Post on 07-Aug-2020
transcript
50 IEEE power & energy magazine july/august 2007
by Hugh Rudnick,Alejandro Arnau,Sebastian Mocarquer,and Efrain Voscoboinik
1540-7977/07/$25.00©2007 IEEE
july/august 2007 IEEE power & energy magazine 51
EELECTRICITY DISTRIBUTION COMPANIES, BEING NETWORK INDUSTRIES, AIMto transport and distribute electric power from specific points in high- or medium-voltage linesto end consumers at appropriate voltage levels for industrial and residential use. This activity isorganized in public service utilities that obtain power supply through contracts with generators.
During the last two decades, many countries and geographical areas of the world have madedrastic transformations in their electrical sectors, both in terms of segmentation and privatizationof state monopolies. Because of these transformations, a major change in the role of the statehas been witnessed. The state has transformed itself from a producer and enterprise-owner agentinto an agent that regulates those stages of the electrical sector that become natural monopolies,such as electricity distribution. The challenge is to stimulate an efficient service in distribution,similar to that which would be achieved in a competitive environment.
The characteristics of the distribution activity vary from region to region, depending on thenature of the demand. In Latin America, the main challenge in distribution is one of large growthsof demand, values around 6% to 8% being common in the region, requiring doubling of capacityevery ten years. This, coupled in some countries to the need to extend networks to increase elec-trification levels, imposes significant challenges to distribution network expansion, different tothose faced in North Amer-ica and Europe. Thus, theneed to stimulate efficiencybecomes of paramountimportance, particularly atthe time of privatization ofthe previously state-ownedcompanies.
To regulate electricaldistribution and stimu-
late efficiency, mostLatin-American
countries thathave started this
transformation have adopted an incentive regulation approach, using the concept ofefficient companies that are adapted to demand and that operate under optimal invest-
ment and operations plans. Under this scheme, to force companies to be efficient, the regu-lator fixes prices according to efficient costs, not necessarily considering actual companies.The actual company will get a normal profitability only if it is capable of emulating the effi-cient performance, reducing its operating and investment expenditure, thus minimizing thepresent value of its costs. In general, this regulation has implied a reduction trend in distribu-tion tariffs.
This article assesses the 20-year experience in Latin America in applying incentive price reg-ulation to its distribution companies.
Distribution in Latin AmericaThe development and typology of distribution networks in Latin America was conditioned bythe historical context in which they were built. In particular, the strategy and technology used todevelop the networks is related to the origin of the concessionaire who started the distribution
Incentive Price RegulationStimulates Efficiency inElectricity Distribution in Latin America
© CARTESIA & STOCKBYTE
service, which could be either North American or European.Countries such as Colombia and Brazil, among others, weredirectly influenced by the United States on account of theircommercial relationships with this country at the beginningof the century. Their electric power distribution networksshow the U.S. topology and distribution, characterized by amedium-voltage network, small transformation centers closeto the final user, and a small low-voltage network. On theother hand, countries such as Argentina, Peru, Uruguay, andChile were influenced by Germany, France, and England,and, therefore, their networks are of the European type, withtransformers with greater capacity and low-voltage three-phase distribution networks.
Optimum economic performance in 60-Hz-frequencies isfound in low-power transformers. This has allowed for thedevelopment of networks of the U.S. type; i.e., distribution inmedium voltage with small transformers near the customer.For 50 Hz, optimum performance is produced in transformerswith greater power, which justifies the use of European-typenetworks.
In Latin America, because of consumption and develop-ment levels in general, there is a greater development of low-voltage networks compared to Europe and the United States.For similar reasons, the percentage of underground networksis also lower compared to the abovementioned countries. Infact, the development of underground networks is morerelated to the compliance with city-planning regulations(e.g., to preserve the cultural heritage) than to the electricservice regulation.
Distribution Activity The distribution activity is characterized by the constant invest-ments needed to render good services and to achieve the vari-ous scale economies that can be attained by companies whendeveloping their facilities and their management and operation.Although economies of scale add up to efficiency, they alsomake the revenues generated through a marginal cost tariff notto be enough to cover these companies’ total costs. Likewise,the strong interdependence of investments and the long capital-recovery period give origin to a costs function that is clearlyunder-additive for the relevant demand range. This makes oneconclude that it is more socially efficient to have a single com-pany instead of several companies operating in a same geo-graphical area. In this manner, and as it is a matter ofguaranteeing maximum coverage, with the highest quality andleast price possible, single distribution companies are justified
and they are allowed to operate as a natural monopoly.Under monopolistic conditions, consumers have no possi-
bilities to choose, and that makes it necessary to regulate theservice in order to prevent unreasonable practices (poor qual-ity, low coverage, high prices). For that purpose, the regulatorestablishes the rights and obligations of distribution compa-nies, assigning concession zones to install, operate, andexploit public service distribution networks. Likewise, theregulator establishes price levels and creates incentives thatallow management improvement since there is no marketcompetition to promote them.
In this manner, the primary goal of any regulatoryscheme is to provide the appropriate incentives to companiesto force them to be efficient and through adequate price sig-nals to make them to be able to transfer, in the long term,part of their benefits to the users given their efficient invest-ment and operating policies. These incentives will be avail-able for the companies only if the regulator can show acertain level of commitment and stability. This commitmentand stability results from ensuring the regulator follows cer-tain general principles on transparency, efficiency, stability,and straightforwardness.
Based on these principles, the regulation of naturalmonopolies is made through different approaches.
Challenges of Distribution RegulationRegarding costs associated with the activity that is intendedto be remunerated, they are associated to the networkexploitation, maintenance, and expansion components. Theycan be grouped into the items indicated in Figure 1, withinwhat is named the value-added distribution. Each one ofthese costs is indistinctly affected by climatic, geographical,and demographic factors that are external to the companies.This reveals that tariff regulation is a very complex anddemanding process, considering:
✔ the need to adequately identify and value the differentcomponents
✔ the need to fairly and transparently weight the influ-ence of factors such as network type (rural or urban),overhead or underground lines, and the type and densi-ty of consumption present in the company’s activity
✔ the convenience of emitting signals to encourage theadoption of more efficient behaviors by the companiesperforming in the concession areas.
These challenges have been treated differently in regula-tions, with some using rate of return and price cap methods,
52 IEEE power & energy magazine july/august 2007
In Latin America, because of consumption and development levelsthere is a greater development of low-voltage networks compared toEurope and the United States.
while others adopted a benchmark scheme, using the conceptof direct comparison with an efficient company. All methodsserve the same final objective, but the context in which theywere created was different, making them more suitable thanothers depending on the general environment of the country.Many regulatory schemes use comparison procedures orbenchmarking as a methodology. In incentive regulation, effi-ciency is measured against a previous benchmark and theresults deliver the information required to compare the com-panies’ operation and allow the identification of the actionsrequired to drive efficiency improvements.
Different Regulatory ModelsThere have been different regulatory models used in LatinAmerica in applying incentive price regulation to its distribu-tion companies. Table 1 summarizes different approaches thatare briefly described afterwards for sample countries.
ArgentinaThe revenue-cap and price-cap tariff scheme in Argentina isbased on enough revenues to render an efficient service, andit determines the tariff value for each supply category. The
tariff is the sum of the VAD (distribution added value, in itsSpanish acronym for Valor Agregado de Distribución) and aquasi-perfect pass-through of the purchase cost of energy andpower in the wholesale electricity market.
The tariff scheme in itself is an incentive for efficiency,since it acknowledges the efficiency values the distributorshould operate with ex-ante the rendering of the service. Thedistributor will obtain the expected profitability for itsinvestors only if it adjusts its costs to the acknowledged val-ues. The system transfers risk decisions to the provider sincemistakes in deciding investments, expenses, indebtednessstrategies, or technological adjustments will be ultimatelypaid by the investor, who will receive less profits for the capi-tal invested in the activity.
With reference to the use of resources, it is a model basedon economic signals, where decisions are taken by agents; bymeans of their tariffs, distributors inform consumers on theefficiency cost of each consumption alternative and they, intheir capacity as demand makers, decide on the use and allo-cation of resources.
Large users freely agree on the price of their contractswith generators or brokers. Regarding the use of networks, if
figure 1. Cost components of the distribution activity.
