【正文】
growth in a relatively large closed economy and a relatively small open economy. The empirical estimation is based on an augmented production function . The Dickey and Fuller and Phillips and Perron unit root test statistics are used to examine the stationarity of the data. The ohansen cointegration method is used to examine the longrun relationship between terms of trade and real GDP per capita. The cointegration method is able to test the existence of more than one cointegrating vector in a generalised forecast error variance deposition and generalised impulse response function are used to examine the rela tionship of variables in a system. The generalised forecast error variance depositions and generalised impulse response functions solve the orthogonalised problem of the forecast error variance depositions and impulse response functions of Sims. The problem is that the latter approaches are sensitive to the order of the variables in the vector autoregressive (VAR) system. This article is organised as follows. Section 2 discusses the literature review of terms of trade and economic growth. Section 3 explains the methodology and data in this study. Section4 presents empirical results and discussions. Section 5 provides some concluding remarks. 2. Literature review of terms of trade and economic growth There are many studies on the relationship between terms of trade and economic growth. Mendoza examines the impact of terms of trade on economic growth in 40 industrial and developing countries using crosssectional data for the period 1971–1991. The results show a positive impact of terms of trade on economic growth. Moreover, terms of trade shocks are reported to account for nearly half the changes in actual GDP. Furthermore, economic growth is found to be slower in countries with higher terms of trade volatility. Bleaney and Greenaway examine the impact of terms of trade on economic growth in 14 subSaharan African countries using crosssectional data for the period 1980–1995. They ?nd that an increase in terms of trade will lead to an increase in economic growth. Hadass and Williamson examine the relationship between terms of trade and economic growth using crosssectional data for the period 1870–1940. They ?nd that although terms of trade movement favours primary product exporters, it reduces their economic growth. Moreover, there is strong evidence of asymmetry in economic growth between the core and the periphery. Generally, the impact of terms of trade on economic growth is very small for both the core and the periphery. In the prewar period, changes in terms of trade explained less than one?fth of economic growth, which is expressed by the GDP per capita growth rate. Noheless, the study covers few of the developing countries that remained poor up to World War II. On the other hand, Blattman et al. analyse more than half a century of preWorld War II data, 1870–1938, for 35 countries. They ?nd that terms of trade have a significant impact on economic growth. Moreover, the impact of terms of trade is greater on the periphery than in the core. A decline