THE IMPACT OF TRADE ON HOUSEHOLD WELFARE IN GHANA

Master thesis I, 15 hp

Master’s in Economics / Master thesis, 15 hp

Spring term 2020

THE IMPACT OF TRADE ON HOUSEHOLD WELFARE IN

GHANA

Author: Sarah Dawu Supervisor: Göran Bostedt

Abstract

The distributional impacts of trade liberalisation across households’ groups vary in its degree.

The objective of this study is to examine the effect of the Common External Tariff (CET) of the Economic Community of West African States (ECOWAS) on domestic price of agricultural goods in Ghana in the year 2017. Also, I analyse the effect of price change on the welfare of rural households from the perspective of households as both consumers and producers. The study analyses the impacts across the regions in Ghana.

The findings indicate that in the year 2017, price of agricultural goods declined due to the reduction in the ECOWAS CET. However, the tariff pass-through rate was relatively low at 14 percent. When there has been an interaction between tariff and distance (trade cost), the findings suggest that the magnitude at which tariff reduction transmits to various regions or geographical zones in Ghana differs. The transmission rate declines the farther away a region is to the harbour. The findings indicate that households as producers have experienced a marginal decrease in income due to the reduction in the price of agricultural goods. On the other hand, households as consumers have benefitted as there have been an increase in their income through the reduction in the cost of their consumption basket.

Keywords

ECOWAS CET, Ghana, Agriculture, Price, Household welfare

Table of Contents

1. Introduction ... 1

2. Trade reforms and Agriculture in Ghana ... 3

3. Previous Studies ... 9

3.1 Theoretical Background ... 11

3.2 Data ... 14

4. Methodology ... 14

4.1 Impact of trade on price of goods ... 15

4.2 The link between price and wages ... 16

4.3 Total welfare impact on households ... 17

5. Results and Analysis ... 18

6. Conclusion ... 25

References ... 26

Appendix ... 29

1 1. Introduction

Despite a robust investment-to-GDP ratio of about 25 percent, GDP growth in Africa remains well below what is needed to reach many Sustainable Development Goals (SDGs) target and keep pace with rapid population growth. (United Nations 2019). Economic integration of countries into global markets is one of the ways that offer opportunity for rapid growth. Due to this, African countries have taken several measures to promote regional integration, a major part of which is intra-regional trade. One of the measures includes the creation of regional blocks such as economic community of West African States (ECOWAS)

, Common Market for East African and Southern Africa (COMESA), Economic Community of Central African States (ECCAS), South African Customs Union (SACU), Arab Maghreb Union (AMU).

Among these regional blocs, ECOWAS is considered one of the pillar blocks of the African continent. ECOWAS was established on 28 May 1975, with the signing of the Treaty of Lagos, with its stated mission to promote economic integration across the region. A revised version of the treaty was agreed and signed on 24 July 1993 in Cotonou. Article 3 of the 1993 treaty states that: “The aims of the community are to promote cooperation and integration leading to the establishment of an economic union in West Africa in order to raise the living standards of its people, and to maintain and enhance economic stability, foster relations among member states and contribute to the progress and development of the African continent”. In order to achieve the aims, the community shall ensure among other relevant provisions; the establishment of a common market through the liberalisation of trade by the abolition, among member states of customs duties levied on imports and exports and abolition among member states, of non-tariff barriers in order to establish a free trade area at the community level. Furthermore, there will be the adoption of a common external tariff (CET) and a common trade policy vis-à-vis third countries, as well as the removal, between member states, of obstacles to the free movement of person and goods, service, and capital, and to the right of residence and establishment.

Hence, CET was inspired by this article and has been applied by the 15 ECOWAS member states.

1 Within ECOWAS, two sub-regional blocks are created: West African Economic and Monetary Union (WAEMU) and West African Monetary Zone (WAMZ). Ghana is part of WAMZ.

2 Member states include Benin, Burkina Faso, Cape Verde, The Gambia, Ghana, Guinea-Bissau, Ivory Coast, Liberia, Mali, Niger, Nigeria, Senegal, Sierra Leone, and Togo

2

Although it is generally accepted that trade liberalisation increases the aggregate welfare of a country, how these welfare gains are distributed among its population remains an important policy question. (Marchand 2012). It is important to know the manner in which trade reforms are transmitted to local levels and how this may affect households in different geographical zones.

Most often, trade policies affect domestic prices and returns to production factors (labour) which in turn have effects on the consumption and production decisions of households. (Nicita 2009, Porto 2006, Marchand 2012). This impact on goods prices and factor prices depends on a set of household characteristics such as household’s expenditure patterns and geographical location. The observable heterogeneity in consumption behaviour and income sources in different regions in Ghana is expected to lead to a different impact. Some households will benefit, others will lose and still others will not experience any kind of effect. The reason for such diverse outcomes can be explained by the mechanism through which trade policies operate. From the perspective of the household, whether trade policy matters are related to these factors. First, it depends on the influence trade policy have on the domestic market where the household operate, second, it depends on the exposure of the households to goods affected by trade. Also, household’s consumption patterns and or if the household like or dislikes for the good in question.

This study will contribute to the understanding of the impact of trade reforms on household welfare using data drawn from various sources. Importantly, the analysis focuses on the impact of the ECOWAS CET on rural households’ in Ghana in the year 2017. This study is important since a change in tariff rate may influence domestic prices. The following questions will be addressed:

• How has the ECOWAS CET impacted domestic prices?

• What is the effect of price change on rural households in Ghana?

It is important that these questions are addressed because in Ghana, rural households can be

both producers and consumers of crops at the same time. For example, a rural household that

grows maize on the farm can both sell and consume maize. In order to assess how household

fare when tariff is pass through to prices, it is important to consider a households’ net position

with respect to production and consumption. Thus, due to the dual nature of a household, it is

necessary to understand the net position of a household whether a household is a net producer

3

or net consumer. A net producer household is defined as a household for which total gross income derived from the crop exceeds total purchase. For the net producer, a decrease in the price of maize due to changes in tariff rate will hurt the household.

A net consumer household is a household for which total gross income derived from the crop is less than total purchase. In this instance, a decrease in the price will be beneficial to such a household.

The next section covers Ghana’s trade history and agriculture, which is the main activity of rural residents. Section 3 comprises of the literature of the study and data. Section 4 gives information on methodology based on micro and macro data, which enables the identification of the different channels through which trade reforms may affect household welfare. Section 5 and 6 provides results and conclusion of the study, respectively.