Distribution Value Added
Fixed Costs Investment Costs, Operation and Maintenance Distribution Losses
• Meter Reading• Billing• Distribution of Bills• Accounting Related to Client• Bill Follow Up• Client Relation• Other Fixed Costs Related to Client
CapitalCosts
OperationalCosts
TechnicalLosses
OtherLosses
High Voltage• Feeders• Control Equipment • Protection Equipment
Low Voltage• Distribution Substations• Low Voltage Networks• Protection Equipment
HV LV
• Storage• Workshops• Labs and Tools• Operational Engineering• Transport• Security• Rent, Insurance• Network Maintenance and Operation• Patents and Property Taxes
The tariff scheme in itself is an incentive for efficiency, since it acknowledges the efficiency values the distributor should operate with ex-ante the rendering of the service.
july/august 2007 IEEE power & energy magazine 53
Arg
entin
a
Nat
ural
mon
opol
y. P
rice
cap
with
pref
ixed
tarif
f stru
ctur
e. F
ree
acce
ss.
Con
cess
ions
(bet
wee
n 35
and
99
year
s) w
ith a
rea
excl
usiv
ityth
at c
an b
e el
imin
ated
by
tech
nolo
gica
l inn
ovat
ion.
Com
petit
ion
for t
he m
arke
tev
ery
15 y
ears
.
Dis
puta
ble
rega
rdin
g th
eun
regu
late
d us
er m
arke
t(c
onsu
mpt
ions
abo
ve 3
0 kW
).In
crea
se in
dis
puta
bilit
y du
e to
decr
ease
in re
quire
men
ts to
beco
me
an u
nreg
ulat
ed u
ser.
Ther
e is
no
sepa
ratio
n of
the
met
erin
g an
d re
adin
g of
the
serv
ice,
exc
ept f
or u
sers
ove
r 1M
W (r
eal-t
ime
met
erin
g).
Full
oblig
atio
n to
pla
ce n
etw
orks
at th
e di
spos
al o
f all
dem
and
with
in th
e co
nces
sion
and
supp
ly a
reas
for r
egul
ated
use
rs.
Net
wor
k ex
pans
ion
isco
ntem
plat
ed u
nto
a ce
rtain
dist
ance
, as
wel
l as
expa
nsio
nw
ith re
fund
able
con
tribu
tions
to u
sers
whe
n th
is d
ista
nce
isex
ceed
ed.
Exis
tenc
e of
the
conc
ept o
fbr
oker
, ind
epen
dent
of d
istri
bu-
tion
and
with
out t
he c
apac
ityto
ser
ve c
aptiv
e us
ers.
The
re a
rebr
oker
s op
erat
ing
with
littl
evo
lum
e at
the
mom
ent.
Ther
e ar
e fu
nds
for r
ural
or
unpr
ofita
ble
elec
trific
atio
n.
It is
not
regu
late
d bu
t is
disc
usse
dfo
r eac
h ca
se. I
n di
strib
utio
nth
ere
is a
tend
ency
to u
se th
eV
NR
of e
ffici
ent f
acili
ties.
Ther
e is
som
e di
scus
sion
ove
rth
is. R
evie
w e
very
five
yea
rs.
Type
of
Regu
latio
n
Dis
puta
bilit
y of
the
Net
wor
kM
arke
t
Dis
puta
bilit
yof
the
Trad
ing
Mar
ket
Supp
yO
blig
atio
n
Exis
tenc
e of
the
Lega
lC
once
pt o
fBr
oker
Subs
idie
s fo
rEx
pans
ion
Cap
ital B
ase
and
Cap
ital
Retu
rn
Boliv
ia
Nat
ural
mon
opol
y. P
rice
cap
with
pre
fixed
tarif
f stru
ctur
e.Fr
ee a
cces
s.
Ther
e is
no
disp
utab
ility
in th
edi
strib
utio
n m
arke
t. C
ontra
cts
are
awar
ded
on a
nex
clus
ivity
bas
is fo
r a 4
0-ye
arpe
riod.
Dis
puta
ble
rega
rdin
g th
eun
regu
late
d us
er m
arke
t(c
onsu
mpt
ions
abo
ve
1 M
W).
Ther
e is
no
sepa
ratio
nof
the
met
erin
g an
d re
adin
g of
the
serv
ice.
Unr
egul
ated
use
rsm
ust a
dapt
thei
r met
erin
gco
nditi
ons.
Obl
igat
ion
to re
nder
ser
vice
tore
gula
ted
user
s w
ithin
the
conc
essi
on a
nd s
uppl
y ar
ea.
The
conc
ept o
f bro
ker d
oes
not
exis
t.
The
exec
utiv
e po
wer
, by
mea
ns o
fth
e N
atio
nal F
und
for R
egio
nal
Dev
elop
men
t, al
loca
tes i
nter
nal
and
exte
rnal
fina
ncin
gre
sour
ces.
No
refu
ndab
leco
ntrib
utio
ns a
re c
onte
mpl
ated
.U
sers
are
resp
onsib
le fo
r the
expa
nsio
ns th
ey c
ause
.
Ass
ets
desi
gnat
ed to
the
conc
essi
on, e
qual
to th
e va
lue
of th
e fix
ed n
et a
sset
plu
s th
ene
t lab
or c
apita
l, le
ss th
e va
lue
of lo
ng-te
rm li
abili
ties
asso
ciat
ed to
fixe
d as
sets
. It i
sfix
ed e
very
four
yea
rs.
Chi
le
Nat
ural
mon
opol
y. Y
ards
tick
com
petit
ion
with
cap
inta
riff s
truct
ure.
Fre
e ac
cess
.
Are
a ex
clus
ivity
. Fin
alco
nces
sion
s w
ithou
t tim
elim
itatio
ns.
Dis
puta
ble
rega
rdin
g th
eun
regu
late
d us
er m
arke
t for
cons
umpt
ions
abo
ve 2
MW
.C
onsu
mpt
ions
of c
lient
sbe
twee
n 0.
5 an
d 2
MW
can
deci
de b
etw
een
the
regu
late
d ta
riff o
run
regu
late
d m
arke
t.
Obl
igat
ion
to re
nder
ser
vice
with
in th
e co
nces
sion
are
ato
who
ever
requ
ests
it. T
here
are
refu
ndab
le c
ontri
butio
nsto
new
sup
plie
s an
dex
pans
ions
.
Ther
e is
no
lega
l con
cept
of
brok
er. T
he tr
adin
g m
arke
t is
take
n by
gen
erat
ors.
No
subs
idie
s ar
e ex
pect
ed fo
rex
pans
ion.
The
re a
re ru
ral
elec
trific
atio
n go
vern
men
tpr
ogra
ms.
It is
cal
cula
ted
by m
eans
of t
heV
NR
of th
e ne
twor
kec
onom
ical
ly a
dapt
ed to
dem
and
in o
rder
to re
nder
the
serv
ice
at m
inim
um c
ost.
Cap
ital b
ase
is re
view
edev
ery
four
yea
rs.
Col
umbi
a
Pric
e ca
p. F
ree
acce
ss.
Ther
e ar
e no
con
cess
ions
or
franc
hise
s. T
here
is n
oex
clus
ivity
. The
re m
ay b
epa
ralle
l net
wor
ks in
the
sam
ear
ea.
Ther
e ar
e no
exp
licit
barr
iers
but t
he u
ser m
ust p
ay fo
r the
met
erin
g co
sts.
The
dist
ribut
or is
not
obl
iged
tom
ake
expa
nsio
ns if
they
are
not p
rofit
able
; tha
t is,
it is
only
obl
iged
if th
e ex
pans
ion
is re
mun
erat
ed b
y th
e ta
riff.
The
conc
ept o
f bro
ker e
xist
s.
The
gove
rnm
ent g
uara
ntee
s th
eim
plem
enta
tion
of p
roje
cts
cons
ider
ed in
the
inve
stm
ent
plan
that
are
not
pro
fitab
lefo
r inv
esto
rs. T
hese
pro
ject
sar
e fin
ance
d by
the
Rura
lEl
ectri
ficat
ion
Supp
ort F
und.
VN
R ac
cord
ing
to th
e el
ectri
can
d no
nele
ctric
ass
etin
vent
ory.
Non
elec
tric
asse
tsar
e lim
ited
to 4
.1%
of t
heV
NR
annu
ity.
Peru
Nat
ural
mon
opol
y. Y
ards
tick
com
petit
ion
with
cap
in ta
riff
stru
ctur
e. F
ree
acce
ss.
Are
a ex
clus
ivity
. Fin
al c
on-
cess
ions
with
out t
ime
limita
-tio
ns. T
hey
term
inat
e by
expi
ratio
n or
wai
ver.
