 (Office of the United States Trade Representative, 2014). This was accompanied by a reduction of duties on some products and an increase in the duties on others.4

As shown in Table 1, the average unweighted applied MFN tariff in 2013 was 12.8 per cent, which represents a slight increase from the 12.7 per cent rate in 2007. The MFN rates on agricultural products were generally higher, at an average of 17.3 per cent, with the highest rates 4 Items for which duties were reduced from as high as 20 per cent to duty-free included fish livers, roe and flour, seeds, clinker and bulk cement, gasoil and related products, fishing yarn and equipment, mosquito nets, and contact lenses. Those items for which tariffs increased included mobile phones, online telephone sets, cordless handsets, rough wood, ferrous and non-ferrous metal scrap, air coolers, and battery chargers.

4

applicable to dairy products and tobacco. Tariffs on non-agricultural products remained lower, averaging 12 per cent. Based on the International Standard Industrial Classification (ISIC)5

definition of sectors, agriculture was the most tariff-protected sector, with an average tariff of 15.1 per cent, followed by manufacturing (12.7 per cent), while imports of mining and quarrying products had the lowest tariffs (11.2 per cent).

Subsequently, as a member of ECOWAS, Ghana endorsed the ECOWAS CET, which was adopted by ECOWAS Ministers of Finance on 20 March 2014, and came into effect on 1 January 2015. The ECOWAS CET is based on the four tariff bands of the WAEMU CET and an additional fifth band that involves a 35 per cent duty on goods in sectors such as poultry and rice that the government sought to protect.

The revision of Ghana’s trade policy triggered by implementation of the ECOWAS CET in 2015 will result in considerable changes in Ghana’s tariff structure for agricultural and non-agricultural products. Overall, there will be a slight reduction in the country’s average unweighted applied MFN tariff from 12.8 to 12.3 per cent. Mining and quarrying will be the sector that will experience the greatest reduction in tariff rates, which could have both positive and negative implications for the country. On the positive side, the reduction may attract more investors into the sector and this might increase government revenue through the payment of taxes that could in turn be used for development purposes. On the other hand, increased mining activity triggered by the tariff reduction may have negative effects on the environment if proper regulations are not adopted and enforced. Agriculture will be the sector to record the second largest reduction in tariffs. Wheat and maize, for instance, could become cheaper for consumers. This could have a positive effect on consumers, who may benefit from lower food prices, but a negative effect on domestic producers, who may not be able to withstand the competition from imports of cheaper foreign food items. Regarding cotton, the lowering of duties to the CET rate of 5 per cent for the least-processed forms of cotton and to 10 per cent for the most commonly traded forms of cotton could help lower costs of fabric and textile production in Ghana, given its dependence on imported inputs (WTO, 2014: 7).

On the other hand, tariffs in seven categories will increase, in some cases by a large margin. Tariffs on petroleum will nearly double. Similarly, tariffs on non-electrical machinery and transport equipment will nearly double. This may adversely affect the country’s efforts to upgrade its economic structure through imports of equipment and machinery and also send a negative signal to foreign investors who may have interest in establishing industries in Ghana. The three food products whose average tariffs will increase by between 25 and 57 per cent are sugar and confectionary, meat products, and fish and fishery products (World Trade Organization, 2014). The increase in tariffs on these commodities may translate into increases in their prices, either directly or indirectly through higher costs of production. This could in turn reduce the welfare of households if they happen to be net consumers of these commodities.

Table 1: Trends in Ghana's most-favoured-nation tariffs, 2007–2015 (per cent)

5 The ISIC is the international reference classification of productive activities developed and maintained by the United Nations Statistics Division. Its main purpose is to provide a set of activity categories that can be used for the collection and reporting of statistics. Details about ISIC Revision 4 can be found in United Nations (2008).

5

Categories Ghana2007

WAEMUCET 2012

Ghana2013

ECOWASCET 2015

T0 T1

6

Total 12.7 12.3 12.8 12.3 1.0 -4.0By Harmonized System category6 Agricultural products 17.5 14.9 17.3 15.6 -1.0 -10.0Animals and products thereof 19.4 18.5 19.0 23.9 -2.0 26.0Dairy products 20.0 14.4 20.0 16.0 0.0 -20.0Fruit, vegetables, and plants 18.9 17.6 18.3 17.6 -3.0 -4.0Coffee and tea 20.0 17.2 20.0 12.0 0.0 -40.0Cereals and preparations 17.8 12.7 16.2 13.5 -9.0 -17.0Oils seeds, fats, oil and their products 14.6 10.5 14.6 14.1 0.0 -3.0Sugar and confectionary 11.1 13.3 11.0 13.8 -1.0 25.0Beverages, spirits, and tobacco 19.8 19.0 19.8 17.0 0.0 -14Other agricultural products 14.4 9.4 15.1 9.5 5.0 -37Non-agricultural products 12.0 11.8 12.0 11.7 0.0 -3.0Fish and fishery products 11.1 15.5 9.8 15.4 -12.0 57.0Minerals and metals 12.2 11.8 12.5 11.7 2.0 -6.0Chemicals and photographic supplies 11.9 7.7 12.1 8.0 2.0 -34.0Wood, pulp, paper, and furniture 16.1 11.3 16.8 11.4 4.0 -32.0Textiles 16.9 16.5 16.8 16.0 -1.0 -5.0Clothing 20.0 20.0 20.0 20.8 0.0 4.0Leather, rubber, footwear, and travel goods 14.3 14.2 15.0 12.9 5.0 -14.0Non-electric machinery 2.8 7.3 3.1 7.0 11.0 126.0Electric machinery 10.3 11.3 10.6 11.2 3.0 6.0Transport equipment 6.0 11.0 5.5 10.2 -8.0 85.0Non-agricultural products n.e.s. 15.6 14.3 15.0 14.3 -4.0 -5.0Petroleum 9.0 7.9 4.3 7.9 -52.2 84By ISIC sectorAgriculture, hunting and fishing 15.7 13 15.1 11.5 -4.0 -24Mining and quarrying 11.2 5.0 11.2 5.1 0.0 -54Manufacturing 12.6 12.4 12.7 12.5 1.0 -2.0Source: Prepared by the author based on data from WTO (2014).Note: T0 and T1 are the percentage changes in tariffs for 2007–2013 (before the ECOWAS CET), and 2013–2015 (after the ECOWAS CET), respectively. WTO (2014) explains that the 2007 tariff is based on HS 2002 nomenclature consisting of 5,969 tariff lines (at the 10-digit tariff line level). The 2013 tariff is based on HS 2012 nomenclature consisting of 6,062 tariff lines (at the 10-digit tariff line level). The WAEMU tariff schedule consists of 2012 tariff rates (5,550 tariff lines at the 10-digit tariff line level) based on the HS 2007 nomenclature, while the ECOWAS tariff schedule is based on HS 2012 nomenclature consisting of 5,899 tariff lines (at the 10-digit tariff line level). According to WTO (2014), the tariff data were obtained from Ghanaian authorities. CET: common external tariff; ECOWAS: Economic Community of West African States; ISIC: International Standard Industrial Classification; n.e.s.: not elsewhere specified; WAEMU: West African Economic and Monetary Union.

