Estimating persistent and transient electricity efficiency in the residential sector of developing country

The massive increase in the consumption of electricity in Pakistan also increases the pressure on energy security and environment. As the residential sector accounts for almost half of the total electricity demand, increasing the residential electricity efficiency is critical. This research is based on stochastic frontier analysis on balanced panel data for eleven electricity distribution regions for Pakistan during 2004 to 2023. The True Fixed Effects Model (TFEM) provides an estimate of the average residential electricity efficiency score at 0.854, which means that consumption is 14.6 percent more than the statistically estimated minimum feasible demand. The model by Kumbhakar and Heshmati (1995) further breaks down inefficiency into transient and persistent inefficiency. Average transient efficiency is 0.947 and average persistent efficiency is 0.470, so long-run structural inefficiency has dominated short-run behavioral inefficiency. Persistent efficiency is especially low in FATA, Frontier and Karachi where structural constraints are pronounced. Distribution losses have a negative elasticity of about −0.140 which reflects the system inefficiencies and recording weaknesses Efficiency estimates are statistical benchmarks with respect to a minimum demand frontier and should be considered as upper limits of reductions in demand and not engineering limits. These results provide support for region specific grid upgrades, appliance standards and demand side programs.