Dis
put-
abili
ty in
exp
ansi
on a
reas
isso
lved
by
mea
ns o
f auc
tions
.
Disp
utab
le re
gard
ing
the
unre
gula
ted
user
mar
ket
(con
sum
ptio
ns a
bove
1 M
W).
Obl
igat
ion
to re
nder
ser
vice
with
inth
e co
nces
sion
are
a an
d up
to10
0 m
of t
he e
lect
ric p
ower
line,
to w
hoev
er re
ques
ts it
.Th
ere
are
refu
ndab
leco
ntrib
utio
ns to
new
sup
plie
san
d ex
pans
ions
.
Ther
e is
no
conc
ept o
f bro
ker.
The
tradi
ng m
arke
t has
not
deve
lope
d.
No
subs
idie
s ar
e ex
pect
ed fo
rex
pans
ion.
The
re a
re re
fund
able
cont
ribut
ions
from
pub
lic a
ndpr
ivat
e in
stitu
tions
for
elec
trific
atio
n in
favo
r of u
sers
.Th
ere
is a
Rur
al E
lect
rific
atio
nA
ct, w
hich
has
not
bee
nre
gula
ted
yet.
It is
cal
cula
ted
by m
eans
of t
heV
NR
of th
e ne
twor
k ec
onom
i-ca
lly a
dapt
ed to
dem
and
inor
der t
o re
nder
the
serv
ice
atm
inim
um c
ost.
Cap
ital b
ase
isre
view
ed e
very
four
yea
rs.
tab
le 1
.Dif
fere
nt
ince
nti
ve r
egu
lato
ry m
od
els
for
dis
trib
uti
on
in L
atin
Am
eric
a.
54 IEEE power & energy magazine july/august 2007
july/august 2007 IEEE power & energy magazine
Not
exp
ress
ly d
efin
ed.
Reas
onab
le c
apita
l cos
t. Th
ere
is a
tend
ency
to u
seW
AC
C/C
APM
.
Eval
uatio
n of
eac
h ca
se b
ym
eans
of s
tudi
es b
y th
ere
gula
tor a
nd th
e co
mpa
ny.
VN
R or
ave
rage
incr
emen
tal
cost
(CIP
) met
hods
are
use
d.A
ckno
wle
dged
inve
stm
ents
are
thos
e re
sulti
ng fr
omsp
ecifi
c st
udie
s. T
hey
are
not
revi
ewed
ex
post
.
O&
M, A
&G
, and
trad
ing
cost
s(n
o m
argi
n fo
r pur
chas
e in
the
WEM
), no
ntra
nsfe
rabl
e ta
xes,
and
tech
nica
l and
nont
echn
ical
loss
es a
reef
ficie
ntly
ack
now
ledg
edde
pend
ing
on th
e ca
se.
Not
exp
ress
ly d
efin
ed. I
n ge
nera
l,it
is c
onsi
dere
d in
abe
nchm
arki
ng a
s a
perc
enta
geof
inve
stm
ent o
r in
a ce
rtain
mod
el c
ompa
ny.
They
are
revi
ewed
in e
very
tarif
fre
view
, and
thos
e th
at re
mai
nco
nsta
nt d
urin
g th
e ta
riff p
erio
dar
e ac
know
ledg
ed. A
nac
know
ledg
ed lo
ss c
oeffi
cien
t,al
so w
orki
ng a
s an
ince
ntiv
e to
redu
ce th
em, i
s allo
cate
d. In
som
e ca
ses,
nont
echn
ical
loss
esar
e ac
know
ledg
ed. A
gree
men
tw
ith th
e go
vern
men
t to
treat
robb
ery.
The
y ar
e va
lued
at
trans
fer p
rice.
Rem
uner
atio
n is
cal
cula
ted
indo
llars
. It i
s ad
just
ed b
yw
eigh
ted
Am
eric
an C
PI/P
PIA
vera
ge. T
here
is n
o cu
rren
t Xfa
ctor
.
It is
not
man
dato
ry a
lthou
gh th
esu
pplie
r's re
spon
sibi
lity
isul
timat
ely
the
cont
ract
ing
sign
al.
Retu
rn R
ate
Ove
r Cap
ital
Eval
uatio
n of
Cap
ital C
osts
Ack
now
ledg
edO
pera
tion
Cos
ts
Evau
atio
n of
Ope
ratio
nC
osts
Trea
tmen
t of
Loss
es
Intra
revi
ewPe
riod
VA
DA
djus
tmen
t
Obl
igat
ion
toC
ontra
ct in
the
Who
lesa
leM
arke
t
Det
erm
inat
ion
of c
apita
l cos
t by
benc
hmar
king
. It i
sde
term
ined
on
the
basi
s of
the
aver
age
retu
rn ra
te o
f the
last
3ye
ars
in A
mer
ican
util
ities
.
Stud
ies
orde
red
by th
e co
mpa
nyto
spe
cial
ized
con
sulta
nts,
pre-
qual
ified
by
the
regu
lato
r,w
ho w
ill p
repa
re th
e te
rms
ofre
fere
nce
and
will
rece
ive
the
stud
ies.
Inve
stm
ent p
lans
are
anal
yzed
on
a ca
se-b
y-ca
seba
sis.
Con
sum
ers c
osts,
taxe
s, op
erat
ion
costs
, mai
nten
ance
cos
ts,ad
min
istra
tion
and
gene
ral
costs
, fin
anci
al c
osts,
and
oth
erco
sts re
late
d to
supp
ly.
Cas
e by
cas
e, c
onsi
derin
g th
eex
pect
ed g
row
th o
f dem
and,
expa
nsio
n pl
ans,
and
oper
atio
n in
dica
tors
, and
that
of u
nit c
osts
def
ined
for a
four
-yea
r per
iod.
A te
chni
cal l
oss
coef
ficie
nt is
ackn
owle
dged
, in
acco
rdan
ce w
ith th
eco
nces
sion
are
a an
d ne
twor
kch
arac
teris
tics.
It is
usu
ally
<10%
. In
addi
tion,
a lo
ssre
duct
ion
fact
or is
app
lied
bym
eans
of a
mon
thly
effic
ienc
y fa
ctor
.
Diff
eren
t X fa
ctor
s fo
r eac
h ty
peof
cos
ts: l
osse
s, O
&M
, A&
G,
com
mer
cial
. Cap
pric
es a
read
just
ed b
y va
riatio
ns in
cons
umer
infla
tion
inde
xes,
tarif
f rat
es, e
xcha
nge
rate
.
They
mus
t hav
e va
lid c
ontra
cts
with
gen
erat
ion
com
pani
esgu
aran
teei
ng to
cov
er 8
0% o
fth
eir p
eak
dem
and
for a
min
imum
thre
e-ye
ar p
erio
d.Th
ere
are
no in
cent
ives
toco
ntra
ct d
ue to
the
ackn
ow-
ledg
ed tr
ansf
er p
rice
sche
me.
10%
ove
r cap
ital b
ase.
Prof
itabi
lity
of th
e in
dust
ry a
s a
who
le s
houl
d be
bet
wee
n 6%
and
14%
.
Eval
uatio
n by
VN
R. A
sset
sco
rresp
ondi
ng to
a n
etw
ork
tech
nica
lly a
nd e
cono
mic
ally
adap
ted
to d
eman
d w
ill b
eac
know
ledg
ed. T
he re
gula
tor
will
def
ine
typi
cal a
reas
and
calc
ulat
e th
eir V
AD
. Stu
dies
are
carri
ed o
ut fo
r eac
h ty
pica
l are
aan
d th
ey c
orre
spon
d to
mod
elco
mpa
nies
.
Stan
dard
ope
ratio
n, m
aint
enan
ce,
adm
inis
trativ
e, a
ndco
mm
erci
aliz
atio
n co
sts
will
be a
ckno
wle
dged
as
wel
l as
ape
rcen
tage
of t
echn
ical
and
nont
echn
ical
loss
es.
They
are
det
erm
ined
eve
ry
four
yea
rs a
nd c
orre
spon
d to
mod
el c
ompa
nies
for e
ach
typi
cal a
rea.
The
VA
D d
eter
min
atio
n pr
oces
sin
corp
orat
es th
e le
vel o
fte
chni
cal a
nd n
onte
chni
cal
loss
es.
Tarif
fs a
re a
pplie
d in
loca
lcu
rren
cy a
nd a
re a
djus
ted
byva
riatio
ns in
: loc
al c
onsu
mer
infla
tion
inde
x, im
porte
dgo
ods
inde
x, a
nd in
dust
rial
pric
e in
dex.
