Power and Irrigation Subsidies An
example for: Andhra Pradesh &
Punjab
Maximo Torero
m.torero@cgiar.org
International Food Policy Research
Institute
Based on paper by: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies
in Andhra Pradesh & Punjab. IFPRI, Washington.
IGC-ISI India Development Policy Conference
Major stylized facts of the current pricing
mechanism and subsidy scheme
• Jump in electric pump use
• Jump in the share of electricity consumption in
agriculture
• Huge deficit with respect to revenue
• Growing imbalances: reduction of cross-subsidy
and Subsidy substantially increased
• Deterioration of supply
• Environmental damage
• Subsidy is regressive
Page 2
Jump in electric pump use
Irrigated Area (in lakh ha) and Sources in AI
600
547
500
400
300
247
217
200
160
104
100
1
0
Total
Canal
1961-62
TW
2000-01
• Bet. 1960 and 2000, irrigated area more than doubled.
• This came mostly from tube well (TW) irrigation.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra
Pradesh & Punjab. IFPRI, Washington
Page 3
Jump in electric pump use
Irrigated Area (in lakh ha) and Sources in AP
50
45
40
35
30
25
20
15
10
5
0
45
30
16
13
11
0.2
Total
Canal
1961-62
TW
2000-01
• In AP, the real changes take place in irrigation through TW.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra
Pradesh & Punjab. IFPRI, Washington
Page 4
Jump in electric pump use
Irrigated Area (in lakh ha) and Sources in PJ
45
40
40
35
30
25
30
29
20
13
15
16
10
10
5
0
Total
Canal
1970-71
TW
2000-01
• The role of TW in irrigation is even more prominent in PJ.
• While the role of TW has been increasing, the canal irrigation
has been declining in PJ.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra
Pradesh & Punjab. IFPRI, Washington
Page 5
Jump in electric pump use
Energization of TW (in million)
AI
AP
14
12
12
10
8
1.0
0.8
0.6
0.4
0.2
0.0
6
4
4
2
0
1981
2.0
1.8
1.6
1.4
1.2
1998
PJ
1.8
0.8
0.7
0.7
0.6
0.5
0.4
0.3
0.4
0.3
0.2
0.1
0.0
1981
1998
1981
1998
• With the increased role of TW in irrigation, the energization of
TW took place in a rapid pace.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra
Pradesh & Punjab. IFPRI, Washington
Page 6
Jump in the share of electricity consumption in
agriculture
Electricity Consumption by Consumer Category, 1960-2001
80
70
(in %)
60
50
40
30
20
10
0
1960
1970
1980
1990
1991
1992
Agriculture
1993
1994
Domestic
1995
1996
Industry
1997
1998
1999
2000
2001
Commercial
• Effects of two – first, an increase in the number of pumps, and
second, an increase in electricity consumption per pump set –
has increased the demand for electricity in irrigation by many
folds.
• By 1998, agriculture emerged as a the largest consumer of
electricity in India.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra Pradesh & Punjab. IFPRI, Washington
Page 7
Huge deficit with respect to revenue
Share of Agriculture in Consumption and Revenue
(as a % of Total)
35
30
25
20
15
10
5
0
1994-95 1995-96
1996-97
1997-98
consumption
1998-99
1999-00
2000-01
2001-02
revenue
• However, the consumption share did not match with the revenue
share, and created a financing gap as a result.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra
Pradesh & Punjab. IFPRI, Washington
Page 8
Growing imbalances
(in Paise/Kwh)
Cost of Supply, Avg. Ind. Tariff, Ag. Tariff, and Gross Subsidy
450
400
350
300
250
200
150
100
50
0
1996-97
1997-98
Cost of Supply
1998-99
Avg. Ind. Tariff
1999-00
2000-01
Avg. Ag. Tariff
2001-02
Gross Subsidy
• Partial reforms have not been helpful since the cost of
supply has been going up while the agricultural tariff has not
been reformed
• A reduction in cross-subsidy has added to the odd further
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra Pradesh & Punjab. IFPRI, Washington
Page 9
Environmental damage
Production Pattern and Ground Water Level Fall
between 1981 and 2000 in Andhra Pradesh and Punjab