2. Trade reforms and Agriculture in Ghana

Ghana’s trade policy has evolved from a fairly liberal one in the 19950s through a significantly controlled regime in the 1970s, after which the economy underwent major trade and economic reforms in the 1980s to give rise to the liberalised trade policy. This happened in the arena of international trade under the General Agreement on Trade and Tariff (GATT) and World Trade Organisation (WTO), trade agreements between Ghana and major trading partners, as well as the country’s own economic development policy. Ghana’s trade policy was also influenced, especially in the 1980s and 1990s by the Bretton Woods Institutions. Trade policy under the 1980s Structural Adjustment Programme (SAP) included tariff adjustments, import liberalisation, liberalisation of foreign exchange, deregulation of domestic market prices and controls and institutional reforms that particularly affected revenue-generating bodies such as the Customs, Excise and Preventive Service (CEPS). The main objectives of the reforms were to restore incentives to produce exports and increase the overall availability of foreign exchange, and to improve foreign exchange allocation and channel it into selected, high- priority areas. (Ackah and Aryeetey, 2012)

In 1993, two programmes were established under the Ministry of Trade and Industry to enhance

the supply capabilities of exporters by assisting them with incentives. These are the USAID

sponsored Trade and Investment Programme (TIP), and the Private Enterprise and Export

4

Developments (PEED) initiative sponsored by the World Bank. The TIP was to help eliminate the various obstacles to export expansion in order to achieve accelerated export growth. It lasted for five years. Following the end of TIP in 1999, USAID initiated the Trade and Investment Reform Programme, (TIRP). Under TIRP, all restrictions on non-traditional exports were removed, paving the way for tremendous increase in the value of non-traditional exports. In February 2004, a new trade policy document was adopted. It was set within the context of Ghana’s strategic vision of achieving middle-income status by 2012 and becoming a leading agro-industrial country in Africa. The policy provided guidelines for the implementation of the government’s domestic and international trade agenda. It was also designed to ensure a consistent and stable policy environment where consumers could operate effectively. The policy placed emphasis on two parallel strategies: an export led industrialisation strategy and a domestic-led industrialisation based on import competition.

These strategies were supported through the promotion of increased competitiveness of local producers in domestic and international market based on fair and equal competition. It introduced an import and domestic trade regime, which promoted domestic trade regime. The objective of the policy was to facilitate trade, which incorporated customs and clearance, and related issues and focused on tariff and non-tariff measures. This was followed by the Trade and Investment Programme for a Competitive Export Economy: (TIPCEE 2004-2009) also aimed at creating an enabling environment for further growth in non-traditional exports. One of the aims of TIPCEE was to provide technical assistance to smallholder farmer groups in the horticultural sector and businesses involved in agro-processing for export. (Ackah and Aryeetey, 2012).

Over the years, the agricultural sector has been one of the most important sectors that drive the growth of Ghana’s economy. The sector provide food, raw materials for the industries and generates income for households from production to the sales of products. It is estimated that smallholder farm households do more than 80 percent of agricultural activities. (Danso- Abbeam, Armed and Baidoo, 2014). Furthermore, over half (51.5 percent) of households in Ghana own or operate a farm. Farming is mostly rural, engaging about 83 percent of rural households.

Among the major cereal foods found in Ghana are maize, rice, millet, wheat, and sorghum. These goods, if not all are traded within Africa. See Table 1 and Figure 1 for a

3 Source: Ghana living standard survey 6 report. Agricultural operations are common in rural savannah with about 93 percent of households involved. Forest and rural coastal are 81.2 and 65.4 percent, respectively.

Consumption of own products take place mostly in rural households

5

description of Ghana’s share from ECOWAS. As Table 1 shows, for rice, wheat and to a certain extent dry onions imports are important relative to domestic production. Overall, as Figure 2 shows, food products are only 12.34 percent of total imports.

Rice has expressly been identified in the Food and Agricultural Sector Development Policy (FASDEP) as an important food crop that should be given special attention for food sufficiency.

(MOFA). In this context, the Rice Sector Project was implemented by the Ministry of Food and Agriculture in collaboration with Agence Francaise de Development (AFD) of France.

This project was in line with MOFA’s strategy to facilitate the production of food crops to attain food-sufficiency, output processing and marketing systems. It was to support lowland rice production of up to 6,000 hectares in the Northern, Upper East, Upper West Region and northern parts of the Volta Region of Ghana and to achieve a production 16,250 metric tons of milled rice annually by the project completion date. The overall objective was to improve livelihoods of poor farmers in the targeted regions, where 2,500 farm families (or 15,000 persons) were to be involved in at the production level. Moreover, 1,000 actors (equivalent to about 5,000 persons) were also to be part of the processing and marketing aspects.

Other food crops of equal relevance in terms of domestic production are cassava, yam, plantain, and cocoyam. Between the years 2007 and 2014, the International Food for Agricultural Development (IFAD) and the Government of Ghana (GoG) funded the Root and Tuber Improvement and Marketing Programme (RTIMP), which was to enhance income and food security in order to improve livelihoods of the rural poor. Its interventions focused especially on improving the outputs of small-scale root and tuber farmers, processors and traders, particularly women. RTIMP operated integrated activities on over 80 districts where it built a competitive market-based Root and Tuber Commodity Chain (RTCC) supported by relevant, effective, and sustainable services that are available to the rural poor. During the period, sixteen commodity chain segments were identified and supported to increase trade volumes as well as initiating partnership deals with Freshmacs Ltd (as off-taker) to cultivate Beauregard, (orange fresh variety) sweet potato for export. Most of these foods, if not all, are traded within the ECOWAS region.

The ECOWAS CET was adopted at a Heads of State Summit in October 2013 in Dakar. It

represents the most-favoured-nation (MFN) rates ECOWAS applies in relation to non –

member countries. The implementation of CET by ECOWAS is to ensure that countries

converge towards a single tariff rate in order to ensure equal contribution from member states

6

to regional integration efforts. ECOWAS CET is made up of five tariff bands where imported goods are taxed based on the category to which they belong. As such, essential social goods (medicine) classed in the zero category are taxed 0 percent, while essential commodities, basic raw material, capital goods, specific inputs classed in the first category, are taxed 5 percent.

Inputs and intermediate products (classed into the second category) are taxed at 10 percent.