Dis
tribu
tors
are
requ
ired
tom
aint
ain
thre
e ye
ars
ofco
ntra
cts
at a
ll tim
es. R
ecen
tch
ange
s in
who
lesa
le m
arke
tsin
trodu
ced
auct
ions
for
regu
late
d co
nsum
ptio
ns.
14.0
6% o
ver t
he re
plac
emen
tva
lue
of th
e as
sets
for l
evel
IV a
nd 1
6.06
% fo
r lev
els
I, II,
and
III.
Eval
uatio
n by
VN
R or
CIP
(but
only
in L
V a
nd w
hen
the
mar
gina
l cos
t exc
eeds
the
aver
age
cost
).
O&
M a
nd A
&G
cos
ts a
nd a
perc
enta
ge o
f los
ses
byvo
ltage
leve
l will
be
ackn
owle
dged
.
They
are
det
erm
ined
as
ape
rcen
tage
of t
here
plac
emen
t val
ue o
f ele
ctric
asse
ts: 2
% fo
r lev
els
III a
ndIV
, 4%
for l
evel
II, a
nd fo
rle
vel I
ther
e is
a c
ostin
gpr
oces
s pe
r act
ivity
.
For e
ach
volta
ge le
vel t
here
isan
ack
now
ledg
ed p
erce
ntag
efo
r non
tech
nica
l los
ses.
For
leve
l I it
was
1.6
8% in
200
3an
d 0.
67%
in 2
007.
Tarif
fs a
re a
pplie
d in
loca
lcu
rren
cy a
nd a
re a
djus
ted
byva
riatio
ns in
infla
tion
inde
xes,
exc
hang
e ra
te.
At f
irst t
here
was
the
oblig
atio
n to
cont
ract
at 8
0%, w
hich
gra
d-ua
lly d
ecre
ased
. Cur
rent
ly,
ther
e is
a co
ntra
ctin
g pr
oce-
dure
for t
he re
gula
ted
mar
ket.
12%
ove
r cap
ital b
ase.
Prof
itabi
lity
of th
e gr
oup
ofco
nces
sion
aire
s sh
ould
be
betw
een
8% a
nd 1
6%.
Eval
uatio
n by
VN
R. A
sset
sco
rres
pond
ing
to a
net
wor
kte
chni
cally
and
eco
nom
ical
lyad
apte
d to
dem
and
will
be
ackn
owle
dged
. The
regu
lato
rw
ill fi
x ty
pica
l are
as a
ndca
lcul
ate
thei
r VA
D. S
tudi
es a
reca
rrie
d ou
t for
eac
h ty
pica
l are
aan
d th
ey c
orre
spon
d to
mod
elco
mpa
nies
.
Stan
dard
O&
M, A
&G
, and
com
mer
cial
izat
ion
cost
s w
illbe
ack
now
ledg
ed a
s w
ell a
s a
perc
enta
ge o
f tec
hnic
al a
ndno
ntec
hnic
al lo
sses
.
They
are
det
erm
ined
eve
ry fo
urye
ars
and
corr
espo
nd to
mod
elco
mpa
nies
for e
ach
typi
cal
area
.
The
VAD
inco
rpor
ates
an
ackn
owle
dged
leve
l of
tech
nica
l and
non
tech
nica
llo
sses
. Eve
ry fo
ur y
ears
, val
ues
are
adju
sted
acc
ordi
ng to
tech
nolo
gica
l adv
ance
s an
d to
man
agem
ent r
esul
ts in
the
prev
ious
tarif
f per
iod.
The
y ar
eth
e go
vern
men
t's p
oliti
cal
deci
sion
. Fix
ing
anac
know
ledg
ed le
vel w
orks
as
an in
cent
ive
to re
duce
them
.
Tarif
fs a
re a
pplie
d in
loca
lcu
rren
cy a
nd a
re a
djus
ted
byva
riatio
ns in
con
sum
erin
flatio
n in
dexe
s, e
xpor
t dut
y,ex
chan
ge ra
te, s
alar
ies,
and
pric
e of
cop
per.
Thro
ugh
auct
ions
up
to 2
5% o
fpe
ak d
eman
d; th
e re
st m
ust b
egu
aran
teed
by
cont
ract
s w
ithge
nera
tion
com
pani
es.
Imba
lanc
es a
re n
egot
iate
d on
the
spot
mar
ket.
fixed
eve
ry fo
ur y
ears
.ev
ery
four
yea
rs.
55
(con
tinue
d)
tab
le 1
.Dif
fere
nt
ince
nti
ve r
egu
lato
ry m
od
els
for
dis
trib
uti
on
in L
atin
Am
eric
a (c
on
tin
ued
).
Arg
entin
a
The
seas
onal
mar
ket p
rice
isac
know
ledg
ed, t
his
is a
quar
terly
pro
ject
ion,
it is
am
arke
t with
a s
tabi
lizat
ion
fund
that
min
imiz
es v
olat
ility
.It
is a
djus
ted
quar
terly
.
Pref
ixed
tarif
fs (n
o ta
riff
optio
ns) b
y vo
ltage
and
usag
e le
vel (
only
in L
V).
Ade
quat
e pr
ice
sign
als.
As
a ru
le, t
hey
do n
ot e
xist
.Th
ere
have
bee
n, h
owev
er,
grad
ual e
limin
atio
nm
echa
nism
s.
As
a ru
le, i
t sho
uld
be n
eutra
lbu
t def
icie
ncie
s in
the
tarif
fpa
ram
eter
s do
not
gen
erat
eth
e de
sire
d ne
utra
lity.
Ther
e ar
e re
gula
ted
caps
. In
gene
ral,
they
pro
duce
neut
ralit
y in
the
dist
ribut
or's
rem
uner
atio
n. T
here
is a
terr
itoria
l jur
isdi
ctio
n co
nflic
t.
Stag
es, e
ach
with
gre
ater
requ
irem
ents
. Pro
duct
,se
rvic
e, a
nd c
omm
erci
alqu
ality
are
con
trolle
d.
Non
supp
lied
ener
gy (N
SE)
resu
lting
from
the
mea
sure
din
dexe
s is
val
ued.
Ever
y fiv
e ye
ars.
Ack
now
ledg
edTr
ansf
erPr
ices
Type
of T
ariff
Stru
ctur
e
Exis
tanc
e of
Cro
ssed
Subs
idie
s
Neu
tralit
y in
the
Tran
sfer o
fSu
pply
Cos
ts
Tolls
for
Unr
egul
ated
Use
rs
Ttyp
e of
Con
trol
Pena
lty S
chem
e
Tarif
f Per
iod
Boliv
ia
It is
bas
ed o
n th
e di
ffere
nten
ergy
and
pow
er p
rices
inth
e no
de a
lloca
ted
to e
ach
dist
ribut
or, w
ith a
pro
ject
ion
of d
eman
d fo
r the
follo
win
g48
mon
ths.
Pref
ixed
tarif
fs (n
o ta
riff o
ptio
ns)
by m
axim
um d
eman
d an
dvo
ltage
and
usa
ge le
vels
(onl
y in
LV
). Th
ere
are
tarif
fsfo
r usa
ge in
sm
all d
eman
dsan
d irr
igat
ion.
The
regu
latio
n sc
hem
e do
es n
otsh
ow c
ross
ed s
ubsi
dies
. But
,in
fact
, the
re a
re s
ome
ince
rtain
cas
es.
Acc
ordi
ng to
the
regu
latio
n, it
is n
eutra
l alth
ough
ther
e ar
eac
tual
ly c
ases
whe
re it
is n
ot.
Ther
e is
a m
echa
nism
tode
term
ine
the
toll.
It is
adju
sted
by
inde
xatio
n to
mai
ntai
n re
mun
erat
ion
inre
al te
rms.
Prod
uct,
serv
ice,
and
com
mer
cial
qua
lity
are
cont
rolle
d. F
our s
tage
s of
incr
easi
ng re
quire
men
ts.
Pena
lties
on
the
basi
s of
NSE
acco
rdin
g to
the
indi
cato
rsin
eac
h st
age
and
type
. The
max
imum
rem
uner
atio
nre
duct
ion
mus
t be
10%
of
the
annu
al tu
rnov
er.
Ever
y fo
ur y
ears
.
Chi
le
Nod
e pr
ices
are
ack
now
ledg
ed.
They
resu
lt fro
mco
mbi
natio
n of
a 4
8-m
onth
proj
ectio
n an
d au
ctio
ns fo
rre
gula
ted
cons
umpt
ions
.
