Social Accounting Matrix for Côte D'ivoire 2015

Social Accounting Matrix for Côte d'Ivoire 2015

Ferreira, V., Almazán-Gómez, M., Nechifor, V., Ferrari, E., Diallo, S. S.

2021

EUR 30784 EN

This publication is a Technical report by the Joint Research Centre (JRC), the European Commission’s science and knowledge service. It aims to provide evidence-based scientific support to the European policymaking process. The scientific output expressed does not imply a policy position of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use that might be made of this publication. For information on the methodology and quality underlying the data used in this publication for which the source is neither Eurostat nor other Commission services, users should contact the referenced source. The designations employed and the presentation of material on the maps do not imply the expression of any opinion whatsoever on the part of the European Union concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.

Contact information Name: Emanuele Ferrari Address: Calle Inca Garcilaso, Sevilla Email: [email protected] Tel.: +34 954488461

EU Science Hub https://ec.europa.eu/jrc

JRC125648

EUR 30784 EN

PDF ISBN 978-92-76-36180-0 ISSN 1831-9424 doi:10.2760/875449

Luxembourg: Publications Office of the European Union, 2021

The reuse policy of the European Commission is implemented by the Commission Decision 2011/833/EU of 12 December 2011 on the reuse of Commission documents (OJ L 330, 14.12.2011, p. 39). Except otherwise noted, the reuse of this document is authorised under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/). This means that reuse is allowed provided appropriate credit is given and any changes are indicated. For any use or reproduction of photos or other material that is not owned by the EU, permission must be sought directly from the copyright holders.

All content © European Union, 2021.

How to cite: Ferreira, V., Almazán-Gómez, M.Á., Nechifor, V., Ferrari, E. and Diallo, S. S., Social Accounting Matrix for Côte d`Ivoire 2015, EUR 30784 EN, Publications Office of the European Union, Luxembourg, 2021, ISBN 978-92-76-36180-0, doi:10.2760/875449, JRC125648.

Contents

Acknowledgements ...... 1 Abstract ...... 2 1 Introduction...... 3 2 Social Accounting Matrices. Concept and general issues...... 6 2.1 Structure of a SAM ...... 7 2.2 Economic agents and accounts ...... 9 2.2.1 Activities and Commodities ...... 9 2.2.2 Factors ...... 9 2.2.3 Households ...... 9 2.2.4 Enterprises ...... 10 2.2.5 Government ...... 10 2.2.6 Investment-Savings ...... 10 2.2.7 Rest of the World ...... 10 3 Côte d’Ivoire SAM 2015 ...... 12 3.1 Home Production for Home Consumption (HPHC) approach ...... 12 3.2 Structure and estimation of the 2015 Côte d’Ivoire SAM ...... 12 3.2.1 Accounts and data sources ...... 15 3.2.2 SAM final adjustment and balancing ...... 19 4 Côte d’Ivoire 2015 analysis using SAM data...... 21 4.1 Economy structure analysis ...... 21 4.2 Linear multiplier analysis ...... 25 4.2.1 Multipliers ...... 25 5 Conclusions ...... 29 References ...... 30 List of abbreviations and definitions ...... 33 List of figures ...... 34 List of tables ...... 35 Annexes ...... 36 Annex 1. Accounts of the Côte d’Ivoire Social Accounting Matrix 2015...... 36 Annex 2. The linear SAM model and multipliers ...... 39 Annex 3. On-line resources to download the Côte d’Ivoire Social Accounting Matrix 2015 ...... 40

Acknowledgements We are grateful to our colleagues Arnaldo Caivano and Javier Castro to improve the diffusion of our work with the online application

Authors Valeria Ferreira Miguel Ángel Almazán-Gómez Centro de Predicción Económica (CEPREDE). Victor Nechifor JRC - Seville Emanuele Ferrari JRC – Seville Souleymane Sadio Diallo Centre Ivoirien de Recherches Économiques Et Sociales

Abstract A Social Accounting Matrix (SAM) is a comprehensive and economy-wide database recording data on all transactions that take place in an economy over a period of time, usually one year. It has two principal objectives. On the one hand, it presents a complete picture of the economy, taking into account the economy's structure and the interrelationships between economic agents. On the other hand, it provides a coherent framework to analyse how the economy works and to predict the effects of policy interventions through its use as a database in multisectoral linear models by calculating multipliers, and in the calibration and exploitation of Computable General Equilibrium (CGE) models. This report presents the Côte d’Ivoire's SAM for 2015, with the main purpose of providing a suitable database for implementing and evaluating the country's own developmental social and economic policies and initiatives. For this purpose, the basic structure of a SAM is presented, explaining the meaning of each account. Then, the accounts included in the SAM of Côte d'Ivoire are explained in detail covering the main aspects of its construction and estimation. This SAM has the advantage of including the Household Production for Household Consumption (HPHC) approach and a high disaggregation of the agricultural and food sector, which is very important for an economy like the Ivorian case. Finally, the SAM is used as a database to perform a descriptive analysis of the Ivorian economy and to obtain results of employment, output and value added multipliers with the application of a linear multiplier analysis. Annex 3 explains how to download the matrix available online.

1 Introduction The European Commission is committed to cooperate with developing countries to provide solutions to problems related to inequality, nutrition and food security. This is carried out by evaluating related policies, facilitating the access of researchers in these countries to analytical tools that allow them to carry out such assessments on their own. In this sense, the Joint Research Centre (JRC), the European Commission's in-house science service, is committed to provide support for: i) improvement of information systems on agriculture, nutrition and food security, ii) policy and economic analysis to support policy decision-making process and iii) scientific advice on selected topics concerning sustainable agriculture and food and nutrition security. In particular, the Economic of Agriculture Unit of the JRC Directorate D, Sustainable Resources, is responsible to elaborate the methodology and tools to provide economy-wide analysis related to sustainability of policies in the sectors of agri-food, sustainable resources and fight against food and nutrition insecurity. The analyses and tools proposed should support the EU institutions, and the partner countries for the elaboration and assessment of policies and demand-driven technical and scientific advice. Among possible scientific tools, economic simulation models represent interrelationships between selected economic variables and provide a simplified representation of economic reality to be used to quantify impacts of policy changes (i.e., ex-ante policy analysis). The construction of databases to provide evidence-base policy support to partner countries is part of the objectives of the Pan-Africa Network for Economic Analysis of Policies (PANAP). PANAP is a network among academic/research and institutional partners collaborating with the JRC in developing research on agricultural economics and policy issues with a focus on Africa. PANAP is part of the Action Agenda of the Political Declaration of the 3rd AU-EU Agriculture Ministerial Conference held in Rome on 21 June 2019 (EC Decision C(2019) 4277). At the launch and first meeting of PANAP (6-8 November 2019, Addis Ababa, Ethiopia) key stakeholders, researchers, data analysists, policy makers, from national and multilateral institutions engaged in fruitful discussion on the role of science in supporting policy decision making in Africa with focus on the farming and food sectors (Morokong & Ferrari, 2020). The objective of PANAP is to strengthen the liaison between researchers/scientists and policy makers in Africa, including relevant multi-lateral African institutions, and to stimulate their cooperation on selected topics linked to policy priorities in Africa. PANAP will also contribute in understanding and resolving scientific issues in the fields of agriculture and food security to support efficient policies. It will also facilitate achieving sustainability of the agri-food sectors to enhance food and nutrition security in alignment with the Malabo Declaration Commitment 3 on ending hunger in Africa by 2025 and SDG 1 and 2. Cote d'Ivoire is a Western African country, being the largest economy member of the Economic Community of West African States (ECOWAS) after Nigeria and Ghana. Therefore, Côte d’Ivoire is one of the countries analysed in this context.1 The main pillar in the macroeconomic analysis is the use of a tailored version of a single-country Computable General Equilibrium (CGE) model to analyse some of the agricultural and rural policy priorities to reduce poverty and inequalities, considering the specificities of the economy in the country.

(1) Similar SAMs databases have been constructed for other African countries, for example for Senegal (Boulanger et al., 2017), Kenya (Mainar-Causapé et al., 2018a), Ethiopia (Mengistu et al., 2019), and Ghana (Ferreira et al., 2021). All these SAMs are freely available in the JRC DataM repository.

In the case of Côte d’Ivoire, the country has suffered a period of political and economic instability, due to the coup d'état in 1999 and the post-electoral crisis in 2011. Since 2012, the country show a period of consistent economic growth, with an increase in the GDP per capita and exports (Ducroque et al., 2018). However, despite a decrease in poverty, the poverty rate remains high (representing 46.3%), especially in rural areas, according to the 2015 household survey (ENV, 2015). The agriculture sector is key in the Côte d'Ivoire's economy. Côte d'Ivoire is a leading producer of many agricultural products, being the world's largest producer and exporter of cocoa. Due to the strong dependence on agriculture and related activities, there is particular interest in the agricultural sectors considering the National Agricultural Investment Program (NAIP II - 2017-2025). Furthermore, there are remarkable differences in employment and welfare between the rural and urban zones of the districts. Hence, most of the employed population lives in urban areas (57.1%). The city of Abidjan has a relatively important economic weight and concentrates 22.1% of the total employed population. Moreover, in Côte d'Ivoire the employment is predominantly informal, representing 91.8% of employed persons and more than 70% of the employed population have at most primary level of education (ENV, 2015). Therefore, the promotion of policies focus on the economic and social development in the country requires the elaboration of specific database that provide a description of the national and regional economy with the interlinkage between them and the disaggregation of the main sectors. The calibration of CGE model requires a complex database system, called Social Accounting Matrix (SAM). The estimation of a new SAM for Côte d’Ivoire is an important achievement itself, because it provides detailed information about the economic structure of the country and serves, also, as main database for linear multisectoral and CGE models. Whereas the latest Côte d’Ivoire SAM for 2015 has been estimated by Diallo (2018), this report presents significant modifications and extensions to the SAM to obtain a database for the analysis of different policies at stake. The main improvements in the SAM presented in this report include the regionalisation of agricultural activities, factors of production and households. This facilitates the identification and analysis of those households that produce both for own consumption and commerce. The SAM developed shows the breakdown of the accounts by district, rural and urban areas for households, and by skill level for the labour factor. The previously constructed SAMs for African countries available in the JRC DataM repository have been used for policy analysis, considering the calculation of multipliers as well as the application of CGE models. To mention a few examples, the SAM for Ethiopia and Kenya were used to analyse the impact of COVID-19 on economic performance and poverty incidence, evaluating the effectiveness of the government recovery measures (Nechifor et al., 2020a; Nechifor et al., 2020b). Linking the CGE model with a microsimulation approach, the Kenya SAM was used to evaluate the food security impacts of expanding fertilizer capacities (Boulanger et al., 2020) and due to macroeconomic outcomes of the pandemic (Nechifor et al., 2021). For Senegal, the SAM was used to analyse policy options for the development of the agriculture sector by applying a CGE model (Boulanger et al., 2018). Furthermore, the multipliers were calculated using SAMs, for example, to analyse the capacity of the primary sector in Kenya suggesting policy recommendations (Mainar-Casupé et al., 2020). In the case of Egypt, Osman et al. (2021) used a mixed multiplier analysis to identify the seasonal agricultural activities with high production and income multipliers suitable for policy promotion. For Ethiopia, Boulanger et al. (2019) analysed the effects of different policy

options to support the creation of rural job opportunities by calculating the employment multiplier and using a CGE model. The SAM developed in this report and freely available in the JRC DataM repository is expected to be a useful database for various types of policy impact analysis and decision making. This report is structured as follows: the concept, general issues and the accounts description of Social Accounting Matrices are presented in Section 2, while Section 3 describes the structure and estimation of the Côte d’Ivoire SAM 2015. This section builds upon an introductory background on the Home Production for Home Consumption issue, a key aspect of this new SAM. Then, the structure and estimations of the Côte d’Ivoire SAM is presented, detailing the accounts and data basis used and the final adjustments, SAM balancing and disaggregation procedures. The Section 4 shows a description of the Ivorian economy structure using SAM data and a simple multiplier analysis is presented to illustrate the usefulness of the SAM linear models in policy impact assessment. Finally, Section 5 includes some conclusions, and the Annexes shows some additional tables and presents the downloadable application.

