The Economics of system loss

Abstract

This paper deals with the problem of electricity theft. In developing countries, individuals and firms illegally underpay their electricity bills in various ways. This results in large amounts of unbilled electricity, and causes significant revenue losses to the electricity distribution utilities and hence, presents a case of compliance similar to tax evasion. We argue however that electricity theft, unlike tax evasion, is easier to measure, as it is captured in what is called system loss,defined as unbilled electricity as a percentage of the electricity supplied. Whereas this measure always contains some technical loss due to transmission and distribution channels, a large part of it also contains the stolen amount, since it remains unbilled. We look at changes in system loss in Indian electricity supplying authorities for different states between 2003 and 2012 in response to the stricter legal environment. We expect that electricity theft should respond to these changes, whereas technical loss should be non-responsive. We use the recent provisions for stronger law enforcement institutions stipulated under the Electricity Act 2003 in India to test this hypothesis. The disparity in the timing of adoption of the Act of different states allows us to design a quasi-experiment. We also consider electricity prices to draw a similar analysis corresponding to the effect of tax on tax evasion. We find that adoption of the Act and thereby the creation of a stronger legal institution reduce system loss whereas increases in price, like increases in tax, increases system loss.