Document Library

Key Text Utility Subsidies as Social Transfers: An Empirical Evaluation of Targeting Performance

Author: Kristin Komives et. al.
Date: 2007
Size: 21 pages (218 KB)

Access full text: available online

Summary

Do utility subsidies actually help the poor? This paper from the Development Policy Review argues that the average targeting performance of water and electricity subsidies is similar to that of other social transfer mechanisms using the same targeting method. The most common utility subsidies are consumption-based. These aim to subsidise low-volume users but primarily benefit the non-poor. Many geographically-targeted and most means-tested utility subsidies are progressive, but still exclude many poor households. Connection subsidies are an attractive alternative in low coverage areas, but they will only reach the poor if utilities extend network access to poor households and if households choose to connect.

Recent years have seen numerous empirical studies of a wide range of social transfer programmes, including social funds, food subsidies, public works programmes and housing subsidies. Many of these evaluations are hampered by their focus on single cases, and almost none have addressed water and electricity subsidies. Nonetheless, almost every utility in the developing world includes a targeted subsidy of some variety in their residential tariff structures. Such subsidies involve costs which are much larger, in many cases, than other social transfer mechanisms.

Conclusions from the authors’ analysis are based on a relatively small number of cases. The strongest conclusions relate to quantity-targeted subsidies. Subsidy programmes using means-testing to allocate benefits have a progressive incidence, whether they are water subsidies, electricity subsidies, or other social-policy instruments. Geographic targeting results in a progressive distribution on average for water subsidies and other instruments, but not for electricity subsidies.

Subsidies based on the quantity of a good or service consumed disproportionately go to the non-poor. In many cases, the poor are excluded from the subsidies entirely because they are not connected to the grid or their connection is unmetered. Increasing block tariffs (IBTs) are an example of quantity targeted subsidies.
Even a random distribution of subsidies would probably benefit the poor more than the current quantity-based system.
Overall, the quantity of water or electricity consumed is a poor measure of poverty. This criterion in fact means that almost all households receive subsidies of one form or another.
When compared to other social transfer mechanisms, current utility subsidies look particularly inefficient. While a clear majority of other such mechanisms have a progressive distribution of benefits, utility subsidies are regressive more than half of the time.
On the other hand, this issue is traceable to particularly inefficient targeting methods for water and electricity.

The use of utility subsidies as an in-kind mechanism for wealth transfer should be treated with caution. Administrative capacity will be crucial for the performance of any administrative targeting needed to put these subsidies on a progressive footing.

Subsidies that are targeted on the basis of the quantity of good/service consumed are the poorest performers. These in-kind transfers all have the problem that they generally provide a greater total subsidy to those who consume more of a particular subsidised good.
The best items for in-kind transfer are likely to be "self-targeted" items which are widely and disproportionately used by the poor. Traditional water and electricity services, which are used across social classes, do not fit the model, and it is difficult to find items that are ideal.
Service-level targeting, an example of self-targeting, is an option worth exploring for the directing of subsidies. As with connection subsidies, an analysis would have to map out household demands, constraints and alternatives before moving forward.

Access full text: available online

Source: Komives, K., Halpern, J., Foster, V., Wodon, Q. and Abdullah, R., 2007, 'Utility Subsidies as Social Transfers: An Empirical Evaluation of Targeting Performance', Development Policy Review, Vol. 25, No. 6, pp. 659-679