6 The Harmonized Commodity Description and Coding System (HS) is a multipurpose international product nomenclature developed by the World Customs Organization. It is comprised of about 5,000 commodity groups, each identified by a six-digit code and supported by well-defined rules to achieve uniform classification. The HS contributes to the harmonization of customs and trade procedures, and the non-documentary trade data interchange in connection with such procedures, thus reducing the costs related to international trade (see http://www.wcoomd.org/en/topics/nomenclature/overview/what-is-the-harmonized-system.aspx).

7

The implications of changes in import tariffs for poverty and household welfare are important policy issues. Poverty indicators based on reports of the last four rounds of the Ghana Living Standard Survey (GLSS) show that from 1991 to 2013, poverty in the country declined considerably,7 although there were some variations across regions and across segments of the population. The distribution of the decline in poverty since 1998/1999 demonstrates it was concentrated mostly in the Central, Western, Eastern, Upper East, and Northern regions of Ghana. Households of farmers in general, the non-farm self-employed, and public sector employees enjoyed the greatest gains in their standard of living, while private sector employees and households with unemployed heads experienced the smallest gains. From a gender perspective, Figure 1 shows that, consistent with the general reduction in the poverty level, female-headed households appear to be better off than male-headed households and are increasingly less impoverished (Ghana Statistical Service, 2007).

Another important observation is that between the 1991/1992 and 1998/1999 periods, the magnitude of the reduction in poverty for male-headed households was 14 per cent compared to 12 per cent for female-headed households. In contrast, between the 1998/1999 and 2005/2006 periods, the percentage of poverty reduction was greater for female-headed households (9.9 per cent) than male-headed households (7.9 per cent), and between 2005/2006 and 2012/2013, the magnitude was greater for male-headed households (9 per cent) than female-headed households (7.7 per cent). Even though there have been significant declines in poverty levels for both male-headed and female-headed households over the past two decades – poverty at the national level has decreased by 52.5 per cent – this reduction has favoured female-headed households more than male-headed ones (54.3 per cent and 52 per cent reductions, respectively). The poverty level remained consistently lower among female-headed households than male-headed households, which is contrary to the “feminization of poverty” hypothesis. This can be partly due to the fact that over these years, Ghana’s economic growth was largely driven by the services and agricultural sectors, where the shares of women's employment are higher than the shares of men's employment.

7 However, there are two caveats in looking at these figures. First, the contribution of the various tariff reforms to this reduction in poverty remains an important policy question. This is because there have been several policy interventions, including the Livelihood Empowerment Against Poverty Programme and the Ghana School Feeding Programme. Their contribution to the reduction in poverty among households will be difficult to disentangle from the effects of the reform. Second, there is the need for a careful consideration of direct comparison of the poverty estimates of the 2012/2013 survey with the estimates of the previous four GLSS rounds. The figures regarding poverty levels may not be fully comparable because of changes in the Consumer Price Index basket and new consumer items that have been introduced onto the market, leading to changes in household consumption. Only the 2005/2006 indicators were adjusted by the Ghana Statistical Service to make them comparable to the 2012/2013 indicators (Ghana Statistical Service, 2014).

8

Figure 1: Poverty distribution in Ghana between 1991 and 2013 (per cent)

Source: Prepared by the author based on Ghana Statistical Service reports.Note: The poverty rates used in carrying out this analysis were obtained from the reports of the Ghana Statistical Service for the corresponding years available at http://www.statsghana.gov.gh/nada/index.php/catalog.

It can be deduced from the above analysis that during the period when Ghana’s tariff was frequently reformed, the incidence of poverty was reduced significantly. However, the extent of the association between the incidence of poverty and the tariff revisions requires more rigorous analysis. Even though there has been a marked reduction in poverty trends, attributing this development to the tariff reduction alone would be misleading, since other non-trade-related factors contributed to the reduction in the poverty level.

3 Gender implications of trade reforms for poverty

The recent proliferation of research on the complex relationship between trade integration, poverty, and welfare, along with the analysis of variations in the impact of trade reforms, can be traced to the mainstream trade theories and the Heckscher-Ohlin model. In a nutshell, these theories hold that production specialization according to each nation’s comparative advantage typically leads to more efficient allocation of resources in the world economy and consequently to higher levels of output and growth in all countries. This growth in turn fosters national development and reduces poverty. The diversity of empirical findings on the effect of trade liberalization on poverty proves how difficult it is to establish a relationship between the two (Le, 2014). The evidence about the distribution of the impact from trade liberalization is also mixed, with some segments of society (e.g. women versus men) coming out as winners and others losing. Adopting the methodological approach of Porto (2006) to assess the welfare impact of trade liberalization, Nicita (2009), Marchand (2012), and Casabianca (2015) observe that trade liberalization benefits households along the entire income distribution.

9

With regard to gender, women are essential in terms of food security in many sub-Saharan African countries. They are typically small farmers or food crop producers who are less able to take advantage of trade reforms than their male counterparts, who are mostly large and medium-sized producers. Since women are typically small farmers, they often lack access to credit, new technologies, and the marketing skills required to take advantage of new and expanding markets. Çagatay (2001) emphasizes that where trade liberalization has provided women with more job opportunities, it has done so at the expenses of job quality, as their “competitive advantage” as workers has been in their relatively lower wages and acceptance of inferior working conditions. Much of these employment effects have been realized in state-promoted export processing zones, which are exempt from local labour laws, and in the informal sector, where work is characterized by long hours, insecure employment, unhealthy working conditions, and low wages. A study by UNCTAD (2011) reveals that although there are only marginal differences between the poverty impact of trade liberalization on female- and male-headed households, the impact tends to be larger for female-headed households. Country-specific case studies suggest that trade liberalization in predominantly agricultural economies may disadvantage female-headed households compared to male-headed households, even when traditional export crop production expands (Cockburn et al., 2008; Chandra et al., 2010).