Final consumer goods (classed in the third category) and specific goods for economic development (classed in the fourth category) are taxed 20 percent and 35 percent, respectively.

Before the adoption of the ECOWAS CET, Ghana’s MFN tariff was modified frequently between 2008 and 2013, which it averaged 12.8 percent. It consists of 0 percent, 5 percent, 10 percent, and 20 percent bands, the last two applicable to more than 40 percent of tariff lines.

Following Ghana’s adoption of the Harmonised System (HS) 2012 tariff classification, the share of the 2013 tariff that is bound under the GATT/WTO increased from 14.7 percent to 16.3 percent of total tariff lines, but it remains rather low. Ghana also bound other duties and charges (ODCs) at zero on non-agricultural and most agricultural goods, but in reality, numerous ODCs are applied on all imports, and their number and incidence have increased.

Meanwhile, widespread tariff and other import duty exemptions schemes available in Ghana’s statutes effectively allow many operators not to pay customs duties, ODCs, VAT or other taxes.

(WT/TPR/S/298 • Ghana). About 25 percent of the otherwise payable import duties is forgone annually in exemptions of various sorts. This reflects inconsistencies between Ghana’s economic structure and its tariff regime. The application of the ECOWAS CET could stabilise Ghana’s tariff regime and address some of these inconsistencies.

As an example of Ghana’s most favourable tariff rate, Table 2 shows tariff rates on certain products charged by Ghana to Nigeria as a trading partner.

4 Source:

https://www.wto.org/english/tratop_e/tpr_e/tp398_e.htm

7

Table 1: Ghana’s import 2010-2017 of certain important agricultural products.

Year 2010 2012 2013 2014 2016 2017

Rice DP

491603 481134 569524 604000 687679 722080 import 320143 400316 644328 276139 698396 819746

export 6 1 0 1430 83 210

millet DP 218952 179684 155131 155000 159017 163484

import 203 79 * 155 5420 940

export 5 1 4 29 2 3

oats DP 37 45 45 40 47 49

import 698 958 923 153 521 23

export 0 0 0 NA 0 NA

maize DP 187169

5

1949897 176447 7

1762000 172191 0

2011179

import 955 113213 3172 2641 72054 40661

export 8720 1253 3406 786 693 3975

wheat DP NA

NA NA NA NA NA

import 315838 190109 261255 445338 527141 1092937

export 0 0 0 0 0 0

cassava DP 135040 86

1454727 9

159899 40

1779821 7

177982 17

1900872 5

import 425 349 333 697 22 183

export 0 0 1091 2003 627 2027

Dry beans

DP 199266 NA NA 152769 143216 170490

import 7455 23243 10718 114 10387 8511

export 101 * * 64 88 65

Dry onions

DP 100000 130000 138188 143982 * 144389

import 51009 72146 56288 10753 69362 53204

export 2805 312 584 6 197 297

Source: FAOSTATS, Food and Agriculture Organisation of the United Nations.

5Indicates domestic production.

6 Values are not available. Notes: all values except import and export values of cassava are in tonnes. Cassava is reported in US dollar (1000). There are other agricultural products traded within and outside the ECOWAS region. For more information, see FAOSTATS.

8

Figure 1: Ghana’s import share share (%) from ECOWAS.

Source: Prepared by author based on data from World Bank World Integrated Trade Solution (WITS).

Table 2: Ghana’s Most Favourable Nation tariff rate to an ECOWAS member state using Nigeria as an example.

Goods Year Total Imports

(thousands of USD)

Weighted Average tariff (percent)

Food products 2010 8637.47 19.83

2013 21806.6 16.94

2017 55526.63 20.61

Animals 2010 573.56 20

2013 1204.68 18.3

2017 7753.84 20.74

Vegetables 2010 19.81 7.61

2013 807.26 13.74

2017 14021.36 13.31

Source: World Bank World Integrated Trade Solution.

9 3. Previous Studies

Trade liberalisation makes some individuals better off and others worse off and this income redistribution can be systematically biased in favour of or against poor individuals. Reductions in tariffs and non-tariff barriers, both at home and abroad, affect consumer and producer prices, which in turn affect household production, household consumption, labour earnings, and transfers. Since the poor and the rich generally consume different bundles and have different sources of income, trade policy changes will affect them differently. (Porto, 2014)

Deaton (1989, 1997) showed how to use budget shares to measure consumption effects when there are barriers to international trade. With respect to income, lower trade barriers cause changes in factor demands that lead to changes in wages, agro-input prices, and household income. The net effect would depend on how much is spent on agro-consumption goods and by how much, the household involves in formal labour markets and in the agricultural sector.

He explained the net benefit ratio as the ratio of household’s net sales of a commodity to household income and used the 1981-2 Socioeconomic Survey of Thailand to examine the effects of rice prices on the distribution of real income across different households. In addition, he described the consumption and production patterns for rice and further related the patterns to households’ living standards and geographical location. He found that higher prices provided direct benefit to rural households at all levels of living, but the main beneficiaries were neither the poorest rural households nor the richest households. Rather, households in the middle of the rural income distribution benefitted.

Porto (2006) developed a methodology to estimate the effects of trade policies on the

distribution of income in developing countries. He used Argentinean households, survey data

to assess the impacts of tariff reforms on household welfare in the 1990s. His methodology was

based on two links; first, he connected trade policies to prices and second, connected prices to

household welfare. The reason being that in small open economies, trade reforms cause prices

of traded goods to change and the price changes tend to affect households as consumers and as

income earners. Porto estimated the direct response of the prices of traded goods and wages to

tariff changes, as well as the endogenous response of non-traded goods, all as separate

components of household welfare. He further showed a distributional analysis which was based

on the estimation of the average compensating variation at different levels of per capita

household expenditure. This analysis was performed across the entire distribution of income

10

of Argentinean households. His findings indicated that the regional trade agreement Mercosur,

has had pro-poor distributional effects and as such, tariff removals would tend to benefit the poor over the rich. Although there has been increase in poverty and income inequality during the 1990s in Argentina, Porto’s findings indicated trade was not responsible for that.

