Who benefits from utility subsidies?
description
Transcript of Who benefits from utility subsidies?
1
Who benefits from utility subsidies?
Caroline van den BergK. Komives, V. Foster, J. Halpern, Q. Wodon and R. Abdullah
September 13, 2006
2
Consumer utility subsidies widespread
Most residential customers are not charged the full cost of the water and electricity service they receiveAverage residential tariffs only cover O&M plus some capital costs in:– 3% of water utilities in low-income countries– 39% of water utilities in upper middle income countries– 50% of water utilities in high income countries
3
Why are these subsidies a concern?
Subsidies can be justifiedConcern about…– Impact of “unfunded” subsidies on customers – Effect of subsidies on incentives
• Inefficient water consumption• Disincentives on utilities to reduce costs, expand service
– Do subsidies really reach the poor?• Some case study evidence suggested problems
4
Assumptions behind subsidizationA policy of subsidizing basic services is conceptually appealing…– Cost recovery prices are not considered “affordable”– Subsidies are assumed to help guarantee service for all– In-kind transfers are seen as easy to implement
And… these services also lend themselves to subsidization…– High fixed costs with long asset lives
• Possible to get away with under-pricing in the short term– High percentage of non-attributable costs
• Hard to assign costs to particular customers– Monopoly services
5
How did we answer these questions?
Systematic comparison of case studies– Existing and simulated subsidies– From 13 water utilities and 27 electrical utilities from Asia, Latin
America, Africa, and E.E./C.A.
Estimation of the financial value of the subsidy: – Avg. cost of water or electricity received – amount paid
Determinants of targeting performance– Benefit targeting indicator (% of benefits going to poor / % of pop that is
poor): <1.00 regressive; > 1.00 progressive– Access and uptake rates, targeting, subsidy per unit, quantity consumed
and cost
6
Who benefits from utility subsidies?
Beneficiary targeting: Do subsidies reach the poor?Benefit targeting: What percentage of the benefits go to the poor?
Why care about targeting?– Lower subsidy budget – More bang for the buck– Fewer distortions in consumption decisions
7
Beneficiary targeting
The evidence about subsidies reaching the poor
0%
20%
40%
60%
80%
100%
Poorest 2nd 3rd 4th Richest
Access to WSS by income quintile
Water supply Sanitation
8
Benefit targeting:Existing quantity based subsidies regressive
% of poor hhs receiving subsidy vs. benefit targeting performance
India, State IBTs, 0.56
Cape Verde, 0.48Sao Tome, 0.41
P eru, 0.82Honduras, 0.49
Guatemala, 0.20
Hungary (S), 0.98
Rwanda (S), 0.35
Kathmandu, 0.56
Bangalore, 0.66Sri Lanka, 0.83
Cape Verde, 0.240.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
70.0%
80.0%
90.0%
100.0%
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Benefit targeting performance indicator
Sh
are
of
po
or
hh
s re
ceiv
ing
su
bsi
dy
Electricity
Water
9
Access, connection, and metering– Many poor households are
simply not eligible
Quantity-consumed is not necessarily a good indicator of poor households
Water consumption by quintile
0
5
10
15
20
25
30
35
40
45
50
1 2 3 4 5
Quintile
Cu
bic
met
ers
per
mo
nth
Why?
10
Why? (2)
Most existing subsidies are general subsidies to all or almost all residential customers– Few households pay average cost or cross-subsidize others
Quantity-targeted subsidies usually provide a greater subsidy per unit to low volume consumers, but… – A smaller subsidy over more units of consumption = a larger total subsidy– If there is a fixed fee, the smallest volume users pay the highest average price
per unit
Can quantity-targeted subsidies be improved by tinkering with the tariff structure?– E.g. reducing the size of the subsidized block of an IBT
11
Can the targeting of quantity-based subsidies be improved?
Tinkering with the tariff won’t help households without access, connections, or meters
Two changes could help marginally improve targeting performance:– Very small first block of an IBT– Moving from an IBT to a “volume-differentiated tariff”
But only if – (1) poor consume significantly less than the non-poor, and – (2) average residential tariffs are at or near average cost
Improvement would be easiest for utilities with a high degree of cost recovery, located in areas with high coverage of connections and meters
12
Other alternative forms of subsidies
Service-level targeting of consumption subsidies– Self-selection: are the poor more likely than non-poor to choose a
lower service level?• Subsidies for public water taps
– Bangalore Ω = 2.14; Kathmandu Ω = 1.54
Are connection subsidies doing better?– Benefit goes beyond the cost of the connection itself– Simulations results are all progressive
• But will the poor be offered connections? And will they connect?• How large is the group of consumers that are poor and not connected?
13
Other alternative forms of subsidiesGeographical targeting
– Pro: better targeting than quantity based subsidies– Cons:
• Can work if poor and non-poor live in different areas• Many customers must pay the cost for geographical targeting
Means testing targeting– Pro: better targeting than previous types– Cons:
• High administrative cost• High errors of exclusion
14
Subsidies as social transfers
Consumption(Quantity)
Geographic Means-testing Self-selection(Service level)
Water
Existing Cons 0.60 1.05 1.36 1.84Sim. Cons 0.78 0.86 1.19 -
Sim. Conn - 1.30 1.71 -
Electricity
Existing Cons 0.63 0.90 1.23 -
Sim. Cons 0.64 - 1.39 -
Other* 1.00 1.33 1.40 1.78
Median targeting performance, by type of subsidy and targeting method
*From Coady et al study on targeting performance of social transfers
15
Parting thoughts:Subsidies as “pro-poor” utility policy
Keeping services affordable for the poor?– Only for the connected poor (with meters), who are
accurately identified by the targeting mechanism
What about low coverage situations?– Connection subsidies are often more likely to reach the
poor, but…• There may be other barriers to connections (tenure status, cost of
fixtures, billing practices, good alternatives)• Connecting more households to a service burdened by
“unfunded” subsidies will only further erode bankrupt utilities
16
Parting thoughts:Subsidies as “pro-poor” utility policy
Our assumptions on the poor and their demand for WSS services are not always very accurate:– Whether subsidies are needed is based on assumptions of
demand and knowledge on household behavior on utility level
17
Parting thoughts:Prices, subsidies, and cost recovery
There is no easy way around the need to increase levels of cost recovery if service is to be improved and expanded.– The removal of existing regressive subsidies is widely unpopular.– Improving the targeting of subsidies won’t change that.
But raising prices or securing alternative sources of subsidies are not the only possible tools:– Improving revenue collection– Reducing operating and especially capital costs – Removing impediments to more flexible service levels, technologies,
and modes of provision