2 Social Accounting Matrices. Concept and general issues. A Social Accounting Matrix (SAM) is a comprehensive and economy-wide database that enables the representation of economic and social information on all the transactions carried out among the agents of a specific economy over a period of time, generally one year. The origins of SAMs are found in the pioneering works of Stone (1947) among others, and subsequent advances in their use as a model for economic analysis (Defourny & Thorbecke, 1984; Pyatt & Round, 1985). A SAM extends the information provided by the Input-Output table by incorporating all economic transactions, using a more disaggregated accounting structure. (2) This allows closing the circular flow of income and expenditure reflecting the interlinkages among production, trade, demand, income generation and its redistribution between institutional sectors (Pyatt & Round, 1985), providing the opportunity to analyse the distribution of wealth and income. The underlying basis of a SAM is the concept of a circular flow of income represented (3) in Figure 1.

Figure 1. The circular flow (simple version)

Source: Mainar-Causapé et al. (2018b).

The elaboration of a SAM requires the use of many statistical sources, mainly the statistical systems of National Accounts, data on trade, as well as socio-economic data related to household income, employment and expenditure that can be obtained from household budget surveys and labour force surveys. With this information, households can be disaggregated using socio-economic characteristics (e.g. income, education, regions, rural- urban division, etc.) to analyse income distribution. The need for further information depends on the disaggregation required for each account depending on the type of analysis to be carried out.

(2) Some problems of I-O frameworks are still present in the SAM framework (i.e. the use of coefficients and fixed prices and no substitution elasticities) (Mainar-Causapé et al., 2018a). (3) It represents a basic version of the circular flow. It does not take into account transactions between institutions, taxes, savings and investment, among other flows.

A SAM represents a complete snapshot of the economy showing the economic structure and the relationships between economic agents, recording data on all transactions between production activities, factors of production, institutions, and the rest of the world, within a specific country or region under analysis. In this sense, a SAM is a very useful tool to understand the structure of an economy and to be used as a suitable database for economic modelling. As such, it can be used as a database for multisectoral linear models allowing the calculation of multipliers or it can also be used as a database for the application of more complex Computable General Equilibrium (CGE) models (Mainar-Causapé et al., 2018a). This makes it possible to analyse how the economy works considering socio-economic issues such as employment, poverty, growth, and income distribution, etc., and to evaluate the impact of different policies interventions.

2.1 Structure of a SAM The SAM is a database represented by a square matrix, organized in a logical framework that provides a visual display of all the transactions that take place between agents in an economy, determining how income is generated and how it is expended. For this purpose, each account is represented by a row that shows the sources of income of each agent and a column, which show the payments made, both in monetary values. Thus, each cell (i,j) shows the transaction between account i and j, in which account i receives income from j. Typically, a Social Accounting Matrix has six basic groups of accounts: ● Activities or Commodities (or both, separated) ● Factors of production ● Private Institutions (Households and Corporations/Enterprises) ● Public Institution (Government) ● (Combined) Capital accounts ● Accounts for the Rest of the World. Figure 2 shows the basic structure of a standard SAM (4); however, the structure of the matrix will be determined by the accounts included in these six basic groups and the disaggregation of these accounts. That along with the geographical scope it represents (national, regional, multi-regional, etc.) will depend on the analysis to be carried out.

(4) For more detail of this general structure see Miller & Blair (2009), Mainar-Causapé et al. (2018b) or Round (2003).

Figure 2. A Social Accounting Matrix (SAM) standard structure

Enterprises / Investment- Rest of the Commodities Margins Activities Factors Households Government Total Corporations Savings World

Transaction Intermediate Household Government Investment and Commodities costs (trade / (inputs) Exports Demand consumption expenditure stock changes transport) consumption

Transaction costs Margins (trade / Margins transport)

Gross output / Domestic Production Activities production (activity income)

Remuneration of Factor income Factors factors / Factor Factor income from RoW income

Factor income (Inter Distribution of Government Transfers to Household Households distribution to Households enterprise income transfers to Households income households transfers) to households households from RoW

Factor income Government Transfers to Enterprises / Enterprise distribution to transfers to Enterprises income Corporations enterprises enterprises from RoW

Factor income to Direct Household Direct Enterprise Transfers to Net taxes on Net taxes on Government Government Government / taxes / Transfers taxes / Transfers Government products production income Factor taxes to Government to Government from RoW

Capital (Capital Household Government transfers from Investment-Savings (Depreciation) Enterprise savings accounts Savings savings savings RoW (Balance transfers) of Payments)

Factor income Household Enterprise income Government Payments to Rest of the World Imports distribution to transfers to RoW to Row transfers to RoW RoW RoW

Costs of Expenditure on Household Enterprise Government Incomes from Total Supply Margins production Investment factors expenditure expenditure expenditure RoW activities

Source: Aragie et al. (2017), Mainar-Causapé et al. (2018b) and Round (2003a).

2.2 Economic agents and accounts

2.2.1 Activities and Commodities The standard structure of the SAM (as shown in figure 2) distinguishes between Activities (that carry out the production of commodities) and Commodities (goods and services). In the Activities, the flows are valued at production costs and in the Commodities, are valued at market prices (including indirect taxes on products and transactions cost). The sum of the values of the Activities represents the domestic production (at production prices). To obtain the supply of Commodities (at purchaser’s prices), net taxes on products, margins and imports must be added. The supply of Commodities is sold domestically (activities, households, government) or exported. Activity accounts represent by columns, the production process including the intermediate consumption of commodities as inputs and the use of factors of production (such as labour, capital, etc.). The sum of the remuneration of factors plus taxes less subsidies on production is the value added by Activities. By rows, Activities cells show the value of commodities produced by each activity (in basic prices). The row sum for an activity is the value of its gross output. By analysing the cells by columns, the Commodity shows the supply of Commodities at purchaser prices, that includes the domestic production by Activities, the imports of rest of the world and payment of taxes less subsidies on products (domestic and imported) to the government (Mainar-Causapé et al., 2018a). The commodity row accounts describe the demand side, composed of intermediate consumption (by Activities) and the final demand consumption by institutional sectors (Household and Government), investment and exports. For each commodity the SAM records the costs associated with marketing and transportation (margins). These transaction costs are associated with the cost of moving goods between producers, markets and national borders, whether for domestic consumption, import or export, and are part of the supply costs of commodities (Mainar-Causapé et al., 2018b).

2.2.2 Factors The factors of production are the resources endowments used in the production process (combined with intermediate inputs to produce goods and services). Traditionally, they include capital and labour, as well as other types of factors, such as land or other natural resources. The different types of factors of production and their disaggregation will depend on the aim of the analysis. For example, labour account can be disaggregated by educational level (e.g., skilled and unskilled workers), and capital factor can be disaggregated in accordance with their use (e.g., agricultural/non-agricultural capital). The production factors collect, by row, the income received (wages, rent, etc.) from the activities and the rest of the world. By columns, these incomes are distributed to the owners of the factors of production (Household, Enterprises and Government) and the Rest of the World.

2.2.3 Households The household accounts describe the income and spending of all the individuals in an economy. In a traditionally SAM, households are often aggregated into Representative Household Group (RHG) perhaps categorized by income class, geographical location or demographic characteristics. This categorization assumes that all individual household in an

RHG are, on average, affected in the same manner by a policy shock. This type of information is crucial for the analysis of socio-economic problems (unemployment, income distribution etc.) allowing the modeler to analyse the distributional effects of an economic shock across different household types (Burfisher, 2016). The Household row account receives incomes, which includes income from factors supplied (as owners of labour, capital, land, or natural resources), transfers from Enterprises (distributed profits and direct transfers), from the Government (direct transfers), from the Rest of the World (usually for labour services and remittances) and from other households (Mainar-Causapé et al., 2018b). By columns, Household expenditure is detailed, showing the use of household income for consumption of commodities, payment of direct taxes to the government, transfers to other households (domestic and foreign); and households’ savings (or financing need if negative).

2.2.4 Enterprises Enterprises represent the institutional part of the productive sector, so generally they do not consume inputs such as activities. The enterprise accounts collect, by rows, the incomes received related to the ownership of asset (capital, land, or natural resources) and income form transfers from other institutions. The transfer of these revenues is shown in columns, towards other institutions, for example households in the form of dividends, the payment of direct taxes (enterprise tax) or savings.

2.2.5 Government The Government participates as a “productive activity” through the public sector and the marketed goods and services resulting from their activities are recorded in the respective Activity and Commodities accounts (Mainar-Causapé et al., 2018a). Generally, the Public Administration institutional sector is represented by the Government account for the sector itself, while tax collection is captured through other accounts specific for each tax category (VAT, income, tariffs, etc.). Government income is shown in rows, coming from taxes, transfers received (domestic and foreign) and the remuneration of factors of production (including any government-owned assets). Government expenditure is broken down by column into the consumption demand of goods and services, transfers to other institutions in form of subsidies or benefits (household and enterprises) or to other countries (e.g. debt services payments). The investment-savings cell shows government savings (government surplus) or government deficit (if negative).

2.2.6 Investment-Savings The row of this account records the savings generated by all domestic institutions (households, enterprises, and government) as well as the balance of foreign trade on capital account with the rest of the world. The column shows the capital investment in goods and services, the Gross Fixed Capital Formation (GFCF) and changes in stocks.

2.2.7 Rest of the World The foreign sector account shows the economic interaction within the country (or region) analysed and the rest of the world. The foreign sector represents two trade accounts (imports and exports) and can be disaggregated depending on the specific interest of the analysis (e.g., ECOWAS, African Union and Rest of the world).

By rows, the account represents the income received by the rest of the world account that include imports of goods and services (commodities), transfers to abroad from institutions (households, businesses, and government), and the remuneration of factors of production abroad. By column, it shows the Rest of the World expenditures, which are divided into the purchase of goods and services (exports), the payments to domestic factors of production used abroad and transfers recorded from other economies (factor payments, foreign loans and aid, remittances, etc.). The balance, which enters the investment-savings cell, reflects the surplus or deficit with the Rest of the World. Consequently, indicating the capital inflow (in case of negative trade balance) or outflow (positive trade balance) between the Rest of the World and the country at stake.