Following the work of Porto (2006), Nicita (2009) analysed the distributive effects of tariff liberalisation in Mexico during the period of 1990-2000. He expanded on Porto’s methodology by linking border measures to domestic prices of goods. He found that Mexican tariff liberalisation reduced consumers’ prices both for agricultural and manufacturing products and increased the wage gap between skilled and unskilled labourers. The results showed urban areas, as well as Mexican states along the United States border experienced higher welfare gains. There was an overall positive effect of tariff liberalisation with significant differences across income levels and geographic regions. Marchand (2012), estimated the price transmission mechanisms in Indian urban and rural households’ separately in order to understand the extent to which trade reforms in the 1988 to 2000 affected domestic prices. Her findings suggested that trade policy is not perfectly transmitted to consumers. Market imperfections and trade cost partially isolate households. Moreover, urban markets can transmit prices with a higher elasticity relative to the rural markets and as such, a higher welfare gains for urban areas at all levels of per capita expenditure. There was a higher gain for poorer households in both rural and urban areas due to their high expenditure share of traded goods.

Several studies have examined the impact of trade policy on welfare in Africa. According to Ackah and Aryeetey’s book on globalisation, trade and poverty in Ghana (Ackah and Aryeetey 2012), trade liberalisation has had both positive and negative medium-term impacts on different categories of Ghanaian households. Households engaging in exporting activities as well as skilled workers benefitted, some poor and unskilled workers appeared to have suffered, at least in the short to medium term. They concluded that, the overall impact of trade liberalisation appeared to have been positive in Ghana and to have contributed to poverty reduction over the longer term. Some have used computable general equilibrium (CGE) models to simulate tariff reductions and conclude that liberalisation lead to aggregate welfare gains but that urban and rural population are affected differently.

Despite the significant body of literature examining the distributional impact of trade liberalisation, very few studies evaluate the trade liberalisation policy of the ECOWAS CET.

7 Includes Argentina, Brazil, Paraguay and Uruguay

11

Orkoh (2018) examined the effects of implementing CET on households’ in Ghana with a special focus on gender. Kareem (2014) analysed the impact of ECOWAS CET on Nigeria’s households through an ex-post econometric analysis using household survey data. He found that, ECOWAS CET had reduced prices of agricultural goods. While he did not find any significant link between wages and prices, his findings indicated a net positive effect on households in Nigeria due to gains from expenditure which outweighed the loses in the households’ purchasing power due to lower income from sales of agricultural products. Ajayi and Osafo-Kwaako (2007), Balogun and Dauda (2012) and Nwafor (2006) looked only at the macroeconomic impact of CET on Nigeria.

Some studies examined the potential distributional impact through ex-ante analyses: Fiamohe et al (2015) used CGE model to examine the potential impact of adoption of EOWAS CET on the rice sector in six ECOWAS countries, which included Ghana. Their findings indicated that CET would reduce both rural and urban poverty in Ghana. Simulation showed 0.35 percent of national poverty reduction which results from Ghana rural household’s poverty reduction by 0.43 percent 0.21 percent urban households’ reduction in poverty. Their findings suggested ECOWAS CET would likely generate a very weak welfare effects in Ghana. They concluded that the CET would have various effects on rice economies which will increase intra-regional trade. Nwafor et al (2005) also examined the potential impact on rural and urban poverty. Their findings indicated that in the first year of ECOWAS trade, there will be a negative effect which will be stronger in rural areas. However, ECOWAS trade regime will experience a positive effect over time which will increase in strength.

3.1 Theoretical Background

Tariffs are tax levied on imports, which could take an ad valorem or specific rate form. It has

the power to change the consumption, production and trade patterns of the domestic country

with respect to world trade. It is a tool that influence the price wedge between the domestic

price and foreign price of goods. Changes in tariffs of imported goods will have different effects

throughout the economy such as effects on government revenue and effects on household

income. For instance, a reduction in a tariff will affect the domestic price of the good and its

relative price in the economy. The change in the relative price will create disequilibrium in the

factor market, thereby resulting in factor income adjustment. Farm workers could lose their

12

jobs or producers who are unable to compete with relatively lower prices would exit the market.

This could affect wages. Changes in prices and wages will induce changes in household disposable income and ultimately household welfare. A household welfare is a function of prices of all goods that the household faces and income. The effect of a single price change on household welfare is proportional to its net supply position in that good expressed at current prices as a proportion of total expenditure. A graphical representation of a reduction of the ECOWAS Common External Tariff can be found in figure 2a. From the figure, a decrease in tariff rate from Ghana tariff to ECOWAS CET decreases price from P1 to P2. This intend increase consumer demand from QD1 to QD2 whiles producer supply is decreased from QS1 to QS2.

Hence, a decrease in the tariff rate will decrease the price of commodity which in tend will cause a change in rural household budget. However, other factors could affect the price change.

For an imported good, its world price, the tariff it faces, and the exchange rate combines to

define its price after tariff liberalisation. When a good arrives at the domestic country’s port, it

passes through channels before it arrives at local markets (see figure 2b). It encounters other

cost especially cost of transporting the good to major distribution centres. Before the good

reaches’ households and individuals, it has accrued other cost which affect the magnitude of

tariff liberalisation.

13

Figure 2a Effects of tariff on imports and domestic production.

Source: Prepared by author.

Figure 2b: Various channels of tariff liberalisation pass-through.

Source: Prepared by author.

Every household is affected by the change in the price of a good. The extent to which household consumption is affected by price changes depend on the expenditure shares of the good and the tariff reduction for the good. Using the patterns of households’ supply and demand, it could be seen that there is a decrease in supply of goods for households as producers and on the other hand, households as consumers has seen an increase in the demand of their consumption basket.

The welfare impact of price changes on households can be measured in monetary terms by

using money metric indirect utility function. With the household budget change, the question

in line is how much money (positive or negative) the household would require in order to

maintain its previous level of living. This is known as the Compensation Variation (CV)

The question is, will a decrease in the tariff rate of goods be beneficial or detrimental to

households? If the tariff is detrimental, can we identify the most vulnerable segments of the

Ghanaian population. The study focuses on Agricultural goods, using rice, maize and meat as

proxy of agricultural goods. These goods are common in every household in Ghana, its

availability makes it effective in the line of study.