Some studies applied the CGE model approach. For example, Annabi et al. (2005) use it to draw conclusions on the distributional impact of trade liberalization on welfare and poverty in Senegal. The authors assert that in the short run, poverty and inequality would decrease, but in the long run income distribution would worsen. According to Thurlow (2006), trade-induced growth benefited male and female workers in South Africa. However, male-headed households stood a better chance of benefiting more from rising factor incomes. The author indicated that trade reforms contributed to an observed decline in the gender wage gap due to rising employment among higher-skilled female workers. In Ghana, no specific study has focused on the impact of trade reforms from the gender perspective. Bhasin and Obeng (2007), Bhasin (2012), and Quartey et al. (2013) conducted closely related studies that involved the use of CGE models to examine the impact of trade liberalization on poverty and income distribution of various categories of households. Though the authors concluded that elimination of import and export duties reduces the incidence, depth, and severity of poverty of all categories of households, the gender element was missing. The focus of the studies was on the spatial and regional distribution of poverty in the country.

This brief review of the existing literature points to the fact that trade liberalization and integration have diverse poverty and welfare effects depending on the type of household and its characteristics, the skills and gender composition of household members, and the geographical location of the household.

4 Methodology

This study applies a top-down approach by combining a macro CGE (top) model and a micro (bottom) household model (Bourguignon and Savard, 2008). Both models are integrated in a sequential fashion by taking parameters from the CGE model and feeding them into the micro model, either through behavioral microsimulations or non-parametric microsimulations, without any further interaction between the macro and micro levels. The top model is similar to standard

10

CGE models where households are aggregated into a small number of representative households. Therefore, the analysis at this stage can only be done at an aggregated level. The bottom model considers all the households in the Ghana Living Standard Survey and models their behaviour. In the top model, the macroeconomic shock is introduced to generate new equilibrium prices of commodities and factors. These new prices are introduced in the bottom model, where individual household income, consumption, factor supply, and other behavioural variables are obtained. In the case of this study, we used the CGE model to simulate the effect of tariff changes between 2013 and 2015 on commodity prices. The rationale for combining these two models is to exploit the advantages of CGE and microsimulation modelling. The approach offsets the respective drawbacks of each type of model, which are primarily the lack of general equilibrium effects in microsimulation models and the limitations arising from the representative household assumptions in CGE models. Even though the CGE model can be applied to represent a given economy and assume optimizing behaviour by economic agents such as households or consumers and firms, the microsimulation model allows a detailed analysis of who gains and who loses from a policy reform while neglecting general equilibrium effects. Therefore, the top-down approach combines the advantages of both models and enhances the analysis at the household level. Another major advantage of the top-down approach is that the analysis and modelling of household and labour market behaviours can be done separately from the economy-wide analysis. Thus there is no need to reconcile household survey data with national accounts and other macroeconomic data. The communication between the two types of models is in the form of information about prices, wages, and employment, and there is no need to reconcile data on levels (Vos and Sánchez, 2010). Another advantage of this top-down model is that macrosimulations typically provide information only about (simulated) changes in welfare distribution among aggregate household groups, hence missing many possible within-group changes. The microsimulation helps to make up for the missing detail.

The CGE model used for the macrosimulation (top model) is based on the dynamic (recursive) computable general equilibrium (DCGE) model developed by Breisinger et al. (2008). The model is an extended version of a static standard CGE model developed in the early 2000s by Löfgren et al. (2002) at the International Food Policy Research Institute (IFPRI) (Diao, 2011). The Ghana DCGE was first developed for analysing the country’s economic transformation and agricultural development in order to support the Comprehensive Africa Agricultural Development Program roundtable in Ghana. It is an economy-wide, multisectoral model that simultaneously and endogenously solves for a series of economic variables, including commodity prices (Breisinger et al., 2008).

The DCGE model, with its general equilibrium features, captures economic activities on both the demand and supply sides. On the supply side, the model defines a specific production function for each economic activity. That economic activity is either agricultural production, for which the functions are defined at the sub-national level, or non-agricultural production, which is defined at the national level. The demand side of the model is represented by a series of consumer demand functions. In the model, the system of consumer demand functions is solved by maximizing a Stone-Geary utility function in which the income elasticity does not need to be equal to one, making it quite different from the Cobb-Douglas utility function. As a result, the

11

marginal budget share for each consumer good departs from the average budget share of this good in the consumer’s total budget (Diao, 2011). The model represents a small open economy that has no influence on international markets, and it is calibrated using Ghana’s Social Accounting Matrix (SAM) for 2005. It covers 143 activity sectors, 60 commodities, 12 factors of production, and nine categories of households. The model is presented in four blocks: production and prices; institutional incomes and domestic demand equations; equilibrium conditions and macroeconomic closure; and factor accumulation and allocation equations (Bhasin and Obeng, 2007). The constituents of the factors are six categories of labour, two categories of capital and four categories of land. The households are made up of five urban households and four rural households, while the macroeconomic balances are made up of the government balance, external balance (the current account of the balance of payments, which includes the trade balance), and savings-investment balance. Effort was made to update the model to 2013, but the needed data were not available. As a result, the model was used to simulate the changes in prices from 2007 to 2013 (before the ECOWAS CET), and from 2013 to 2015, after implementation of the ECOWAS CET.

Once the changes in tariffs introduced in the CGE model and the simulated prices of goods and services were obtained, these prices were used in the microsimulation model to get the resulting changes in welfare at the household level. This approach used the two frameworks sequentially: first, the policy reform was simulated with the CGE model, then the simulated percentage changes in prices were passed down to the microsimulation model, taking into consideration the gender of the household head, as shown in Figure 2. In linking the parameters from the CGE to the microsimulation model to assess the consumption and poverty effects, we matched the commodities in the SAM with the same commodities in the household survey data, and then applied the first-order approach as described in Deaton (1989). This approach consists of calculating the share of household consumption expenditure and income (where the household is also a producer in the case of farmers) related to the commodities for each household. These shares were multiplied by the changes in prices obtained from the CGE model, and added to obtain the total change in welfare.

12

Figure 2: The top-down computable general equilibrium approach

Source: Adapted from Bourguignon and Savard (2008)..

Following Deaton (1997), the function for the net welfare effect of the changes in prices for each commodity can be specified as:

, (1)

where and are, respectively, the income and budget shares of commodity and is the

compensating variation associated with a change in the price of good . The compensation variation is the revenue that the social planner (government) would have to provide to the household to compensate for the effects of the price change. It is imperative to stress that the household can be both a consumer and a producer of the commodity. Assuming that the price increases and that the household is a net producer , the household will benefit from this

price change. On the other hand, if the household is a net consumer , then a price increase will make the household worse off. We also acknowledge that although the first-order effect methodology does not capture labour income from increased competition and the subsequent increase in wages, it can be used to explore the distributional effects of the changes in price across the entire income distribution, or over specific regions or groups within the population, such as urban versus rural households, poor versus non-poor households, and male- versus female-headed households. In this study, the household level distributional analysis was conducted using non-parametric regressions of the welfare effect on income (in log).