3 Côte d’Ivoire SAM 2015

3.1 Home Production for Home Consumption (HPHC) approach The classic Representative Household Group (RHG) includes the household behaviour as consumer of goods and services and as providers of factors of production (and receptor or providers of transfers). However, the dual role of households as producers and consumers is often a typical characteristic of developing countries with economies that have a large proportion of subsistence agriculture. Therefore, there is a need to define and estimate the home production for home consumption (HPHC) within a SAM. This Home Production for Home Consumption (HPHC) approach is introduced in the SAM by assuming that households act as units of production of commodities consuming part or all of their output. Hence, it is necessary to extend the SAM structure by including additional columns and rows for the commodity and activity accounts. Considering the Commodities accounts, additional rows and columns are required to separate the commodities produced by these households for own consumption (HPHC as input or final product) and other marketed commodities (produced both by conventional productive activities and by households). The rows of these commodity accounts show the HPHCs use as intermediate inputs in the productive activities of households and their consumption in final demand of household (RHG). The column summarizes the contribution of the activities of households to each of these goods. HPHC commodities can only be produced by the RHGs that consume those commodities, so each RHG must be a household and an activity. The columns of the household’s activities show the cost structure of their production (combination of inputs-own produced and marketed- and value added), and the rows show the destination of their production as inputs (own-consumption goods or marketed commodities). The taxes, trade and transport costs incurred should be allocated only to the marketed commodity (Mainar-Causapé et al., 2018a). The inclusion of the HPHC approach in a SAM requires more data to distinguish consumer demand between marketed and HPHC commodities and to identify the cost structures used in production. Typically, this will require reconciling data from different sources, especially data on household income and expenditure, labour force and agricultural (production) collected through surveys, and also some additional estimations.

3.2 Structure and estimation of the 2015 Côte d’Ivoire SAM This study estimates a new SAM for Côte d’Ivoire (base year 2015) based on the standard structure outlined above also including some peculiarities. Thus, the new SAM includes specific accounts for the treatment of HPHC and a high regionalisation differentiated across districts and divided within rural and urban areas. Consequently, this new framework would allow addressing specific issues related for example to production and productive factors for each district, in order to reduce poverty and inequality within and between regions. The regional breakdown in the SAM for Côte d’Ivoire 2015 correspond to the 14 administrative districts of the country (table 1). For each district, the household account is disaggregated into rural and urban zones. Finally, 27 different Representative Household Group (RHG) are detailed in the SAM. Furthermore, 13 of the districts (each one except the city of Abidjan) has a specific agricultural activity producing 10 “subsistence commodities” destined for home consumption, and 17 marketed commodities. The remaining marketed

commodities are produced by the specific activities aggregated at a national level (37 in total). The Côte d’Ivoire SAM distinguishes between activities and commodities and the HPHC approach mentioned above. The structure and a short version of the SAM is summarised in Table 2. The SAM further contains 96 accounts of production factors. There are 3 types of labour, 1 land factor, 1 livestock factor and 1 capital factor. Labour and land factors are disaggregated across the districts, and labour factors are also separated between rural and urban areas. In summary, the 2015 Côte d’Ivoire SAM contains 247 accounts:  50 activities (13 household regional agricultural activities and 37 national activities)  64 commodities (10 HPHC and 54 marketed commodities)  96 factors of production (81 labour accounts, 13 land factors and 2 capital factors)  27 households (disaggregated by rural/urban and by districts)  5 tax accounts  1 transaction costs (margins)  3 other institutional accounts: enterprises, government, and the rest of the world  1 savings and investment account. All the accounts considered in the 2015 SAM survey in Côte d'Ivoire are detailed in Annex 1 (Table A1). Annex 2 comprises the on-line resources to download the complete 2015 Côte d’Ivoire SAM. This base SAM database for Côte d’Ivoire was also disaggregated for a specific analysis on the cocoa sector where cocoa output is detailed across the 14 districts separate from the other agricultural production. This version of the Côte d’Ivoire SAM is available upon request.

Table 1. District and regional description in Côte d’Ivoire SAM 2015.

SAM Districts Regions included Autonomous District of Abidjan Abidjan Bas-Sassandra Gbôklé/ Nawa/ San-Pédro Comoé Indénié-Djuablin/ Sud-Comoé Denguélé Folon/ Kabadougou Gôh-Djiboua Gôh/ Lôh-Djiboua Lacs Bélier/ Iffou/ Moronou/ N'Zi Lagunes Agnéby-Tiassa/ Grands-Ponts/ La Mé Montagnes Cavally/ Guémon/ Tonkpi Sassandra-Marahoué Haut-Sassandra/ Marahoué Savanes Bagoué/ Poro/ Tchologo Vallée du Bandama Gbêkê/ Hambol Woroba Béré/ Bafing/ Worodougou Autonomous District of Yamoussoukro Yamoussoukro Zanzan Bounkani / Gontougo Source: Own elaboration (based on ISO 3166-2: CI).

Table 2. Côte d’Ivoire SAM 2015 expressed in millions of FCFA- West African CFA francs (aggregated values).

Source: Own elaboration

3.2.1 Accounts and data sources

3.2.1.1 Activities and Commodities The Côte d’Ivoire SAM 2015 (and in some cases the 2006 SAM) (Diallo, 2018; Fofana, & Diallo, 2015) 5 provides the initial values of rows and columns for commodities and activities, To introduce the HPHC, estimates are required to disaggregate the additional commodity and activity accounts. Households have to be divided between the classic Representative Household Group (RHG) (consuming good and services, providing factors of production and receiving or performing transfers) and the household presented as units of production of commodities. The latter accounts represent the economic behaviour of households as producers of food commodities (agricultural and livestock products) and are disaggregated by districts. It is also necessary to separate accounts for commodities produced by the households for their own consumption and other marketed commodities (produced both by households and by conventional productive activities). The microdata that enables such disaggregation in the SAM are collected from the Household Standard of Living Survey (named ENV 2015) conducted by the National Statistical Institute for 2015 (INS 2015). This survey provides information on the activities carried out by households, agriculture and livestock, and household consumption, among others, which are essential for the disaggregation mentioned above. The SAM for Côte d’Ivoire is characterized by a high disaggregation of agricultural and food commodities. This is evident since out of the 64 commodities (for own consumption and marketed), 39 are part of the agricultural and food products group. In summary, the 2015 Côte d’Ivoire SAM includes 50 activities (13 of them are households as producers), producing 10 HPHC commodities and 54 marketed commodities. Table 3 and 4 summarised the breakdown of commodities and activities, and they are also show in detail in Annex 1 Table A1 (Accounts of the Côte d’Ivoire Social Accounting Matrix 2015).

3.2.1.2 Households The Côte d’Ivoire SAM 2015 distinguishes between rural and urban households and disaggregates each one by districts, using information from the 2015 Household Standard of Living Survey conducted by the National Statistical Institute (Enquete sur le niveu de vie des menages en Côte d’Ivoire: ENV 2015) (INS 2015). As a result, the Côte d’Ivoire SAM 2015 contains 27 representative household groups (RHG). Such classification will allow for a good analysis of redistributive aspects and the specific impact of different policies.

3.2.1.3 Factors of production The disaggregation of factors of production is very important depending on the objective of the analysis. The Côte d’Ivoire SAM 2015 separates factors into four broad categories: labour, land, livestock, and capital (Table 5). Outside the use of employment in the HPHC activities, the matrix does not present a disaggregation of employment considering the formal and informal sector due to the limited availability of data, but can be considered as a future modification for further analysis. The SAM classifies the labour factors into three types: skilled, semi- skilled and unskilled labour. The definition of a skilled worker is one with a post-

(5) The SAM for 2015 was developed by Souleymane Sadio Diallo for the Ministère de l’Economie et des Finances – Côte d’Ivoire (Diallo, 2018). The SAM for 2006 was constructed by Ismael Fofana and Souleymane Sadio Diallo for the International Food Policy Research Institute (Fofana & Diallo, 2015).

primary education and, therefore, a semi-skilled worker is one who has a first-degree education (Fofana and Diallo, 2015). Each labour factor is also regionalized given the 14 districts and further separated into rural and urban areas. This information is disaggregated using the data derived from the ENV 2015. Also, there is one land factor for each district, one account for livestock and other account for capital.

3.2.1.4 Other accounts The Côte d'Ivoire SAM has one account for the government and five different types of taxes and subsidies: tax and subsidy on production (indtax), taxes on products (vattax), import taxes (imptax), export taxes (exptax) and direct taxes (dirtax). Taxes on production are paid by production activities based on their output. Taxes on products are paid by domestic firms on their intermediate input purchases, and by consumers and investors on their purchases of final goods and services. Direct taxes are paid by enterprises and households (based for example on income earned from wages and rents). The transaction cost account includes the cost associated with marketing and transportation. The margins include the costs of trade and transport from moving goods between producers, markets and/or national borders, either for domestic, import or export trade. There is one account for savings and investment that includes by row the domestic private savings by enterprises and households and the fiscal surplus (or deficit) for the government. By columns, it includes the combination of gross fixed capital formation (investment on commodities as machinery, vehicles, and equipment) and change in stocks. The description and codification of all accounts is presented in the Annex 1, Table A1.

Table 3. Activities disaggregated in Côte d’Ivoire SAM 2015

Representative Households Groups Activities as activities Abidjan Description in English Description in French Bas-Sassandra Forestry Sylviculture, exploitation forestière Comoé Fish Pêche et pisciculture Denguélé Crude oil Industrie petrolière Gôh-Djiboua Other mining Autres industries extractives Lacs Meat - Fish processed Production de viande et poisson Lagunes Grain milling Travail des grains et fabrication de produits Montagnes Cocoa products, coffee Transformation du cacao et du café Sassandra-Marahoué Oilseed industry Industrie des oléagineux Savanes Bakery Boulangerie, pâtisserie et pâtes alimentaires Vallée du Bandama Dairy products and fruits products Industrie laitière, industrie des fruits Woroba Beverages Industrie des boissons Yamoussoukro Tobacco (processed) Industrie du tabac Zanzan Textile & clothing Industrie textile et de l'habillement Leather & footwear Industrie du cuir et de la chaussure Wood Travail du bois et fabrication d'articles en bois Paper & Printing and publishing Industrie du papiers et du cartons, imprimerie Petroleum Raffinage et cokéfaction Chemicals Industrie chimique Rubber and plastic Industrie du caoutchouc et des plastiques Fabrication d'autres produits minéraux non Non-metal minerals métalliques Metals and metal Produits métalliques de base Machinery and equipment Fabrication de machines, d'équipements et tranport Furniture and other manufacturing Fabrication de meubles, industries diverses Electricity Production et distribution d'électricité de gaz Construction Construction Wholesale and retail trade Commerce de gros et de détail Maintenance / Repair Réparations Accommodation and food services Hôtels et restaurants Transportation and storage Transports et communications Information and communication Postes et télécommunications Finance and insurance Activités financières Real estate activities Activités immobilières Business Services Services fournis aux entreprises Public administration Administration publique et de sécurité social Education Education Health and social work Santé et action sociale Other services Services collectifs, sociaux et personnels Source: Own elaboration.