14 3.2 Data

To be able to estimate the impact of trade on different types of households, household survey data are needed. This study uses data from survey conducted by the Ghana Statistical Service (GSS). The latest round (the Ghana Living Standard Survey, GLSS 7) was conducted in October 2016 and ended on September 2017. It provides detailed information on patterns of households’ consumption and expenditure at a lower level of disaggregation. The household questionnaire contains modules covering information on demographic characteristics of household members, education, health, employment, migration, housing, agriculture sources of income and expenditure. A two-stage stratified sampling design was adopted. At first stage, 1,200 Enumeration Areas (EAs) were selected to form the primary sampling units (PSUs). The PSUs were allocated into 10 regions using probability proportional to population size. The EAs further divided into urban and rural localities of residence. A complete listing of households in the selected PSUs was undertaken to form the secondary sampling units (SSUs). At the second stage, 15 households from each PSU were selected systematically. Hence, the total sample size came to of 14,010 households nationwide. Of the 14,010 households, 7,991 were rural households while the remaining 6,019 were from urban households. Tariff data were sourced from the World Bank World Integrated Trade Solution (WITS) database, while the world prices were taken from the World Bank Commodities Price data at nominal USD per kilo of the commodity. The value of the commodity was converted into Ghana Cedis at the prevailing 2017 exchange rate for the purpose of the study. The prices cover agricultural items such as maize, rice, meat, etc. The values of distance used as proxy for trade cost were calculated from the Global feed distance calculator website.

Distance in kilometres was calculated from either the Takoradi or the Tema harbour to the capital town of each regions.

4. Methodology

The theoretical framework for this study follows the approach of Porto (2006) and, Nicita (2009) to estimate the impact of trade on the welfare of rural households in Ghana. The model

8 Source: https://distancecalculator.globefeed.com/Country_Distance_Calculator.asp?more=1

15

focuses on the effect of changes on domestic price that could be attributed to changes in the ECOWAS CET.

4.1 Impact of trade on price of goods

In order to estimate the effect of trade on prices, it is important to know the channels through which trade policies affect prices.

• Direct transmission into prices of imported goods through border tax imposition

• Indirect transmission into prices of domestic goods through changes in the domestic producers’ mark-up, due to competition between domestic producers’ and imports.

In an instance where domestic goods dominate the domestic market, i.e. when foreign goods are price takers, trade policy would have no effect on the good’s market price. However, when imports fully dominate the domestic goods i.e. when domestic goods are price takers, the trade policy would have the greatest impacts. This implies that the impact of tariffs on domestic prices depends not only on the tariff change but also on the extent to which the change in price is absorbed by foreign or domestic mark-ups.

𝑃

= 𝑃𝑃

(𝑃𝐼

(1 + 𝑇

)𝑇𝐶

)

(1) where market price encountered by households for consuming a good x at time t in region r (𝑃

) is a function of producer prices (𝑃𝑃

), the international price in local currency (𝑃𝐼

), the tariff level (𝑇

) and the trade cost

𝑇𝐶

. Parameter 𝛼, which takes the values of 0 to 1, is a measure of the extent to which domestic goods dominate over imported ones and (1 − 𝛼) is the pass-through, which indicates the extent to which tariffs, international prices, and trade cost affect domestic prices. When 𝛼 = 0, it implies there is a complete pass-through and the full extent of change in border price is transmitted to consumer prices. However, if 𝛼 = 1, then the pass-through is zero, implying a movement in border price has no effect on prices paid by consumers.

9 Cost incurred in transporting and marketing goods. It tends to affect imported goods relative to local varieties. This means that consumer prices at the local level depend not only on border prices and tariffs (which are uniform across local markets), but also on local producer prices and transport cost (which are heterogeneous across local markets).

16

In this the model, it is assumed that: consumer goods are not differentiated at by origin and that their prices are at average levels of imported and domestic producer prices. Moreover, trade cost only affects imported goods.

Equation (1) can be rewritten in logs form as:

ln 𝑃

= 𝛼 ln 𝑃𝑃

+ (1 − 𝛼) ln 𝑃𝐼

+ (1 − 𝛼) ln(1 + 𝑇

) + (1 − 𝛼) ln 𝑇𝐶

(2) Following Nicita (2009), and Goldberg and Knetter (1997), the unrestricted form of equation (2) is assumed, as such, the set of coefficients (1-𝛼) can differ both in sign and in magnitude.

ln 𝑃

= 𝛽

+ 𝛽

ln 𝑃𝑃

+ 𝛽

ln 𝑃𝐼

+ 𝛽

ln 𝑇𝐶

+ 𝛾 ln(1 + 𝑇

) + 𝜀

(3) In equation (3), the shortest distance (in kilometres) to the nearest main port entry is used as proxy for trade cost. In this case, the Tema and Takoradi harbours are used. With the use of distance, it is assumed that trade cost is constant in the time period of the analysis. Because the study’s objective is to estimate the effect of tariff liberalisation on domestic prices at the regional level, an interaction term between distance and the tariff rate is included in the final equation. The reason is to isolate the domestic impact of tariff changes on the pass-through.

The estimating equation is given by:

ln 𝑃

= 𝛽

+ 𝛽

ln 𝑃𝑃

+ 𝛽

𝑇𝐶

+ 𝛽

ln 𝑃𝐼

+ 𝛾 ln(1 + 𝑇

) + 𝛾

ln(1 + 𝑇

)𝑇𝐶

+ 𝛾

(ln(1 + 𝑇

)𝑇𝐶)

+ 𝜀

(4)

Where 𝛾 represent the tariff elasticity of the pass-through; that is the percentage increase in domestic price resulting from a one percent increase in the tariff. If 𝛾 = 1, the pass-through is complete, which implies that all changes in tariff is transmitted into retail price and if 𝛾 < 1, the pass-through is incomplete. 𝛾

and 𝛾

capture the effect of trade cost on the tariff pass- through. In an instance where trade cost has an identical effect across all geographic areas, 𝛾

= 𝛾

= 0. On the other hand, if local prices vary depending on distance from the harbour, then, 𝛾

≠ 0 and or 𝛾

≠ 0

4.2 The link between price and wages

This equation follows the framework proposed by Nicita (2009) to measure trade liberalisation

on labour. It is well known that households receive part of their income by selling their labour,

therefore if trade liberalisation passes through a domestic price, the factor market will be in

disequilibrium, thereby resulting in factor income adjustment. In connecting price and wages,

17

the wage-price elasticities are estimated. The equation takes into consideration the role of product prices and worker characteristics.

ln 𝑊

= ∑ 𝛳

𝑥,𝑟,𝑠

ln 𝑃

ẟ

+ 𝑙

+ 𝐻

+ 𝜀

(5) where 𝑊

is the observed wage of individual j in time t in region r, 𝑃

is the price of good x, 𝑙

is a vector of individual characteristics, and 𝐻

is a vector of household characteristics, 𝛳

and 𝜀

represent a dummy variable for worker skill and an error term respectively. The wage responses to price ẟ

are limited to vary between skilled and unskilled workers.

