Tariff-Adjustments and Intra-African Trade
Description
The literature on the tariff – trade nexus currently provides limited or no guidance on whether tariff changes have symmetric effects on trade, and whether economic size affects trade benefits of tariff changes. An increase in tariffs may produce different trade effects from decreases in tariffs. If such asymmetry is proven, trade projections and policy decisions based on symmetric interpretation of tariff variations are bound to be unstable and misleading. Beyond the possible asymmetric effects of tariff changes, the smooth running of the AfCFTA also depends on the extent to which the trade-effect of tariff changes is contingent on economic size. On one hand, if smaller economies enjoy much smaller trade-gains from tariff reductions than larger economies, then smaller economies may be dis-incentivized from full participation in free trade area activities. On the other hand, if smaller economies gain more from tariff reductions than larger economies, then creation of the free trade area can be expected to reduce inequality and enhance trade integration across Africa. Overall, it is important to consistently track the effect of tariff reductions on AfCFTA member states even post implementation, to provide information for appropriate adjustment of trade rules and bring the best out of the agreement. This study obtains trade-effects of tariff variations, as well as the interaction effects of tariff changes and economic size on bilateral trade. The overall effect of tariff adjustments is measured in terms of gross trade improvement without segregating this into trade diversion, and trade creation components as the focus is on overall trade improvement in Africa. The data covers bilateral trade information on 49 African reporting countries from 1992 to 2016. Variables covered include, the Gross Domestic Product (GDP) of both Reporter and Partner countries (in billions of USD), tariffs, contiguity-whether the trading partners share a common border or not, whether trade partners share a common official language or not, and whether a colonial relationship exists between the pair of trading countries (e.g South Africa and Namibia). Other variables include number of HS6 products imported, and value of imports (in Billions of USD), exchange rates, and institutional quality. The institutional quality variables used are rule of law ratio, and regulatory quality ratio.
Files
Steps to reproduce
This data covers bilateral trade information on 49 African reporting countries from 1992 to 2016. Variables covered include, the Gross Domestic Product (GDP) of both Reporter and Partner countries (in billions of USD), tariffs, contiguity-whether the trading partners share a common border or not, whether trade partners share a common official language or not, and whether a colonial relationship exists between the pair of trading countries (e.g South Africa and Namibia). Other variables include number of HS6 products imported, and value of imports (in Billions of USD), exchange rates, and institutional quality. The institutional quality variables used are rule of law ratio, and regulatory quality ratio. Data on trade variables including imports, tariffs, contiguity, common official language, colonial relationship, and exchange rates were taken from the World Integrated Trade Solutions (WITS) of the World Bank. Data on GDP and exchange rates came from the World Development Indicators (WDI) database of the World Bank, while data on institutional quality came from the World Bank’s Worldwide Governance Indicators database. The rest of the data, covering distance, colonial relationship, contiguity, and common official language came from the CEPII database. Distance presents two measures of the simple distance between two countries; distance between the most populated cities (in km), and distance between the capital cities of two trading countries. Given that the most populated cities tend to be the capital cities, these two variables are actually strongly correlated. We retain distance between capital cities as the measure of distance between pairs of countries. The tariff rate used in this study is the Effectively Applied Tariff (EAT) rate. EAT is defined as the lowest available tariff, and combines information on the preferential tariffs and the Most Favoured Nation (MFN) tariff. If a preferential tariff exists, it is used as the EAT. Otherwise, the MFN applied tariff is used. Importing countries apply the MFN tariff if the product fails to meet rules that determine the product's country of origin. The MFN is the normal non-discriminatory tariff charged on imports. Table 1 summarizes information on trading activity between countries included in the data. As an example, Ghana imported a lot more products from South Africa than it did from any other African country, and exported to Nigeria more than it exported into any other African country. There is some indication that proximity promotes trade as can be seen amongst the countries from Southern Africa, and those from West Africa. For instance, South Africa dealt with Namibia and Botswana more than any other African country.