Table 4. Commodities disaggregated in Côte d’Ivoire SAM 2015

HPHC Marketed commodities

Description in English Description in French Description in English Description in French Maize Mais Maize Mais Rice Riz Rice Riz Yam Igname Yam Igname Cassava Manioc Cassava Manioc Plantain Banane plantain Plantain Banane plantain Other food crops Autres vivriers Other food crops Autres vivriers Fruits and nuts Fruits et noix Cocoa Cacao Other industrial products Autres produits industriels Coffee Café Plants and seeds Plants et semences Cotton Coton Livestock Élevage et chasse Fruits and nuts Fruits et noix Palm Paume Other oilseeds Autres oléagineux Hevea Hévea Cashew Anacarde Other industrial products Autres produits industriels Plants and seeds Plants et semences Livestock Élevage et chasse Forestry Sylvicoles Fishing Pêche et pisciculture Crude oil Industrie petrolière Other mining Autres industries extractives Meat and fish Viande et poisson Grain milling Travail des grains et produits am Cocoa products, coffee Produits du cacao, du café Oilseed Industrie des oléagineux Bakery Boulangerie, pâtisserie et pâtes alimentaires Dairy and fruits products Produits laitiers et fruits Beverages Boissons Tobacco Tabac Textiles Industrie textile et de l'habillement Leather and footware Cuirs et chaussures Wood products Travail du bois et articles en bois Paper and printing Papiers et cartons; édités et imprimerie Petroleum products Produits du raffinage, de la cokéfaction Chemicals products Produits chimiques Rubber and plastic Caoutchouc et plastique Non-metalic minerals Autres produits minéraux non métalliques Produits métalliques de base et ouvrages en Metals and metal products métal Machines, appareils électriques et matériels Machinery and other equipment transport Furniture and other manufacturing Meubles, produits des industries diverses Electricity, gas and steam, water Electricité, gaz, eau et glace alimentaire supply and sewage Construction Travaux de construction Trade services Vente en gros et en détail Repair services Réparations Accommodation and food services Hôtellerie et restauration Transport services Transports et communication Postal and telecommunications Services des postes et télécommunications services Financial services Services financiers Real estate Services immobiliers Business Services Services aux entreprises Services d'administration publique et de Public administration and defence sécu Education Education Health and social work Services de santé et d'action sociale Other services Services collectifs, sociaux et personnels Source: Own elaboration.

Table 5. Description of factors in Côte d’Ivoire SAM 2015

Description Notes Unskilled labour No education Semi-skilled labour Primary education Skilled labour Higher and secondary education level (post-primary education) Land Harvested crop land Livestock For livestock (e.g., live animals, paddocks, beehives) Capital Other non-Agriculture Capital (e.g., mineral resources, equipment, others) Source: Own elaboration based on ENV (2015).

3.2.2 SAM final adjustment and balancing For the construction of the SAM presented in this report, the 2015 Côte d’Ivoire SAM (and in some cases the 2006 SAM) was used (Diallo, 2018; Fofana, & Diallo, 2015). Furthermore, the 2015 Household Standard of Living Survey conducted by the National Statistical Institute (Enquete sur le niveu de vie des menages en Côte d’Ivoire) was used to carry out the aforementioned disaggregation (INS, 2015). The main objective of the Household Living Standards Survey is to collect information to improve the planning and evaluation of economic and social policies in Côte d'Ivoire. To this end, it provides basic data on the standard of living and living conditions of households and it includes detailed information on health, education, employment, farming, income- generating activity, housing expenditures, activities, transport, etc. The ENV (2015) data are nationally representative. The survey has a sample size of 12,900 households with a response rate of 99.9%, and the data was collected between 23rd January and 25th March 2015. The data collected is representative of the 33 regions (31 regions plus the city of Abidjan and the Autonomous District of Yamoussoukro) classified in the 14 districts and separated into both urban and rural areas. Regarding the regions, it was necessary to distinguish between the city of Abidjan which has a relatively important economic weight and the rest of the districts. Furthermore, the distinction between urban and rural areas was necessary because of the differences in average consumption, employment, and income. For these reasons, the 2015 national SAM for Côte d’Ivoire is disaggregated into 14 districts that guarantee the representativeness of the 33 regions. The district disaggregation has been applied to households (as productive units (activities) and as institutional units), and for labour and land factors (see table 1). The microdata provided by the survey are necessary to disaggregate the sub-accounts specified in this SAM. Since the construction of the SAM uses many different sources, it was also necessary to apply techniques to reconcile and balance the data. To this end, the RAS and the Cross Entropy methods are useful to balance a matrix when there is not enough information but macroeconomic targets or targets for specific accounts and cells or sub- matrices are available (McDougall, 1999; Robinson et al., 2001). The RAS procedure, also knows as a “biproportional” matrix balancing technique, is based on a process of iterations. These iterations occur until the balance between the rows and columns is reached (Bacharach, 1970). The cross-entropy (CE) is a method that minimizes the distances between the priors and the resulting value, subject to the row and column sum restrictions. The CE approach represents an extension of the standard RAS method and provides a flexible method for estimating a SAM when dealing with scattered and inconsistent data (Robinson, et al., 1998).

In the construction of this matrix, the RAS method (or the generalized version named GRAS) has been applied whenever it has been necessary to balance the matrix. More specifically, the RAS technique is used to reconcile the data in a matrix with certain previously defined row and column values. Hence, it is based on iteratively adjusting the initial matrix A0, with its row sums (u0) and column sums (v0), in order to generate a new matrix A1 satisfying the defined target of certain row sums (u1) and column sums (v1). Using this technique, the obtained matrix differs as little as possible from the previous matrix. In this solution, the zero-value entries keep their zero values and the positive entries keep their positives values. The generalized alternative (GRAS) emerged in order to have the possibility of updating a given matrix with negative entries (Junius & Oosterhaven, 2003; Umed et al., 2013; Miller & Blair, 2009).

4 Côte d’Ivoire 2015 analysis using SAM data

4.1 Economy structure analysis A SAM provides the macroeconomic data that allows the description of the economic structure of a country. It also presents the microeconomic data that describes a nation’s economic activity in detail. The data analysis of the Côte d’Ivoire SAM for the year 2015 provides a snapshot of the characteristics of the Ivorian economy. The macroeconomic analysis shows that domestic absorption represents 101.8% of the GDP and the foreign sector shows that exports represent 40.2% and imports 42% of the GDP (Figure 3). Regarding the composition of domestic absorption, private household consumption represents 67.4%, followed by investment representing 19.7% and government expenditure 12.9% (Figure 4).

Figure 3. Domestic absorption, imports, and exports as % of the GDP. Côte d’Ivoire 2015.

101.8%

40.2% 42.0%

Exports Imports Domestic absortion Source: Own elaboration with Côte d’Ivoire SAM 2015.

Figure 4. Domestic absorption composition. Côte d’Ivoire 2015.

19.7%

12.9%

67.4%

Private consumption Government consumption Investment

Source: Own elaboration with Côte d’Ivoire SAM 2015. By analysing the households' consumption of commodities, the primary sector and food industry stand out with 50.8% of expenditure. The services sector is the second most important with 25.8% of household consumption (Figure 5). Rice, yam, and plantain (within

agriculture) and meat and fish, grain milling, oilseed, dairy, and beverages (within the food industry sector) are the most demanded products by households. For the spending on services, 28% represent the postal and telecommunications services and 20% the transport services. The expenditure in the manufacturing commodities represents 16.4% destined mainly for chemicals (5%) and machinery and equipment (3.8%). Finally, 5.7% of household expenditure is allocated to petroleum products.

Figure 5. Household consumption pattern. Côte d’Ivoire 2015.

Services; 25.8%

Construction; 0.4%

Primary sector and Utilities; 0.9% food industry; 50.8%

Manufactured ; 16.4%

Petroleum and mining; 5.7%

Source: Own elaboration with Côte d’Ivoire SAM 2015. The total values of household consumption on commodities are similarly distributed between rural and urban households (48.4% and 51.6% respectively). The in-depth analysis of household consumption patterns shows certain non-significant differences in consumption by group of commodities, for rural and urban regions, as well as across each district. This can be seen in the Figure 6 that shows the distribution of household consumption of group of commodities, considering the consumption patterns of households classified into rural and urban areas and also by districts. It represents the proportion that each commodity or group of commodities represents in the consumption of each group. Thus, rural households have a slightly higher consumption of primary sector and food industry commodities, while urban households also consume slightly more manufactured goods, including textiles and clothing, and petroleum and mining products. These minor differences are also reflected in the analysis of the distribution of commodities consumption across districts. Considering that Abidjan is the main city of Côte d’Ivoire and Yamoussoukro is the capital, the consumption patterns that stands out for urban areas is clearly visible for both autonomous districts. In the urban area, the city of Abidjan predominates, responsible for 26.9% of consumption. In the other districts, in addition to the primary sector and the food industry, the use of transport and communication services stands out. In addition, rural households consume their own produced commodities, highlighting the consumption of yam (23.3%), rice (16.3%) and cassava (12.8%).

Figure 6. Household consumption patterns (by rural/urban and districts). Côte d’Ivoire 2015.

Urban Rural Zanzan Yamoussoukro Woroba Vallée du Bandama Savanes Sassandra-Marahoué Montagnes Lagunes Lacs Gôh-Djiboua Denguélé Comoé Bas-Sassandra Abidjan

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Primary sector Food industry Petroleum and mining Textile and clothing Other manufactured Transport and communications Other services Utilities

Source: Own elaboration with Côte d’Ivoire SAM 2015. After explaining the household consumption pattern, it is important to explain the commodities composition of exports and imports. In the case of exports, the primary sector and food industry represent the 53.20%, of which 33.76% refers to cocoa exports (including cocoa and cocoa products). Then, crude oil and petroleum represent 20.83%, and manufactured products 19.41% (of which 6.35% refers to machinery and other equipment). For imports, the manufacturing sectors stand out with 51.22%, due to imports of machinery and equipment (24%) and chemicals (11.56%) (Figure 7).

Figure 7. Exports and imports composition. Côte d’Ivoire 2015. Export Imports Utilities; 0.76% Services; 5.79% Services; 11.18% Construction; 0.01% Construction; 0.04% Primary sector and Utilities; 0.02% food industry; 18.53% Manufactured ; 19.41% Primary sector and food industry; 53.20%

Petroleum and Petroleum and mining; 20.83% Manufactured ; 51.22% mining; 19.01%

Source: Own elaboration with Côte d’Ivoire SAM 2015.

Focusing on productive factors, labour, land, livestock, and capital represent 55.56%, 9.27%, 0.13% and 33.45% of the value added, respectively (Table 6). Analysing employment, unskilled labour represents almost half of the employment (46.67%), mainly focused on the primary sector and services. Considering both semi-skilled and skilled labour, their representation in the primary sector is low, as opposed to services and to some extent to manufacturing. In the case of capital, in addition to services (52.24%), the petroleum and mining sector stands out (19.38%).

Table 6. Distribution of factors by aggregate activities. Côte d’Ivoire 2015.

Labour Land Livestock Capital Unskilled Semi-skilled Skilled Primary sector 16.93% 7.83% 1.15% 100% 100% 5.33% Food industry 2.54% 1.35% 0.49% - - 13.03% Petroleum and mining 1.25% 0.75% 0.41% - - 19.38% Manufactured 4.63% 4.42% 1.70% - - 7.56% Utilities 0.31% 0.43% 0.11% - - 2.45% Construction 2.97% 2.22% 1.82% - - 0.01% Services 18.04% 16.44% 14.21% - - 52.24% Total 46.67% 33.43% 19.89% 100% 100% 100% % of the Value added 55.56% 9.27% 0.13% 33.45% Source: Own elaboration with Côte d’Ivoire SAM 2015. Continuing with the analysis of the distribution of value added by activities (Figure 8), services stand out representing 45.86%, followed by home production. Households as activities (focus on agricultural activities) represent 25.56% of the value added, which includes all the land and livestock factor, 25.9% of labour factors and 5.3% of capital factors. In the case of services, capital and labour factor are particularly important with 52.2% and 48.7% respectively.

Figure 8. Distribution of factors and Value Added by aggregate activities. Côte d’Ivoire 2015.

