East Africa Literature Review

CHAPTER 3

LITERATURE REVIEW

3.1 Introduction

In the past, a number of studies have attempted to unravel the dynamics behind trade and development in the East African region but the results have been mixed and in some instances, confusing. Thus, whereas most of the studies reviewed so far reveal East African intra-regional trade to be low (i.e. less than 10 percent of total trade flows for most East African countries except for Kenya), studies sponsored by USAID-REDSO and carried out by Ackello Ogutu and Protase Echessa (1996) and (1998) show the opposite. The Ogutu et.al.,1996, study entitled “Unrecorded Cross Border Trade between Kenya and Uganda”, and “Unrecorded Trade Between Tanzania and Neighbouring Countries, 1998”, respectively, show

Alemayehu and Kibret (2002), for instance, carried out a study that determined that bilateral trade flows in the East African region are best explained by standard Gravity Model variables such as GNP and GNP per capita, infrastructure variables e.g. road length per 1000 people, number of telephones per 100 people, policy variables (such as FDI flows, parallel market premium and capital deepening), political factors (occurrence of war, coup revolution), and cultural and geographical factors (language, sharing borders and landlocked states). The study methodology utilized in this study was a censored Tobit Gravity model similar to those used by Longo and Sekkat, 2001, Elbadawi, 1997, Forotutan and Prichett 1993. The main findings of this study entailed a demonstration that good macro-economic policies such as financial deepening and infrastructural development are important determinants of bilateral trade in East Africa. Interestingly, regional integration arrangements were found to have had no or limited impact on intra-regional Trade flows. This was rather surprising given that the key objective of the Study was “Regional Economic Integration in Africa: A Review of Problems and Prospects with a Case Study of COMESA.” Similarly, results showed that the proxy used to measure political instability, war, especially, did not have the expected sign. Indeed intra-COMESA trade was found to be not

Key among these is Niringiye et al., (2010) that was designed to determine factors that impact export trade participation of agricultural manufacturing firms in the East African region. This study was based on the premise that in order for East African agricultural manufacturing companies to obtain a global competitiveness (comparative advantage), it is imperative for them to have an indication of the factors that influence their export participation in the global market. The study utilized a Probit Regression Analysis model (Baltagi, 2005) that was designed based on the fact that a firm’s decision to export was governed mainly by its level of “sunk” costs. This was in line with similar models developed by Dixit (1989) and Krugman (1989). Similar studies had also been carried out by Robert and Tybout (1997) and Bernard and Wagner (2001). The Probit Regression Analysis Format is based on the strong and often defensible observation that a firm’s decision to export is quite often governed by its magnitude of “sunk” costs or past investments. Thus, as propounded by the economy-wide Harrod-Domar model of investment, it is assumed that an export firm will seek to determine a unique rate of investment that will result in an equilibrium rate of growth at which there will be no excess capacity. This