(a) Andhra Pradesh
(b) Punjab
Page 10
Who are the beneficiaries?
% of HHs owns Electric Pump
Land Possession and Electric Pump Ownership
80
70
60
50
40
30
20
10
0
0
<0.1
<0.4
<0.6
<1
<2
<3
<5
<10
>=10
Land Possession (in Ha)
AI
AP
PJ
Source: Data from 54th round
• There is a very strong link between land possession
and electric pump ownership in AI, and in AP and PJ.
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra Pradesh & Punjab. IFPRI, Washington
Page 11
Distribution of subsidy
All India
35
100
90
30
25
70
60
20
50
15
40
30
10
Total Irr. Area
% of Total Irr. Area
80
20
5
10
t
R
Po
ich
es
9
8
7
6
5
4
3
or
e
2
0
st
0
% of Total Irr. Area
Total Irr. Area
Source: Data from 55th round
•
Not surprisingly, the distribution of irrigated land is extremely
skewed in the case of all India.
Page 12
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra Pradesh & Punjab. IFPRI, Washington
In AP?
Andhra Pradesh
100
90
25
80
70
20
60
15
50
40
10
30
20
5
Total Irr. Area
% of Total Irrigated Area
30
10
% of Total Irr. Area
Ri 9
ch
es
t
8
7
6
5
4
3
0
2
Po
or
es
t
0
Total Irr. Area
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra Pradesh & Punjab. IFPRI, Washington
Page 13
And in Punjab?
100
35
90
80
30
70
25
20
60
50
15
40
30
10
% of Total Irr. Area
Ri 9
ch
es
t
8
7
6
5
0
4
0
3
20
10
2
5
Total Irr. Area (cumulative)
40
Po
or
es
t
% of Total Irrigated Area
Punjab
Total Irr. Area
• And it is worse in Punjab!
Source: Chowdhury, S and Torero, M; (2009). Power and Irrigation Subsidies in Andhra Pradesh & Punjab. IFPRI, Washington
Page 14
Objective of this paper
• Examine the general setting of subsidy in
power for irrigation and canal irrigation
• Identify alternative institutional mechanisms
to rationalize the subsidies
Page 15
From a Vicious Circle
Exogenous Shocks:
Green revolution
Public objective:
Low tariff
Market
outcome:
Excess
demand
Imbalances:
Showtfall in
cross-subsidy Gap between
tariff and cost
Supply outcome:
low quality, not
reliable and
limited
availability
SEBs: Low
Supply
availability
Page 16
To a Virtuous Circle
Exogenous Shocks:
Green revolution
Public objective:
Low tariff
Market
outcome:
Equilibrium
Price
discrimination:
optimal
consumer
plans
Supply outcome:
high quality,
reliable and
availability
SEBs: Supply
availability
Page 17
Methodology
We follow a four step methodology:
• Step 1: Estimate rural households demand for
electricity
• Step 2: Measure consumer welfare
• Step 3: Explore alternative price schemes based on
price discrimination theory
– to better assign the current subsidy,
– to identify ways through which it can be funded through
the market.
Page 18
The Data
• Secondary sources:
– Data aggregated at the level of the state and
district
– Household data
• NSS 54th round (CPR)
• NSS 55th round (Consumption)
AI
AP
PJ
54th Round 55 th Round
78990
70818
5721
5109
2533
2142
Page 19
Step 1: Estimate rural households demand for electricity
• Estimate rural households demand for electricity using the almost ideal
demand system (AIDS) developed in Deaton and Muellbauer (1980)
N good demand system:
X
wi   i   i log    j  ij log p j
P
are constant parameters and X is the representative expenditure on the
system of goods given by:
X  i pi qi
where qi is the quantity demanded for ith good
P is the overall price index derived from:
log P   0  i log pi  i  j  ij log pi log p j
Page 20
AIDS Elasticity's
The own price elasticity for Andhra Pradesh and Punjab
together is -0.5192, and the price elasticity for each of
the consumer groups based on the size of their land
possession can be summarized in:
Own Price Elasticities of power consumption
Andrah Pradesh
All
-0.67
Marginal/ Small holders
-0.69
Medium holder
-0.55
Large holders
-0.50
Punjab
-0.85
-0.91
-0.91
-0.86
Page 21
Step 2: Measure consumer welfare
• Our welfare measure for a given socioeconomic level j, we define
S ( pit ,.) 
j
it
Pmax