45.86% 52.2% Services 48.7%

3.81% 0.01% Construction 7.0%

0.92% 2.5% Utilities 0.9%

8.77% 7.6% Manufactured 10.7%

7.99% Petroleum and 19.4% 2.4% mining 7.09% 13.0% Food industry 4.4%

25.56% 5.3% 25.9% Primary sector 100% 100% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Value Added Capital Labour Livestock Land

Source: Own elaboration with Côte d’Ivoire SAM 2015. Finally, the distribution of households' income by factors of production and by transfers from enterprises, the government, and the rest of the world, is represented in Table 7. Almost

56.45% of households’ income comes from compensation to labour and 28.54% from capital incomes. According to the information provided in the SAM, 28.6% of the income received from compensation to labour and 34.8% of the income received form compensation to capital corresponds to households from Abidjan. Hence, land and livestock factor income are concentrated for the districts Bas-Sassandra, Lacs and Sassandra-Marahoué. Focusing the analysis on each district, capital represented more than 50% of the incomes in Autonomous district of Yamoussoukro and more than 35% for Abidjan, Lagunes and Savanes. The transfers received by the government are more concentrated in Denguélé (representing 14.3% of its income). Mora than 80% of the income received by Bas-Sassandra and Zanzan come from labour and land factors. Both labour and land are key factors in income in the rural zones, representing 72% of the incomes.

Table 7. Distribution of households’ income. Côte d’Ivoire 2015.

Factors income Transfers Labour Land Livestock Capital Government Rest of the World Côte d’Ivoire 56.45% 9.47% 0.13% 28.54% 4.08% 1.33%

Urban 60.47% 35.68% 2.90% 0.95% Rural 51.97% 20.04% 0.27% 20.57% 5.39% 1.76% Abidjan 61.86% 38.07% 0.05% 0.02% Bas-Sassandra 63.16% 17.37% 0.30% 14.57% 3.47% 1.13% Comoé 51.66% 6.96% 0.09% 33.02% 6.23% 2.04% Denguélé 50.38% 15.10% 0.20% 15.35% 14.30% 4.67% Gôh-Djiboua 49.87% 8.39% 0.12% 33.86% 5.85% 1.91% Lacs 58.74% 18.03% 0.23% 16.29% 5.06% 1.65% Lagunes 53.58% 3.70% 0.05% 35.29% 5.56% 1.82% Montagnes 59.15% 11.53% 0.13% 20.22% 6.76% 2.21% Sassandra-Marahoué 54.87% 18.46% 0.21% 22.73% 2.81% 0.92% Savanes 43.84% 11.41% 0.17% 36.71% 5.93% 1.94% Vallée du Bandama 53.60% 5.40% 0.06% 34.91% 4.49% 1.47% Woroba 50.48% 15.72% 0.21% 20.89% 9.57% 3.13% Yamoussoukro 38.69% 0.38% 0.01% 54.64% 4.73% 1.55% Zanzan 58.77% 21.23% 0.27% 9.38% 7.80% 2.55% Source: Own elaboration with Côte d’Ivoire SAM 2015.

4.2 Linear multiplier analysis The linear SAM models are a simple way to explore the information provided by a SAM regarding the structure of an economy. This methodology enables to analyse the linkages between sectors through the calculation of multipliers, which is useful for studying the impact of different policies (Round, 2003b). In this section the output, value added, and employment multipliers are calculated. A brief explanation of the calculation of the three multipliers is included in Annex 2.

4.2.1 Multipliers The analysis of the multiplier effect shows those commodities with capacity to generate output, employment and value added above average, and therefore, identifies which ones are suitable to be promoted through policies. The analysis of multipliers is a useful tool for ex-ante policy evaluation, however, the exact results obtained for each multiplier should be taken with some caution due to some limiting assumptions of the SAM linear models (Mainar-

Causapé et al., 2018a). These limitations are mainly related to the excess capacity in all sectors and unemployed factors of production (no supply constraints) and fixed prices (not taking substitutions effects into account) (Miller & Blair, 2009; Round, 2003b). The values of the multipliers calculated for the Ivoirian economy in 2015 are presented in Tables 8 and 9. The cells are shaded to compare those below or above the average. Each value of the output multiplier shows the increase in gross output in sectors, due to one-unit exogenous injection into the final demand of a commodity (i.e. an exogenous increase in exports demand). As for example, the output multiplier of yam indicates that a one-unit increase in exogenous demand in yam leads to 2.65 of output increase in the economy. Likewise, the value added multiplier indicates the new value added created by the additional production in responses to an exogenous shock in demand. Finally, the values of the employment multipliers measure the increment in the number of jobs generated by the exogenous increase in demand. Taking the multipliers of Table 8 into account, the primary sector group showed a higher backward income generation capacity than the economy average. Within agriculture, the calculation of the output multiplier highlights the importance of several products, with major values for forestry (2.98), plantain (2.73), cassava (2.69), yam (2.65), hevea (2.65) and maize (2.62). On the other hand, rice, fishing, and other oilseeds have a weaker impact on the economy, both in production, value added and employment. Regarding the employment multiplier, the values are clearly above average for most primary sector products, with highest impact for plantain, cotton, maize, cashew, and yam. Analysing this group of commodities, the distribution of employment generated by the multiplier is concentrated with over 70% in “unskilled workers”, while the impact on “semi-skilled workers” represents between 18% and 30% and the impact on “skilled workers” is weak. Moreover, the jobs generated through the multiplier are predominantly in rural areas of Côte d’Ivoire, representing about 70% of the impact. For some commodities, the employment multiplier impact predominates in specific regions, as in the case of maize and cotton, where 34% of the multiplier effect is distributed in Savanes. For example, the case of Yam where the impact on employment is concentrated in Sassandra-Marahoué (37% of the multiplier effect), the case for plantain and fruits predominated in Montagnes (37% and 28%), or the case of cashew in Zanzan (34%). Following on with the multiplier analysis, it shows that many of the commodities classified under the food industry sector, such as meat and fish, grain milling, dairy, beverages and tobacco have very low multiplier values in output, value added and employment. Within this group, only cocoa products and oilseed have an output and value added multipliers above average. For the food industry sector, the employment multiplier has very low values and the jobs generated by an exogenous impact of demand are distributed similar to the agriculture group but decreasing the impact on “unskilled workers” and slightly increasing the impact on “skill workers”. The employment multiplier effects within this group of products are similarly distributed between rural and urban areas. However, for cocoa and grain milling the impact of the jobs generated in the rural region is slightly higher.

Table 8. Linear multipliers of primary sector and food industry commodities for Côte d’Ivoire 2015.

Multipliers Description Output Value added Employment Maize 2.62 1.65 0.89 Rice 1.85 1.19 0.58 Yam 2.65 1.74 0.87 Cassava 2.69 1.74 0.81 Plantain 2.73 1.74 0.92 Other food crops 2.42 1.56 0.70 Cocoa 2.42 1.53 0.76 Coffee 2.42 1.53 0.82 Cotton 2.63 1.64 0.91 Primary Fruits and nuts 2.39 1.51 0.81 sector Palm 2.63 1.66 0.84 Other oilseeds 2.01 1.27 0.64 Hevea 2.65 1.74 0.62 Cashew 2.56 1.67 0.89 Other industrial 2.40 1.52 0.79 products Plants and seeds 2.59 1.66 0.81 Livestock 2.56 1.64 0.81 Forestry 2.98 1.34 0.72 Fishing 1.89 0.97 0.52 Meat and fish 1.53 0.68 0.37 Grain milling 1.56 0.81 0.43 Cocoa products, coffee 2.64 1.38 0.53 Oilseed 2.77 1.34 0.49 Food industry Bakery 2.72 1.03 0.51 Dairy and fruits 1.70 0.98 0.37 products Beverages 2.00 1.09 0.54 Tobacco 1.37 0.72 0.30 Average 2.27 1.31 0.61 Source: Own elaboration with Côte d’Ivoire SAM 2015. Shaded cells: value greater than the average Focusing on Table 9, the results indicate that among the manufactured and utilities sectors, only wood products and furniture and other manufacturing stand out in terms of job creation. Even though some products show output multipliers above the average, they do not generate a high value added multiplier (for example rubber and plastic products, wood products and utilities). The crude oil and mining sector show low values for the three multipliers, with only other mining having a slightly above average value added multiplier. The construction has output multiplier above the average and employment and value added multipliers a slightly above average. In the case of services, they show above-average values in terms of output, value added and employment for trade, repair services, accommodation and food services, education and health and social work. Other commodities within this group have an output multiplier above the global average, but significantly lower in terms of employment (such as postal and telecommunications services, real estate, and public administration).

For those that stand out with higher employment multipliers, the analysis of the distribution of their impact shows that, although unskilled workers predominate, services also distribute their impact towards semi-skilled workers (approx. 30%) and skill workers (approx. 8%). Moreover, the distribution is higher for urban regions, and is particularly more concentrated in Abidjan.

Table 9. Linear multipliers of manufactures, utilities, services, and other commodities for Côte d’Ivoire 2015.

Multipliers Description Output Value added Employment Crude oil 1.48 1.00 0.28 Petroleum Other mining 2.17 1.47 0.39 and mining Petroleum products 1.76 0.77 0.25 Textiles 2.19 1.05 0.56 Leather and footwear 2.02 1.12 0.64 Wood products 2.77 1.33 0.74 Paper and printing 1.61 0.89 0.30 Chemical products 1.14 0.58 0.25 Manufacture Rubber and plastic products 2.40 1.23 0.50 Non-metallic minerals 1.97 0.96 0.60 Metals and metal products 1.53 0.65 0.31 Machinery and other equipment 0.48 0.29 0.16 Furniture and other manufacturing 2.46 1.43 0.76 Electricity, gas and steam, water Utilities 3.19 1.40 0.53 supply and sewage Construction Construction 2.78 1.32 0.73 Trade services 2.67 1.66 0.92 Repair services 2.57 1.51 0.86 Accommodation and food services 2.76 1.63 0.86 Transport services 2.21 1.17 0.57 Postal and telecommunications 2.51 1.54 0.47 services Services Financial services 2.05 1.24 0.32 Real estate 2.75 1.58 0.48 Business Services 2.07 1.38 0.36 Public administration and defence 2.61 1.61 0.54 Education 2.58 1.81 0.98 Health and social work 2.77 1.56 0.75 Other services 1.64 1.04 0.56 Average 2.27 1.31 0.61 Source: Own elaboration with Côte d’Ivoire SAM 2015. Shaded cells: value greater than the average

5 Conclusions This report presents a detailed Côte d’Ivoire SAM for 2015 expanded to include district-level households, factors of production and agricultural production. It explains its accounts and the main data sources for its construction. In summary, the new Côte d’Ivoire SAM 2015 presented in this report contains 247 accounts: 50 activities (13 for household as producers), producing 54 marketed commodities and 10 ’home-production for home-consumption (HPHC) commodities’. It considers 3 type of labour (skilled, unskilled, and semi-skilled) in 14 districts and disaggregated in rural and urban (81 labour accounts), 1 type of land for each district (13 land factor accounts) and 1 livestock factor and 1 capital factor. Regarding households, each of the 13 districts is divided in urban and rural areas, and there is one urban account for the city of Abidjan (27 households’ accounts). The disaggregation across Côte d’Ivoire's districts is very important to promote policies focused on equality and development in the country. Finally, there is one account for margins, one corresponding to enterprises, other for the government, the rest of the world, and one for savings and investment. In Annex 1, Table A1 shows all accounts considered in the Côte d’Ivoire SAM 2015. The SAM can be freely downloaded from DataM portal, as explained in the Annex 3. For Côte d'Ivoire, there is a 2006 and a 2015 SAM, which have a lower level of disaggregation and fewer account types than the one built in this report (Diallo, 2018; Fofana, & Diallo, 2015). The disaggregation and particularities of this new SAM make it a novel database suitable for future research work with a focus on agriculture and food security. The main objective is to have an appropriate database that can be used for the analysis of policies and incentives to improve the social and economic conditions of the country. Previous studies in developing countries used SAMs as databases to analyse policies and impacts related to food security, trade, agricultural development, among others (Boulanger et al., 2018; 2019; Mainar-Casupé et al., 2020; Nechifor et al., 2021; Osman et al. 2021;). The output, value added, and employment multipliers are calculated in this report to showcase the usefulness of the SAM for agricultural and development policies. Although their interpretation should be taken with caution due to the limitations of the methodology, they are useful for an initial impact analysis (Mainar et al., 2020). To this end, the results obtained show the significance of the primary sector, which has a higher impact on Côte d'Ivoire's economy, both in terms of output generation, value added and job creation. Therefore, these sectors would be the most suitable to be promoted through specific policies. At the same time, the SAM documented in this report is expected to be used in other ex-ante policy evaluations notably through the application of country-level CGE models.