Tarif
f opt
ions
in L
V a
nd M
Vre
stric
ted
to m
eter
ing
cond
ition
s an
d vo
ltage
leve
ls. T
here
is a
tarif
f cha
rtfo
r eac
h ty
pica
l are
a w
ithin
each
com
pany
.
As
a ru
le, t
hey
do n
ot e
xist
.
Neu
tral.
Tolls
are
cal
cula
ted
so th
eypr
oduc
e ne
utra
lity
indi
strib
utor
's re
mun
erat
ion
and
are
set b
y re
gula
tor.
Con
cess
iona
ires a
re o
blig
ed to
guar
ante
e qu
ality
and
relia
bilit
y de
term
ined
inby
law
s, w
hich
is su
perv
ised
byEl
ectri
city
Sup
erin
tend
ence
Pena
lties
can
be
impo
sed
byEl
ectri
city
Sup
erin
tend
ence
and
cons
umer
s co
mpe
nsat
edif
qual
ity a
nd re
liabi
lity
mea
sure
d in
dexe
s do
not
mee
t sta
ndar
ds re
quire
d.
Ever
y fo
ur y
ears
.
Col
umbi
a
Acc
ordi
ng to
dis
tribu
tor's
aver
age
purc
hase
s [b
oth
forw
ard
(tran
spar
ence
and
free
com
petit
ion)
and
spo
t]an
d to
the
aver
age
pric
e in
the
Col
ombi
an m
arke
t.
Post
age-
stam
p ta
riffs
by
volta
gele
vel.
Tarif
fs a
ffect
ed b
y fa
ctor
dest
ined
to s
ubsi
dize
low
-in
com
e co
nsum
ers.
Max
imum
20%
cos
t for
serv
ice
rend
erin
g.N
ot e
ntire
ly n
eutra
l. M
ay c
reat
elo
sses
or p
rem
ium
s.
Regu
late
d to
lls fo
r all
user
s (fr
eean
d ca
ptiv
e).
Qua
lity
regu
latio
ns w
ere
issu
edsu
bseq
uent
to ta
riff s
anct
ion
to c
ontro
l pro
duct
and
ser
vice
qual
ity. T
here
are
com
pens
a-tio
ns fo
r use
rs, a
nd q
ualit
ygo
als
are
curr
ently
bei
ngre
view
ed.
NSE
resu
lting
from
the
mea
sure
d in
dexe
s is
appr
aise
d.
Ever
y fo
ur y
ears
.
Peru
Nod
e pr
ices
are
ack
now
ledg
ed.
They
resu
lt fro
m a
48-
mon
thpr
ojec
tion
and
mus
t be
+/-
10%
of t
he a
vera
ge p
rice
ofen
ergy
agr
eed
with
unr
egu-
late
d us
ers.
Oth
erw
ise,
they
are
adju
sted
.Ta
riff o
ptio
ns in
LV
and
MV
rest
ricte
d to
met
erin
gco
nditi
ons
and
volta
gele
vels
. The
re is
a ta
riff c
hart
for e
ach
typi
cal a
rea
with
inea
ch c
ompa
ny. P
repa
ymen
tis
a ta
riff o
ptio
n.A
s a
rule
, the
y do
not
exi
st.
In g
ener
al, i
t is
neut
ral.
Regu
late
d to
lls fo
r fre
e us
ers.
Con
cess
iona
ires
are
oblig
ed to
guar
ante
e th
e qu
ality
fixe
dun
der c
ontra
ct. T
here
is a
tech
nica
l reg
ulat
ion
(sub
sequ
ent t
o th
e la
w a
ndth
e co
nces
sion
con
tract
s) to
cont
rol p
rodu
ct, s
ervi
ce,
com
mer
cial
, and
pub
liclig
htin
g qu
ality
.
Com
pens
atio
ns to
use
rs o
n th
eba
sis
of N
SE a
nd th
edi
ffere
nce
betw
een
ratio
ning
cost
and
tarif
f.
Ever
y 4
year
s.
56 IEEE power & energy magazine july/august 2007
they do not agree with the transporter or distributor, a regu-lated tariff is applied, which must coincide with the VAD.
Another issue, of a circumstantial nature and exogenous toelectric power regulation, is that originated by the passing ofthe Emergency Act (Law No. 25561) dated 9 January 2002,which abandoned the convertibility system in force in thecountry since 1991 and unilaterally decided to implement asubstantial modification of certain basic conditions in conces-sion contracts. All these measures have had a negative impacton the electric power industry in general and on the compa-nies’ profitability.
BoliviaUnder the local regulations, maximum distribution prices arefixed according to the average distribution cost and, taking intoaccount operation and unit-cost indicators, the evolution of theassets designated for concession, taxes, and enhancements inthe distribution company’s efficiency. Base tariffs are deter-mined by considering average values representing supply costsfor a period of four years and a profit margin on equity definedby law. Intratariff period indexation formulas basically reflectthe retail inflation indicator and the efficiency indicators.
Regulation is completed with a series of resolutions by theregulator, which establishes methodological guidelines todefine detailed review parameters. It is important to mentionthat when Bolivia’s regulation fixes the capital cost rate intariff reviews, it makes a difference between an owner’s capi-tal (equity) and a third-party’s capital, with regulatory differ-ences in the manner of estimating each rate.
The Superintendence of Electricity of the Government ofBolivia has a regulatory system for electric power distributorsthat requires statistical cost analysis. The system is detailed inthe price and tariff rules (RPT).
Under the RPT, distributors must operate under price con-trol schemes lasting four years. Four different types of refer-ence tariffs must be fixed: peak demand, off-peak demand,energy, and customer services. These tariffs may vary withhigh-, medium- and low-power consumers. The off-peak con-sumption tariff covers the cost of electric power distributionand general and administrative costs.
Maximum reference prices depend on base tariffs that areadjusted monthly by means of indexes. Indexation formulasattempt to reflect variations in the price of supplies and inoperative efficiency. The impact of inflation is measured bythe consumer price index.
Efficiencies in electric power loss control are reflectedin calculation formulas through “X factors” called “loss
reduction indexes.” X factors for energy and demand lossesare calculated independently. Changes in the efficiency tomanage other supplies different from capital are reflectedin calculation formulas through X factors called “costreduction indexes.” Factors for distribution, consumption,and administrative and general expenses (A&G) are calcu-lated independently.
The RPT establishes that tariff bases must be calculatedby using the projected average cost of the service during thefour years of the planning study. These cost projections mustbe approved by the Superintendence through resolutions.
For cost projections for the four-year period, the Superin-tendence will establish a set of indicators relating costs toother parameters such as: asset value, number of clients, ener-gy sales, length of lines, etc. Such indicators will mark levelsof efficiency that include the analysis of the fulfillment of theindicators in the previous period, and which cannot be lowerthan those resulting from the actual operation of the companyin such period.
ChileIn Chile, distribution networks are those whose voltage isunder 23 kV.
The regulated distribution price corresponds to the meanadded value by this activity determined from model firmsoperating in the country. The final price paid by a regulatedconsumer integrates the regulated generation-transmissionprice, with which the generators supply the distributing firmsand an added value for the distribution service.
The regulation mechanism determines its distribution tar-iffs from the optimization of a real firm that serves as a refer-ence for the construction of a model firm, and such modelfirm is benchmarked with all the distribution concessionairefirms. Thus, this scheme corresponds to an incentive tariffsmodel of the yardstick competition type, where the relativeperformance of the industry is assessed, assuring in theory aspecific minimum return to those firms that have a perform-ance similar to the model firm.
A core element to determine distribution tariffs is thedimensioning of the model firm. In the international applica-tion of the yardstick competition mechanism, the regulationof monopolistic activities is determined through the compari-son of costs and performance of similar firms or mirror firmsor the reduced comparison of heterogeneous firms correctedfor differences. In the Chilean distribution monopoly regula-tion model, there is a hybrid benchmarking scheme betweendifferent firms. On one hand, groups of firms of similar
In Chile, the regulated distribution price corresponds to the mean added value of the activity determined from model firms operating in the country.
july/august 2007 IEEE power & energy magazine 57
characteristics are compared, identified through typical areas,with a model firm. Then, the performance of heterogeneousfirms is compared in an integrated manner, with an assessmentof the global adequacy of the industry with a single standard.In the former case and through a theoretical model andthrough direct comparison, efforts are made to provide theefficiency signal of similar firms and in the latter case effortsare made to produce a horizontal comparison that fits the theo-retical model with the average reality of heterogeneous firms.