13

5 Data and description of household statistics

The CGE model used in this study was built using the 2005 SAM for Ghana, which was constructed by IFPRI based on the fifth round of the Ghana Living Standard Survey.8 The model is a comprehensive dataset that encapsulates all the information contained in the national income and product accounts and the input-output table, as well as the monetary flows between institutions in the country. The SAM estimates the structure of the Ghanaian economy in 2005 and includes detailed information on 56 production sectors, six factors of production, income and expenditures of rural and urban households, the government budget, and the balance of payments (Breisinger et al., 2007). The data on tariffs (presented in Table 1), obtained from WTO (2014), were based on calculations of the WTO Secretariat using data provided by Ghanaian authorities. Table 2 shows the import structure based on the SAM.

Table 2: Ghana’s imports of selected commodities as a percentage of total importsCommodity Import

shareCommodity Import

shareMaize* 0.2 Clothing 4.5Rice 3.4 Footwear 0.9 Other grains* 0.1 Pulp and paper 0.4Other crops* 0.2 Oils* 9.6Chicken 1.5 Fuel 4.7Beef* 0.7 Fertilizers 2.6Goat* 0.2 Chemicals 6.4Other livestock* 0.4 Metals 2.7Formal food processing 8.2 Capital goods 43.9Dairy 0.2 Electricity* 0.1Meat 2.8 Other services* 4.9Textile 1.4

Source: Author’s calculations based on Ghana’s 2005 Social Accounting Matrix (SAM).Note: The structure of the SAM shows that there is not an import share for all commodities. * Though the SAM reports imports for these commodities, no import tariffs are reported.

In building the microsimulation model, the study relied on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6). The survey was designed to provide nationally and regionally representative indicators covering a broad range of topics such as education, health, employment, housing conditions, migration, tourism, poverty, household agriculture, access to financial services, and asset ownership. In order to address the needs of Savannah Accelerated Development Authority (SADA) areas and also to provide nationally representative quarterly labour force statistics, the numbers of primary sampling units and households were increased from 580 and 8,700 to 1,200 and 18,000, respectively. This represents an increase of about 107 per cent compared to the GLSS5 (Ghana Statistical Service, 2014). Table 3 describes the characteristics of the household heads in the survey data. It shows that in the household survey

8 The SAM dataset was obtained from the IFPRI website, and the GLLS6 dataset from the Ghana Statistical Service. The SAM can be downloaded from IFPRI at http://www.ifpri.org/publication/ghana-social-accounting-matrix-2005 and the GLSS6 from the Ghana Statistical Service at http://www.statsghana.gov.gh/nada/index.php/catalog/72.

14

data used for the micro-level analysis, there were 16,772 household heads. The proportion of male-headed households (69.5 per cent) was higher than that of females (30.5 per cent). The distribution of the households by place of residence (rural and urban) indicates that most of the female household heads live in urban areas.

Table 3: Distribution of households by gender of household head and place of residence

Gender and place of residence Number of households Share (per cent)Male household head 11,652 69.50Female household head 5,120 30.50Total 16,772 100.00Female household heads in rural areas 1,950 11.63Female household heads in urban areas 5,532 32.98Male household heads in rural areas 3,170 18.90Male household heads in urban areas 6,120 36.49Total 16,772 100.00

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

The distribution of expenditure per capita from a gender perspective (Figure 3) depicts an inversed U-shape, implying that expenditure increases, gets to the optimum, and declines. Figure 3 shows that male-headed households are poorer than female-headed ones because the density curve for the former shifts to the left. An inference that can be drawn from this result is that even though male-headed households may have higher income per capita than their female counterparts, female household heads spend a greater proportion of their income on commodities consumed within the household, making them appear to be richer than male household heads. This could also explain why findings of the various rounds of the GLSS have consistently showed that female household heads are better off than male household heads in terms of poverty incidence.

Figure 3: Distribution of total expenditure by gender of household head

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

15

Figure 4 compares the income and expenditure mean of households in rural and urban areas. The results show that the average income and expenditure levels of urban households are higher than those of rural households. It also shows higher expenditure per capita for male-headed households in urban areas, meaning that male household heads in urban areas spend more than female household heads and male household heads in rural areas. However, these figures are likely to be influenced by a few extremely high values, since they are averages. With regard to per capita expenditure on food, female heads in both rural and urban areas spend a greater proportion of their income on household upkeep than male heads. The picture becomes clearer with the analysis of non-food expenditure, where male heads spend more on non-food items than female heads. The implication is that female-headed households stand to gain from the reduction in the tariff on imported food items more than male-headed households. On the other hand, the reduction in the tariff on non-food items will favour male-headed households more than female-headed households. Nonetheless, the net effect of the tariff reform on both food and non-food items depends on the net position of households as either producers or consumers.

Figure 4: Mean income and expenditure by gender of household head and place of residence

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

6 Simulations and results

Following the methodology described in Section 4, we used the CGE model to simulate the changes in prices of commodities after the implementation of the ECOWAS CET. Then we introduced the resulting changes in commodity prices in the microsimulation model to simulate the changes in welfare. This section considers the gender dimension in the analysis by disaggregating the dataset based on the gender (male and female) of the household heads. In order to identify household heads by their place and region of residence, we further disaggregated the data into female-headed and male-headed households in rural and urban areas as well as across the 10 regions of the country. This allowed for the analyses of potential winners

16

and losers from the reform based on gender and from the perspective of place and region of residence.

6.1 Computable general equilibrium results

This sub-section presents the simulated results on prices from the CGE model. In simulating the changes in prices, we introduced the changes in tariffs (Table 1) as the trade shocks. The simulated prices (Table 4) involved 60 food and non-food commodities and services. The simulated results for the 2007/2013 and 2013/2015 periods show that most of the commodities whose prices decreased were non-food items. These include pulp and paper, fertilizers, chemicals, clothing, textiles, and metals. Among the 33 food items, only the price of rice decreased. This reduction may have a positive impact on households as consumers, since rice is the second most widely consumed cereal by Ghanaian households, after maize. Available statistics suggest that in 2014, Ghanaians consumed 754,698 metric tons of rice and imported 52 per cent of that. This reduction will therefore benefit consumers and may further increase demand for rice. On the flip side, the price reduction of rice can have adverse consequences on local rice production, since domestic producers may not be able to withstand foreign competition.