Despite the many relevant information in the household data, it does not provide information on individual wages. Due to this, this study will only focus on the changes in price on household welfare.

4.3 Total welfare impact on households

As stated earlier, in Ghana, most rural households are simultaneously producers and consumers of same crop. Thus, sales from agricultural products often constitutes a larger part of household income. For this reason, in estimating the overall impact of trade on household welfare, a farm household model which considers the household both as producers and consumers of goods is used. The indirect utility function of the household is written as:

𝑈

= 𝑉

[𝑌

, 𝑃] (6)

where household utility 𝑈

is expressed as a function of a vector of prices P faced by the

household and the household’s income 𝑌

which includes the income from farm activity and

non-farm activity. The changes in the utility 𝑑𝑢

of household h in region r depends on the

changes in domestic prices (of goods and factors) and on household’s specific labour income,

agricultural production, and consumption.

18 𝑑𝑢

∑ 𝛳

𝑑𝑤

+ ∑ 𝛳

𝑥 𝑠

𝑑𝑝

− ∑ 𝛳

𝑥

𝑑𝑝

(7)

where 𝛳

is the share of household h’s income spend on good x, 𝛳

is the share of income received from selling good x at price P by household h and 𝛳

is the share of income earned from selling labour by household h. 𝑑𝑝

and 𝑑𝑤

are changes in prices and wages

respectively. Both changes in prices and wages are expressed in percentages.

It is possible to analyse the effect of price changes on rural households’ welfare after using equation 4 to estimate the effect of tariff liberalisation on good prices. The empirical approach used in this study follows Nicita (2009).

5. Results and Analysis

In estimating the results in table 3, equation 4 was used. Three commodities, maize, rice and meat are used. Average prices per kilo of these goods are sourced from the Ghana Living Standard Survey 7 (GLSS7) and are matched to the prices of world substitutes and their respective tariff rates. The world price is converted into domestic price (Ghana Cedis) at the 2017 exchange rate. It should be noted that the nominal values of price in GLSS7 are used as both consumer and producer prices. As the supply of agricultural food products in a raw state produced in Ghana are exempted from Value Added Tax (VAT).

The table presents a regression result on consumer price as the dependent variable and producer price, world price, tariff rate as independent variables. It is an estimate of a model taking into consideration trade cost (i.e. the tariff distance interaction) as an independent variable.

10 Is derived by the total differentiation of equation (6) and the application of Roy’s identity. Nicita (2009)

19

Table 3: Regression results of tariff liberalisation on consumer price in Ghana.

Dependent Variable - Log Price

Producer price 0.913***

(0.0213)

World price 0.0259

(0.0136)

Tariffs 0.144***

(0.0376) Tariff*Distance 0.000102***

(0.0000) Tariff*Distance squared -5.32e-08***

(0.0000)

constant -0.364***

(0.1101) Number of Observations 878

All variables are in log and standard errors corrected for heteroscedasticity are presented in parenthesis. Variables that are significant are denoted by ***, **, and * at 1, 5 and 10 percent significant levels, respectively.

The result indicates that 1 percent reduction in ECOWAS CET reduces domestic consumer price by 14 percent on agricultural goods. This result is low as compared to Kareem (2012) who had 74 percent on agricultural goods in Nigeria. The reason for the low result could be that consumers in Ghana tend to prefer locally produced agricultural goods to imported varieties, which leads to a larger supply of local goods in the market. As Nicita (2009) states in his empirical study, “the magnitude to which trade policies affect the price paid by consumers depends ultimately upon the level of integration of local markets into the economy.”

From equation 4, there has been reduction in domestic price of agricultural goods due to tariff

reduction which agrees with the theory in Figure 2a. Furthermore, the findings indicate that

transmission of the reduction rate differs across regions in Ghana. The coefficient of the tariff

distance interaction is statistically significant; it gives the change in tariff pass through per

20

every kilometre of distance from the harbour. The coefficient of the tariff distance and tariff distance squared indicates that trade cost affects domestic price. In this study, when the distance variable is added, the pass through is found to decline with distance from the Tema and Takoradi harbours to the other geographical areas. The coefficients of the tariff distance and tariff distance square differs, which implies price of agricultural goods differs across regions in Ghana. The tariff distance square coefficient suggests that the effect declines on the margin.

This could explain that change in trade policy is not perfectly transmitted to the consumer.

Market imperfections may partially isolate households from tariff changes in geographical area. The world price variable has had no significant impact on agricultural goods in the year of this study.

Table 4a: Regression results on household welfare.

Dependent Variable - Welfare

National Economy Rural Area

Household Wage 0.101***

(0.0040)

0.0827***

(0.0046)

Agric Income -0.0544***

(0.0053)

-0.0254***

(0.0044)

Consumption 0.415***

(0.0095)

0.296***

(0.0087)

All standard errors are in parenthesis. Variables that are significant are denoted by ***, **, and * at 1, 5 and 10 percent significant levels, respectively.

As expressed in section 3.1, changes in price changes welfare of households but the change

can affect each household differently. The theory suggests that a decrease in price increase

demand by consumers and decrease supply by producers. (see figure 2a). The statistical results

in table 4a is obtained from equation 7. The results show the percentage change in aggregate

welfare. The report is on the overall consumers’ welfare gains, producers’ welfare loses, and

21

household wages in the whole economy of Ghana and in the rural households in Ghana. Per the focus of this study, I discuss only results on consumers and producers in rural households in Ghana.

Table 4a reports that 1 percent increase in consumption significantly increases consumers’

welfare by 29.6 percent. It also indicates that 1 percent decrease in Agricultural income of producers decrease producer’s welfare by 2.5 percent.

In the welfare effect from the perspective of consumers, households who spend the greater part of their income on food have benefitted more. Figure 3b shows that rural households in all the regions except Greater Accra spend a greater share of their income on food. Because of the greater share of agricultural goods in their expenditure basket, these households have had an increase in their real income due to reduction in the price of agricultural goods through the reduction in the ECOWAS CET. Of course, because Takoradi harbour is in Western region, and the proximity of Central and Eastern regions to Takoradi and Tema harbours respectively, consumers in these three regions also, have seen greater share in increase in their income, due to reduction in their expenditure basket.

