Oil and Electricity in Small Island Developing States: Unmasking the Risks and Opportunities

Abstract

In the first study of this thesis, I analyse the effects of oil price and global demand shocks on Small Island Developing States (SIDS). To do this, I employ a global vector autoregression (GVAR) framework for 26 SIDS using annual data over the period 1980 to 2015. A key innovation associated with this research is the use of remittance weights to capture the close financial linkages between SIDS and advanced economies such as the United States. I find negligible negative effects on economic growth in SIDS on average from a negative oil price shock, though each country reacts differently. In terms of a negative demand shock to US GDP, SIDS appear vulnerable with a marginal positive effect observed for the Pacific while small negative effects emerge for the Caribbean; Atlantic, Indian Ocean, Mediterranean and South China Sea (AIMS); and SIDS as a whole.

Since oil prices influence electricity prices, the second study looks at how electricity prices influence consumer behaviour in a particular country, using Jamaica as an example. To analyse how consumers respond to prices, I use the autoregressive distributive lag bounds testing approach to cointegration to obtain long-run elasticity estimates covering the period 1970 to 2014. The analysis focuses on aggregate electricity demand and three categories of consumers: residential, commercial, and industrial. The findings suggest that residential and industrial consumers are most responsive to price changes with long-run price elasticities of demand of –0.82 and –0.25, respectively. Price-based approaches are likely to be more successful in slowing electricity demand growth in these sectors.

We can use demand elasticities to analyse the behaviour of alternative electricity pricing schemes under different scenarios. Using Jamaica as a case study in the third study, my analysis confirms that in contrast to a price cap plan that uses Ramsey quantity weights, prices under a revenue cap increase as demand in a particular market becomes more elastic relative to the others. In this specific setting, a revenue cap does encourage energy conservation through reductions in electricity use, but this is less likely when marginal cost is very large in the more elastic market relative to other markets. In contrast to a price cap plan, these overall results show that revenue cap schemes are welfare-reducing.

To analyse the efficiency of electricity sectors in SIDS, I apply a two-stage data envelopment analysis (DEA) method to a sample of 32 electricity distribution utilities operating in the Caribbean and Pacific regions. Using 2010 data for three inputs and one output, the overall results show that six utilities are on the efficiency frontier and utilities in the Caribbean are significantly more efficient than those in the Pacific region. I also do not find any statistically significant evidence that the efficiency of utilities is positively related to population density and inversely related to criminality. Notwithstanding the paucity of publicly available data, these results advance the idea that utility benchmarking is possible and can be used to set feasible efficiency targets for electricity networks in SIDS.