The enclave hypothesis and Dutch disease effect: A critical appraisal of Mali's gold mining industry

The main purpose of this paper is to test whether there are significant backward linkages from gold mining to manufacturing in Mali over the period 1982–2017. The analysis also examines the relevance of the ‘Dutch disease’ effect in Mali, where the discovery and exploitation of natural resources lead to a real exchange rate appreciation and a subsequent contraction of the manufacturing sector. Related studies on Mali are scarce, but those that do exist find support for Singer's enclave hypothesis, which states that mining is disconnected from the rest of the economy. A novel feature of the empirical strategy adopted in this paper relative to previous research is that we use both survey data collected from fieldwork and a vector error-correction model (VECM) to test for linkage effects and the Dutch disease hypothesis. The evidence shows that there are significant and persistent backward linkages from gold mining to manufacturing in Mali, thus providing a more optimistic view than the enclave hypothesis of previous research. The survey data, backed up by the econometric evidence, find that both gold mining companies and local suppliers are eager to do business with each other, and that both expect to strengthen their business ties in the near future. At the same time, however, the evidence suggests that there remains ample scope to strengthen backward linkages. We draw on the survey data analysis to suggest several practical policy measures. The econometric evidence rejects the Dutch disease hypothesis. The significant long-run effect of the real exchange rate on manufacturing output, however, provides another channel through which natural resource windfalls could boost industrial development in Mali. Overall, the evidence suggests that the natural resource sector can play a significant role to initiate industrial development, and that government policy is a key conduit to establish and strengthen backward linkages.