j
it
q ( p,.) dp,
pt
Pmax as the maximum price the consumers will observe, which is
instrumentalized by assuming different subsidy regimes
• then including the flat installation charge as an annual installment, the
total net surplus for all services is:
~
j
Sit ( pit , rit )   Sit ( pt )  rit
j
Page 22
Step 3: Explore alternative price schemes based on
price discrimination theory
Non-linear tariffs based on second degree price
discrimination
Assumptions:
• Asymetric information: the firm and the regulator
don’t know the value assign by individuals (farmers)
• The firm and the regulator know the distribution of
probabilities
q  [ q , q* ]
q ~ f(q )
===> Disign of mechanisms
Page 23
Step 3: Explore alternative price schemes based on
price discrimination theory (Cont 1)
1. Discount for quantity:
payment T

t y


 p0

t   p1


 p
n

si
si
si


0  y  x0 

x 0  y  x1 


x n  1  y 

2. Optional Plans:
payment Ti

ti y
(t,T)
 ( t 1 T 1 , ) si

 ( t 2 T 2 , ) si

...

 ( t n T n , ) si

q1 
q 2 
q
n




Page 24
Step 3: Explore alternative price schemes based
on price discrimination theory (cont 2)
• The unit cost (Cj) will depend on the quantity demanded by each farmer
(y), the monthly rental (R), the rate (t) and the number of free Kwh (L),
and is defined for household j as:
cj 
Ri  ti [ y*j  Li ]
y*j
• When consumer plans are introduced, the j-th farmer will choose plan k
that minimizes its expenditure given yj*:
Rk  tk [ y*j  Lk ]  Rm  tm[ y*j  Lm ]; m
• However, consumption plans may change the amount demanded at the
equilibrium point and then farmer will re-adjust
Page 25
Simulation of three progressive pricing
schemes
• The first price scheme is a simple two part payment schedule that
established an initial quantity (q1) priced at p1 (q1 is what
smallholders consume so they will keep same level of subsidy);
while demand exceeding q1 units is priced with marginal cost, i.e.
p2.
• This second mechanism considers a fixed rate (F), under which the
household receives q1 units of electricity. Consumption exceeding
q1 is charged with a marginal cost v1 for households demanding less
than q2 units, and households with consumption exceeding q2 will
pay v2 for additional units.
• The third consumption plan has a variable first part and two
marginal rates. This scenario will eliminate the burden of the
subsidy to the government given small-holders will also pay the first
part of the tariff based on their consumption.
Page 26
Simulation 1: Results of a simple two part tariff
Andhra Pradesh
Information on land size
Potential expenditure
% of sample Average Ha (total) Average Ha Motor (HP) Average HP kwh / Ha
0.36
]0-3]
1.5
500.8
71.33
0.75
0.38
]0-3]
1.5
1725.6
Actual exp
Charge Rs/HP/Year Actual exp
Land size (Ha)
] 0 - 1.8 ]
] 0 - 1.8 ]
Crop
Wheat
Paddy
] 1.8 - 3.64 ]
] 1.8 - 3.64 ]
Wheat
Paddy
16.63
2.45
1.45
1.00
]3-5]
]3-5]
4.0
4.0
] 3.64 + [
] 3.64 + [
Wheat
Paddy
12.05
6.24
3.30
2.94
] 5 - 7.5 ]
] 5 - 7.5 ]
6.3
6.3
kw
First part
price
Exp
kw