References Aragie, E.; Dudu, H.; Ferrari, E.; Mainar-Causapé, A.; McDonald, S & Thierfielder, K, STAGE_DEV. A variant of the STAGE model to analyse developing countries, EUR 28627 EN, JRC Technical Reports, Publications Office of the European Union, Luxembourg, 2017, doi:10.2760/90737. Bacharach, M. Biproportional matrices and input-output change (Vol. 16), CUP Archive, 1970. Boulanger, P., Dudu, H., Ferrari, E. & Mainar Causape, A. Matrice de comptabilité sociale désagrégée de l'économie sénégalaise en 2014, EUR 28979 EN, Publications Office of the European Union, Luxembourg, 2017, doi:10.2760/563430. Boulanger, P., Hasan, D., Emanuele, E., Mainar-Causapé, A., Angelucci, F., Baborska, R., & Meilland, T. Allocations budgétaires optimales et options de réformes pour le secteur agricole dans le Plan Sénégal Emergent 2019-2023, Publications Office of the European Union, Luxembourg, 2018, doi:10.2760/729645. Boulanger, P., Ferrari, E., Mainar Causapé, A., Sartori, M., Beshir, M., Hailu, K. & Tsehay, S. Policy Options to support the Rural Job Opportunity Creation Strategy in Ethiopia, EUR 29949 EN, Publications Office of the European Union, Luxembourg, 2019, doi:10.2760/76450. Boulanger, P., Dudu, H., Ferrari, E., Mainar-Causapé, A. J., & Ramos, M. P. Effectiveness of fertilizer policy reforms to enhance food security in Kenya: a macro–micro simulation analysis. Applied Economics, 2020, pp. 1-21. https://doi.org/10.1080/00036846.2020.1808180 Burfisher, M. E. Introduction to computable general equilibrium models (2n Edition), Cambridge University Press, Cambridge, 2016. Defourny, J., & Thorbecke, E. Structural Path Analysis and Multiplier Decomposition within a Social Accounting Matrix Framework. The Economic Journal, Vol. 94, No 373, 1984, pp. 111– 136. https://doi.org/10.2307/2232220 Diallo, S. S. Matrice de Comptabilité Sociale de la Côte d’Ivoire. Rapport. Direction de Prévision, des Politiques et Statistiques Economiques (DPPSE), Direction Générale de l’Economie, Ministère de l’Economie et des Finances – Côte d’Ivoire, 2018. Ducroque, H., Tillie, P., Louhichi, K. & Gomez-Y-Paloma, S. L'agriculture de la Cote d'Ivoire a la loupe, JRC Working Papers, Publications Office of the European Union, Luxembourg, 2018, JRC107214. Ferreira, V., Almazán-Gómez, M.Á., Nechifor, V. & Ferrari, E. Social Accounting Matrix for Ghana 2015, EUR 30720 EN, Publications Office of the European Union, Luxembourg, 2021, doi:10.2760/432014. Fofana, I & Diallo, S.S. La Matrice de Comptabilite Sociale de la Côte d'Ivoire : Exercice 2006, AGRODEP Data Report N0 6, 2015. INS. Enquête sur le Niveau de Vie des Ménages en Côte d’Ivoire (ENV 2015). Institut National de la Statistiques, Côte d’Ivoire, 2015. Junius, T., & Oosterhaven, J. The solution of updating or regionalizing a matrix with both positive and negative entries. Economic Systems Research, Vol. 15, No 1, 2003, pp. 87-96. Ministère de L’Agriculture et du Développement Rural. Programme National D'Investissement Agricole de Deuxième Génération (2017-2025) - National Agricultural Investment Program (PNIA II - 2017-2025). 2017.

Mainar-Causapé, A..; Boulanger, P.; Dudu, H.; Ferrari, E & McDonlad, S. Social Accounting Matrix of Kenya 2014, EUR 29056 EN, JRC Technical Reports, Publications Office of the European Union, Luxembourg, 2018a, doi: 10.2760/852198. Mainar-Causapé, A..; Ferrari, E & McDonlad, S. Social accounting matrices: basic aspects and main steps for estimation, EUR 2929 EN, JRC Technical Reports, Publications Office of the European Union, Luxembourg, 2018b, doi: 10.2760/010600. Mainar‐Causapé, A., Boulanger, P., Dudu, H., & Ferrari, E. Policy impact assessment in developing countries using Social Accounting Matrices: The Kenya SAM 2014. Review of Development Economics, Vol. 24 No. 3, 2020, pp. 1128-1149. McDougall, R.A. Entropy Theory and RAS are Friends, GTAP, Working Paper, 300, Department of Agricultural Economics, Purdue University, 1991. Mengistu, A.; Woldeyes, F.; Dessie, E.; Ayalew, Z.; Yeshineh, A.; Mainar-Causapé, A. & Ferrari, E. Ethiopia Social Accounting Matrix 2015/16, EUR 29902 EN, Publications Office of the European Union, Luxembourg, 2019, doi:10.2760/974668. Miller, R & Blair, P. Input-Output Analysis: Foundations and Extensions (2nd edition), Cambridge University Press, Cambridge, 2009. Morokong, T & Ferrari, E. The launch of the Pan-African Network for Economic Analysis of Policies (PANAP) in Addis Ababa, Ethiopia. AgriProbe, Vol. 17, No 1, 2020, pp. 40-41. https://hdl.handle.net/10520/EJC-1cdb9a8ec7 Nechifor V., Boysen O., Ferrari E., Hailu K, & Beshir M. Socioeconomic COVID-19 impacts and recovery in Ethiopia, EUR 30484 EN, JRC Technical Reports, Publications Office of the European Union, Luxembourg, 2020a, doi:10.2760/827981. Nechifor V., Ferrari E., Kihiu E., Laichena J. Omanyo D., Musamali R. & Kiriga B. COVID-19 impacts and short-term economic recovery in Kenya, EUR 30296 EN, Publications Office of the European Union, Luxembourg, 2020b, doi:10.2760/767447. Nechifor, V., Ramos, M. P., Ferrari, E., Laichena, J., Kihiu, E., Omanyo, D., ... & Kiriga, B. Food security and welfare changes under COVID-19 in Sub-Saharan Africa: Impacts and responses in Kenya. Global food security, Vol. 28, 2021, pp. 100514. Osman, R., Ferrari, E., Mainar Causapé, A. & Jimenez Calvo, S., Can the Nile generate output, income and employment in Egypt A mixed multiplier analysis, NEW MEDIT, Vol. 20, No. 1, 2021, pp. 3-17. Pulido, A & Fontela, E. Análisis Input-Output: modelos, datos y aplicaciones. Pirámides, Madrid, 1993. Pyatt, G & Round, J. Social Accounting Matrices: a Basis for Planning, The World Bank, Washington, 1985. Round, J. Constructing SAMs for Development Policy Analysis: Lessons Learned and Challenges Ahead, Economic Systems Research, Vol. 15, No 2, 2003a, pp. 161-183. Round, J. Social accounting matrices and SAM-based multiplier analysis. The impact of economic policies on poverty and income distribution: Evaluation techniques and tools, Vol. 261, 2003b, pp. 276.

Robinson, S., Cattaneo, A., & El-Said, M. Estimating a Social Accounting Matrix Using Cross Entropy Methods. Trade and Macroeconomics Discussion Paper No. 33. International Food Policy Research Institute, 1998. Robinson, S.; Cattaneo, A & El-Said, M. Updating and Estimating a Social Accounting Matrix Using Cross Entropy Methods. Economic System Research, Vol. 13, No 1, 2001, pp. 47-64. Stone, R. Measurement of national income and the construction of social accounts, Naciones Unidas, Ginebra, 1947. Temurshoev, U., Miller, R. E., & Bouwmeester, M. C. A note on the GRAS method. Economic Systems Research, Vol. 25, No 3, 2013, pp. 361-367.

List of abbreviations and definitions A Activities accounts AH Representative households’ groups as activities C Commodities accounts CH HPHC Commodities CEM Cross Entropy Method CGE Computable General Equilibrium CIRES Centre Ivoirien de Recherches Economiques Et Sociales DataM JRC data portal of agro-economic modelling DIRTAX Direct taxes ENTERP Enterprise ENV Côte d’Ivoire 2015 Household Standard of Living Survey EXPTAX Export taxes FCAP_NA Capital factors (non-agriculture) FLAB Labour factors FLAND Land factors FLIVST Livestock factor GOVERT Government HPHC Home Production for Home Consumption HH Households accounts I-S Investment- Savings INDTAX Tax and subsidy on production IMPTAX Import taxes PANAP Pan-Africa Network for Economic Analysis of Policies RoW Rest of the World accounts RHG Representative Household Group SAM Social Accounting Matrix VATTTAX Indirect taxes on products Sales taxes TRCOST Margins

List of figures Figure 1. The circular flow (simple version) ...... 6 Figure 2. A Social Accounting Matrix (SAM) standard structure ...... 8 Figure 3. Domestic absorption, imports, and exports as % of the GDP. Côte d’Ivoire 2015...... 21 Figure 4. Domestic absorption composition. Côte d’Ivoire 2015...... 21 Figure 5. Household consumption pattern. Côte d’Ivoire 2015...... 22 Figure 6. Household consumption patterns (by rural/urban and districts). Côte d’Ivoire 2015...... 23 Figure 7. Exports and imports composition. Côte d’Ivoire 2015...... 23 Figure 8. Distribution of factors and Value Added by aggregate activities. Côte d’Ivoire 2015...... 24 Figure A1. QR code – DataM URL: https://datam.jrc.ec.europa.eu ...... 40 Figure A2. QR Code – direct bulk data download in English: https://datam.jrc.ec.europa.eu/datam/perm/dataset/d6d3504f-ace0-4211-8714- 3862268d5e25/download/Dataset_JRC_-_Social_accounting_matrix_-_Ivory_Coast_-_2015.zip ...... 40 Figure A3. QR Code – direct bulk data download in French: https://datam.jrc.ec.europa.eu/datam/perm/dataset/3411da02-b5d2-470e-a8dd- 92373a476fde/download/Dataset_JRC_-_Matrice_de_comptabilit__sociale_-_C_te_d_Ivoire_-_2015.zip ...41 Figure A4. Bulk download of the matrix in flat table format ...... 41 Figure A5. QR Code – direct link to the data warehouse page of the dataset in English: https://datam.jrc.ec.europa.eu/datam/perm/dataset/d6d3504f-ace0-4211-8714-3862268d5e25 ...... 42 Figure A6. QR Code – direct link to the data warehouse page of the dataset in French: https://datam.jrc.ec.europa.eu/datam/perm/dataset/3411da02-b5d2-470e-a8dd-92373a476fde ...... 42 Figure A7. Data warehouse page of the Ivory Coast SAM 2015 ...... 43 Figure A8. Visualising the SAM on the screen ...... 43 Figure A9. Exporting only part of the SAM in flat format ...... 44 Figure A10. QR Code – direct link to the interactive dashboard: https://datam.jrc.ec.europa.eu/datam/mashup/SAM_CI_2015 ...... 44 Figure A11. Navigating within the sheets ...... 44 Figure A12. Selecting language ...... 45 Figure A13. A generic dashboard ...... 45 Figure A14. Making an interactive selection ...... 46 Figure A15. Result of an interactive selection ...... 46 Figure A16. Instructions to change selections ...... 47 Figure A17. How to download the SAM in traditional matrix format ...... 47 Figure A18. Traditional matrix outcome ...... 48 Figure A19. Downloading a SAM with codes ...... 48