Table 4 shows that for those commodities that are imported and that reported import tariffs in the SAM (Table 2), changes in prices are positive or negative depending on the change in the tariff: an increase in tariffs results in higher prices and a decrease in tariffs results in lower prices, which is what we expected. For all other commodities, the changes in their prices come from an indirect effect, given the general equilibrium nature of the CGE model.

Table 4: Simulated prices of commodities from 2007 to 2015Commodity Log

Sim12007–2013

Log Sim22013–2015

Change in price (per cent)

Commodity Log Sim12007–2013

Log Sim22013–2015

Change in price (per cent)

Maize 0.444 0.445 0.064 Meat 0.035 0.041 0.577Yams 0.225 0.226 0.080 Pulp and paper 0.119 0.067 -5.195Groundnut 1.024 1.026 0.215 Fertilizers 0.093 0.061 -3.240Export vegetables 1.450 1.465 1.513 Water 0.149 0.151 0.172Chicken -0.001 0.035 3.637 Communication 0.018 0.013 -0.492Forest 0.001 0.001 0.000 Education 0.001 0.006 0.498Cocoa processing 0.001 -0.001 -0.200 Other grains 0.001 0.001 0.000Footwear 0.053 0.030 -2.303 Soya beans 0.684 0.685 0.151Diesel 0.457 0.456 -0.063 Export fruits -0.128 -0.121 0.68Capital goods 0.625 0.642 1.698 Other crops 0.985 0.984 -0.037Other nuts 0.022 0.026 0.391 Goat 0.001 0.000 -0.100Plantain 0.002 0.000 -0.200 Formal food

processing0.093 0.103 0.907

Rice 1.320 1.299 -2.025 Textile 0.103 0.093 -0.907Cocoyam -0.313 -0.313 0.000 Oil 0.001 0.001 0.000

17

Other nuts -0.499 -0.496 0.329 Chemicals -0.088 -0.102 -1.429Plantain -0.276 -0.273 0.263 Electricity 0.001 0.007 0.598Eggs 0.001 0.001 0.000 Business services 0.115 0.112 -0.268Fish 0.001 0.007 0.598 Health 0.001 0.005 0.399Dairy 0.108 0.099 -0.902 Cassava -0.728 -0.726 0.207Wood products 0.107 0.104 -0.270 Palm oil 0.913 0.918 0.480Fuel 0.001 0.001 0.000 Domestic

vegetables0.903 0.902 -0.122

Construction 0.003 0.002 -0.100 Other export crops 1.707 1.718 1.155Transport 0.025 0.023 -0.195 Other livestock 0.001 0.000 -0.100Public administration

0.001 0.002 0.100 Local food processing

0.001 -0.001 -0.200

Sorghum and millet 1.206 1.206 0.030 Clothing 0.032 0.033 0.097Cowpea 1.204 1.206 0.120 Petrol 0.491 0.490 -0.061Domestic fruits 0.808 0.81 0.134 Metal -0.666 -0.669 -0.390Cocoa 2.342 2.339 -0.241 Trade 0.154 0.154 0.086Beef 0.001 0.000 -0.100 Real estate 0.002 -0.004 -0.601Mining 0.001 0.001 0.000        

Source: Prepared by the author using the computable general equilibrium model for Ghana.Note: The variables labelled Log sim show the simulated prices of the commodities. For instance, Log Sim1 is the simulated price of the commodities in 2013. These values were used as the base values for the simulation of the prices in 2015 (Log Sim2), which represents the period in which Ghana switched from its own tariff to the ECOWAS CET. The third column (change in price) is the difference between the first two columns, the log of prices in 2013 (before the ECOWAS CET) and 2015 (after the ECOWAS CET).

6.2 Non-parametric regression results

This sub-section presents the analysis of the effects of changes in commodity prices on household welfare. The analysis is carried out for female- and male-headed households separately and also considers regional (urban and rural) and geographical disparities. Non-parametric regressions are estimated. This type of regression is useful because it does not require specific assumptions on the distribution of the data or any econometric specification of the functional form of the relationship between the variables of interest (Deaton, 1989; Calvo, 2014). In this analysis, the dependent variable is the change in welfare due to changes in prices. The explanatory variable is the log of per capita expenditure. The objective of the analysis is to assess the welfare effect of the CET on households. We divide the analysis into three steps: the welfare effect on households as consumers (through their expenditure), on households as producers (through their income), and the net welfare effect.

First, we calculate the welfare effects of implementation of the CET on households as consumers by multiplying the budget share of each consumed item by its change in price9 simulated by the CGE model (Table 4). Figure 5 shows the results of the non-parametric regression. The downward sloping curve suggests a positive welfare effect for the poor and a negative welfare

9 This corresponds to the expression in Equation 1. The negative sign indicates that an increase in price results in a decrease in welfare for households as consumers.

18

effect for the rich, indicating a pro-poor effect of implementation of the CET for households as consumers. Figure 5 also shows that implementation of the CET favours poor female-headed households more than poor male-headed households. The expected improvement in the welfare of poor households as consumers may be due to the reduction in the price of the commodities that are consumed most within these households.10 Moreover, female-headed households stand to be better off than male-headed households because of the heterogeneity of consumption within the household. The results suggest that the budget share of items whose price decreases is higher for female-headed households than for male-headed households.

Figure 5: Change in welfare of households as consumers

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

Note: The consumption shares of the commodities, used as an indicator for welfare changes due to changes in expenditure, were obtained by dividing each household’s expenditure by total household expenditure and multiplying the result by the change in price obtained from the macrosimulation (computable general equilibrium).

Figure 6 presents the change in welfare of households as consumers based on their region of residence. The curves for both rural and urban areas follow the same downward sloping shape as for the whole population, with the regression curve for female-headed households lying above the one for male-headed households, indicating a pro-poor and pro-female effect of implementation of the CET. The only exception is for very poor urban households, where female-headed households benefit less than their male counterparts. In both urban and rural areas, the welfare gap between male-headed and female-headed households is larger at the extremes of the expenditure distributions and much narrower in the middle, which can be due to a more homogeneous consumption structure across middle-income households. Figure 6: Change in welfare of households as consumers by area of residence

10 For instance, Table 4 shows that the price of rice decreases by more than 2 per cent and the budget share of rice is higher in poor households.

19

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

We now move to the analysis of the welfare effects of the CET on households as producers. As indicated in the methodology, some households are not only consumers, but also producers who earn income from producing some of the commodities analysed in this study. Figure 7 presents the result on the relationship between the change in welfare of households as producers11 (

in Equation 1) and the level of expenditure. The figure shows a positive relationship between the two variables and an overall negative change in welfare at the national level. This means that implementation of the CET will reduce the welfare of both poor and rich households, but poor households are the most disadvantaged. Male-headed households will be the most affected, while the effect on female-headed households is almost zero at all income levels. This could possibly be due to the fact that most producers are poor male-headed households, such as rural farmers, for example, for whom the prices of their products have decreased (e.g. rice or cocoa).