Table 4b shows a household’s real income change using results from Table 4a. It could be seen

that the overall change in income for all regions is positive. Despite the decline in agricultural

income, the benefit from decrease in expenditure due to lower price is large enough to offset

the decrease in agricultural income. Some households in regions farther away from harbour

like Brong Ahafo region and Northern part of Ghana have seen a greater decline in agricultural

income for producers as well as well a decline in the cost of the consumption basket for

consumers. The reason could be that these regions operate within their own market and do not

have any impact from ECOWAS CET or there is a different entry point of trade like the

northern border with Burkina Faso and Brong Ahafo border with Cote d’Ivoire which is being

operated as an entry points for trade.

22 Table 4b: Changes in Income by region.

Region Changes in real income due to change in price Overall changes in real income Agricultural sales (GHS and %) Consumption expenditure (%)

(GHS and %)

Ashanti -428.3 -0.2 10008.1 2.9 2.7

Brong Ahafo -914.3 -0.43 10306.9 3.1 2.67

Central -294.2 -0.14 10154.3 3 2.86

Eastern -774. -0.37 11841.2 3.5 3.13

Greater Accra

-188.4 -0.09 2345.2 0.7 0.61

Noerthern -823.8 -0.4 13339.1 4 3.6

Upper East -379.2 -0.17 9999.2 3 2.83

Upper West -340.6 -0.16 7446.6 2.2 2.04

Volta -570.2 -0.27 11015.1 3.3 3.03

Western -645.2 -0.31 12573.2 3.8 3.49

Total -2.54 29.5 26.96

Source: Prepared by author.

Figure 3a presents the average share of income by households in various regions in Ghana. As observed, rural households in Greater Accra, Brong Ahafo, Eastern, Northern and Volta regions receive a greater share of their total household income from sales of agricultural goods.

As observed, in Brong Ahafo region and Northern region, the share of any other income

(wages) is not half the share of agricultural income. This shows that Brong Ahafo and Northern

regions depend more on Agricultural income and as such, a reduction in price of goods due to

the reduction in the ECOWAS CET, will have a greater effect on their household income as

23

compared to households in other regions. Furthermore, the findings in table 4 indicate that households in Regions where Agricultural income is higher than any other income are affected most since the coefficient of Agricultural sales is negative.

Figure 3a: Structure of average income by region.

Source: Prepared by author based on data from the Ghana Living Standard Survey 7.

The analysis of consumption patterns across regions based on Figures 3a and 3b could imply

that low income earners spend more on food items than high income earners. Due to this,

households as consumers who receive low income could benefit more from the reduction in

the price of agricultural goods. Despite the location of Tema harbour in the Greater Accra

Region, rural households as producers in the region has seen a relatively low impact on their

income. Table 4b shows that there has been 0.09 percent depletion in the agricultural income

of producers in the Greater Accra region. The low effect could be due to the reason that

agricultural producers in Greater Accra region are more into fish farming and not crop farming.

24

Also, Table 4b indicates that the effects of the change in price on rural households vary across regions. This could be a result of different endowments and behaviours of households. Table 4b shows that there has been an increase of 26.96 percent in overall income of rural households.

In general, there has been gains due to the 29.6 percent reduction in expenditure basket which outweighs the 2.54 percent losses due to the depletion in the sales of agricultural goods.

Figure 3b: Structure of average consumption by region.

Source: Prepared by author based on data from the Ghana Living Standard Survey 7.

25 6. Conclusion

The implementation of the ECOWAS CET is to ensure member states converge towards a single tariff rate within the region to enhance regional trade and economic growth. This study has examined the impact of the ECOWAS CET on consumers’ price of agricultural goods in the year 2017. The Ghana living Standard Survey 7 is used for this study. The study has taken into consideration the distributive impacts through change in price of agricultural goods from the perspective of households as both producers and consumers.

First, I link the reduction of the ECOWAS CET to domestic price and, second, I analyse the impact of change in price on household welfare across the regions in Ghana. The findings suggest that the ECOWAS CET reduced domestic consumers’ prices for agricultural goods.

This has caused a reduction in the consumers’ expenditure basket and lead to a positive consumer welfare effect on rural households. Consumers who spend a greater share of their income on food benefitted more than those who spend more of their income on non-food items.

The findings indicate that households who receive a lower total income spend a larger share of their income on food than on non-food items. This could imply that lower income earners benefitted more in the reduction of the ECOWAS CET.

However, producers of agricultural goods have seen a marginal reduction on their income.

There has been a reduction in the purchasing power of producers due to decrease in income.

Households with a greater share of their total income from agricultural sales have been affected negatively mostly due to the depletion in the larger part on their household total income. There has been redistribution of income from producers to consumers in all the regions.

Moreover, the study suggests that the margin of tariff pass through to regions decline the farther away a region is to the harbour. This has been seen in Western Region which has a harbour, (the Takoradi harbour) and Central region which is closer to it. Also, Eastern and Volta regions which are closer to the Tema harbour have seen significant effect both as producers and consumers. The Brong Ahafo region and the northern part of Ghana has also been impacted by the change in price. Although the Tema harbour is in Greater Accra Region, rural households as producers in Greater Accra have seen the lowest percentage decline in their income.

Although the results are to some extent in line with economic theory, this research has

limitations. There are other channels of household welfare that can potentially be affected by

trade policy which has not been captured by this study. The channels may include wages

26

(skilled labour or unskilled labour), as well as welfare effects like changes in education. Adding these components could help shed more light on welfare in rural households.

Acknowledgement

This study would not have been possible without the support of the Swedish Institute through their Scholarship for Global Professionals (SISGP). I am grateful to the Swedish Institute for their financial support throughout my study period at Umeå University.

I would like to express my appreciation to my supervisor Göran Bostedt, for his constructive guidance during the planning and development of this study. I am thankful to Amin Karimu for his assistance. I would like to extend my thanks to my family for their support and encouragement throughout my studies.

References

Ackah, C., & Appleton, S. (2007). Food price changes and consumer welfare in Ghana in the 1990s (No. 07/03). CREDIT Research Paper.

Ackah, C., Morrissey, O., & Appleton, S. (2007). Who gains from trade protection in Ghana:

a household-level analysis? (No. 07/02). CREDIT Research Paper.

