Estimation of detailed macroeconomic data using general equilibrium modeling for Ethiopia

Abstract

There is an increasing demand for a detailed database for policy analysis and the development of new economic models. Thus, detailed macroeconomic data i.e., social accounting matrix (SAM) is useful in providing detailed information on all current economic activities within the economy. Hence, there are several works on constructing social accounting matrix based on various methods. However, constructing a new SAM regularly is challenging and costly which is the other major issue discussed in the literature. This is because the construction of SAM is supplemented by data collected from several sources such as national accounts, Input-Output tables, labor force surveys, household surveys, and government accounts. Hence the availability of all the data within the same time frame is one of the main challenges. So, updating the existing SAM based on the available information is mostly used to estimate the recent social accounting matrix. Hence there is vast literature dedicated to estimating a recent SAM by employing several methods such as RAS, Cross entropy as well as updating the GTAP data baseline. These updating approaches are purely based on either iterative algorithm or information theory to minimize the entropy motivated by maximum likelihood estimation. However, economic theory can also be used to update the existing database. For instance, several papers make use of general equilibrium modeling to update the existing baseline with recent data or projections. This approach is mainly based on introducing an exogenous shock to major macroeconomic variables and relying on economic theory such as macroeconomic identities and calibrated elasticity parameters, to estimate a new detailed database. Hence, this thesis aims to implement an analogous experiment to estimate the recent detailed macroeconomic database for Ethiopia with the reference year of 2019. Mainly, by considering real changes to major economic drivers such as value-added by sector, private & government household expenditure, trade balance, and demographic changes. The specific objectives of this study are -1) to estimate a detailed macroeconomic database disaggregated into ten production activities, five factors of production, and two regions, and 2) to provide a descriptive assessment of consequent changes on key macroeconomic variables by comparing it with the initial 2014 SAM for Ethiopia. To estimate detailed macroeconomic data this experiment will make use of the GTAP 10 database with the reference year of 2014. This database depicts values of the economic variable which are presented on aggregate terms and are suitable for computable general equilibrium modeling. Hence, the GTAP 10 database with 141 regions, 65 sectors, and 5 factors will be updated to estimate new detailed macroeconomic data. The database will be updated based on major macroeconomic drivers with the reference year of 2019. Mainly, recent data on the demographics, value-added by sector, import and export by sector, and private and government consumption expenditure will be collected from the national account, world economic outlook 2019, and world governance indicators from the world bank 2019. The data collected from WDI (2019), and WEO (2010) are measured in USD currency, thus the official exchange rate will be used to convert the values into local currency. The updating experiment will be based on the available standard CGE model which is offered by GTAP along with simulation software i.e., RunGTAP. Finally, this thesis will be organized based on five sections. The first section includes an introduction followed by a literature review in the second section. Third section of the thesis will consist of the structure and estimation methodology of the social accounting matrix for Ethiopia. The fourth section will include a discussion of the results followed by a conclusion in the last section.