Energy efficiency and financial depth nexus revisited: does the choice of instrumental variable and measure of financial depth matter?

This study examines the effect of finance on energy efficiency in Africa, addressing two fundamental empirical issues in the energy-finance literature: (1) simultaneous modelling of efficiency estimates and determinants of efficiency and (2) two-way endogeneity problem with income and financial depth. I apply the endogenous stochastic frontier method. Life expectancy at birth instruments for income while religion, latitude of capital city and legal system origin instrument for financial depth. Estimated efficiency in electricity consumption is between 9 and 16%, which is quite low, suggesting greater potential to improve electricity consumption efficiency. Without considering the level of development, the result shows that finding a statistically significant negative effect of financial depth on energy inefficiency depends on the choice of instrument and not the indicator of financial depth. Using country latitude and legal system origin as instruments produce robust statistically significant negative effect for all the indicators of financial depth, but high monetisation of the economy and supply of loanable funds exert the greatest effect, recommending between 0.069 and 0.096% reduction in inefficiency for every 1% increase in these indicators. However, both the choice of instruments and indicator of financial depth matter, when I consider income level. I discuss the policy implications.