840.47
0.40
337.50
0.00
1.18
0.00
337.50
total kw
225
338
840.47
438.8
1511.6
375
1500
2144.60
383
1319
475
2969
5138.60
Two-part tariff
Second part
price
Crop
Wheat
Paddy
] 1.8 - 3.64 ]
] 1.8 - 3.64 ]
Wheat
Paddy
840.47
0.40
337.50
1304.13
1.18
1538.88
1876.38
] 3.64 + [
] 3.64 + [
Wheat
Paddy
840.47
0.40
337.50
4298.13
1.18
5071.80
5409.30
Simulated q
% change in
q
Simulation
% change
Elasticity
price
Exp
Total expenditure
Land size (Ha)
] 0 - 1.8 ]
] 0 - 1.8 ]
Weighted
Avg price
Price t=0
0.40
0.40
0.0%
-0.69
840.47
0.00
Land size (Ha)
] 0 - 1.8 ]
] 0 - 1.8 ]
Crop
Wheat
Paddy
] 1.8 - 3.64 ]
] 1.8 - 3.64 ]
Wheat
Paddy
0.87
0.70
25.1%
-0.55
1847.20
-0.14
] 3.64 + [
] 3.64 + [
Wheat
Paddy
1.05
0.58
82.2%
-0.50
3045.45
-0.41
Page 27
Simulation 1: Results of a simple two part tariff
Punjab
Information on land size
Potential expenditure
% of sample Average Ha (total) Average Ha Motor (HP) Average HP kwh / Ha
0.69
]0-3]
1.5
500.8
30.73
0.94
0.24
]0-3]
1.5
1725.6
Actual exp
Charge Rs/HP/Year Actual exp
Land size (Ha)
] 0 - 1.9 ]
] 0 - 1.9 ]
Crop
Wheat
Paddy
] 1.9 - 3.98 ]
] 1.9 - 3.98 ]
Wheat
Paddy
26.55
2.76
1.81
0.95
]3-5]
]3-5]
4.0
4.0
] 3.98 + [
] 3.98 + [
Wheat
Paddy
42.73
8.21
5.16
3.05
] 5 - 7.5 ]
] 5 - 7.5 ]
6.3
6.3
kw
First part
price
Exp
kw
767.38
1.06
810.00
0.00
1.18
0.00
810.00
total kw
540
810
767.38
438.8
1511.6
540
2160
2224.16
383
1319
720
4500
6002.27
Two-part tariff
Second part
price
Crop
Wheat
Paddy
] 1.9 - 3.98 ]
] 1.9 - 3.98 ]
Wheat
Paddy
767.38
1.06
810.00
1456.77
1.18
1718.99
2528.99
] 3.98 + [
] 3.98 + [
Wheat
Paddy
767.38
1.06
810.00
5234.89
1.18
6177.17
6987.17
Simulated q
% change in
q
Simulation
% change
Elasticity
price
Exp
Total expenditure
Land size (Ha)
] 0 - 1.9 ]
] 0 - 1.9 ]
Weighted
Avg price
Price t=0
1.06
1.06
0.0%
-0.91
767.38
0.00
Land size (Ha)
] 0 - 1.9 ]
] 0 - 1.9 ]
Crop
Wheat
Paddy
] 1.9 - 3.98 ]
] 1.9 - 3.98 ]
Wheat
Paddy
1.14
0.97
17.1%
-0.91
1877.69
-0.16
] 3.98 + [
] 3.98 + [
Wheat
Paddy
1.16
0.75
55.3%
-0.86
3154.04
-0.47
Page 28
Simulation 1:Concentration curves for electricity
consumption (Kw), actual and two-part tariff simulation
Punjab
1
1
0.9
0.9
0.8
0.8
cum dist of electricity consumtion (kw)
cum dist of electricity consumtion (kw)
Andhra Pradesh
0.7
0.6
0.5
0.4
0.3
Simulation
0.2
0.6
0.5
0.4
0.3
simulation
0.2
actual
0.1
Actual
0.1
0.7
0
0
0
0.1
0.2
0.3
0.4
0.5
0.6
cumulative distribution of irrigated Ha
0.7
0.8
0.9
1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
cumulative distribution of irrigated Ha
Page 29
1
Simulation 2: Impact of an optimal consumption plan
with a fix rate and two marginal rates
This second mechanism considers:
• A fixed rate (F), under which the household receives q1 units
of electricity.
• Consumption exceeding q1 is charged with a marginal cost v1
for households demanding less than q2 units.
• Households with consumption exceeding q2 pay v2 for
additional units. In this sense, household’s expenditure can be
represented by
EXPi 
F
if qi  q1
F  v1 (qi  q1 )