11 In this case, the welfare effect is given by , as shown in Equation 1. The expression has a positive sign, indicating that an increase in prices increases the welfare of producers.

20

Figure 7: Change in welfare of households as producers

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6). Note: The share of commodity income used as the measure of welfare due to changes in income was obtained by dividing the share of income obtained by households from the sale of commodities by total income of the household. The results were further multiplied by the change in price of the commodities from the macrosimulation.

Adding the regional dimension to the analysis, Figure 8 shows that households in rural areas see higher reductions in welfare than those in urban areas: the average reduction in welfare is 0.028 per cent in rural areas and 0.014 per cent in urban areas. In both rural and urban areas, male-headed households are more affected than female-headed households. These differences in welfare loss between rural and urban households as well as male-headed and female-headed households can be partly explained by the nature of the economic activities in which these households are engaged. Rural households in Ghana rely heavily on agriculture as their source of income, while urban households are largely involved in non-farm enterprises. Agriculture is the main occupation for about 51.5 per cent of households in Ghana. About 83 per cent of these households are in rural areas, and they are largely male-headed. About 3.2 million households, representing 46 per cent of all households, operate non-farm enterprises, with 52 per cent of them in urban areas. Almost half (49.5 per cent) of all businesses involve trading, while the rest involve some kind of manufacturing activity. Women operate 72 per cent of these businesses (Ghana Statistical Service, 2014).

21

Figure 8: Change in welfare of households as producers by area of residence

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

We now assess the net welfare effect of the CET on households by adding the welfare effect on households as producers and as consumers as shown in Equation 1. Figure 9 depicts the relationship between the welfare effect and household per capita expenditure. The curve resembles the welfare effect on households as consumers (Figure 5), which could be expected because the welfare effect on households as producers (Figure 7) was much smaller than the one on households as consumers. The figure shows that in the case of male-headed households, the net welfare effect will be around zero for the poor and negative for the rich. In the case of female-headed households, the net welfare effect will instead be positive for those at the lower and middle ends of the income categories, but negative for the rich.

Figure 9: Net welfare effect

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

22

In a nutshell, the main finding of this study is that implementation of the CET will lead to a decrease in prices of most items consumed by poor households, especially female-headed households, resulting in an improvement in the welfare of these households. At the same time, it will reduce the welfare of households that are net producers. This conclusion stands in contrast to the findings of a similar study conducted in Nigeria that found that implementation of the CET produced net welfare gains due to a reduction in prices of most agricultural products (Kareem, 2014). Although the results show a pro-poor and pro-female welfare effect, the obtained variations in welfare are small (less than 0.1 per cent). This can be due to the fact that some prices increase and some others decrease after implementation of the CET, resulting in a small net effect.

In order to better understand these dynamics, we further disaggregate the data to look into commodities for which prices increase and those for which prices decrease. We first analyse the relationship between the net welfare effect and the log of per capita expenditure for commodities for which prices are expected to increase. The results, presented in Figure 10, show a downward sloping curve, meaning that the increase in prices of the commodities decreases the net welfare of all households. In addition, male-headed households are affected more than female-headed households. This can be due to differences in the consumption patterns of households, also reflecting the income status and the gender of the head of the household. As expected, the obtained values are much higher than in the previous analysis (net welfare changes above 0.27 per cent in absolute value).

Figure 10: Net welfare effect: Commodities for which prices increase

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6). Note: Among the commodities for which prices will either increase or decrease, only those with consistent observations in both datasets were used for the analysis.

When we look at the commodities for which prices decrease, Figure 11 presents a similar downward sloping welfare function. This indicates that the price decrease is likely to improve the purchasing power of households and eventually improve their welfare. The expected net welfare gain is higher for poor households and for female-headed households. Changes in welfare are also higher than those obtained previously.

23

Figure 11: Net welfare effect: Commodities for which prices decrease

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

Considering the net welfare effect from the regional perspective, Figure 12 indicates that for urban areas the regression curve is almost identical to the one that considers only expenditures (Figure 5).12 This confirms the pro-poor, pro-female effect of implementation of the CET, except for the very poor, among which female-headed households benefit less than their male counterparts. In rural areas, the net welfare effect is negative for all male-headed households and lower for the richest ones. When considering only the effect on households as consumers (Figure 5, there was a positive welfare effect on the poorest households, but the negative welfare effect on households as producers (Figure 7) eliminates that small gain. In the case of rural female-headed households, on the other hand, the net welfare effect is positive for the poor and negative for the rich, as in Figure 5

12 This might be because the welfare effect on households as producers is very small in urban areas compared to the welfare effect on households as consumers (Figure 8).

24

Figure 12: Net income (welfare) by place of residence

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

This same analysis was also performed for different geographic locations in Ghana to explore the regional dynamics of the welfare effects of the CET. The analysis is structured based on the country’s 10 regions: Western, Central, Greater Accra, Volta, Eastern, Ashanti, Brong Ahafo, Northern, Upper East, and Upper West. The results (Figure A1 in the annex) reveal the same structure as that for the national level in some locations (Greater Accra, Volta, Ashanti, and Upper West). In these cases, therefore, implementation of the CET is expected to have pro-poor and pro-female effects. However, the results are different for other regions. For example, in the Western region, where oil exploitation has been under way for a little over four years, the effect is pro-female but not pro-poor, since the net welfare function first decreases, then increases, and then decreases again as per capita expenditure increases (a sort of a U-shaped curve). In the Central region, the effect of the CET is pro-poor, but not pro-female. In the Brong Ahafo region, where agriculture is the predominant occupation, the results indicate a net welfare loss for all categories of households, regardless of the income status or the gender of the household head. These variations in the net welfare effect across the 10 regions could be explained by the heterogeneity in the production and consumption structures of households.

To conclude, this analysis has shown that implementation of the CET is likely to affect households in different ways, depending on their positions as either net producers or net consumers. Other determinants are the gender of the household head, geographical location, and changes in the prices of the commodities. Overall, female-headed households stand to be better off than male-headed households. Likewise, poor households will gain, while rich households will lose marginally. The gain will favour households in coastal regions and urban areas more than those in non-coastal regions and rural areas. Moreover, the increase in commodity prices is expected to reduce welfare, while the opposite holds for the commodities for which prices are expected to decrease.

25

6 Conclusions

In 2015, ECOWAS members including Ghana agreed to implement a common external tariff in order to harmonize the tariff structure and foster regional trade and economic growth. The objective of this study has been to assess the impact of the new tariff system on prices and the resulting effect on household welfare, with particular attention to gender differences.

