The present global economic scenario has rekindled interest on natural resources, conflict and growth issues. While for a vast body of literature natural resources are a curse rather than a blessing, different explanations have been attributed as the likely causes. The Dutch Disease explanation has two variants: it may be via the low saving and investment route. Or, alternatively, it may be via the exchange rate mechanism. Some studies have emphasised the negative impact emanating from rent seeking and poor governance. While there is a whole body of literature that attributes the slow growth of resource abundant economies to the absence of the rule of law, property rights on account of institutional decay, etc. Conflict has also been found to be an important factor explaining the dismal performance in resource abundant economies.
In order to study the impact of natural resources and conflict on growth I undertook this study to investigate the impact of natural resources and conflict on growth for a panel of 97 countries for the period 1980 to 2009. These 97 countries included both developed countries (DCs) and the less developed countries (LDCs). Several variables affect GDP differently in the DCs as compared with the LDCs. For example, conflict can have a destabilising effect on growth in the LDCs on account of death, destruction, loss of infrastructure, uncertainty, etc. But since most conflicts do not place in rich countries, their work force and infrastructure are not affected as a result of conflict. And as Kidron stated 47 years ago, conflict increases the demand for armaments, and since arms constitute a large component of the GDP in the developed world , increase in conflict through increasing arms exports increases GDP in the DCs. Conflict, therefore, is expected to have a positive impact on growth in the DCs on account of increase in GDP through increase in armament exported by the DCs and a negative impact on growth in the LDCs on account of death, destruction of infrastructure, instability, etc in the latter. On account of these differential impacts emanating from many of our explanatory variables to the dependant variable, we estimated the models separately for the DCs and the LDCs.
Our dependant variable GDP is Gross Domestic Product in constant (year 2000) US $. Labour is measured by the percentage of secondary school enrolment in total enrolment. Capital is measured in terms of gross fixed capital formation in constant dollars in the year 2000. The impact of government expenditure on growth, measured in constant US$ in 2000, is the percentage of final government consumption as a percentage of GDP. Trade openness is exports + imports as a percentage of GDP. Our natural resource variable is oil, gas and coal production and not export of the same as a percentage of GDP. Coal and oil production in million tons while gas production given in the World Development Indicators (WDI) was in billion cubic meters, the latter was converted into million cubic meters. While ore and metals are export of ores and metals/GDP for individual LDCs. These data have been taken from World Development Indicator (WDI) CD ROM 2009, 10 and 11.
For the rich countries we find that the coefficient for physical capital is large and significant at the 100 percent confidence level, whereas the coefficient for human capital is negative but insignificant. Government expenditure is positive and significant at the 6 percent level, while natural resources and openness have a positive though insignificant impact. The really interesting finding is that conflict captured by world deaths has a positive and significant impact on rich countries output. The R-square is almost 73 percent, reflecting that 73 percent of the variations in rich countries GDP are explained by the model.
For the LDCs both physical and human capital have a large and highly significant impact on LDC GDP. Government expenditure also has a positive impact, although it is not significant at any of the conventional levels. Openness has a negative and significant impact on GDP in the LDCs. While the contribution of natural capital becomes positive and significant, contradicting the predominant ‘resource curse’ view. And the really interesting finding is that conflict has a negative and significant impact on GDP in the LDCs.
We find that natural resources are not a curse per se. That the transmission mechanism of resource abundance on output is through conflict. It is only when resource abundant high performing countries are engaged in conflict that their performance nose dives. These findings lend support to earlier writers who observed that Botswana had a remarkable growth performance due to the absence of conflict and Sierra Leone remained poor as it remained bogged down with conflict. Our stance that the transmission mechanism is via conflict supports the stance that increase in defence capabilities of a country reduces conflict. Our finding lends credibility to the importance of external factors in lowering growth in resource abundant countries.