if q1  qi  q 2
F  v1 (q 2  q1 )  v 2 (qi  q1  q 2 )
if qi  q 2
Page 30
Profits of electricity industry under different combinations of v1
and v2 (v1<v2)
Andhra Pradesh
Punjab
Page 31
Summary of Simulation 2, selected values of v1 and v2
Andhra Pradesh
1
2
3
4
5
6
1.18
337.5
840.47
66.0%
1.18
337.5
840.47
66.0%
1.18
337.5
840.47
66.0%
1.18
337.5
840.47
66.0%
1.18
337.5
840.47
66.0%
1.18
337.5
840.47
66.0%
0.00
0.00
0.40
0.75
0.60
1.55
1.10
2.20
2.00
3.45
2.70
5.00
2.19
1.00
0.75
0.50
0.33
0.27
1.04
0.60
1.04
1.32
1.04
2.56
1.04
4.65
1.04
-23.49
1.04
-10.43
Assumptions
Marginal Cost
Fixed rate (F)
Units included in fixed rate (q1)
Subsidy fixed rate (% MgC)
Parameters
v1
v2
Impact
Simul Subs / Actual Subs
Subs smallholders / largeholders
Actual
Simul
Page 32
Summary of Simulation 2, selected values of v1 and v2
Punjab
1
2
3
4
5
6
1.89
810
767.38
44.2%
1.89
810
767.38
44.2%
1.89
810
767.38
44.2%
1.89
810
767.38
44.2%
1.89
810
767.38
44.2%
1.89
810
767.38
44.2%
0.00
0.00
0.15
0.80
0.35
1.05
0.60
1.60
0.90
2.10
4.90
5.00
2.64
1.00
0.75
0.50
0.33
0.02
0.05
0.04
0.05
0.07
0.05
0.10
0.05
0.17
0.05
0.28
0.05
-16.79
Assumptions
Marginal Cost
Fixed rate (F)
Units included in fixed rate (q1)
Subsidy fixed rate (% MgC)
Parameters
v1
v2
Impact
Simul Subs / Actual Subs
Subs smallholders / largeholders
Actual
Simul
Page 33
Summary of Simulation 2, Distribution of electricity subsidy, selected values
of v1 and v2 (% of total subsidy): Andhra Pradesh
Page 34
Summary of Simulation 2, Distribution of electricity subsidy, selected values
of v1 and v2 (% of total subsidy): Punjab
Page 35
Summary of Simulation 3: Impacts of an optimal consumption plan with a variable
first part and two marginal rates – Andhra Pradesh
sim/actual
2
1.5-2
loss
1-1.5
1.5
1.4-1.6
sim/actual 1.6
loss
1.2-1.4
1.4
1-1.2
0.5-1
1.2
0-0.5
1
1
0.8-1
0.6-0.8
0.8
0.4-0.6
0.6
0.2
v2
2.7
0-0.2
v2
0
Subs first segment=49%
1.4-1.6
1.2-1.4
1-1.2
0.6-0.8
0.6-0.8
0.8
0.4-0.6
0.4-0.6
0.6
0.2-0.4
0.2-0.4
0.4
4.8
0.4
0.2
2.4
0-0.2
4.8
0.2
v2
2.4
0
0
v1
4.5
2.7
1.8
3.6
75% increase in F
0.9
0
0
0-0.2
0
100% increase in F
Subs first segment=40%
v1
4.5
0.6
3.6
0.8
0.8-1
1
0.8-1
1
1-1.2
1.2
2.7
1.2
loss
0.9
1.4
1.2-1.4
sim/actual 1.4
1.8
sim/actual 1.6
v2
0
Subs first segment=32%
1.2-1.4
sim/actual 1.4
1-1.2
1.2
0.8-1
1-1.2
sim/actual 1.2
loss
0.8-1
1
0.6-0.8
1
0.6-0.8
0.8
0.4-0.6
0.8
0.4-0.6
0.6
0.2-0.4
0.6
0.2-0.4
0.4
0.4
0-0.2
0.2
0.2
Subs first segment=23%
0
v2
2.4
Subs first segment=15%
v1
4.5
150% increase in F
1.8
0
3.6
-0.2
0
4.5
3.6
v1
2.7
0.9
1.8
0
125% increase in F
-0.2-0
4.8
2.7
2.5
0
0-0.2
y
0.9
loss
v1
4.5
50% increase in F
3.6
0
0
Subs first segment=57%
loss
2.4
0
4.5
2.7
v1
3.6
25% increase in F
1.8
0.9
0
0
4.8
0.9
2.4
0.2-0.4
0.4
4.8
1.8
0.5
v2
0
Page 36
Summary of Simulation 3: Impacts of an optimal consumption plan with a variable
first part and two marginal rates – Punjab
2-2.5
sim/actual 2.5
1.5-2
loss
2
1.5
loss
2
1-1.5
0.5-1
1.5
0.5-1
0-0.5
0-0.5
1
-0.5-0
0.5
4.8
v1
2-2.5