List of tables Table 1. District and regional description in Côte d’Ivoire SAM 2015...... 13 Table 2. Côte d’Ivoire SAM 2015 expressed in millions of FCFA- West African CFA francs (aggregated values)...... 14 Table 3. Activities disaggregated in Côte d’Ivoire SAM 2015 ...... 17 Table 4. Commodities disaggregated in Côte d’Ivoire SAM 2015 ...... 18 Table 5. Description of factors in Côte d’Ivoire SAM 2015 ...... 19 Table 6. Distribution of factors by aggregate activities. Côte d’Ivoire 2015...... 24 Table 7. Distribution of households’ income. Côte d’Ivoire 2015...... 25 Table 8. Linear multipliers of primary sector and food industry commodities for Côte d’Ivoire 2015...... 27 Table 9. Linear multipliers of manufactures, utilities, services, and other commodities for Côte d’Ivoire 2015...... 28 Table A.1. Accounts of the Côte d’Ivoire Social Accounting Matrix 2015 ...... 36

Annexes Annex 1. Accounts of the Côte d’Ivoire Social Accounting Matrix 2015

Table A.1. Accounts of the Côte d’Ivoire Social Accounting Matrix 2015

Code Description Code Description ah_f_BaS Bas-Sassandra (Activities-Households as producers) flab_urb_USK_Abi Urban Unskilled labour Abidjan (Factors) ah_f_Com Comoé (Activities-Households as producers) flab_urb_SSK_Abi Urban Semi-skilled labour Abidjan (Factors) ah_f_Dng Denguélé (Activities-Households as producers) flab_urb_SKL_Abi Urban Skilled labour Abidjan (Factors) ah_f_GDj Gôh-Djiboua (Activities-Households as producers) flab_urb_USK_BaS Urban Unskilled labour Bas-Sassandra (Factors) ah_f_Lac Lacs (Activities-Households as producers) flab_urb_SSK_BaS Urban Semi-skilled labour Bas-Sassandra (Factors) ah_f_Lgn Lagunes (Activities-Households as producers) flab_urb_SKL_BaS Urban Skilled labour Bas-Sassandra (Factors) ah_f_Mnt Montagnes (Activities-Households as producers) flab_urb_USK_Com Urban Unskilled labour Comoé (Factors) Sassandra-Marahoué (Activities-Households as ah_f_Sma producers) flab_urb_SSK_Com Urban Semi-skilled labour Comoé (Factors) ah_f_Sav Savanes (Activities-Households as producers) flab_urb_SKL_Com Urban Skilled labour Comoé (Factors) ah_f_VdB Vallée du Bandama (Activities-Households as producers) flab_urb_USK_Dng Urban Unskilled labour Denguélé (Factors) ah_f_Wor Woroba (Activities-Households as producers) flab_urb_SSK_Dng Urban Semi-skilled labour Denguélé (Factors) ah_f_Yam Yamoussoukro (Activities-Households as producers) flab_urb_SKL_Dng Urban Skilled labour Denguélé (Factors) ah_f_Zan Zanzan (Activities-Households as producers) flab_urb_USK_GDj Urban Unskilled labour Gôh-Djiboua (Factors) a_fore Forestry (Activities) flab_urb_SSK_GDj Urban Semi-skilled labour Gôh-Djiboua (Factors) a_fish Fish (Activities) flab_urb_SKL_GDj Urban Skilled labour Gôh-Djiboua (Factors) a_coil Crude oil (Activities) flab_urb_USK_Lac Urban Unskilled labour Lacs (Factors) a_omin Other mining (Activities) flab_urb_SSK_Lac Urban Semi-skilled labour Lacs (Factors) a_meatfish Meat - Fish processed (Activities) flab_urb_SKL_Lac Urban Skilled labour Lacs (Factors) a_gmll Grain milling (Activities) flab_urb_USK_Lgn Urban Unskilled labour Lagunes (Factors) a_cocoacoff_ind Cocoa products, coffee (Activities) flab_urb_SSK_Lgn Urban Semi-skilled labour Lagunes (Factors) a_oilseed_ind Oilseed industry (Activities) flab_urb_SKL_Lgn Urban Skilled labour Lagunes (Factors) a_bakery_ind Bakery (Activities) flab_urb_USK_Mnt Urban Unskilled labour Montagnes (Factors) a_dair Dairy products and fruits products (Activities) flab_urb_SSK_Mnt Urban Semi-skilled labour Montagnes (Factors) a_beve Beverages (Activities) flab_urb_SKL_Mnt Urban Skilled labour Montagnes (Factors) a_ptob Tobacco (processed) (Activities) flab_urb_USK_Sma Urban Unskilled labour Sassandra-Marahoué (Factors) a_text Textile & clothing (Activities) flab_urb_SSK_Sma Urban Semi-skilled labour Sassandra-Marahoué (Factors) a_leat Leather & footwear (Activities) flab_urb_SKL_Sma Urban Skilled labour Sassandra-Marahoué (Factors) a_wood Wood (Activities) flab_urb_USK_Sav Urban Unskilled labour Savanes (Factors) a_papr Paper & Printing and publishing (Activities) flab_urb_SSK_Sav Urban Semi-skilled labour Savanes (Factors) a_petr Petroleum (Activities) flab_urb_SKL_Sav Urban Skilled labour Savanes (Factors) a_chem Chemicals (Activities) flab_urb_USK_VdB Urban Unskilled labour Vallée du Bandama (Factors) a_ruber_ind Rubber and plastic (Activities) flab_urb_SSK_VdB Urban Semi-skilled labour Vallée du Bandama (Factors) a_othnmet Non-metallic minerals (Activities) flab_urb_SKL_VdB Urban Skilled labour Vallée du Bandama (Factors) a_metl Metals and metal (Activities) flab_urb_USK_Wor Urban Unskilled labour Woroba (Factors) a_mach Machinery and equipment (Activities) flab_urb_SSK_Wor Urban Semi-skilled labour Woroba (Factors) a_furniture_ind Furniture and other manufacturing (Activities) flab_urb_SKL_Wor Urban Skilled labour Woroba (Factors) a_energy Electricity (Activities) flab_urb_USK_Yam Urban Unskilled labour Yamoussoukro (Factors) a_cons Construction (Activities) flab_urb_SSK_Yam Urban Semi-skilled labour Yamoussoukro (Factors) a_trad Wholesale and retail trade (Activities) flab_urb_SKL_Yam Urban Skilled labour Yamoussoukro (Factors) a_repairs Maintenance / Repair (Activities) flab_urb_USK_Zan Urban Unskilled labour Zanzan (Factors) a_hotl Accommodation and food services (Activities) flab_urb_SSK_Zan Urban Semi-skilled labour Zanzan (Factors) a_tran Transportation and storage (Activities) flab_urb_SKL_Zan Urban Skilled labour Zanzan (Factors) a_comm Information and communication (Activities) flab_rur_USK_BaS Rural Unskilled labour Bas-Sassandra (Factors) a_fsrv Finance and insurance (Activities) flab_rur_SSK_BaS Rural Semi-skilled labour Bas-Sassandra (Factors) a_real Real estate activities (Activities) flab_rur_SKL_BaS Rural Skilled labour Bas-Sassandra (Factors) a_bsrv Business services (Activities) flab_rur_USK_Com Rural Unskilled labour Comoé (Factors) a_padm Public administration (Activities) flab_rur_SSK_Com Rural Semi-skilled labour Comoé (Factors) a_educ Education (Activities) flab_rur_SKL_Com Rural Skilled labour Comoé (Factors)

a_heal Health and social work (Activities) flab_rur_USK_Dng Rural Unskilled labour Denguélé (Factors) a_osrv Other services (Activities) flab_rur_SSK_Dng Rural Semi-skilled labour Denguélé (Factors) ch_maiz Maize (home consumed commodities) flab_rur_SKL_Dng Rural Skilled labour Denguélé (Factors) ch_rice Rice (home consumed commodities) flab_rur_USK_GDj Rural Unskilled labour Gôh-Djiboua (Factors) ch_yam Yam (home consumed commodities) flab_rur_SSK_GDj Rural Semi-skilled labour Gôh-Djiboua (Factors) ch_cass Cassava (home consumed commodities) flab_rur_SKL_GDj Rural Skilled labour Gôh-Djiboua (Factors) ch_plantain Plantain (home consumed commodities) flab_rur_USK_Lac Rural Unskilled labour Lacs (Factors) ch_othcrops Other food crops (home consumed commodities) flab_rur_SSK_Lac Rural Semi-skilled labour Lacs (Factors) ch_fruits Fruits and nuts (home consumed commodities) flab_rur_SKL_Lac Rural Skilled labour Lacs (Factors) Other industrial products (home consumed ch_othind commodities) flab_rur_USK_Lgn Rural Unskilled labour Lagunes (Factors) ch_plants Plants and seeds (home consumed commodities) flab_rur_SSK_Lgn Rural Semi-skilled labour Lagunes (Factors) ch_livstk Livestock (home consumed commodities) flab_rur_SKL_Lgn Rural Skilled labour Lagunes (Factors) c_maiz Maize (commodities) flab_rur_USK_Mnt Rural Unskilled labour Montagnes (Factors) c_rice Rice (commodities) flab_rur_SSK_Mnt Rural Semi-skilled labour Montagnes (Factors) c_yam Yam (commodities) flab_rur_SKL_Mnt Rural Skilled labour Montagnes (Factors) c_cass Cassava (commodities) flab_rur_USK_Sma Rural Unskilled labour Sassandra-Marahoué (Factors) c_plantain Plantain (commodities) flab_rur_SSK_Sma Rural Semi-skilled labour Sassandra-Marahoué (Factors) c_othcrops Other food crops (commodities) flab_rur_SKL_Sma Rural Skilled labour Sassandra-Marahoué (Factors) c_cocoa Cocoa (commodities) flab_rur_USK_Sav Rural Unskilled labour Savanes (Factors) c_coff Coffee (commodities) flab_rur_SSK_Sav Rural Semi-skilled labour Savanes (Factors) c_cott Cotton (commodities) flab_rur_SKL_Sav Rural Skilled labour Savanes (Factors) c_fruits Fruits and nuts (commodities) flab_rur_USK_VdB Rural Unskilled labour Vallée du Bandama (Factors) c_palm Palm (commodities) flab_rur_SSK_VdB Rural Semi-skilled labour Vallée du Bandama (Factors) c_othoils Other oilseeds (commodities) flab_rur_SKL_VdB Rural Skilled labour Vallée du Bandama (Factors) c_latex Hevea (commodities) flab_rur_USK_Wor Rural Unskilled labour Woroba (Factors) c_cashew Cashew (commodities) flab_rur_SSK_Wor Rural Semi-skilled labour Woroba (Factors) c_othind Other industrial products (commodities) flab_rur_SKL_Wor Rural Skilled labour Woroba (Factors) c_plants Plants and seeds (commodities) flab_rur_USK_Yam Rural Unskilled labour Yamoussoukro (Factors) c_livstk Livestock (commodities) flab_rur_SSK_Yam Rural Semi-skilled labour Yamoussoukro (Factors) c_fore Forestry (commodities) flab_rur_SKL_Yam Rural Skilled labour Yamoussoukro (Factors) c_fish Fishing (commodities) flab_rur_USK_Zan Rural Unskilled labour Zanzan (Factors) c_coil Crude oil (commodities) flab_rur_SSK_Zan Rural Semi-skilled labour Zanzan (Factors) c_omin Other mining (commodities) flab_rur_SKL_Zan Rural Skilled labour Zanzan (Factors) c_meatfish Meat and fish (commodities) fland_BaS Land capital Bas-Sassandra (Factors) c_gmll Grain milling (commodities) fland_Com Land capital Comoé (Factors) c_cocoacoff Cocoa products, coffee (commodities) fland_Dng Land capital Denguélé (Factors) c_oilseed Oilseed (commodities) fland_GDj Land capital Gôh-Djiboua (Factors) c_bakery Bakery (commodities) fland_Lac Land capital Lacs (Factors) c_dair Dairy products and fruits products (commodities) fland_Lgn Land capital Lagunes (Factors) c_beve Beverages (commodities) fland_Mnt Land capital Montagnes (Factors) c_ptob Tobacco (commodities) fland_Sma Land capital Sassandra-Marahoué (Factors) c_text Textiles (commodities) fland_Sav Land capital Savanes (Factors) c_leat Leather and footwear (commodities) fland_VdB Land capital Vallée du Bandama (Factors) c_wood Wood products (commodities) fland_Wor Land capital Woroba (Factors) c_papr Paper and printing (commodities) fland_Yam Land capital Yamoussoukro (Factors) c_petr Petroleum products (commodities) fland_Zan Land capital Zanzan (Factors) c_chem Chemicals products (commodities) flivst Livestock capital (Factors) c_ruber Rubber and plastic products (commodities) fcap_na Other Capital (Factors) c_othnmet Non-metallic minerals (commodities) hh_urb_Abi Abidjan Urban (Institutions) c_metl Metals and metal products (commodities) hh_urb_BaS Bas-Sassandra Urban (Institutions) c_mach Machinery and other equipment (commodities) hh_urb_Com Comoé Urban (Institutions) c_furniture Furniture and other manufacturing (commodities) hh_urb_Dng Denguélé Urban (Institutions) Electricity, gas and steam, water supply and sewage c_energy (commodities) hh_urb_GDj Gôh-Djiboua Urban (Institutions) c_cons Construction (commodities) hh_urb_Lac Lacs Urban (Institutions)