Asmah, E. E. (2011). Rural livelihood diversification and agricultural household welfare in Ghana. Journal of Development and Agricultural Economics, 3(7), 325-334.

Breisinger, C., Diao, X., Thurlow, J., & Al-Hassan, R. M. (2008). Agriculture for development in Ghana: New opportunities and challenges. Intl Food Policy Res Inst.

Cherkaoui, M., Khellaf, A., & Nihou, A. (2011, October). The price effect of tariff liberalization in MOROCCO: Measuring the impact on household welfare. In Politics and economic development, EF 17th annual conference.

Danso-Abbeam, G., Armed, M., & Baidoo, F. (2014). Determinants of consumer preference

for local rice in Tamale Metropolis, Ghana. International Journal of Education and Sosial

Science, 1(2), 114-122.

27

De Melo, J., & Laski, A. (2014). Reflections on Adapting to the ECOWAS CET.

Deaton, A. (1985). Panel data from time series of cross-sections. Journal of econometrics, 30(1-2), 109-126.

Deaton, A. (1989). Rice prices and income distribution in Thailand: a non-parametric analysis. The Economic Journal, 99(395), 1-37.

Deaton, A. (1997). The analysis of household surveys: a microeconometric approach to development policy. The World Bank.

Fiamohe, R., Diallo, S., Diagne, A., & Agossadou, A. (2015). Impact of the ECOWAS Common External Tariff on the rice sector in West Africa (No. 1008-2016-80153).

Frankel, J., Parsley, D., & Wei, S. J. (2012). Slow pass-through around the world: a new import for developing countries? Open Economies Review, 23(2), 213-251.

Glewwe, P., & Twum-Baah, K. A. (1991). The distribution of welfare in Ghana, 1987- 88. World Bank Living Standards Measurement Study Working Paper, 75.

Goldberg, P. K., & Pavcnik, N. (2003). The response of the informal sector to trade liberalization. Journal of development Economics, 72(2), 463-496.

Goldberg, P. K., & Pavcnik, N. (2005). Trade, wages, and the political economy of trade protection: evidence from the Colombian trade reforms. Journal of international Economics, 66(1), 75-105.

Goldberg, P. K., & Pavcnik, N. (2007). Distributional effects of globalization in developing countries. Journal of economic Literature, 45(1), 39-82.

Goldberg, P. K., & Pavcnik, N. (2016). The effects of trade policy. In Handbook of commercial policy (Vol. 1, pp. 161-206). North-Holland.

Kareem, O. I. (2014). The welfare impact in Nigeria of the ECOWAS Common External Tariff:

A distributional effects analysis. Trade Policies, Household Welfare and Poverty Alleviation, 273.

Kayizzi‐Mugerwa, S., Anyanwu, J. C., & Conceição, P. (2014). Regional integration in Africa:

an introduction. African Development Review, 26(S1), 1-6.

Marchand, B. U. (2012). Tariff pass-through and the distributional effects of trade

liberalization. Journal of Development Economics, 99(2), 265-281.

28

Minot, N., & Goletti, F. (1999). International Food Policy Research Institute. Effects of transaction costs on supply response and marketed surplus.

Nicita, A. (2004). Who benefited from trade liberalization in Mexico? Measuring the effects on household welfare. The World Bank.

Nicita, A. (2009). The price effect of tariff liberalization: Measuring the impact on household welfare. Journal of Development Economics, 89(1), 19-27.

Nicita, A., Olarreaga, M., & Porto, G. (2014). Pro-poor trade policy in Sub-Saharan Africa. Journal of International Economics, 92(2), 252-265.

Nwafor, M., Ogujiuba, K., & Adeola, A. (2005). The impact of trade liberalization on poverty in Nigeria: Micro simulation in A CGE Model. Interim report to the PEP Research Network.

Orkoh, E. (2018). Gender welfare effects of regional trade integration on households in Ghana. Trade and poverty reduction: New evidence of impacts in developing countries, 36-57.

Osakwe, P. N. (2015). Transformative regionalism, trade and the challenge of poverty reduction in Africa. UNCTAD Trade and Poverty Paper Series, (1).

Oyejide, T. A. (2000). Policies for regional integration in Africa. African Development Bank.

Porto, G. (2006). Using survey data to assess the distributional effects of trade policy. Journal of International Economics, 70(1), 140–160. https://doi.org/10.1016/j.jinteco.2005.09.003

Winters, L. A. (2002). Trade liberalisation and poverty: what are the links? World Economy, 25(9), 1339-1367.

Winters, L. A., McCulloch, N., & McKay, A. (2004). Trade liberalization and poverty: the evidence so far. Journal of economic literature, 42(1), 72-115.

Winters, L. A., McCulloch, N., & McKay, A. (2004). Trade liberalization and poverty: the evidence so far. Journal of economic literature, 42(1), 72-115.

Ackah, C., & Aryeetey, E. (Eds.). (2012). Globalization, trade and poverty in Ghana. IDRC.

29 Appendix

Appendix A: Summary statistics of variables.

Variables Observations Mean Standard Deviation

Minimum Maximum Ln Domestic Price 878 1.375311 .9897474 -.0833816 2.995732 Ln Producer price 878 1.375311 .9897474 -.0833816 2.995732 Ln World Price 878 .8538798 1.345263 -.3903183 2.956378

Ln Tariff 878 3.150501 .2466669 2.890205 3.632309

Trade Cost 878 202.1704 191.3826 .79 579.62

Ln Tariff*Distance 878 625.2327 583.1775 2.413815 1987.634 Ln Tariff*Distance

squared

878 730624.5 1034960 5.826504 3950688

Source: Prepared by author based on Ghana Living Standard Survey (GLSS7).

Variables Observations Mean Standard Deviation

Minimum Maximum

Welfare 7,991 3060.892 3085.606 39.02839 63069.84

Wage 7,991 2199.979 6896.651 0 132480

Agric Income 7,991 2639.979 7635.767 0 259950

Consumption 7,991 4186.684 3897.83 14.6 86967.82

Source: Prepared by author based on Ghana living Standard Survey (GLSS7).

30

Apendix B: Food prices in various regions in Ghana. Prices are in log form.

Source: Prepared by author based on data from Ghana Living Standard Survey 7.

31

Appendix C: Ghana’s import and export flow in the world.

Source: World Bank World Integrated Trade Solution.