The electric legislation determines the distribution tariffsbased on the VAD value that is based in a model firm and thatconsiders the three main components that form part of thedistribution business cost: infrastructure and equipment costs,energy and power losses, and operating expenses such asadministration, operation, and maintenance expenses. Thelaw groups them as follows:
✔ fixed costs for administration, invoicing, and user serv-ice expenses, independent from their consumption
✔ mean distribution losses in power and energy✔ standard investment, maintenance, and operating costs
associated to the distribution by unit of power sup-plied. The annual investment costs will be calculatedconsidering the new replacement value (VNR, theSpanish acronym for Valor Nuevo de Reemplazo), thefacilities adapted to the demand, and a discount rateequal to a real 10% per year.
The components indicated are calculated for a specificnumber of typical distribution areas defined by the NationalEnergy Commission, with a previous consultation with thefirms. The process to determine the VNR has the objective ofcalculating the “cost to renew all the works, facilities, andphysical goods dedicated to provide the distribution service inthe respective concessions.” The concept of VNR used by theChilean legislation to be applied to distribution activities hasbeen a hybrid between the substitution and replacements costs.
The law requires that when the model company is to becalculated, two independent studies must be done, one by thedistribution company and a second one done by the NationalEnergy Commission. The results of these two studies must beaveraged considering a weight of two thirds for the govern-ment and one third for the distribution company. Tariffs arethen cross checked so that the industry as a whole has prof-itability between 6–14%.
ColombiaIn Colombia, distribution networks are those whose voltage isunder 220 kV. There are no concessions or franchises. Distri-bution only includes the transmission of energy through thenetwork. Electric power distribution in Colombia is separatedfrom the sale and purchase of energy, which are part ofanother activity (trading) together with metering, reading,billing, and collecting. Therefore, distribution is limited by
regulation to the “network business,”whose remuneration is determined by theso-called “usage charges.” These “usagecharges” of the local or regional distribu-tion system are the regulated retributionthat the distribution companies receive torun their business. They consist of theannual revenues for efficient operation,following the theoretical criterion ofremuneration of the distribution activitypresented above.
Trading is a different business and isopen to free competition. There is com-petition in the distribution business. Sincethere are no concessions or franchises,there can be more than one distributor ineach area (parallel networks). The dis-tributor has no obligation to expand, butit must grant free access to its networksin case it has surplus capacity. If the
In Columbia, efficient management focuses on investmentmanagement, administrative costs, operation and maintenance, and losses.
figure 2. Evolution of Edesur’s losses.
Evolution of Annual Index of Energy Losses
0%
5%
10%
15%
20%
25%
30%
1992 1993 1994 1995 1996 1997 1998 1999
Year
Loss
es In
dex
(%)
58 IEEE power & energy magazine july/august 2007
figure 3. Evolution of quality of service indicators.
FMIK
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
1° Six-
Mon
th P
eriod
E1
2° Six-
Mon
th P
eriod
E1
3° Six-
Mon
th P
eriod
E1
4° Six-
Mon
th P
eriod
E1
5° Six-
Mon
th P
eriod
E1
6° Six-
Mon
th P
eriod
E1
1° Six-
Mon
th P
eriod
E2
2° Six-
Mon
th P
eriod
E2
3° Six-
Mon
th P
eriod
E2
4° Six-
Mon
th P
eriod
E2
5° Six-
Mon
th P
eriod
E2
6° Six-
Mon
th P
eriod
E2
7° Six-
Mon
th P
eriod
E2
Num
ber
of P
ower
Inte
rrup
tions
/Sem
este
r
Internal Total
TTIK
1° Six-
Mon
th P
eriod
E1
2° Six-
Mon
th P
eriod
E1
3° Six-
Mon
th P
eriod
E1
4° Six-
Mon
th P
eriod
E1
5° Six-
Mon
th P
eriod
E1
6° Six-
Mon
th P
eriod
E1
1° Six-
Mon
th P
eriod
E2
2° Six-
Mon
th P
eriod
E2
3° Six-
Mon
th P
eriod
E2
4° Six-
Mon
th P
eriod
E2
5° Six-
Mon
th P
eriod
E2
6° Six-
Mon
th P
eriod
E2
7° Six-
Mon
th P
eriod
E2
0
2
4
6
8
10
12
14
H/S
ix-M
onth
Per
iod
Internal Total
Greater fairness in efficiency sharing is an objective that, although partially achieved, should be revised, in order to allow a larger transfer of benefits of the scheme to end users.
july/august 2007 IEEE power & energy magazine 59
expansion is not profitable, it is not obliged to make it. In thiscase, the expansion is made by the distributor who finds itprofitable to expand the network and serve that demand.
The new user connecting to the network pays the entrancecosts and the facilities become the users’ property. Tariffs areof the postage-stamp type by voltage level. The VNR is cal-culated with an actual profitability of 14.06% for level IV and16.06% for levels I, II, and III, and operation and mainte-nance (O&M) expenses of 2% for levels III and IV. Level IIhas 4%, and in the case of level I, there is a costing processper activity.
The nontechnical losses acknowledged in the tariff are0.67% in level I in 2007. For each voltage level there is a per-centage of acknowledged losses. Once the VNR has been cal-culated with profitability and O&M expenses, thepostage-stamp is determined by dividing the cost accumulat-ed in each voltage level by the useful energy considered bythe regulator. That is the actual energy running through thenetwork according to the acknowledged loss factor.
From a conceptual point of view, payments to third par-ties work as a pass-through of costs beyond company man-agement. In addition, the assets profitability or capital
opportunity cost is determined by regulation. Therefore, effi-cient management focuses on the remaining aspects: invest-ment management, administrative costs, operation andmaintenance, and losses.
PeruPeruvian regulation has many elements in common withChile. In order to fix electric power tariffs, the regulation stip-ulates that the VAD must be calculated for each of the typi-cally defined areas, taking a model company as a reference.The VAD includes costs associated to the user, standard loss-es in the distribution system, and standard investment, opera-tion, and maintenance costs. The investment cost implies theVNR annuity of the economically adapted system, consider-ing its useful life and the restatement rate fixed at 12%.
Tariff regulation is indirect. A tariff is established byadding up bus prices with their acknowledged losses plus aVAD. The VAD is calculated and recalculated every 4 years,and it is restated within each period by means of polynomialformulas. It includes capital returns based on a reference net-work economically adapted to demand and the costs ofexploitation and trading of a model company.
Electric power distribution in Colombia is separated from thesale and purchase of energy, which are part of another activity(trading) together with metering, reading, billing, and collecting.
figure 4. Evolution of number of clients per employee at Edesur.
Evolution of Number of Customers and Employees
0
1,000
2,000
3,000
4,000
5,000
6,000
1993 1994 1995 1996 1997 1998 1999
Year
Num
ber
of C
usto
mer
s (T
hous
and)
/Num
ber
ofE
mpl
oyee
s
0
100
200
300
400
500
600
700
800
Cus
tom
ers
Per
Em
ploy
ee
Number of Customers (Thousand)
Number of Employees
Number of Customers/Employee
IEEE power & energy magazine july/august 200760
The tariff system itself is an incentive for efficiency as itacknowledges the efficiency values the distributor mustoperate with ex ante the rendering of the service; the distrib-utor will obtain the profitability expected by its investorsonly if it adjusts its costs to the acknowledged values. Thesystem transfers decision risks to the provider since mis-takes when deciding investments, expenses, indebtednessstrategies, or technological adaptations will be ultimatelypaid by the investor, who will receive less profit for the des-ignated capital.
Regarding the use of resources, it is a model based oneconomic signals where decisions are made by agents.Through their tariffs, distributors inform consumers on theefficiency cost of each consumption alternative and they, intheir capacity as demand makers, decide on the use and allo-cation of resources.
Results of RegulationThe application of incentive price regulation to the distribu-tion companies has been a successful process in LatinAmerica, with difficulties experienced that vary from countryto country. Specific examples of positive results are given forthe same sample countries.
ArgentinaThe case of the Edesur distribution company can beregarded as representative of regulatory incentives in met-ropolitan distributors. In the period between the beginningof the company’s privatization (September 1992) and theend of 2000, total investment was US$1,061.7 million,which corresponds to an annual investment value of 118million dollars/year. This investment allowed for a reduc-tion of losses (Figure 2), an enhancement of quality ofservice, and the introduction of new technologies. Losscontrol programs changed from a loss index of 26% at thebeginning of private management to 8% in 1997. With ref-
erence to quality of service, regulation generated appropri-ate economic signals to improve indicators. Figure 3shows the evolution of the average interruption frequencyper kVA (FMIK) and the average interruption time perkVA (TTIK) since the beginning of private management.