1.5-2
2
1-1.5
1.5
0.5-1
sim/actual
1
0-0.5
-0.5-0
0.5
0
4.8
v2
v1
4.5
3.6
100% increase in F
2.7
0
1.8
-0.5
0
4.5
3.6
2.7
1.8
v1
4.8
2.4
0.9
-0.5
0.5-1
1
0.5
2.4
1-1.5
1.5
-0.5-0
0
1.5-2
2
loss
0-0.5
0.9
v2
0
Subs first segment=16%
sim/actual 2.5
0
3.6
0
50% increase in F
2.7
-0.5
v2
0
Subs first segment=30%
75% increase in F
2.4
4.5
v1
4.5
3.6
2.7
0.9
25% increase in F
1.8
0
0
4.8
0
1.8
2.4
0.9
0.5
v2
0
Subs first segment= -11%
Subs first segment=2%
sim/actual
1.5-2
1-1.5
1
loss
2-2.5
sim/actual 2.5
1.5-2
2
1-1.5
loss
0.5-1
1.5
sim/actual
1.5-2
2
1-1.5
loss
0.5-1
1.5
0-0.5
0-0.5
1
1
-0.5-0
0.5
0.5
4.8
2.4
Subs first segment= -39%
v1
4.5
150% increase in F
3.6
0
2.7
-0.5
0
4.5
3.6
1.8
v1
2.7
0.9
0
Subs first segment= -25%
v2
4.8
1.8
2.4
-0.5
0
0.9
0
125% increase in F
-0.5-0
v2
0
Page 37
Conclusions
•
The use of electricity in agriculture for irrigation following the green revolution has significantly
contributed to agricultural productivity growth in India.
•
However, there is an inbuilt inefficiency in the current pricing mechanism and measuring system of
power for irrigation in India and specifically in Andhra Pradesh and Punjab.
•
One of the mechanisms used to cover the subsidy for agricultural (and domestic) power
consumption was cross-subsidy from industrial and commercial consumers. In fact, the tariff
charged to industrial and commercial consumers in India has been one of the highest in the world.
•
Both of this problems have generated what we called along this report a vicious cycle which results
in poor supply outcome in form of low quality of power, unreliable supply, unavailable to many
potential users and high transmission and distribution (T&D) losses.
•
In addition, and most seriously, it had exacerbated the fall of the ground water level.
•
A price discrimination strategy is proposed based on the size of the farmers plot and on the
implementation of a two part tariff mechanism.
•
In summary, in all these three price schemes, the major result is that the subsidy will be more
progressive and resources will be used more efficiently. If low-demand consumers or high-demand
consumers want to consume more electricity, they will need to pay a charge over the marginal costs
for each unit above their fixed charge.
Page 38
Future research
• Implement proposed tariff schemes implemented based on the
simulated electricity consumption by the farmers according to
their plot size and what they produce
• Open the option for self selecting into pre-paid meters if they
belief price schemes are not capturing real consumption and use
it as a way to manage the price discrimination mechanism
• Measure the difference between two schemes in terms of:
– progressiveness of the distribution of the subsidy with respect to the
first two part tariff mechanism;
– reduction or elimination of the burden of the subsidy to the government
by cross subsidizing small holders with the revenues from large holders;
– changes in the quality of the service
Page 39