Energy efficiency as a sustainability concern in Africa and financial development: How much bias is involved?

This study contributes to the literature on whether financial development stimulates technical energy efficiency (TEE) or not, by addressing core biases that creep into the relationship and thereby reducing the ability to draw causal inferences from financial development to TEE. Our approach is based on the instrumental stochastic frontier technique, where biases in the frontier and inefficiency equations are dealt with using external instrumental variables. The legal system origin of the country and life expectancy at birth were used as instruments for financial development and income, respectively. The current study demonstrates substantial bias in income elasticity, the estimate of energy efficiency, and the effect of financial development on energy efficiency. Both income elasticity and energy efficiency estimate risk upward bias. Equally, the effect of financial system development and financial institution development on TEE risk downward bias. Other results show that all aspects of financial institution development stimulate TEE, but access to financial institutions is more important. These results raise caution about future studies' estimates of the effect of financial development on TEE. Though this study has demonstrated the potency of external instruments in dealing with the bias in the coefficient estimates, we consider this might prove to be a luxury solution in some cases due to data limitation, context differences, and theory. In those circumstances, reliance on internal instruments might prove to be the second-best option.