c_trad Trade services (commodities) hh_urb_Lgn Lagunes Urban (Institutions) c_repairs Repair services (commodities) hh_urb_Mnt Montagnes Urban (Institutions) c_hotl Accommodation and food services (commodities) hh_urb_Sma Sassandra-Marahoué Urban (Institutions) c_tran Transport services (commodities) hh_urb_Sav Savanes Urban (Institutions) c_comm Postal and telecommunications services (commodities) hh_urb_VdB Vallée du Bandama Urban (Institutions) c_fsrv Financial services (commodities) hh_urb_Wor Woroba Urban (Institutions) c_real Real estate (commodities) hh_urb_Yam Yamoussoukro Urban (Institutions) c_bsrv Business Services (commodities) hh_urb_Zan Zanzan Urban (Institutions) c_padm Public administration and defence (commodities) hh_rur_BaS Bas-Sassandra Rural (Institutions) c_educ Education (commodities) hh_rur_Com Comoé Rural (Institutions) c_heal Health and social work (commodities) hh_rur_Dng Denguélé Rural (Institutions) c_osrv Other services (commodities) hh_rur_GDj Gôh-Djiboua Rural (Institutions) trcost Transaction costs hh_rur_Lac Lacs Rural (Institutions) dirtax Direct taxes hh_rur_Lgn Lagunes Rural (Institutions) indtax Tax and subsidy on production hh_rur_Mnt Montagnes Rural (Institutions) vattax Indirect taxes on products hh_rur_Sma Sassandra-Marahoué Rural (Institutions) imptax Import taxes hh_rur_Sav Savanes Rural (Institutions) exptax Export taxes hh_rur_VdB Vallée du Bandama Rural (Institutions) enterp Enterprises / Companies hh_rur_Wor Woroba Rural (Institutions) govert Government hh_rur_Yam Yamoussoukro Rural (Institutions) row Rest of the World hh_rur_Zan Zanzan Rural (Institutions) i-s Investment-savings Source: Own elaboration.

Annex 2. The linear SAM model and multipliers The linear SAM models are a simple way to analyse the information provided by a SAM database regarding the structure of an economy and the linkages within its account. This type of analysis is useful to study the impact of different policies through the calculation of multipliers.

6 The starting point is the Leontief's equilibrium equation 풚풏 = 푨풏풚풏 + 풙 ( ), applied to the case of a SAM obtaining the SAM Leontief inverse (Pyatt & Round, 1985). The standard representation is as follows: 푀 = (퐼 −A), where the matrix 퐴 is the coefficient matrix (calculating dividing each element of the SAM by the total of their corresponding column), and each element 푚 in 푀 shows the output requirements of account 푖 to increase the final demand of account 푗 by one unit (Mainar-Causapé et al., 2018a). This matrix is used as a tool to evaluate the capacity of each economic sector to generate output and employment in the rest of the economy through the analysis of multipliers. Output multiplier

The sum of the multiplier values of 푴 (corresponding to the commodities columns and the rows of productive activities of 푀), shows the output multiplier. This multiplier indicates the final increase in gross output of all production activities generated due to a unitary exogenous shock in exogenous values for the corresponding commodity. A high value of this multiplier indicates an account with a large backward income expansion influence on the rest of the economy, given its interdependence with other sectors (Pulido & Fontela, 1993). Employment multiplier The employment multiplier measures the impact in the number of jobs that would be generated by an exogenous shock in final demand. For its calculation it is necessary to have the vector of employment 풆, which represents the ratios between the number of jobs and the output of each activity (per million of output value). The elements of 풆 in diagonal form the employment matrix 푬, that is multiplied by the part of the multiplier matrix (푴) with rows corresponding to the productive accounts and the columns corresponding to commodities. The employment multiplier is defined as: 푴풆 = 푬 × 푴풂, of which element 풎풆풊풋 is the increment in the number of jobs in sector 푖 when there is a unit exogenous injection into the endogenous final demand account of 푗. The sum of the columns in the matrix shows the global effect on employment produced by the exogenous increase in demand (number of jobs per million of additional output from each activity (Mainar-Causapé et al., 2018a). Due to the model assumptions mentioned above, the results should not be interpreted as an accurate forecast of job creation resulting from exogenous shocks. Furthermore, the results do not take into account social variables, such as the quality of employment. However, the results can be useful as an indicator of the commodities in the economy with the greatest potential for employment generation. Value added multiplier The value added multiplier relates the new value added created in each sector by the additional production in response to an exogenous shock in demand (Miller & Blair, 2009). Similar to the calculation of the employment multiplier, a value added vector can be used to calculate the value added multiplier. The vector 풗 contains the ratios between the value added and the output of each activity.

(6) Assuming Leontief technologies (i.e. with fixed prices and no substitution elasticities), the result should be taken with caution.

Annex 3. On-line resources to download the Côte d’Ivoire Social Accounting Matrix 2015

Figure A1. QR code – DataM URL: https://datam.jrc.ec.europa.eu

Source: JRC, 2021.

Bulk download Using DataM, users can make a bulk download of SAM in a ZIP file containing a CSV file in English or French. The hyperlinks for the direct bulk download is in Figure A2 and Figure A3.

Figure A2. QR Code – direct bulk data download in English: https://datam.jrc.ec.europa.eu/datam/perm/dataset/d6d3504f- ace0-4211-8714-3862268d5e25/download/Dataset_JRC_-_Social_accounting_matrix_-_Ivory_Coast_-_2015.zip

Source: JRC, 2021.

Figure A3. QR Code – direct bulk data download in French: https://datam.jrc.ec.europa.eu/datam/perm/dataset/3411da02- b5d2-470e-a8dd-92373a476fde/download/Dataset_JRC_-_Matrice_de_comptabilit__sociale_-_C_te_d_Ivoire_-_2015.zip

Source: JRC, 2021.

In the bulk download, the SAM is presented in a standard flat format as CSV file with header row. Conceptually, it contains a column for the spending agent, a column for the receiving agent and a column for the value in Millions of FCFA. See figure A4.

Figure A4. Bulk download of the matrix in flat table format

Source: DataM, provided by the European Commission – Joint Research Centre. Dataset: JRC – SAM Ivory Coast - 2015, accessed on 03/06/2021.

In fact, the file contains also columns for the codes internally used in GAMS for the agents and the Year (always 2015). The extra columns help for using the data in modelling tools, and for characterizing this file among other SAM's that are published by JRC.

Interactive download

— DataM includes also a function for interactive download, which allows filtering the only part of interest of the datasets and to preview results on the screen. Find the direct link for the SAM in the figure A5.

Figure A5. QR Code – direct link to the data warehouse page of the dataset in English: https://datam.jrc.ec.europa.eu/datam/perm/dataset/d6d3504f-ace0-4211-8714-3862268d5e25

Source: JRC, 2021.

Figure A6. QR Code – direct link to the data warehouse page of the dataset in French: https://datam.jrc.ec.europa.eu/datam/perm/dataset/3411da02-b5d2-470e-a8dd-92373a476fde

Source: JRC, 2021.

— The links give access to the screen in the figure A7.

Figure A7. Data warehouse page of the Ivory Coast SAM 2015

Source: DataM, JRC, 2021.

— After specifying some filtering and pushing on the "Next" button, the data is visualised on the screen, see figure A8.

Figure A8. Visualising the SAM on the screen

Source: DataM, JRC, 2021.

— The "Download results as CSV" option would produce a file similar to the one obtained in the bulk download, but only with the records reflecting the selection operated. See figure A9.

Figure A9. Exporting only part of the SAM in flat format

Source: DataM, provided by the European Commission – Joint Research Centre. Dataset: JRC – SAM Ivory Coast - 2015, accessed on 03/06/2021.

Interactive dashboard

— Finally, users may explore and analyse the data through an interactive dashboard placed in the “PANAP network” section of the website (Figure A10). This dashboard includes both English and French languages.

Figure A10. QR Code – direct link to the interactive dashboard: https://datam.jrc.ec.europa.eu/datam/mashup/SAM_CI_2015

Source: JRC, 2021.

— The interactive dashboard allows users to undertake their own analysis of the dataset. — It consists of a number of sheets that allow analysing data from different perspectives.

Figure A11. Navigating within the sheets

Source: DataM, JRC, 2021.

The language can be selected using the component showed in Figure A12:

Figure A12. Selecting language

Source: DataM, JRC, 2021.

Each sheet consists of a screen with a number of different visualisations (tables, charts and maps) and some filtering boxes. The key strength of the tool is that all these visualisations are interactive and interrelated. This allows users to study the data by means of simple mouse gestures. The DataM visualisation framework is quite intuitive; some basic guidelines to facilitate its use will follow. All DataM dashboards are similar to the example shown in Figure A13.

Figure A13. A generic dashboard