The descriptive analysis shows that female-headed households are relatively less poor than male-headed households, and that the former spend more on food items than the latter. The macrosimulation analysis (done through a CGE model) shows that implementation of the CET is likely to lead to mixed effects on commodity prices, given that some tariffs were scheduled to increase and others to decrease following implementation. When we introduce the changes in prices from the CGE into the microsimulation for the welfare analysis, the results reveal that implementation of the CET will have a positive consumption welfare effect on poor households, but a negative effect on rich households. Moreover, the CET will reduce the welfare of both poor and rich households as producers, with poor households being the most affected.

From a gender perspective, female-headed households will be relatively better off as consumers than their male counterparts. As producers, male-headed households will be the most affected by the reduction in welfare, while the effect on female-headed household will be almost zero. When we consider only commodities for which prices increase, there will be a reduction in household welfare. However, for commodities whose prices decrease, there will be an improvement in the welfare of households, meaning that the dominant effect is the one on households as consumers. The net welfare analysis shows that implementation of the CET will lead to a net loss for all income categories of male-headed households and rich female-headed households. However, there will be a positive effect on female-headed households in lower- and middle-income categories. Households in urban areas stand to gain relatively more than their rural counterparts. .

In order to close the welfare gaps identified, the government may consider designing compensatory policies such as income transfers that will target poor male-headed households that are producers. The analysis by geographic location identifies Brong Ahafo as the region where all categories of households lose the most, probably due to the fact that the region depends mainly on agriculture (and households are mainly net producers). In this context, the government could also streamline the Livelihood Empowerment Against Poverty Programme, which is currently being implemented to help poor male-headed households. For instance, the government could improve infrastructure in the regions where there will be a net welfare loss in order to make them more competitive.

This study used the top-down approach. Given its general equilibrium nature, this approach has the advantage of better capturing the direct and indirect effects of tariff on prices than other methodologies such as pass-throughs. However, we deem it important to add some caveats. First, the feedback effects from household behaviour are not taken into account in the CGE model, as in the case of a top-down bottom-up methodology (Bourgignon and Savard, 2008). Second, the CGE model used data from Ghana’s 2005 Social Accounting Matrix to simulate the prices for 2013. We then ran a second simulation with the changes in tariffs after CET implementation (2013–2015). Having an updated SAM for 2013 may produce more accurate results. A step

26

beyond this study would be to include production factor effects in the analysis, given that the CGE model also simulates changes in wages and capital. However, this would require additional effort to match the survey with the SAM, an analysis that was not within the scope of the present work.

27

Annex

Table A1: Summary statistics of the net welfare effect by commodity

No. Commodity No. of observations

Mean Standard deviation

Minimum Maximum

1 Cassava 5,750 0.000146 0.000828 -0.00152 0.0062 Yams 5,750 0.000062 0.000825 -0.00206 0.0083 Plantains 5,750 4.84E-05 0.000752 -0.0042 0.0074 Oils 5,750 -1.8E-05 0.000406 -0.00167 0.0055 Vegetables 5,750 -0.00016 0.000461 -0.004 0.0046 Domestic fruits 5,750 3.61E-05 0.000522 -0.00192 0.0077 Maize 6,643 0.001613 0.002319 -0.00252 0.0068 Rice 6,643 -0.00021 0.002597 -0.015 0.0132749 Cocoa beans 6,647 0.000942 0.001619 0.0000 0.00410 Processed cocoa 16,750 -6.64E-06 2.17E-05 -0.001 0.000011 Sorghum 6,643 0.000143 0.000723 -0.00222 0.00612 Groundnuts 6,643 0.00082 0.001898 -0.00144 0.00713 Goats 7,145 0.000482 0.00153 -0.00269 0.00814 Other livestock 7,145 0.000422 0.001489 -0.0065 0.00715 Palm oil 5,750 -1.3E-05 0.000738 -0.009 0.00916 Chicken 7,145 0.00215 0.009269 -0.02122 0.04217 Fishing 7,145 6.72E-05 0.000416 -0.00098 0.00318 Cocoyam 5,753 1.05E-05 0.000158 0.0000 0.00519 Wood 6,647 1.29E-06 0.000068 0.0000 0.00420 Other crops 6,647 9.09E-05 0.000661 0.0000 0.00721 Other nuts 6,647 0.000219 0.001297 0.0000 0.0122 Beef 16,750 -5.6E-05 0.000128 -0.00392 0.00023 Dairy products 16,750 -0.00039 0.000383 -0.005 0.00024 Eggs 16,750 -3.8E-05 7.14E-05 -0.0032 0.00025 Petrol 16,750 -4.1E-05 0.00016 -0.00375 0.00026 Transport 16,750 -0.00014 0.000221 -0.00463 0.00027 Other services 16,750 -4.1E-05 8.77E-05 -0.004 0.00028 Clothing 16,750 -0.00184 0.000836 -0.004 0.00029 Electricity 16,750 -0.00038 0.000597 -0.005 0.00030 Fuel 16,750 -3.3E-05 7.62E-05 -0.00095 0.00031 Furniture 16,750 0.00061 0.002368 0.000 0.03802232 Textile 16,750 0.000119 0.000172 0.000 0.00440533 Fertilizers 16,750 0.0003 0.001678 0.000 0.02426234 Footwear 16,750 0.000581 0.001044 0.000 0.01635 Formal processed food 16,750 -4.44E-07 1.86E-06 -.0000793 0.000Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

28

Figure A1: Net income share by gender and region of residence

Source: Prepared by the author based on the 2012/2013 round of the Ghana Living Standard Survey (GLSS6).

30

References

Ackah C, and Aryeetey E (2012). Globalization, trade and poverty in Ghana. International Development Research Centre. Ottawa.

Adams C (1993). For Good and Evil: The Impact of Taxes on the Course of Civilization. Rowman and Littlefield.

Annabi N, Cissé F, Cockburn J, and Decaluwe B (2005). Trade liberalisation, growth and poverty in Senegal: A dynamic microsimulation CGE model analysis. CEPII Working Paper 2005-07. Centre d’Etudes Prospectives et d’Informations Internationales. Paris.

Bhasin VK, and Annim S (2005). Impact of elimination of trade taxes on poverty and income distribution in Ghana. Revised Report submitted to the Global Development Network (January).

Bhasin VK, and Obeng C (2007). Trade liberalization, foreign aid, poverty and income distributions of households in Ghana. Ghana Policy Journal 1: 5–22.

Bird K (2004). A framework to analyse linkages between trade policy, poverty reduction and sustainable development. Paper for the Africa Trade and Poverty Programme of the Department for International Development. London.