Function outsourcing programs allowed contractors toperform those tasks Edesur believes are not part of the com-pany’s business core, which resulted in staff reduction andits corresponding cost savings. As a result of these programsand a change in the working method, the ratio of the num-ber of clients per employee increased from 374 in 1993 to748 in 1999, with a substantial growth in the distributor’sproductivity (Figure 4).
BoliviaDuring a ten-year period of regulation of the distribution sec-tor (1996–2005), investment reached US$283.7 million (seeFigure 5). Total investment during this period exceeded thevalue of all investment performed until 1995, which showsthe strong incentive of the new regulatory framework toinvest. Figure 6 shows the evolution of investment in distribu-tion during the 1999-2005 period, after the process of reformwas initiated.
Quality control in the service was introduced through sev-eral six-month periods. The first six were called transitionperiods and the following were scheme periods with stricterindicators than the previous ones. Figure 7 and Figure 8show the evolution of interruption frequency and averagetime and customer service average time, respectively. Theyshow that all these indicators have had a descending trendand that there has been constant improvement during theentire 1999–2005 period.
The new investment in transmission, generation, and dis-tribution projects allowed for the introduction of new tech-nologies, especially in substation equipment, control systemequipment, software, SCADA, generator automation, etc.
july/august 2007 IEEE power & energy magazine 61
figure 5. Investment during the period 1996–2005.
Transmission Company
Distribution Company
Power Plant/ Power GenerationCompany
Total Investment
121,8
283,7
424,9
830,4
0 100 200 300 400 500 600 700 800
ChileChile has had 20 years of experience in applying benchmarkregulation to its 36 distribution companies. As Figure 9 illus-trates for the low-voltage segment of the largest Chilean dis-tribution company, the remuneration of the distributionbusiness has followed a downward pattern, with an overallreduction of 44% since 1984.
Despite the cost reductions, returns for the distribution compa-nies have been very favorable, concentrating between 10–20%, ascan be seen in Figure 10. This is what leads to question if fairnesshas been achieved and if there has been an adequate transfer ofbenefits to the final consumers, given the distribution companiesefficient investment and operating policies.
It is clear that distribution companies have been able togain efficiency throughout time, reflectingthe clear incentives of the benchmarkmethodology. Reductions of technical andnontechnical losses for the largest Chileandistribution company are shown in Figure11. Reduction of losses has been achievedover time, reducing them more than 50% inless than ten years.
ColombiaReforms in Colombia were introduced inall utility services, electricity included. Thenew framework has had a very positiveimpact on coverage, company sustainabili-ty, and quality of service. Figure 12 showscoverage response to the introduction of
The application of incentive price regulation to the distributioncompanies has been a successful process in Latin America, withdifficulties experienced that vary from country to country.
figure 7. Evolution of interruption frequency and average time.
Six-Month Period
Average Frequency Average Time TC
T: Transition PeriodR: Scheme Period
T1 T2 T3 T4 T5 T6 R1 R2 R3 R4 R5 R6 R7 R8 R90.0
2.0
4.0
6.0
8.0
10.0
Tc
7.06.05.04.03.02.01.00%
Fc
figure 6. Evolution of investment in distribution.
Distribution Investment
Approved Total Performed Total
Cumulative Performed TotalCumulative Approved Total
US
D
200,000
180,000
160,000
120,000
140,000
100,000
80,000
60,000
40,000
20,000
01999 20052000 2001 2002 2003 2004
62 IEEE power & energy magazine july/august 2007
reform and the regulatory frame-work for different utility services.In Bogotá and Medellin, electricpower companies reached almostglobal coverage (100%). In Bar-ranquilla, Cali, and Bucaramanga,coverage reported for 2002 was86%, 92%, and 96%, respectively.
Service interruption frequency(SIF) and service interruptionduration (SID) indicators measurethe duration and frequency ofservice interruptions. According to the Superintendence ofResidential Utilities (SSPD), the SID indicator, accumulat-ing electric power service interruption hours, was reducedto almost 50% between 2000 and 2001, falling from 38 to21. Likewise, the SIF indicator, which accumulates thenumber of interruptions in electric circuits, fell from 43 to20 in the same period.
The results of reform in company efficiency are conclu-sive. The indicator showing employeesper 1,000 connections has definitelyfallen in the five companies in the sam-ple (Table 2), with especially strongadjustments in those that were in worseconditions and whose management wastransferred to the private sector. Thoseon the coast changed from an index of8.7 in 1998 to 1.8 in 2001.
A pending issue in relation to bro-kers is what is known in regulationtexts as “market skimming.” Incomingbrokers concentrate on users withhigher consumption and lower tradingunit costs. The incumbent broker can-not compete for these customers.Since under regulation prices cannotbe discriminated, a competitive offerto a user of the incoming brokerwould imply a decrease in charges forall its market. Skimming may causethe incumbent’s bankruptcy, the lossof its market, and the appearance ofinefficient brokers.
PeruAfter eight years of operation of thereform, results can be summarized asfollows. The electrification coefficientin the country has grown from 56% to73%, with electric power service cover-age of 3.3 million users. Electricity losslevels have decreased from 22% to10%. Market agents have receivedprice signals that have allowed the
Peruvian electric power system to develop by means of com-petitive tariffs, compared to other countries with similar reg-ulatory, economic, and geographical frameworks, such asArgentina, Chile, and Bolivia. The regulatory frameworksignals have impacted on company efficiency; for example,the cost of staff and third-party services over total revenueshas fallen from 26.5% to 13.7% in the distribution sector(Figure 13).
figure 8. Evolution of customer service average time.
Six-Month Period
121086420
151413121110987654321
Number of Employees Employees/1,000 Customers
Company 1998 1999 2000 2001 1998 1999 2000 2001
EE.PP.M 2,735 2,149 1,476 1,668 3.5 2.6 1.7 1.9ESSA 1,044 1,018 852 836 2.7 2.5 2.1 1.9EMCALI 658 572 563 524 1.5 1.3 1.3 1.2ELECTRICARIBE 2,314 1,251 1,300 1,224 8.7 1.8 2.0 1.8CODENSA 1,904 1,213 996 841 1.4 0.7 0.6 0.5
SOURCE: SSPD (Superintendence of Residential Utilities) Energy Representative.
Table 2. Evolution of the number of employees per customer.
figure 9. Costs recognized for low-voltage distribution.
1984 1988 1992 1996 2000 2004
Year
Evolution Low Voltage Cost Area 1Pesos of July 2004
9.000
8.000
7.000
6.000
5.000
4.000
3.000
2.000
1.000
0
$/kW
/Mon
th
july/august 2007 IEEE power & energy magazine 63
The average collection period has also been optimized; inthe distribution sector it has fallen from 115 days to 61.1days, as shown in Figure 14.
Regarding service coverage, there is a significant increasein the service, from 55.9% at the end of 1993 to 72.2% in1999. When the new regulatory framework was implement-ed, companies received adequate incentives to achieve theseresults, as is shown in Figure 15.
Regarding losses in distribution companies, which wentthrough a rather difficult situation in years such as 1993, whenlosses hit a record of 21.8%, companies have managed toreduce losses to 10.4% in the 1993–2000 period (Figure 16).This is due to the loss reduction schedule of the regulator aswell as to the tariff mechanisms companies can apply as aresult of profits resulting from cost reductions.
In summary, as seen in the different country examples, the
balance of incentive regulation in distribution has been posi-tive, for the following reasons:
✔ increase in levels of electrification and supply coverage✔ increase in efficiency of the distribution service/reduc-
tion of costs (investment, operation, and management)✔ reduction of losses✔ introduction of new technologies✔ increase in quality of service.
Challenges of Incentive Price RegulationThe application of incentive price regulation has not beenexempt from difficulties over the years, with conflicts arisingover its application between the regulators and the companiesinvolved. While the regulators in general aim at reducing tar-iffs as much as possible, the companies aim at increasingtheir revenues. Some of the areas of conflict are described.
TechnologiesThe search for efficiency in distributionrequires evaluating different technologi-cal alternatives that are not necessarilyin use by the concessionary company,leaving aside the historical practices. Inthis sense, the evaluation of conductortechnologies is a matter of permanentdiscussion; e.g., the use of copper ver-sus aluminum, the determination of dis-tribution voltage, the range for optimaluse of the conductors, and the comparedlength of networks of low and high volt-age. The studies show that an efficientexpansion process would install practi-cally the same amount of network oflow voltage, but with a greater transportcapacity, mainly correlated with thestreet layout, and install a high-voltagenetwork of smaller length, conditionedby the optimization of the secondarytransformers.