Bourguignon F, and Savard L (2008). Distributional effects of trade reform: An integrated macro-micro model applied to the Philippines. In: The Impact of Macroeconomic Policies on Poverty and Income Distribution: Macro-micro Evaluation Techniques and Tools, Bourguignon F, Bussolo M, and Pereira da Silva A, eds. Palgrave-Macmillan. Houndmills, United Kingdom.

Breisinger C, Thurlow J, and Duncan M (2007). A 2005 social accounting matrix (SAM) for Ghana. International Food Policy Research Institute. Washington, DC.

Breisinger C, Diao X, Thurlow J, and Al-Hassan RM (2008). Agriculture for development in Ghana: New opportunities and challenges. IFPRI Discussion Paper 00784. International Food Policy Research Institute. Washington, DC.

Çagatay N (2001). Trade, Gender and Poverty. United Nations Development Programme. New York.

Calvo PA (2014). Welfare impact of wheat export restrictions in Argentina: Non-parametric analysis on urban households. In Trade Policies, Household Welfare and Poverty Alleviation: Case Studies from the Virtual Institute Academic Network. UNCTAD/DGS/2014/3. United Nations Conference on Trade and Development. New York and Geneva. Available at: http://vi.unctad.org/tap/docs/chapters/argentina.pdf.

Casabianca EJ (2015). Distributional effects of multilateral and preferential trade liberalisation: The case of Paraguay. The Journal of International Trade and Economic Development 25(1): 80–102.

Chandra AC, Lontoh LA, and Margawati A (2010). The gender implications of trade liberalization in Southeast Asia. International Institute for Sustainable Development. Geneva

Cockburn J, Decaluwé B, and Robichaud V (2008). Trade liberalization and poverty: A CGE analysis of the 1990s experience in Africa and Asia. Poverty and Economic Policy (PEP) Research Network, Inter-American Development Bank Trade and Integration Sector. Washington, DC.

De Melo J, and Laski A (2014). Reflections on adapting to the ECOWAS CET. FERDI Policy Brief B-107. Fondation pour les Études et Recherches sur le Développement International. Available at: http://archive-ouverte.unige.ch/unige:46919.

Deaton A (1989). Rice prices and income distribution in Thailand: a non-parametric analysis. The Economic Journal 99: 1–37.

Diao X (2011). Economywide impact of Avian influenza in Ghana–A dynamic computable general equilibrium (DCGE) model analysis. International Journal of Livestock Production 2(10): 145–58.

Ghana Statistical Service (2007). Pattern and trends of poverty in Ghana, 1991–2006. Ghana Statistical Service. Accra.

Ghana Statistical Service (2014). Ghana Living Standards Survey round 6 (GLSS 6): Main report. Ghana Statistical Service. Accra.

Harrison A, and Tang H (2005). Trade Liberalization: Why so Much Controversy in Growth in the 1990s: Learning from a Decade of Reforms. Edited by N. Roberto Zagha, World Bank, 2005.

Kareem OI (2014). The welfare impact in Nigeria of the ECOWAS common external tariff: A distributional effects analysis. In: Trade Policies, Household Welfare and Poverty Alleviation: Case Studies from the Virtual Institute Academic Network. UNCTAD/DGS/2014/3. United Nations Conference on Trade and Development. New York and Geneva. Available at: http://vi.unctad.org/tap/docs/chapters/nigeria.pdf.

Le MS (2014). Trade openness and household welfare within a country: A microeconomic analysis of Vietnamese households. Journal of Asian Economics 33: 56–70.

Löfgren H, Harris RL, and Robinson S (2002). A standard computable general equilibrium (CGE) model in GAMS (vol. 5). International Food Policy Research Institute. Washington, DC.

Marchand BU (2012). Tariff pass-through and the distributional effects of trade liberalization. Journal of Development Economics 99(2): 265–81.

32

Martinez MG, and Poole N (2004). The development of private fresh produce safety standards: Implications for developing Mediterranean exporting countries. Food Policy 29(3): 229–55.Nicita A (2009). The price effect of tariff liberalization: Measuring the impact on household welfare. Journal of Development Economics 89(1): 19–27.

Office of the United States Trade Representative (2014). 2014 National trade estimate report on foreign trade barriers: Available at: https://ustr.gov/sites/default/files/2014%20NTE%20Report%20on%20FTB.pdf.

Osakwe PN (2015). Transformative regionalism, trade and the challenge of poverty reduction in Africa. UNCTAD Trade and Poverty Paper Series No. 1. United Nations Conference on Trade and Development. New York and Geneva.

Porto GG (2006). Using survey data to assess the distributional effects of trade policy. Journal of International Economics 70(1): 140–60.

Quartey P, Aidam P, and Obeng CK (2013). The impact of trade liberalization on poverty in Ghana. Available at: https://www.gtap.agecon.purdue.edu/resources/download/2665.pdf.

Roquefeuil QD, Plunkett D, and Ofei F (2014). The impact of closer regional economic integration on food security in West Africa: Focus on the ECOWAS common external tariff. ECDPM Discussion Paper 154. European Centre for Development Policy Management. Maastricht.

Thurlow J (2006). Has trade liberalization in South Africa affected men and women differently? IFPRI Development Strategy and Governance Division Discussion Paper No.36. International Food Policy Research Institute. Washington, DC.

UNCTAD (2010). Virtual Institute teaching material on trade and poverty. United Nations Conference on Trade and Development. New York and Geneva.

UNCTAD (2011). Who is benefiting from trade liberalization in Cape Verde? A gender perspective. UNCTAD/OSG/2011/2. United Nations Conference on Trade and Development. New York and Geneva. Available at: http://unctad.org/en/PublicationsLibrary/osg2011d2_en.pdf.

UNCTAD (2012). Virtual Institute teaching material on contemporary issues in international macroeconomics, trade and finance. UNCTAD/GDS/2011/2. United Nations Conference on Trade and Development. New York and Geneva.

United Nations (2008). ISIC Revision 4.0. International Standard Industrial Classification of All Economic Activities. Available at: http://unstats.un.org/unsd/cr/registry/isic-4.asp (accessed on 4 March 2015).

33

Vos R, and Sánchez MV (2010). A non-parametric microsimulation approach to assess changes in inequality and poverty. DESA Working Paper No. 94. United Nations Department of Economic and Social Affairs. New York.

Winters LA, McCulloch N, and McKay A (2004). Trade liberalization and poverty: the evidence so far. Journal of Economic Literature 42(1): 72–115.

WTO (2014). Trade policy review report by the secretariat, Ghana. Report No. WT/TPR/S/298. Geneva. World Trade Organization.

34