In relation to secondary transform-ers, the total installations are evaluated,considering both location and capacity.The studies show that an efficient com-pany would install a smaller number oftransformers of greater capacity that arebetter located, which neverthelessresults in a greater total installed capaci-ty with a lower cost than the concession-ary company, essentially due toeconomies of scale.
ManagementCertainly an area that causes conflict inthe efficiency analysis bears relation tothe design of the organization andfigure 11. Reduction of distribution losses.
19.8 18.816.1
13.6 13.312.0
10.69.3 9.3
199519941993199219911990198919881987
20
15
10
5
0
%
figure 10. Return of distribution companies in Chile.
Returns of Distribution Companies (Chile)35%
30%
25%
20%
%
15%
10%
5%
0%
CGERIO MAIPOCHILECTRA
ELIQSAEMELARICONAFEEMELATLITORAL
EMELECTRIC
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
Year
64 IEEE power & energy magazine july/august 2007
infrastructure necessary to administer, operate, and maintainthe distribution network of the company. Aspects such asorganization design and the number of workers and level ofoutsourcing of functions (such as maintenance or commercialareas) are permanently discussed and subject to differentviews between the regulator and the distri-bution companies.
Economies of Scale and PricesWith the objective to determine thevalue of the efficient distribution activi-ty, it is necessary to have studies of mar-ket prices of the necessary items for theinstallation and operation of the distribu-tion network, as well as studies of wagesfor the employees of the company. Thereis a permanent discussion on differentissues, such as the concept of marketprice, given the difficulty to obtain pricereferences of specific equipment, whichoften forces use of actual informationfrom the company, and the level ofwages adapted for the distribution com-pany, which is generally obtained from asurvey of prices and benchmarks withother companies.
Another subject of discussion is the price to considerfor commodities at the time of the study, such as steel,copper, and aluminum, with the discussion arising in rela-tion to valuing at a cost that does not correspond to the onehistorically incurred by the monopoly. Possibly, it can mean
figure 13. Evolution of the cost of staff and services over total revenues.
15,116,5
13,410,8
16,1 15,913,1
21,3 21,919,5
18,7 18,4 17,815,7
26,5
20,519,6 18,7
19,517,7
13,7
30
25
20
15
10
5
0Power Plants Transmission Companies Distribution Companies
1994 1995 1996 1997 1998 1999 June 2000
figure 12. Evolution of utility coverage.
Server SystemAqueduct
Electric PowerNatural GasLocal Telephone System
100%
80%
90%
70%
60%
50%
40%
30%
20%
10%
0%1970 1975 1980 1985 1990 1995 2000
The application of incentive price regulation has not been exemptfrom difficulties over the years, with conflicts arising over itsapplication between the regulators and the companies involved.
july/august 2007 IEEE power & energy magazine 65
figure 16. Reduction of distribution losses.
25%
20%
15%
10%
5%
0%1993
Standard Losses Acknowledged Losses Real Losses
9.0%
21.8%20.6%
19.7%
17.1%
14.6%
12.4%11.5% 10.4%
7.9% 7.6%
8.6% 8.5% 8.4% 7.7%4.8% 4.2% 3.6%
7.2%7.2%7.3%7.3%7.4%7.5%
1994 1995 1996 1997 1998 1999 June 2000
figure 15. Evolution of customer coverage.
3,500,000
3,000,000
Num
ber
of C
usto
mer
s
2,500,000
2,000,000
1,500,000
1,000,000
500,000
Customers Electrification Rate
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
80%
70%
60%
50%
40%
30%
20%
10%
0%
Ele
ctrif
icat
ion
Rat
e
figure 14. Evolution of collection period.
169,0
71,357,6
51,648,955,7
42,4
86.0
126,2116,4
42,543,751,8
42,7
115,0
74,9 77,7
61,169,266,771,7
180
160
140
120
100
80
60
40
20
0Power Plants Transmission Companies Distribution Companies
1994 1995 1996 1997 1998 1999 June 2000
66 IEEE power & energy magazine july/august 2007
july/august 2007 IEEE power & energy magazine
a subvaluation of the company if the commodities at thetime of the fixation correspond to historical lower of prices,or vice versa.
In general, the incentive regulation process needs to fol-low general guidelines that may also be difficult to achieve.For example, it is very important to assure regulator inde-pendency from monopoly pressure and avoid regulator cap-ture. This is a key issue in any of the schemes that havebeen formulated to tariff the distribution segment. Theincentive regulation exacerbates conflictive interestsbetween regulator and monopoly. On the other hand, thereis always the risk that the regulator could manipulate theprices with other objectives, such as political interests tolower rates.
Also, it is very important to consider reliability and qualityof supply, especially since the companies have to necessarilycomply with current reliability and quality regulations.Therefore, there has to be a consistency between the stan-dards of quality and reliability that are required and the levelof income set by the resultant tariff.
ConclusionsThe Latin America experience of incentive regulation for dis-tribution companies has resulted in a sustained evidence ofefficiency, through clear incentives for cost reductions, andattraction to investors, given adequate returns to investmentcapital. Nonetheless, greater fairness in efficiency sharing isan objective that, although partially achieved, should berevised, in order to allow a larger transfer of benefits of thescheme to end users.
There are no major implications in the application of regu-latory reforms in network development technology, exceptfor specific cases such as commercial loss reduction. In fact,regulation signals for loss reduction have motivated distribu-tors to adopt distribution models of the U.S. type even incountries with European distribution, since they have provedto be efficient against energy theft.
For Further ReadingH. Rudnick and R. Sanhueza, “Benchmark regulation andefficiency of electricity distribution in a restructuredpower sector,” in Proc. 2004 IEEE Int. Conf. Electric Util-ity Deregulation, Restructuring and Power Technologies,Apr. 2004, pp. 9–12.
L. Barroso, T. Hammons, R. Varela, and H. Rudnick,“Panel Session on Latin America -Price cap regulation:
Stimulating efficiency in electricity distribution in LatinAmerica,” in Proc. IEEE Power Engineering Society Gener-al Meeting, Tampa, Fl, June 2007.
E. Lerner, “Distribution perspectives in Colombia,” Tech-nical Report-Project CIER—COCIER 02, June 2000.[Online]. Available (in Spanish): www.cocier.org
E. Voscoboinik and C. Zillie, “Application of efficientoperative cost models to fix tariffs for electric power distribu-tion and transmission companies,” (in Spanish) in Proc. 12thIberoamaerican Regional Conf., CIGRE, Foz do Iguaçu,Paraguay, May 2007.
BiographiesHugh Rudnick is a professor of electrical engineering atPontificia Universidad Católica de Chile. He graduated fromthe University of Chile, later obtaining his M.Sc. and Ph.D.from Victoria University of Manchester, United Kingdom.His research and teaching activities focus on the economicoperation, planning, and regulation of electric power systems.He has been a consultant with utilities and regulators in LatinAmerica, the United Nations, and the World Bank. He is aFellow of the IEEE.
Alejandro Arnau graduated as an electrical engineer withan M.B.A. from the University of La Plata, Argentina. Hisexpertise in economic and technical projects was developedin numerous studies on different regulation environments fordistribution companies in Latin America, particularly inArgentina, Chile, Uruguay, Bolivia, Peru, Colombia,Venezuela, Guyana, Nicaragua, Panamá, El Salvador, theDominican Republic, and Guatemala.
Sebastian Mocarquer graduated as an industrialelectrical engineer from Pontificia Universidad Católicade Chile. He is currently the development manager atSystep Ingeniería y Diseños. He has directed severaltariff studies in Chile and has made regulatory studieswith utilities, regulators, and investment banks in Chileand abroad.
Efrain Voscoboinik is an electrical engineer who graduat-ed from the University of La Plata, Argentina. He also took apostgraduate course in maintenance management at the sameuniversity. He has been an assistant professor of electric facil-ities at the University of La Plata and is a professor of engi-neering at the Universidad Tecnológica Nacional, Argentina.Since 1999, he has been working for Mercados Energéticos,actively participating in several regulatory studies and duediligence processes.
67
In incentive regulation, efficiency is measured against a previousbenchmark and the results allow the identification of the actionsrequired to drive efficiency improvements.
p&e